Athena Swan 2018 and WGRiITN 2019

The Women Graduate Research in IT Network first formed in 2015. The Network has experienced a revival this year and will start 2020 energised and full of impressive ideas.

There are new representatives in place as well; at Caulfield  Caitie Doogan, at Clayton Fariha Jaigirdar, and at Malaysia, Najini Arachchige. Network activities this year included:

  • Hosting an International Women’s Day event;
  • Network meeting to discuss initiatives;
  • Hopper Down Under representation; and
  • Sponsoring the attendance of five GR students at the Women in Tech Conference in Sydney in September. 

Those attending the Women in Tech Conference said taking part in an industry focused event rather than a research conference was a new experience and provided them with opportunities to network with CIOs, CEOs and IT Industry leaders they would not normally have direct access to. This was invaluable for their future careers.

J. Holden, C. Doogan, N. Anantharama and W. Buntine.

Athena Swan Annual Event was held on the 30th of August 2018 as part of Monash University’s Diversity and Inclusion week to celebrate Monash’s diverse community and strengthens inclusive culture.

Caitie Doogan, Nandini Anantharama, Julie Holden , Anne Nicholson and Wray Buntine attended the Athena Swan event  on Untangling unconscious bias. All five said the learn more than expected about the challenges and benifits of diversity within STEMM. Caitie commented:

I was pleased that they talked about what to do when we have heterogeneity and how this does not equate to diversity in STEMM.

Caitie Doogan

MONASH MASTER OF DATA SCIENCE GRADUATE WINS $500K IN 2019 OZ MINERALS EXPLORER CHALLENGE

Michael Rodda has won the lucrative crowdsourcing competition aimed at predicting where to drill for high-value mineral tenements. He and his two team mates, Jesse Ober (environmental scientist) and Glen Willis (process engineering) have each pocked a cool $167,000 for their efforts. Michael and I went through our masters together, and I caught up with him over the weekend to ask what his strategy was.

I fused historical drilling records for the states and used them as targets. Then I used national geophysics datasets, hyperspectral imaging and national surface geology spatial datasets to predict anomalous values of minerals.

Michael thinks outside the box and is in tune with the need for interpretable Algorithms.

I took a realistic view of data science, so instead of going for a convNet (which would make sense given all the data was image based), I combined lightGMB with SHAP (SHapley Additive exPlanations) values. It’s based on fairly recent research into model interpretability as a convNet (Convolutional Neural Network) would provide zero insights to the exploration geologists as to how my model was predicting stuff.

Michael is now a Graduate Data Scientist working at NAB. He tells me it wasn’t easy juggling his work and the gruelling competition.

I would finish at 6 pm, go home and work on this until 1 am, for three months. But yeah, all it takes is to be curious with data science.

Michael Graduated in 2018. His thesis was ‘Improving Performance of Recurrent Neural Network Forecasting Models Through the Use of Data Augmentation Techniques’.

Well done mate!


Clever kids are all about Machine Learning

The LDS editor, Caitie, this week received a request from James, a high school student in WA asking a few questions about Machine Learning (ML). It was passed to them from their supervisor. The questions were incredibly well formed and our editor wanted to share the responses to the student….wistfully wishing their own students would ask such good questions.

BTW, this kid is 17!

Briefly, what area(s) of ML do you work with?

My work involves designing and optimising novel research methods which make use of computational text mining. I look at Natural Language Processing (NLP) pipelines and a technique called topic modelling to conduct research on social media data and discover insights about different events and conversations happening across the world. I blend these insights with known facts and other datasets to answer particular research questions that traditional methods can’t explain.

Are you currently working on a project involving ML? If so, could you explain what it is, your goals, its purpose, etc.?

In my current project, I am taking real-world social media data sets (Twitter) and using a topic model called MetaLDA build here by our Machine Learning Group, to determine what sorts of discourses are present across 10 years of chatter about the ‘right to be forgotten’.

The right to be forgotten is new data protection right, that’s now built into the General Data Protection Regulation (GDPR) and applies to any company that processes data from EU citizens – That’s pretty much most countries in Australia. 

The challenge of Twitter data sets is that they are huge (1 million+ tweets) and what we call sparse and noisy, which I will explain next. Topic models are sensitive to these problems, and so we need to use a sophisticated set of Machine Learning techniques to overcome these issues. The problem is that our ML algorithms don’t fix these problems by themselves. So, we have to look at other aspects of the ML pipeline including:

  • What data we collect
  • How we clean or wrangle it
  • How we process it, and
  • How we optimise the algorithm.

We need to get this right in order to:

  • Find out the topics people are talking about
  • Who those people are
  • How those conversations have changed over time
  • What events affected them.

Additionally, we want to know about how the interactions between tweeters have influenced these changes. So, we need to know how to integrate traditional social science methodologies into a partly automated process of knowledge discovery.

A Topic model is a generative probabilistic model which represents documents as a mixture of topics where the allocation of a document to a topic is dependent on the words within that document. The Latent Dirichlet Allocation (LDA) is the go-to topic model for most studies (Blei, Ng, Jordan & Lafferty, 2003). However, LDA is challenging to optimise when modelling tweets which are very short documents. The shortness of tweets reduces the efficiency of the LDA inference algorithm which reduces the accuracy of their topics. MetaLDA which was developed by our group, is designed specifically for modelling short documents like tweets. 

If applicable, what kind of problems do you encounter in relation to datasets? How you mitigate against these issues?

The challenge of Twitter data sets is that they are huge (1 million+ tweets) and are what we call sparse and noisy. When you look at a tweet, we see it has lots of characters and links in it that we need to remove, because they confuse the algorithm. #hashtags and @usernames are examples. We also need to remove ‘stop words’ such as ‘at’, ‘and’, and ‘the’. But when we remove these, the algorithm becomes even shorter. You can see the difference between an unprocessed tweet and a pre-processed tweet below.

This is a big problem for us, and we have to think carefully about how much noise to remove without overly sanitising our data, which may give inaccurate results.

We also have to handle lots of spam and, tweets that are unrelated to what we are interested in. For example, we were working with a group of Immunologists at UWA and they wanted to know what Rheumatoid arthritis patients thought about the medications they were using and how they talked about their illness on Twitter.

We had figure out how to remove the tweets from doctors, marketers and the media. To do this, we looked at three classification algorithms, Naive Bayes, SVM and a simple neural network. The neural networked performed the best at this task. But we need to do more work.

Other issues we face with Twitter data are malicious spam bots that are becoming harder and harder to detect. I have been working analysis of the hashtag political group #auspol in the lead up to the 2019 election and encountered this problem. Malicious bots try to sway the opinions of Australian voters with ‘Fake news’. This happened in the 2016 US Election too and is was thought to contribute to President Trumps win. It is an ongoing problem in social media analytics (The bots, not Trump).

All data is biased. But unlike what you might read in the news, it’s not always because of the researchers own personal biases. Sometimes it’s because we have no choice but to add or omit certain features to the data set. In the GDPR data set, I’ve had to remove all non-English tweets as the algorithm will get confused. Additionally, none of us speak al 160 languages detected. But since the GDPR was enacted in European countries, a lot of people tweeting about it are tweeting in a language other than English. By removing the non-English tweets, we lose the opinions of those people. The data set then becomes bias towards English speaking users primarily from the UK, US, Canada and Australia.

Do you believe ML systems are appropriate to use in surveillance or law enforcement contexts? How can the negative impacts of ML systems in such contexts be mitigated against?

Hmmm, this is a bit of a hot potato topic, and your question is a little leading, so I’m going to answer, but with a caveat. They should be used, but not before proper ethical, legal and data governance procedures are put in place, with oversight from independent bodies.

Predictive policing has a long history of going wrong. This is an instance of the runaway feedback loop. When a particular neighbourhood, let’s say Armadale has higher drug-related crime statistics then say, Swanbourne, then the police will be sent to Armadale* to look for drug-related activity more than they would Swanbourne. With the higher crime rate and increased police presence, there will be more arrests in Armadale than ever before. These crime statistics are fed back into the model for new protections, but of course, Armadale will come back as even more of a likely target for drug-related crime and they cycle continues regardless of the relative crime rate compared to the benchmark suburb of Swanbourne.  This isn’t great, obviously, but luckily, most police departments have realised this already.

*I’m allowed to pick on Armadale since I lived there for a while.

Alternatively, there are some truly amazing examples of when ML systems have been used in Law enforcement. At Monash University, Janis Dalins, who is a PhD student and Australian Federal Police (AFP) Officer, is working with Prof Campbell Willson on AI systems that help identify criminal activity on the Darkweb. The idea is that AI crawlers could make it faster to track down images of crimes against children which currently has to be done manually at significant cost to the AFP’s time, resources and mental health of the officers who have to view these images. 

