Mining tweets with R: Uncover the secrets spoken in the twitter universe

Twitter is currently ranked in the top 10 for most-visited website, and averaging on 500 million tweets a day. It is not just a microblogging outlet used by individuals and celebrities, but also for big commercial organisations, such as Telstra, NAB and Qantas, as a communication channel. However, few companies have deployed data analytics in this space due to the challenges in mining unstructured data. And hence, it is unclear what value can be achieved from mining twitter data. I will embark on the journey to explore some of the data mining techniques that can be applied on tweets to uncover potential gems for business or personal use.

Eu Jin Lok is a data analyst for iSelect, a graduate from Monash University with a Masters in Econometrics and has been using R for more than 3 years now both professionally and for causal purposes (eg- Kaggle). This will be his 2nd talk for the MelbURN group and in this talk, he will embark on the task of applying data mining techniques on tweeter feeds using a real example.

 

Join or login to comment.

  • Eu J.

    Hey guys

    The presentation is now up on slideshare (http://www.slideshare.net/EuJinLok/data-analytics-on-twitter-feeds). The organizers will host files on the site sometime next week. The codes will be ready on GitHub next week possibly. I'll put a notice on this thread once its up and loaded. Or you can follow me on Twitter @ejlok1 (I usually announce it on Twitter first before anywhere else)

    1 · April 5, 2014

    • Mark A.

      Hi Eu Jin. Is the code available yet?

      May 22

  • A former member
    A former member

    Great discussion on experimentation with Twitter and Visualization. Thanks guys!

    April 2, 2014

    • Eu J.

      Hey Mark, I think you're referring to the "space" plot? Its from rgexf package: http://cran.r-project...­

      April 5, 2014

    • Eu J.

      If you refer to page 12 on the link above, run the code snippet and you'll get an example plot like I did. Enjoy! Let me know if you have any questions

      1 · April 5, 2014

  • Abiodun M.

    i can we have presentation slid

    April 5, 2014

  • Preeti J.

    I agree with Jason. Generalising trending basis the available data to the entire population is not correct here simply coz the tweets are not random and can be biased. However, if we get a good sample then the analysis which was done can be interesting and reliable. So i guess important in today's presentation was the ways the data can be looked at and not the results which came from the unreliable sample. Secondly I think in any inference it is important to tell about the population for which one is going to collect the sample from. In todays case the population definitely wasnt the overall tweeta in the world since the text mining was limited to english language.

    April 3, 2014

    • Mark A.

      and multiple sources over one source is always better

      April 4, 2014

  • Khurram

    Hi Eu Jin , Could you please share us yours presentation pack?

    2 · April 3, 2014

    • Mark A.

      and the GitHub account :)

      April 4, 2014

  • Mark A.

    Great intro into tweets! Questions to ponder - are tweets a resemblance of public opinion? What is the geo demographic skew? Do tweets predict?

    April 2, 2014

    • Khurram

      You can use streaming technology to land all the gegrophical dispersed data..

      April 3, 2014

    • Mark A.

      Interesting article Jason. The first thing I looked for was Turkey. I would also agree with you Nick, tools are used to analyse the data but to bring the insights to life a little flare of imagination on logic is needed. I agree that Eu Jin did a great job!

      April 4, 2014

  • Khurram

    Great work EU Jin , it gave us an insight that how can we uncover social media data to an information.

    April 3, 2014

  • Eu J.

    Hey guys. Thanks for all the comments. Just to re-iterate again that the presentation is focused on possible technical applications, not making final conclusions about @MAS. It merely serve as a good example, and if there's something you can bring home from the presentation, it would be the techniques I've applied, not the conclusions about @MAS

    Regards
    Eu Jin

    3 · April 3, 2014

    • Mathew K.

      You did a fantastic job Eu Jin and your presentation did indeed provide examples of the different ways data can be mined and compared to a hypothesis. This was a great example of your scientific approach. You also clearly identified the limitations you encountered along the way, proving your transparent approach to what you were wanting to get out if your project and 'your' interpretation of the results. You have also indicated and outlined to the audience the required data and method they can use and tools. If they want to draw their 'own' conclusions if the studies. Job well done. From a teacher and newbie in the field wishing to start learning more.

      April 3, 2014

    • Jason T W.

      Definitely - thank you for taking time to build the presentation and sharing your thinking Eu Jin. It was insightful and as a beginner to R I definitely benefitted from it. Much appreciated :)

      April 3, 2014

  • Parth P.

    Seems like I missed a lot by being absent! Was this recorded? Any slides?
    Thanks
    Parth

    April 3, 2014

  • Andrew M.

    I couldn't make it unfortunately. Did anyone video the session?

    April 3, 2014

  • Nick S.

    A bit more than expected or advertised.
    But gReat.

    April 2, 2014

  • Alida A.

    Wow I'm feeling like Alice in Wonderland!

    March 31, 2014

Our Sponsors

People in this
Meetup are also in:

Create a Meetup Group and meet new people

Get started Learn more
Henry

I decided to start Reno Motorcycle Riders Group because I wanted to be part of a group of people who enjoyed my passion... I was excited and nervous. Our group has grown by leaps and bounds. I never thought it would be this big.

Henry, started Reno Motorcycle Riders

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy