Analyzing and modeling complex and big data

Mind the Gap
Mind the Gap

The following video raised some interesting points that I will investigate further when considering learning analytics  and big data this week:

  • How big is big data?
  • What to be careful not to miss when using big data?
  • Why do patterns emerge from big data and do we address them and learn from them?

Professor Maria Fasli reminds us that that we should ‘mind the gap’ between big data and hypotheses to avoid missing the opportunity to discover new knowledge.

 

 

Week 8 – Weekly Synthesis

I’ve been studying all things algorithms this week and found it to be a massively complex yet fascinating topic. It almost feels as if it would be impossible to fully comprehend the scale and spread of algorithms and the influence they have on our daily lives.

To that end, this week’s content on my Lifestream blog has helped me to start make sense of it.

My ‘How algorithms rule the world‘ post helped me gain some perspective about how computer based algorithms can affect the physical lives of every day users. I firstly considered this from an educational point of view however my thinking expanded somewhat after considering the policing example within the article. I now feel that algorithmic culture has a direct influence on societal culture.

I am fascinated not only with the use of algorithms to benefit large volumes of people, but also their role in predicting the future based on likelihood and probability. This theme was touched upon in my cyberpunk-related post with a reference from Red Dwarf.

My final two entries explored social factors (podcast) and big data influence (lecture) when experiencing algorithms on the internet. It was exciting to then have the opportunity to extend this knowledge into my final task.

My studying for the week concluded with a mini-experiment that I conducted in partnership with Chenée. This was a great opportunity to learn first-hand about the amalgamation of social and material factors in influencing an online experience. Our findings complimented Enyon (2013) in that our options are often influenced by the trends set by the wider, global population.


References

Enyon, R. (2013). The rise of Big Data: what does it mean for education, technology, and media research? Learning, Media and Technology 38(3): pp. 237-240.

 

Collaborative filtering

I wanted to learn a little more about how sites such as Amazon compile their suggested product lists based on other people’s spending habits.

I came across this lecture delivered  via a MOOC by the University of Washington. The lecture clearly explains that it (broken down into basic form) is simply a case of counting the number of shoppers that bought any combination of products and offering the most popular items as last minute add-ons.

I found this very helpful when considering the building blocks of algorithms. It would appear that in this instance “Big Data” is being used for creative analysis for the benefit of the masses (Enyon 2013). However there is certainly scope for acknowledging that there are other behaviours within an online shopping experience that may not be identified by spotting trends.

I hope to cover such behaviours and trends by conducting a small experiment and documenting my findings before the end of this week.


References

Enyon, R. (2013). The rise of Big Data: what does it mean for education, technology, and media research? Learning, Media and Technology 38(3): pp. 237-240.

Weapons of Math Destruction: how big data and algorithms affect our lives by Guardian Science Weekly

http://ift.tt/2mbyy2k

“Algorithms are trying to service us based on a computer profile of what we do online, that only has to do with what we do online. It has nothing to do with what we do offline”

This quote has been ringing in my ears since listening to this podcast.

I define an algorithm as a computed process that generates outputs based on trends, statistics and behaviours. Trends and statistics can often be clean cut. However I wonder how accurate digital representations of human behaviour can ever really be?

The consequences of getting it wrong will only continue to grow as algorithms are increasingly embedded into our daily lives.
via IFTTT & Soundcloud

Liked on YouTube: Red Dwarf Redux – S10E04 – Fathers and Suns

 

Viewing points – between 8 mins 40 seconds and 10 minutes

This video popped into my head when I was reading the contents of my previous blog post ‘How algorithms rule the world’.

I drew comparison between the idea of allocating police resources based on the output of algorithms and the actions of the on-board computer in the above video.

Both sources suggest that algorithms can be used to predict future behaviors based on past behaviors, probability and recent trends.

The video also links the cyyberpunk themes (covered in Block 1) to algorithmic theme that we are currently studying.

By: cpsaros

Hi Stuart,

Thanks for such positive feedback!

I think it’s an interesting point you make about the LMS. Something I noticed very early on in my second MOOC was how differently the tool was used and I certainly think that it played a big role in inhibiting interaction.

I was really good to be able to discuss this MOOC with you behind the scenes. The interaction we had , help me identify the key differences in regards to community development and community participation.

from Comments on: Tweet: Micro-ethnography http://ift.tt/2myOcYA
via IFTTT

By: Renee Furner

Hi Stuart – and thanks for your comment and generous review!

I’ve added a more wordy, written format of my micro-ethnography, because, as you allude to, there are quite possibly more factors that were influential in the low uptake of communication. The short version:
-it’s a new ‘group’ – it’s probably unrealistic to expect norm formation and relational exchanges, as per Kozinets’ community progression model (2010);
-the MOOC only lasts for 6 weeks at any rate, so it is unlikely that participants anticipate future interaction, and therefore they may remain task oriented (Walther, 1997, cited in Kozinets, 2010, p. 24)
-the course is, like many MOOCs, information oriented, and prescriptive about what needs to be learned. Perhaps real participation needs to be student driven: students deciding what and how they learn (and with whom).

Heading to your blog just now.. Thanks again!
Renée

from Comments on: micro-ethnography http://ift.tt/2mOnnjR
via IFTTT