COMPAS predicts the likelihood of recidivism for criminals based on various factors. A criticism of the tool is that it oversimplifies the judicial process, is weighed too heavily by judges and can lead to faulty convictions. Do you believe the fault lies with the nature of the tool itself as opposed to the methodology of judges ?

I wrote an opinion piece on COMPAS from my perspective as a Data Scientist. I believe that the COMPAS developers tried their best to create a holistic instrument for the benefit of the individual and the courts efficiency. However, it is being misused. This is outside of the developer’s hands. The fault never lies primarily with the tool, but developers must be diligent in updating the software. An example of where this has gone horribly wrong is the ingrained racial or gender-based biases in HR analytics.

Also have a look at this paper for reference.

Caliskan-Islam, A., Bryson, J. J., & Narayanan, A. (2016). Semantics derived automatically from language corpora necessarily contain human biases. arXiv preprint arXiv:1608.07187, 1-14.

Presumably you believe that ML has the potential to be beneficial towards society, could you illuminate some specific contexts or applications where this is the case, perhaps within your own body of work?

Well, I’ve given you a few examples but here is another. Turning point is Australia’s leading addiction treatment and research centre and together with Monash University has been successful in receiving a $1.2 million grant through the Google AI Impact Challenge. Prof Wray Buntine and Principle investigator Prof. Dan Lubman from Turning Point aim to adapt AI methodologies to streamline coding of national ambulance suicide-related attendance data. The resulting data would play a central role in informing public health prevention, policy and intervention, as well as identifying emerging trends, hidden populations and geographical hotspots for targeted responses relating to suicide. The ultimate aim would be to develop a national suicide monitoring system with the potential to set international standards for informing suicide prevention efforts.

As you can imagine, there are some serious ethical and legal implications for this work, but it ultimately aims to reduce the number of lives lost each year in these tragic circumstances.

Teenagers….scary smart.

The COMPAS Algorithm – It’s not that black and white

Looking closer at the COMPAS recidivism algorithm and application in the US justice system.

The following is an opinion piece written by an LDS contributor.

Algorithmic decision-making is purported by the mainstream media to reduce the impact of human bias in our social systems. Yet the insidious and secretive nature of algorithms allows them to replicate and amplify the social inequalities brought about by ingrained systemic oppression. As a society, we seem to be content with the knowledge that we are not only failing to resolve systematic inequalities such as racism but promoting them. Such technology excuses us from addressing these complex issues by retreating behind the facade of objectivity. Far from advancing equity, the disturbing reality is that algorithms are racist unless you are white.

The above is an example of the sensualised rhetoric promoted by certain groups that ignore the complexity of these nuanced technologies. An infamous example of this is Equivant’s COMPAS software which generates recidivism risk scores claimed by the US criminal justice system to promote fairness and efficiency in pre-trial, sentencing and parole decisions. This is not the case. The use of COMPAS is indeed unfair as indicated by the racial disparities in predictive accuracies that reinforce the entrenched racial biases which see black defendants sentenced far more harshly than their white counterparts (Angwin et al., 2016.). But is it the algorithm that is unfair? Or the way it is being used?

For predictive parity, the output must be equal for both black and white defendants. To achieve predictive parity, a higher rate of false positives for black defendants will occur even though race is not a direct variable in the equation. This is because a defendant will be more likely to have previously offended (a direct variable) if they are black. But while mathematical logic is sound, the bias data use does not satisfy our social notion of fairness due to the implications of ignoring the context of the data chosen.

We see then how the algorithm is claimed to be biased, but the fact is that it was never meant to satisfy both notions of fairness (Hardt, Price, Srebro, 2016; Kleinberg, Mullainathan, Raghavan, 2016). Equivant warns that the recidivism risk score should not be used to classify individuals without considering a comprehensive assessment of their social situation to identify their rehabilitation needs (Northpointe, 2015). The racial bias it seems is not entirely the fault of the algorithm, but rather the misuse of it by the US criminal justice system.

Design and use of data and algorithms must be accompanied by an understanding of the complexities of the social context. Continued use algorithms and data to make decisions about individuals that are manifestly perverse is irresponsible. This misuse and unaligned understanding of fairness have produced devastating long-term social effects that must be accounted for (Angwin, Larson, Mattu & Kirchner, 2016). Although different groups have attempted to address the issues of algorithmic bias, they are failing. This must be resolved.

I got into Uni – Where the F@#k am I going to live?

Recently, LDS has been getting a lot of messages asking for an overview of the Melbourne rental market for students. In a previous article, LDS revealed the absolute shit storm that is student housing in Australia. If you haven’t done so already done, have a read. But it doesn’t have to be this way.

If you want to know what far and reasonable prices are, where to go to find legitimate rentals, and the ‘gotchas’ of the Melbourne market, read on and get the run down. A couple of notes before we start:

We chose the Melbourne market in the south-east as that was the majority of what incoming students requested. The map below indicates this area and caters for several universities and other high education institutions. Each price is listed as per week (pw). Remember, if you are sharing it’s less. We have included this price in the description as well.

Most rentals are through real estate agents don’t come furnished. These are rental properties not share houses. Furniture isn’t that expensive to get, especially second hand.

Inner South East Melbourne

Rental properties through real estate agents

1 brm apartment: $280 – $350 pw

Now these are rare as hen’s teeth for the inner south east. Particularly after the inner ring. For $280 pw you can get a very small apartment which may be a bit older or not so well kept. But there are some great flats about if you happen to get lucky.

Per week low: $280 pw
Recommended occupancy: 1
Rarity: Super rare

If you must live by yourself (understandable) then you’ll may have to go up in price towards the $320 pw mark.

Per week low: $320 pw
Recommended occupancy: 1
Rarity: Rare-ish

2 brm units: $330 – $400 pw

This would be a good choice if you can’t get or don’t want a single apartment. These are also called units. For as low as $330 pw you can get an older two bedroom flat or a renovated unit. The photos below show two separate 2 bedrooms $330 pw units.

Per week low: $330 pw
Per week low per person: $165pw

Per week high: $400 pw
Per week high per person: $200pw

Recommended occupancy: 2
Rarity: Common

If you are happy to pay a bit more, you can get a comfortable 2 bedroom place like the one below priced at $400 pw.

3 to 4 brm townhouse or house $600 to $800 pw

If you find yourself with a good group of friends and want to move in together, a share house is the way to go. It is by far the cheapest option after set up costs. The picture below shows a $500 4 bedroom unit. As you can see from the yellow stickers, this place has just been renovated.

It’s rare to see Australians with more than 5 people in houses. Most students find it quite stressful, particularly if they like a clean house! Real estate agents also are reticent to rent to this many people on a lease as it gets really difficult for them when people are more likely to move on and off the lease.

Per week low: $600 pw (units/townhouses)
Per week low per person: $200 pw for 3 people, $150 pw for 4 people

Recommended occupancy: 3-4
Rarity: Common

Per week high: $800 pw (townhouses/houses)
Per week high per person: $267 pw for 3 people, $200pw for 4 people

Recommended occupancy: 3-4
Rarity: Common

The best time to get in on the rental market is about a 6-4 weeks before orientation week. Students finish degrees in November and June. They start again in March and July. This means you can get in on the glut of houses that become available when students graduate and return home. This is easier in summer than in winter as more students finish in summer.

What if I don’t have a rental history?

Most of the time, students don’t have a rental history, and this makes it hard to rent a house from a real estate. They often then opt for two options. A share house with tenants that are already in there and on a lease with a real estate agent, or a rooming or boarding house. LDS wants to make this very clear that in our opinion WE DO NOT RECOMMEND GOING INTO A ROOMING OR A BOARDING HOUSE. If you want to know what the difference is between living arrangements head to our friends at flatmates.com for an overview of what the different legal definitions of shared housing are.

There are four reasons we don’t recommend doing this:

  • You are likely to be shoved into a house that is too small for the number of people in it, that is not well kept and is potentially dangerous.
  • You will not be able to choose who you share a room with.
  • You will likely be sharing a house with more than ten people. There is a complete lack of privacy.
  • They suck! (Ok not a good reason, but still a reason).
  • Rooming houses are favoured by students who can’t get rentals due to no rental history but also poor rental history. There are many reasons for this but some common ones are alcohol or drug abuse, mental health issues, domestic violence and time in jail. As much as LDS believes inequality and ending certain stigmas, we also believe in safety.
  • The house can get dirty when you have hygienic housemates. There’s always one. Some people tell us that with ten people in these places, you don’t even get the opportunity to cook.

Here is the list of minimum standards proprietors must abide by. If they don’ they can be fined under the Residential Tenancy Act 1997.

If you must stay in one of these properties, please choose a licensed house. Here is a list of registered boarding houses in Victoria. Just google the addresses as the names aren’t on the list.

Although the proprietor does not need to give you a Standard Residential tenancy form for a rooming house, we strongly recommend you get one.

They may offer you an alternative agreement, but the reality is that it is difficult to take complaints to consumer affairs without one. If they refuse, this is a bad sign. You should look for a room elsewhere. The standards for rooming houses are found here. In saying that, there are some good rooming houses that are for students and treat their tenants well. Always have a look at the comments on Google review to get a feel for the quality of the place.

What about share houses?

Another option is to share a house with people who already have a lease with a landlord. This would be our recommendation over a boarding house. But beware of dodgy renters trying to treat a place like a boarding house.

A share house is a term Aussies use to describe a home where the tenants are renting from a real estate agent and either all have their names on the lease (meaning they have all signed a standard residential tenancy agreement) or are sub-letting from the head-tenant (with a standard tenancy agreement). The furniture is usually shared, and different people own different things. In this scenario, you and your housemates would share bills for utilities but not necessarily food or extras like Netflix, for example.

These are great if you are looking for a social environment and a homey feel. Most of the times, you will be ‘interviewed’ by the current housemates. They will want to get to know you and whether you fit the vibe of the house. For example, a group of young professionals in their early 30s might not be interested in an 18-year-old uni student. If you don’t hear back from them in a couple of day’s it’s generally an indication they have gone with someone else. Don’t take it personally. It’s better to be in a house where you all understand each other and get along than one where the dynamics aren’t quite right.

Housemates with common interests have a better experience.

In a lot of share house situations, some people don’t go on the lease straight away. We don’t recommend this, but it’s sometimes because they want to make sure they aren’t living with a sociopath. This protects the other housemates as it’s easier to kick out a tenant who isn’t on the lease. Given this, we always recommend some form of a rental agreement, such as this one. This will protect you in case it all goes south, and you need to get the authorities involved.

Preferably though you should ask for a standard tenancy agreement to be written up to sub-let from the head-tenant, the person holding the lease with the real estate. This protects you as well. A landlord must permit the renter to sub-let, and that’s what the standard tenancy agreement form proves. Otherwise, they are likely trying to make money off of you and could kick you out at any time.

International students really should use share houses a lot more and rely less on rooming houses, which are more expensive on average. Take a look at the prices here. Often share houses are with students near your university, and you can even use that search function in some sites like flatmates.com.au .

More importantly, share houses will give you an opportunity to absorb different cultures and Australians are often keen to share with internationals for the same reason.

Where to find places

Now that you have an overview heres a list of place to look for somewhere to live:
flatmates.com.au
gumtree.com.au
flatmatefinders.com.au
realestate.com.au – share houses
realestate.com.au – rental houses
domain.com.au – rental houses
sharehouses.com.au
Fairy Floss Real Estate – A much loved Facebook Group

Our most recommended website though is this one, dontrentme.com.au
If a property is on there, do not rent it.

We hope this helps you get an overview of rentals in Melbourne. Though some of this is very applicable to other Australian states. In our next post, we are going to cover some very important points about paying, agreements, condition reports and bond. Stay tuned.

Slumlord Millionaire – The Truth About Student Housing in Australia

This time of year is hectic! For students, the massive ‘assignment crush’ is hitting like a tsunami and tutors are frantically trying to mark assignments on time while stemming the flow of emails. Young professionals are finding out the dark side of EOFY and all of us are getting colder, especially those experiencing their first Aussie winter, slowly freezing, unprepared for an actual cold snap.

But spare a thought for those who haven’t started their postgrad journey yet. While these next few months are a time of nervous yet excited anticipation, it’s also a time of angst and uncertainty. As the butterflies settle and the congratulations from friends and family trickle away, an uncomfortable thought might be forming …. ‘where the f#@k am I going to live?’

Most students are prepared to start studying when they get to their university. they have their laptops, readings, books. All they need to dive right in. Except for one major element. Housing.

Every year it’s the same. Students think that two weeks is enough time to get a reasonable place to live and then they end up having to fork out way too much money because they can’t find something by the time uni starts. Or, more often, they settler for sub par accomodation. In the wake of the ABC ‘cash cows‘ investigation, LDS wants to provide new students our take on what student housing really looks like in Australia. A nightmare.

This was what happened to PJ. She arrived in Melbourne from India a fortnight before classes started and four-person share house for $160 per week (pw). While reasonably close to uni, the house is run down, with no cooling or heating. All the bedrooms have 2-3 single beds in them, and everyone shares one bathroom. PJ doesn’t get to choose who she rooms with or who shares the house with her. Recently, a housemate who was constantly intoxicated assaulted another housemate. He was asked to leave but not before the landlord shifted him next door into his other property where he could still access PJs house.

When asked why she rented the room in the first place she said, “they [real estate agents] need to see previous rental history, and we [international students] don’t have that”.

So why didn’t she move out? She told us that “we aren’t earning, and we are just covering food and rent, how are we supposed to save thousands of dollars to get a rental bond?”.

In Melbourne and Sydney, peak house prices have made the rental market tough for anyone, but students are the hardest hit. PJ is yet another student being exploited by both Australian and international landlords who cram as many students as possible into a house. Desperate for a roof over their heads, the new arrivals take what they can get and put up with the harsh and expensive conditions. Not every private landlord or sharehouse, student housing company is like this. But there is an increasing number of opportunistic properties owners looking to cash in on the mass migration of international students to Australian Universities.

These so-called ‘slumlords’ own million dollar properties set to be demolished and redeveloped, but will use to house students until the cost of upkeep is too high. Others own multimillion-dollar property portfolios, which can be as little as 2 apartments in the inner city. They then shove as many people as possible in there to maximise their profits. For example, in PJs house, the expected rent is approximately $160 pw. It’s a four-bedroom house which should have no more than five people in it to remain safe. So the rent would be $800. But the landlords are using it as a rooming house, and as there is only one bathroom, the house can now legally house ten people. The maximum amount the landlord receives now is $1600 pw — a 65% profit margin after costs of upkeep and utilities.

We met PJ’s landlord who unabashedly tells LDS he has several houses all over Melbourne just like this one and that he looks after is students. What we saw though was misery and delusion like he thinks he’s doing the students a favour.

Australian media are highlighting the problem

Students living in squalor

The reality is that incoming students are quite naive thinking that it will be easy to secure housing once they get here. They end up in these types of scenarios and don’t understand it’s not ok. Sally, a member of a Melbourne Universities teaching staff, is concerned as these opportunistic companies and private landlords are looking to make money of arguably quite poor students and to their detriment. “It makes me mad. The poor things don’t say anything because they don’t know it’s illegal. I’ve had them come in [to class] sick and feverish because their houses have no heating and it’s 2 degrees at night.”

Sally’s primary concern is overcrowding. Cleanliness was also mentioned, with so many students in these houses, the kitchens and bathrooms are hard to keep clean. There is so much wear and tear that dirt and crud accumulates that can’t be shifted. Properties are old and damp, and carpets are frayed and crusty with years of caked-on dirt. The big concern is seasonal mould growth, known to cause severe respiratory illness and be aggravating to those with asthma, running their immune systems down throughout their tenancy. We asked Sally if she had seen some of these houses “Oh yeah, last week I saw one where they can’t use the kitchen because it’s so dirty, you know, broken titles with mould, not the sort of thing you can wipe up”. Sally made a disgusted facial expression when telling this story, “some students say they don’t cook for fear of getting ill, or there are too many people, they don’t get the chance to cook food as the kitchens never free”.

Dangerous black mould

“It makes me mad. The poor things don’t say anything because they don’t know it’s illegal. I’ve had them come in sick and feverish because their houses have no heating and it’s 2 degrees at night.”

– Sally, University employee

Overcrowding fears

Overcrowding, where more than the legal or recommended number of people are in a dwelling, is becoming worse for students. Some states have laws against overcrowding, but others don’t, which makes it hard to deter landlords from doing it. The main problem is that landlords are not registering dwellings as rooming or boarding houses which must abide by maximum occupancy standards. For example, the general rules of occupancy in these dwellings are that a bedroom must be more than 7.5m2 for one person. If two people are sharing, the room must be 12m2 or bigger, and for every additional person, 4m2 must be added. A recent study of the Sydney rental market by VincentCare Victoria found that more than a quarter of people were living in overcrowded dwellings. Inner suburbs, like those around universities, were the worst, with 40% or places being classed as overcrowded.

Overcrowding, a larger number of people are in a dwelling than recommended for health and safety reasons, is becoming worse for students. Some states have laws against overcrowding, but others don’t, which makes it hard to deter landlords from doing it. Even worse, states like Victoria make it ok to board four people or more in a room and don’t specify minimum space per occupancy.

To be safe, the standards of occupancy in these dwellings are that a bedroom must be more than 7.5m squared for one person. If two people are sharing, the room must be 12m squared or bigger, and for every additional person, 4m squared must be added. A recent study of the Sydney rental market by VincentCare Victoria found that more than a quarter of people were living in overcrowded dwellings. Inner suburbs, like those around universities, were the worst, with 40% or places being classed as overcrowded.

The other issue is that rogue landlords are not registering dwellings as rooming or boarding houses. These rouge operators may ask for cash rather than providing secure BPay or depositing accounts.

This double bed bedroom is 8 square meters

Many students say they don’t mind as long as the place is close to uni and is cheap, but there are hidden dangers too putting up with dodgy rentals that students don’t think about, fire safety is the most dangerous aspect for example. In 2014 the Lacrosse high rise in Docklands, an inner-city Melbourne apartment block popular with Chinese internationals, narrowly avoided going up in flames as dozens of international students were crammed into two-bedroom apartments. Authorities say it’s just a matter of time before we something as horrific as the 2017 Grenfell Tower fire in London which killed 58 people.

The 2014 Lacrosse high rise fire caused by overcrowding

For students, having their bonds’ stolen’ is the biggest fear. Students hand over security deposits to a landlord but are given a poor excuse as to why it is not returned when they leave. This fear traps students. Noise is also an issue. Studying can be almost impossible, and grades are compromised. Living situations are a common reason why students are struggling to pass classes. As with PJs story, students can’t always feel safe in their own home, and this causes stress for students, particularly young women.

This broom closet sized single bed bedroom is 7 square meters

Not just internationals

The issue isn’t only affecting international students, though they seem to be the hardest hit by the housing crisis. Australian students are also struggling, with many of them couch surfing, squatting or ‘sleeping rough’. Raj, a master of data science graduate applied for 70 rental properties before getting a lease, even though he had landed a well-paying role with a large tech company and had a solid rental history.

And while Australian students are generally sympathetic to internationals, they don’t have to deal with the same issues. By the time they are at postgraduate level, they have gotten wise to the illegitimate landlords or have built up income and rental history to get leases with friends. They also are less like to be preyed upon as international slum lords see Aussies as a liability, they know the legalities and are likely to report. But for most Aussies, the problem is hidden and they don’t see any problems.

Ben, a Masters of Communications student at an Inner-city Melbourne University was asked if he thinks international students are preyed upon. He says that “the ones that I see are actually living in secure apartments with one or two other people. Better than where I live now, right in the CBD with other students near the university. They all stay around that one area and it’s close to everything. Mostly Chinese, they can afford it”. This is a common attitude from domestic students, because that’s whats portrayed in the media and that’s what they see of their class mates. Ben said he doesn’t see an issue but knows that’s because the students at his university have the money to pay for housing, but he knows not all students are as affluent as the ones in his course, he just doesn’t see this side.

He’s not wrong. For the privileged few, university accomodation is comfortable and secure. But it is expensive, costing upwards of $33,000 pa including meals and utilities. That’s $635 pw! More than most families spend on rent.

The beautiful rooms at an exclusive Melbourne Uni college

Raj, on the other hand, as an Indian-Australian who has studied in Canada and Australia, thinks that international students need to take responsibility. He says that some students are compromising on the wrong aspects of their budget, and has some advice for incoming students. His tough love attitude is that they need to be less lazy and deal with more travel time because they can’t afford not too.

“You can’t be close to the uni, you will be priced out of the market. You can’t just take a run down place with people you don’t know. You will be stuck. That’s you for the next while, nowhere comfortable to sleep or study. No privacy. You have to compromise. You have to look further out from the uni and take public transport. You will fail your units otherwise.”

Raj, Data science graduate

The reality of the Australian housing market for students isn’t what they are sold by the glossy photos in Uni course guides, because it isn’t the responsibility of the unis. Raj has a point. While rouge operators are predatory and dangerous, students need to understand that planning and research is key to succeeding in their studies, and a reasonable living situation is part of this. Sally says “reach out, we want to help you. Talk to your tutors, lecturers, advisors. Keep talking, scream out if you have to. Some one will hear you”.

Finding a place to live before starting your degree is daunting, but necessary. Start early to make sure you aren’t left out in the cold. Get to Australia as early as you can to beat the rush.

This post is not affiliated with any Australian University or governing body. It is the opinion of the LDS writing team and cannot be considered legally binding advice regarding your living situation. For more information, please visit your state consumer affairs or tenancy advocacy website.

Vacation and Summer Internships for International Students 2019

March Madness is here again, and for International students, this can be a frustrating time.

“Applicants must be Australian Citizens or an Australian Permanent Resident.”

Right?

So LDS has got your back. Here’s a list of vacation and summer internships as well as graduate programs that are closing fairly soon. All of these programs either have a named data science stream or have a heavy focus on implementing big data analytics.

If you come across any more, please leave the link in the comments so we can add them to the list. At the bottom of this post is a list of companies that are known to hire International students. Their applications have now closed for the 2019/20 season. But remember you have 12-24 months to apply after you graduate.

Vacation and Summer Internships

Deloitte

Technology Internship Program
“You’ll spend three to eight weeks in one of our business units whilst getting a taste for what it’s really like to work in a leading professional services firm. At the end of the placement you will leave with the prospect of securing a graduate position, skipping the queue for a final year graduate placement


Closes: 17th March 2019, 5:00pm
Location: Canberra, Paramatta, Sydney, Alice Springs, Darwin, Brisbane, Adelaide, Hobart, Launceston, Melbourne, Perth
Internationals: Yes
Length: 3-8 weeks
Starting: Early January 2020
More information

PwC

PwC Vacation Program
“Gain valuable experience and develop the skills and networks needed to create a meaningful career. Did you know that most of our vacationers receive an offer for a full-time graduate position after completing the program? If this isn’t reason enough to apply, then read on…If you’re in your second last year of university in an undergraduate or postgraduate degree, our summer Vacation Program offers you first-hand experience at the heart of our business. The program runs over three to eight weeks (length varies per state) and includes full-time work experience during your university break, allowing you to get a head-start in your career before graduating.”

Closes: 17th March 2019, 5:00pm (Perth, Sydney, Melbourne, Brisbane) 24th March 2019 (Adelaide, Canberra and Newcastle).
Location: Perth, Sydney, Melbourne, Brisbane, Canberra, Adelaide and Newcastle
Internationals: Yes
Length: 12 weeks
Starting: Early November 2019 – January 2020
More Information – Technology

South32

Technology Internship Program
“South32 is a globally diversified mining and metals company. We produce [various metals] at our operations in Australia, Southern Africa and South America.We are also the owner of a high-grade zinc, lead and silver development option in North America and have several partnerships with junior explorers with a focus on base metals.”

Closes: 31st March 2019, 5:00pm
Location: Perth
Internationals: Yes
Length: 12 weeks
Starting: 25th November 2019 – 14th February 2020
More Information.

KPMG

Vacation Program
“Our Vacation Program allows you to experience day-to-day-life at KPMG. You can feel what it’s like to be part of the team, working directly alongside our people. Want an insider’s view in to the world of KPMG? As part of our Vacation Program, you’ll discover first hand what’s it like to join one of the world’s leading professional services firms.”

Closes: 7th April 2019, 5:00pm
Location: Adelaide, Brisbane, Canberra, Gold Coast, Melbourne, Parramatta, Perth, Sydney, Hobart, Launceston and Wollongong.
Internationals: Yes
Length: 6 weeks in Adelaide, 8 weeks everywhere else.
Starting: Summer 2019/20

Google

Information Technology Intern
“This program is designed to provide exposure to the technology industry for students who are historically under-represented in this field. Behind Google’s simple search box is one of the most complex technology infrastructures in the world. And behind that infrastructure is a diverse group of Googlers who design, build and keep the lights on for our users. We believe that in addition to hiring the best talent, hiring for a diversity of perspectives, ideas and cultures leads to the creation of better products and services.”

Closes: 13th May 2019, 6:00pm
Location: Sydney
Internationals: Yes
Length: 12 weeks
Starting: Summer 2019/20

Telstra

Telstra Summer Vacation Program
“At Telstra, we know development is a key ingredient for a successful career, so we’ll support you with a dedicated manager to help and guide you, as well as a graduate buddy to share their own experiences with you. We give our Summer Vacationers continual support and formal development sessions to build skills and technical knowledge. Your bi-monthly training and development sessions will cover everything from commercial acumen and managing your career, networking and building your personal brand, to practical training on presentation skills, time management and up skilling on the use of key Microsoft Office tools.”

Opens: June/July 2019
Location: Australia Wide
Internationals: Yes
Length: 12 weeks
Starting: Summer 2019/20
More Information.

Graduate Programs

Deloitte

Technology Graduate Program
“Each new employee is paired up with a guide and a coach to help with a smooth transition into Deloitte life. We provide the right kinds of tools and training to ensure all our graduates can achieve career success through mentoring, peer support and e-learning. Depending on your business unit, we offer the chance to develop further through professional study such as the Chartered Accountants Program and Masters of Tax.”

Closes: 17th March 2019, 5:00pm
Location: Canberra, Paramatta, Sydney, Alice Springs, Darwin, Brisbane, Adelaide, Hobart, Launceston, Melbourne, Perth
Internationals: Yes
Starting: Early March and early July.
More Information.

PwC

PwC Graduate Program
“As one of Australia’s leading professional services firms, we have the opportunity to create meaningful societal change. From growing trust in food production to building hope in artificial intelligence, our people create trust in society and solve important problems each day. By creating a Graduate career with us, here you’ll be provided with the opportunities you need to contribute to a ‘bigger picture’.”

Closes: 17th March 2019, 5:00pm (Perth, Sydney, Melbourne, Brisbane) 24th March 2019 (Adelaide, Canberra and Newcastle)
Location: Perth, Sydney, Melbourne, Brisbane, Canberra, Adelaide and Newcastle
Internationals: Yes
Length:12 weeks
Starting: Early November 2019 – January 2020

KPMG

Graduate Program
“At KPMG, we want you to discover the path that excites you most. After all, when you find a practice area you love, you’ll reach your potential that much faster. It goes without saying that we’re looking for graduates who are passionate and curious – smart individuals who are seeking a challenging and rewarding career. Here, you’ll be part of an outstanding team working across a diverse client list. You’ll be dealing with complex issues that will stretch your thinking. And you’ll have opportunities to learn and grow unlike anywhere else.”

Closes: 24th March 2019, 5:00pm
Location: Brisbane, Perth, Adelaide & Regions (Melbourne and Sydney closed on the 10th March)
Internationals: Yes
Starting: Early February.

Telstra

Telstra Graduate Program 
“The program offers you 12 months of challenging, hands-on and rewarding work that will give you the confidence and know-how to launch your career and achieve your goals. If you’re successful, you’ll work in an environment that’s embracing ways of working like Agile, DevOps, Human Centred Design and Lean. You’ll also be able to collaborate with the best people in their field and have the knowledge that your ideas will always be welcome.”

Closes: 24th March 2019, 12:55 pm
Location: Brisbane, Sydney, Canberra and Melbourne
Internationals: Yes
Starting: Early February
More Information – Data Analytics and Management Program

Procter & Gamble

Supply Chain Network Operations ( Logistics & Supply Chain) -2020 Graduate Role
“At P&G we want you to get your career off to a fast start. That’s why we don’t have a traditional Graduate program with rotations or gradual ramp-up periods. We hand you real responsibility from the moment you walk through the door. We believe it’s the best way to learn. From day one, you’ll help develop or support household name brands you know & trust while working in roles that have a direct impact on our global, US$65 billion-dollar business. You could be overseeing a brand with multimillion dollar sales, supporting the launch of a new product in to the market or work on strategic business priorities to enable us to WIN. There’s no limit to what you can achieve here.”

Closes: 29th March 2019, 11:59 pm
Location: North Ryde, NSW
Internationals: Yes
Starting: Early February

South32

Technology Graduate Program
“As graduate in our Technology team you get to explore different technologies to find solutions to complex problems within the business. You will learn to plan, convert and deliver technical problems into real-life business solutions that will help South32 to improve safety and productivity. You’ll be part a team that’s excited by new and upcoming technologies and welcoming of new and innovative ideas.


Closes: 31st March 2019, 5:00pm
Location: Perth and other operations in WA
Starting: Early February
More Information

RioTinto

Graduate Opportunities 2020 – Australia
“Graduate careers rarely impact everyday life. But at Rio Tinto, your innovation will deliver the building blocks humankind relies on. Like the aluminium in our cars. The iron in our buildings. And the salt that seasons our meals. This is complex work that calls for natural leadership and the ability to quickly develop new skills. Work that thrives from creative thinking and demands astute environmental stewardship. Work that ultimately shapes society and drives human progress”

Closes: 31st March 2019, 5:00pm
Location: Perth and other operations in WA and Brisbane, QLD
Internationals: Yes
Starting: Early February

Assessment centres in Perth and Brisbane will take place during the week commencing 20th May 2019.

REA

Data Science and Analytics Graduate
“REA Group is one of the world’s largest property and lifestyle media companies. Best known in Australia for realestate.com.au, the market-leading property site and app, we’re also a fast-growing global business spanning three regions. Our Data Science and Analytics program will immerse you in the full cycle of how we use technology to become more responsive to our customer needs. You will gain exposure to identifying and converting raw data into highly scalable data sets. Alongside how we conduct modelling & profiling analysis to develop appropriate models to predict the behaviour, usage patterns and trends of our consumers.”

Closes: 1st April 2019, 5:00pm
Location: Melbourne
Internationals: Yes
Starting: Early February
Length: 24 months.

carsales.com

2020 Graduate Program – Technology
“Not only are we the largest online automotive, motorcycle and marine classifieds business in Australia, we’re growing across Latin America and Asia. You’ll have the chance to create an impact in roles and projects that really matter to our business. We’ll surround you with a support crew of motivated, creative people who will help you grow your career. You’ll be learning our tech stack, the languages that we code in and how to work in an agile environment. We have a lot of what we like to call “tribes” and each one is working on a different element of the business. As you work through your rotations you’ll apply your skills to the design, development, and implementation of new complex tech systems that each of these teams are working on.”

Closes: 5th April 2019, 5:00pm
Location: Melbourne
Internationals: Yes
Starting: Early February
Length: 18 months
More Information

City West Water

City West Water 2020 Graduate Program
“City West Water is excited to announce the launch of its 2020 Graduate Program. Be part of a dynamic and customer focused business that puts the customer first and benefits the community. A role with us will see you join 450 people including engineers, planners, Information Technology and various environmental experts working across our sites in the western suburbs of Melbourne.”

Closes: 7th April 2019, 11:59pm
Location: Melbourne
Internationals: Yes
Starting: Early February
More Information.

Digital Alchemy

Graduate Program
“Digital Alchemy (DA) is Asia Pacific’s leading data-driven marketing services provider that helps companies all around the region achieve their marketing goals by implementing world class solutions to optimise the use of customer data. Digital Alchemy services Australia’s largest companies across all industry sectors including financial services, telecommunications, retail, media and other industry sectors. Our headquarters are in the heart of Sydney, with offices in Canada, China, Hong Kong, Singapore, Thailand, New Zealand and Melbourne with over 300 employees.”

Closes: Open now
Location: Sydney
Internationals: Yes
Starting: Early February
More Information.

Atlassian

2020 Graduate Analytics Associate
We are blazing new trails to enable meaningful growth decisions, so we’re constantly growing, learning, and trying to do things differently. We’re data unicorns that distill complex analytical terminology and concepts into concise language, and we interface directly with Product leaders to transmit that knowledge effectively. We’ve been described as fearless, but thorough, because we’re dedicated to an agile methodology that enables moving fast while still ensuring high-quality output. We also know all too well the importance of validating assumptions in our data, and we use various types of testing to prove those assumptions right (or wrong).”

What you’ll be working on:

  • Developing quantitative models of user behaviour and product performance.
  • Communicating the modelling, analysis and results via writing and visualisations.
  • Identifying areas for further investigation and creating innovative analysis methods.

Closes: Open now
Location: Sydney
Internationals: Not explicitly stated – though appears so
Starting: Late January
More Information.

  • AstraZeneca
  • Carlton & United Breweries
  • EML
  • FDM Group
  • Grand Metal
  • IRESS
  • Annalect
  • SAS
  • Schneider Electric
  • Tibra Capital
  • Strategy&
  • LEK
  • TAL
  • SIG
  • WPP Consulting
  • ThoughtWorks

Wrangling your data means getting your hands dirty

This is a quick post about why you need to adopt a mindset of iterative and interactive data wrangling.

Iterative data wrangling is the idea that you use your analysis output as a source of information to ‘redo’ you wrangling. Interactive data wrangling is the idea that you manually intervene in the process.

Some schools have the idea that data mining and cleaning should be as automated as possible. A lot of software has been purpose built for this process in the last couple of years. Because the DS community is quite friendly there are a few free software among the proprietary software like the exxy Alteryx and Trifacta.

These include Googles Openrefine and Quadrient DataCleaner.

The problem with data wrangling software is that it misses things. Things you don’t even know were there until they mess up your output. If you aren’t aware of what your data looks then you will have a hard time fixing issues or even diagnosing them. Often the conclusion is ‘ oh I better add some gradient boosting, my F1-score is a little low’. Alternatively, just clean your data better.

The other problem with automatic wrangling software is that the schema is defined for you. This means that you may end up over cleaning your data and therefore overfitting your models.

Iterative data wrangling is a skill though. The more iterations you do, the more likely it is that you are cleaning up your own mess after you spot what you have done to your data on output. Kind of like how you only see the stains on your favourite top after you ironed it and put it on. Shoulda spot cleaned.

For those new to the field, it is a process of failing. The more times you do dumb things, the more you learn and don’t do them again.

I’ll give you an example.

I have a set of tweets that use ‘coz’, ’cause’ and ‘cuz’ instead of ‘because’.

I thought that these words needed to be normalised, I could have probably just thrown these away but I was trying to be overly thorough. So I wrote a script to do this.

I didn’t check my output to often and only really looked at the top 50 words for each of the topics in the output. Later when I was visualising things I saw this:

Jabecausezi bebecause becausey

I realised that because I had manually altered these terms before I had tokenized (and, therefore, created a list, not a string to alter) I had inserted these words into ‘because’ rather than replace those words with ‘because’. I never picked it up because these words were made rare-ish. Not infrequent enough to be removed, but not frequent enough to turn up in the top 50 words for each topic.

Rare-ish is not a real word btw.

A silly example sure, but just one of many dumb mistakes i have made and never repeated. It actually took me a few weeks to pick up on this and it is the perfect example why you need to iteratively and interactively clean your data. My coherence scores (I was topic modelling at the time) jumped right up and even though I don’t really pay much attention to them now (because I review my topics qualitatively), my metric-happy collaborators got a lot more relaxed about the results.

Even Mark Wahlberg recommends it.

Marky is a neat freak

As always retweet this post @data_little for happy cleaning vibes.

Photo by averie woodard.Thanks Averie for the use of your work at Unsplash.

Every Masters of Data Science Degree in Australia – James Cook University

Over the next couple of months, we’re going to be rolling out the most comprehensive summary of every Masters of Data Science (MDS) offered in Australia.

We are so over reading posts from people who have no idea about the courses. You need to be informed about what career path is best for you, data science is a very broad field. When these sorts of posts flood forums like Whirlpool and Quora, the level of confusion and misinformation makes it difficult for prospective students to figure out what’s real and what isn’t.

Just because your mate’s sister’s boyfriend went there doesn’t mean you know if it’s good or not

– Caitie, every Janurary as she filters through DMs from prospective students.

The other reason is that some MDS (not all) are merely money makers for the universities, and that’s not a good thing for unsuspecting students. With thousands of international students coming to Australia per year to study MDS then finding it won’t help them get PR or even the job they were after, we thought it was time for some transparency. We know that the universities reputation is a critical factor in picking where students do their MDS, but the fact is that rankings don’t necessarily translate into the progressive, rewarding career in Data Science that you are after.

Finally, you need to be comfortable where you are studying. A standard MDS is two years. The University you choose needs to be the right fit for you.

LDS cannot guarantee all details here are up to date and complete. Universities often make changes to their courses and so we strongly encourage seeking clarification from the universities themselves. This article should be taken as an opinion piece.

This first summary focuses on James Cook University (JCU) in Queensland, the Sunshine State. Each article will focus on a different MDS ending in a final summary of all MDS in Australia.

James Cook University

James Cook University (JCU) offers an MDS that seems to be have been created for and with industry in mind. For a quick overview check out the promo video from 2017, when the degree commenced.

JCU MDS promo

Aside from Ron’s top acting skills, he makes it very clear that JCU is heavily aligned with industry. So for those of you who want to go into research, you should probably give this one a miss.

Delivery mode

The JCU is geared toward students who aim to go straight into the industry as well as those already working in the industry. The course is offered 100% online meaning you don’t have to move, you can access the course from anywhere. A great option if you have a family, you are already working or you are looking to start working while studying. We don’t know if that means you can get out of doing assignments though.

It’s not unheard of for students to start getting job offers and internships after their 1st year of study in an MDS. Being able to do your units online provides the flexibility to do this which isn’t possible for those doing a traditional contact-based degree.

JCU online students conduct their work via live chat sessions, recorded lectures and the materials provided on the aptly named JCU learning platform, ‘LearnJCU’. Online students are also are offered the support of a curiously titled ‘success advisor’.

Along with your tutors, your Success Advisor is there for you from your
first day as a student with JCU Online. They are readily available to help you navigate your online degree and keep you motivated throughout the duration of your study.

– JCU MDS brochure

Given it is a data science degree we wonder if the success advisor is a chatbot. You may in-fact prefer a face to face delivery and the company of other students for friendship and comradery throughout your course. If this is the case JCU isn’t for you.

Unit set up

Ok, stay with us here, this bit a little confusing. The JCU MDS is not a normal arrangement, which isn’t necessarily a bad thing.

JCU runs on a ‘carousel’ timetable which means that the units, which JCU formally call subjects, are shorter than the traditional 13-week semester.

Each subject will go for seven weeks and you will complete six subjects a year on a full-time load. These subjects commence in January, March, May, July, September and October.

We are massive fans of this set up as you would only do one subject at a time. This mode of study is far less stressful and avoids the enviable ‘assignment crush’ at the end of the semester followed by the hell that is exams of which there are none.

Yes, it’s true. The JCU MDS has no exams. Most of their subjects have a very similar assessment format:

  • 20% tests or quizzes online
  • 60% assignments
  • 20% computational laboratories/log books

Entry requirements

JCU ha no WAM or GPA requirements. So if you slacked off in your undergrad, this might be the course for you. You will require a Bachelor degree that is equivalent to AQF level 7. Preferably, JCU wants to see evidence of ” high numeracy skills equivalent to senior level mathematics that includes algebra and elementary differential calculus”. We assume the senior level mathematics is a reference to your final year of high school.

If you do not have this (math) you can enter the course as long as you can demonstrate at “least five years of relevant work experience in an IT or Data Science related industry. Industry experience will need to include some background in computing, data analysis or programming.”. This is probably >1% of applicants, and it’s a weird entry requirement alternative.

Now if you don’t come from a STEM background, which many MDS students do not, then you can still pursue the MDS at JCU.

If you have absolutely no mathematics and are petrified of the prospect of learning to code then first you need to ask yourself why you think data science is actually compatible with your life choices, but also be aware that JCU has you somewhat covered.

Students are offered a subject called MA5801:03 Essential Mathematics for Data Scientists – which they take before they even start the course! While it seems obvious, this is revolutionary. Time and time again MDS students come into these degrees and fall over straight away in their first mathematics, statistics or programming unit. Failing a unit can be devastating. Aside from the blow to your self-esteem, a fail can be a hard hit to your bank account, job prospects and for international students, your visa status.

Offering an elementary mathematics course is really quite unique and should, in our opinion be adopted by all universities offering an MDS to non-cognate (not form computer science or mathematics backgrounds) students.

JCU allows you to build your qualifications as you go. So if you don’t meet the entry requirements or only some of them then you start off with the subjects built into the Graduate Certificate or Graduate Diploma then continue to the MDS subjects. No, you can only exit with one qualification, not all three.

JCU qualification progression

If you had to start at the beginning, your MDS would take you 32 months instead of the advertised 24. Sneaky JCU. Always speak to your enrolment advisor before you start your classes about what’s most appropriate for you.

Be aware though, just because you can get away without doing a unit, doesn’t mean you should. Often those approving credit for prior learning, aren’t data scientists (what? no way). Frankly, you could lie through your teeth and say you have three years of programming experience. We know a lot of people who get fake reference letters from their former ‘managers’ and buff up their CVs to get out of one unit. I mean, who’s going to check? The problem though is that these students fail. Oh, how they fail, it is spectacular. If they finally graduate, their rock WAM is going to hinder their chances of getting a decent job.

What you will study

So the first downside to the JCU degree is that it’s quite rigid. There are no electives and you would be required to pass every subject to graduate on time. But unlike traditional semester-based units, if every unit is offered in every teaching block, then you should only be delayed by a few months due to the short teaching periods. Have a look at the handbook here.

JCU MDS students will gain membership of the SAS Academy of Data Science if they complete additional subjects. There appear to be two streams of the JCU-SAS Joint Certificate.

JCU-SAS Joint Certificate in Introductory Data Science JCU-SAS Joint Certificate in Advanced Data Science
MA5800 – Foundations of Data ScienceMA5821 – Advanced Statistical Methods for Data Scientists
MA5820 – Statistical Methods for Data ScienceCP5806 – Data and Information: Management, Security, Privacy and Ethics
CP5804 – Database SystemsMA5831 – Advanced Data Processing and Analysis using SAS
MA5830 – Data Visualisation MA5851 – Big Data: Processing and Analysis

It’s a bit confusing, but it looks like the certification is built into the degree. Be aware that the Data Science master classes and projects are not offered in 2019. This makes sense since the JCU-SAS joint certificate has now been implemented.

SAS is a dominant industry analytics platform and employers highly prize proficiency with it. Many graduates have reported that they are required to use it despite knowing other programming languages. In fact, a couple of us have been asked in interviews if we are ‘whizz bang with SAS’, yes that actually happened. So experience with SAS is great but since the material is can be accessed directly through SAS at a cost of $420 AUD per month (more than enough to get familiar with it), it hardly seems worth giving up valuable units that could have been spent on interesting electives.

As fancy as this all sounds, we are still a little sceptical.

Another reason being the level to which teaching staff are proficient with SAS. Most tutors and lecturers you will have are researchers. We know very few Data Scientists who lecture or tutor who are experts with SAS. This could prove problematic for the new JCU program.

Looking at the unit themselves, we see that they are all run out of the college of engineering. They foundations units are fairly standard and nothing here looks very different from other MDS courses. There is a mixture of Python, R and SQL at this level. It’s difficult to tell due to the carousel timetable, but it looks like students will be exposed to R before they take their foundation programming unit CP5805 – Programming and Data Analytics using Python. Although we hope, JCU hasn’t made this common mistake, those who have had no exposure to programming should consider doing a MOOC or two so they don’t find the experience to jarring.

Looking into MA5810- Introduction to Data Mining we read “Software packages will be adopted for hands-on data mining in real data sets.” Since there is no specific programming language mentioned, we wonder if this means that you would be using SAS rather than coding your own algorithms. After reading through all of the subject outlines we couldn’t see any more mention of programming languages other than SAS. This is not to say that there are no further programming requirements but it is safe to say that SAS is the core ‘language of this degree’. After getting our hands on some of the lectures and tutorial exercises for JCU MDS units we were interested to see just how much this degree is catered for industry. 110%. To new and commitment to SAS, it seems, is integrated seamlessly then into these materials.

Costs

Good news if you are a domestic student then you are going to love this. Full fees for domestic students are $52,800 AU which is on the lower end for domestic MDS fees but the best news is that JCU has CSP places. If you are awarded a CSP then you will only contribute $18,704 AU making it amongst the cheapest MDS in Australia.

International student fees are moderate at $63,000 AU. By comparison, Monash University domestic students pay $64,000 AU and international students at The University of Queensland pay $88,102 AU for a two year MDS degree.

English Language Requirements

English language requirements are standard at all universities and vary within a few points of each other dependent on the exam.

Academic IELTS – 6.5 (no component lower than 6.0)
TOEFL (paper-based) – 570 (with a minimum Test of Written English score of 4.5)
TOEFL (internet-based) – 90 (minimum writing score of 21)
Pearson (PTE Academic) – 64.

University standing

Full disclosure, we don’t really buy into rankings. While it’s true that Australian universities do vary in quality, financial ‘freedom’, facilities and research output, these characteristics are not correlated. Our editor has attended and worked for six universities and says

The worst teaching I ever received was at a Go8 university, the cruellest of research environments was too. The most supportive was at a private university and the most disorganised was a rural one.

LDS editor

Key statistics

JCU has 15.2k students of which 20% are international and 38% are postgraduate. JCU does have a gender imbalance, with 63 women studying for every 37 men. The staff to student ratio is 22.3 students to every staff member. We take these stats from the QS rankings.

The World University Rankings

JCU scores in the 201-250 band for the top 1000 universities in the world as ranked by The World University Rankings. Let’s look at that in detail. JCU is in 20.1% – 25% of universities ranked. JCU is considered a ‘young university’ it was established in 1961 university and ranks 28th in the world when compared to other universities of the same level of maturity. Why is this important? Well, older universities have more money.

More money, better everything.

It’s all relative. But when we look at what is relevant to MDS we can see a different story and not a good one.

The rankings are roughly worked out based on teaching, research, citations, industry income and international outlook. See our post on University Rankings for an explainer.

Based on Teaching, which we think is the most important score, JCU receives 23.5 which is not great. Their Industry income is modest at 41.9 and their international outlook is very good at 75.4

Looking at the rankings for Engineering and technology specifically. Teaching received an abysmal score of 19.9, an okay Industry income at 32.8 and a reasonable international outlook at 72.7.

However, their ranking here has slipped into the 301-400 band, down from the 251-300 band in 2018. Not a good sign.

QS Top Universities

Again JCU has slipped from equal 367th to equal 369th place. However, they rank 43rd in the top 50 universities under 50 years old. JCU is ranked 18th in Australia (out of 43rd). Unfortunately, we can’t give more granulated statistics here as they aren’t available.

Affiliations and accreditations

JCU is a member of the Innovative Research Universities Network which is a group of seven universities that undertake advocacy on issues related to higher education, research and university students.

Because we always get asked, no, JCU is not ACS accredited.

Summary

We feel that JCU is not actually offering an MDS. A better title would be business analytics. This is frustrating because it will produce students who are not equipt for data science roles and may either miss out on graduate opportunities or fail when they enter industry.

With so much focus on SAS students will miss out on other languages. Learning different languages is hard but over two years you become adaptable and with practice become proficient. Practice means every day for several hours. But it seems that JCU will not offer this.

We are impressed by the pre-degree mathematics unit but did not see enough probability and statistics throughout the course. The structure, however, is fantastic. Having one unit over seven weeks will allow students to fully concentrate on that unit. The traditional four units per 13 semesters are out of date and make it hard for working students to study, particularly when unit assessments overlap to an unreasonable extent.

Although the lack of exams seems appealing, we question their absence for units that are theory based such as those concerning ethics and policy.

However, hands-on units work better without exams, so scrapping them all together is the lesser of two evils. Practice is more important than an examination where you have to smash out as much as you can remember. Exams don’t have StackOverflow and so are not a reflection of working life. JCU has replaced exams with log books for programming, lab time and exercises.

The organisation of the website is a bit confusing. Changing unit names and not updating them uniformly is the number one way to irritate and confuse new students. Get it together JCU marketing team.

Oh and just a note on the JCU MDS webpage “One of the fastest Masters degree in this field in Australia” isn’t accurate, two years is standard and other MDS offer early exit based on credit.

Pros

  • Accessible to non-cognate students
  • 100% Online delivery
  • CSP places for domestic students. Low fees in general
  • No-exams
  • 7-week teaching blocks of one unit only

Cons

  • Minimal programming and statistics
  • Overreliance on SAS
  • Possibly not reflective of required industry skills
  • No research stream
  • Poor teaching reputation

And there you have it. JCU MDS may or may not be the right fit for you but that is now for you to make yo your mind about.

Retweet this article for happy enrolment vibes @data_little

Thanks to Nicole Honeywill for providing the feature image for LDS to use. Find Nicole on Unsplash.

Are you in or out? Deadline draws closer for your My Health Record creation

The Australian Government’s new $2 billion AU health record information system, My Health Record is set to improve how medical practitioners to access, share and create information about your medical history. The initiative aims to improve the flow of accurate and timely information to GPs and other specialists in order to improve health outcomes for all Australians.

Here’s Dr Caroline Yates providing is a great summary of why medical practitioners will be able to improve their quality of care if their patients are using My Health Record.

My Health Record is a fantastic initiative and could change the meaning of continuity of Care in Australia, at least in theory.
But I, like many people, have some significant concerns over privacy, security and data access. And, as a data scientist, I’m concerned about data use, accuracy, integrity and completeness.

If you haven’t decided whether to opt out, you only have a few days from today. So let’s unpack the ins and outs of My Health Record before you make your mind up.

What are the benefits of having a My Health Record?

The official My Health Record website states that the benefits of creating a record are as follows.

Better Connected Care

Do you see one doctor all the time?

I don’t.

When I’m sick, with say a chest infection I can’t shake, I don’t care who I see at my local practice as long as it is ASAP. I know that doctors who work within that practice can access my previous doctor’s notes as long as my regular doctor entered them on the practice system.

My Health Record is said to be helpful to patients as it reduces their life admin including transferring medical records when they move. I’ve moved six times in 6 years and had to transfer practices a few times. It’s a simple process. I call my old GP and ask for my medical records to be released to my new GP, that’s it! My medical record follows me. But some people don’t do this, and that’s where the My Health Record would benefit them, as doctors at all practices will be able to see their medical history.

Where it may be of benefit is if I’m travelling interstate and I can’t see my regular GP. I mean, who transfers their records for a random one-off visit?

But the scenario where this would have the most impact is in the event of a life-threatening emergency. If you are unconscious, you can’t tell the doctors trying to help you what you think is wrong. Another instance My Health Record would have a critical impact in this scenario is to prevent catastrophic drug reactions.

I went into anaphylactic shock and then my blood pressure bottomed out because they didn’t know I was on beta blockers. If they had known they would have given me other drugs as well to stop that.

– Patient who had an adverse reaction when given treatment

This is where My Health Record is helpful for adults in general. But there are specific groups of people where My Health Record will be even more beneficial.

Parents

minutes is difficult for parents, I can’t even do it for my health history in that time, let alone for a sick child. My Health Record is aimed at providing a holistic picture of a child’s health for this reason. Many a diagnosis has been delayed or missed due to incomplete childhood records, and My Health Record could prevent this. The autonomy of the child has also been taken into consideration. I’m quite impressed by this and children will be able to take control of their record after the age of 14.

Older Australians

The older you get, the denser your medical record becomes. As we age, our bodies are more susceptible to acute infections and chronic disease. For older Australians, My Health Record affords connected care while easing the burden of remembering all of these details. In cases where the individual is suffering cognitive decline, most likely in older age, this system will be beneficial to them and their health care practitioners.

New Australians

Non-English or non-fluent individuals may use My Health Record to overcome the issues stemming from the language barrier that could significantly impair their quality of care. This is all well and good, but I question the integrity of the data in instances where there may be information that is lost in translation but still entered. Additionally, I want to know how this information is made accessible to these patients online. Is it translated into their language?

People with Chronic disease

Up to 21% of drug-related hospital admissions are due to drug interactions in Australia. That’s pretty high. I have a family member with a fairly serious set of chronic diseases. They are meticulous with the information they give to their doctors and keep great track of the large volume of medications they have to take. But for them, their history and medications being accessible to all specialists would have saved them the pain of having to deal with the fallout from drug interactions as they have had various specialists prescribed counter-indicated drugs at the same time.

Can I control my own My Health Record?

Yep, you control your record. This means that you can choose what to delete and who can see your information. If you don’t want a doctor seeing this, you can change your access settings, but that still makes for an awkward conversion in an appointment. You can check who has accessed your record and even get automated notifications via email or SMS. Be aware that in an emergency your access controls may be overridden to you know, save your life. This is the interface you would see if you were tailoring your My Health Record to your preferred level of access control.

Setting up a My Health Record through MyGov

Is it secure?

Depends on who you ask. The system is said to have:

A multi-layered and strong safeguards in place to protect your information including encryption, firewalls, secure login, authentication mechanisms and audit logging.

– My Health Record website

I’m beyond sceptical of this statement.

In 2018 there were 35 breaches of the My Health Record system, which rose to 42 violations in 2018. The Australian Digital Health Agency (ADHA) said that none of these was malicious….ah?

The ADHA insists that these breachers were the result of situations such as the wrong parent having access to a child’s record when they do not have custody of the child and fraud against the Medicare system through individuals accessing records which are not theirs. I’m pretty sure these could be seen as malicious. Most concerningly the majority of breachers were where the Department of Human Services used the same record as Medicare which is a data integrity issue. In summary, there are system integrity and access control problems to overcome.

ADHA may say that My Health Record is secure, but this doesn’t mean all integrated systems are. The healthcare sector in Australia has a pretty shocking history of poor record management. For example, Family Planning NSW, a service that offers reproductive and sexual health services was hacked in April 2018. Up to 8,000 records were held captive under ransomware, and the agency only became aware of this after clients contacted them and journalist Lauren Ingram notified people via twitter.

Would you want your last STD check made publically available?

Lauren Ingram’s Twitter post

Given the sensitivity of this data, I’m not sure I would be forgiving of such a scenario.

Who can access my health Record?

People who can access your record include

  • GPs
  • pharmacies
  • pathology laboratories
  • hospitals
  • specialists
  • allied professionals
  • secondary providers

That’s not vague at all.

At this point, your brain should be blaring, Danger, Danger! Abort, abort!

Not all employees are as ethical as they should be. Even though there are safeguards which flag when a hospital employee accesses a family record or a specific record too frequently, people can easily override this, and they do. One way this happens is because departments don’t adopt their EMR software or lack training in how to use it, making circumnavigating the system much more attractive.

I have seen thousands of records be transferred into excel spreadsheets and become accessible by all in a department. This meant that to get to one record, the practitioners would scroll through the thousands of others and be able to read that sensitive information with no safeguards in place. The one password to the file, located on a shared drive wasn’t changed in 2 years. In small country towns, this becomes problematic as you learn information about your neighbours or even colleagues, that they would have never divulged to you. Friday night at the pub just became 100% more interesting (or awkward depending on how you see it).

Insurance agencies and employers are legally not allowed to access information on your My Health Record. One reason that the last deadline was extended was to allow for this legislation to pass through the Australian Parliament under the My Health Records Amendment (Strengthening Privacy) Bill 2018. Unauthorised civic access may result in a fine of $315,000, criminal conviction and up to 5 years jail time. Big woop, people do it all the time. Again, a systems issue.

This legislation also prevents privatisation of the My Health Record system. In saying that there seem to be some fair dodgy dealings going on that indicate some commercial conflicts of interest.

Law enforcement and government agencies access are a little less clear with the official My Health Record website stating

To date, the Agency’s official operating policy has been that no information within My Health Record can be released without an order from a judicial officer. The Agency has never received such a request and has never released information. Under new laws, no information can be released to  law enforcement or a government agency without your consent or an order from a judicial officer

– My Health Record

So I guess law enforcement can access your record.

The My Health Record website states that researchers are currently unable to access this data, but policies and frameworks are being drawn up to oversee this. It is likely that data will be used for public research after de-identification as early as 2020. However, media reports state that secondary use for research “and other purposes” is going ahead as well.

My Health Record information can be used for research and public health purposes in either a de-identified form, or in an identified form if the use is expressly consented to by the consumer

Department of Health spokesperson

The most concerning aspect of this is the “other purposes” part. This is actually a code for organisations such as pharmaceutical agencies. You need to make your mind up about how you feel about that. And no, this is not a conspiracy theory. Think about the 2017 Melbourne Datathon where 3 million rows of Pharmaceutical sales via banner pharmacies was given to participants. There is nothing wrong with this but in reality, you could do a lot of damage with this if you really wanted to.

Why did they delay the deadline?

The opt-out period, where you can choose not to have a record generated, was initially set to end in November 2018 but Federal Health minister Greg Hunt postponed this deadline until the 31st of January 2019 among concerns about Data Privacy and Security. Minister Hunt sought to reassure the public that all was well with the system and that even if they missed the deadline, they could still opt-out at any point. What Minister Hunt neglects to mention here is that a record would be created for you after the 31st of January, although you can permanently delete this at any time.

The decision came after legislation strengthening privacy protections for the electronic health record system was amended in the Senate to include the extension.

SBS News

There have also been rumblings that a ‘small’ technical glitch from 2016 that has not yet been fixed. This bug could leave patients information inaccessible or out of date. If it’s still around, and I’m not saying it is, then it needs fixing.

Then there’s that ‘small’ matter of two total system crashes in November 2018. The media attention prompted people to try and opt out on the deadline in November 2018, and the unprecedented surge in website traffic crashed the system. Given the Federal Government’s failure to load test this and other previous initiatives, shout out to #CensusFail, it is likely we’ll see this issue again on the 31st of January. I don’t think #MyRecordFail has the same ring to it through.

Get in early to avoid this

Is my information accurate?

This is probably my biggest issue with the My Health Record system. It’s not the system itself but the users. The sad reality is that some doctors are just bad at their job. Because of this, the accuracy of the information could be quite low. If another doctor is using it, then there could be severe implications for a patients health. Fortunately, the Australian Medical Association is treating My Health Record with a healthy dose “professional scepticism” stating that doctors would likely be handling the records the same way as hospital discharge notes, by planning for a 10% margin of error. Because that’s not concerning at all.

How do I opt out?

If you want to opt out, you need to visit the My Health Record page. It’s actually a pretty easy process. You’ll need your Medicare card and drivers licence (or passport). Here is a helpful video to guide you through.

Trials of nationally accessible EMRs began in 2015 when the system was called the Personally Controlled Electronic Health Record (PCEHR), and the eHealth record. Unfortunately, Australians didn’t respond well to this and location-specific trials recommenced in October 2016 with the system being renamed My Health Record. A little under 1 million Aussies registered in the opt-out participation trials. So if you lived in the Nepean region of the NSW Blue Mountains or Northern Queensland be aware that you may already have a My Health Record.

Hopefully, you have a little more understanding of some of the ins and outs of the My Health Record. I have opted out, but that may not be the best choice for you. Seek out further reading and make your mind up for yourself. You have two days.

Subscribe to LDS for more data related joy. Retweet us on Twitter and tag @data_little for happy recordkeeping times. Yes, recordkeeping is one word.

As always LDS cannot absolutely guarantee the correctness of this information and you should seek out further details through the many links in this article to the official My Health Record website.