Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Clickstream Analytics with Markov Chains


Published on

Covers key concepts of clickstream analysis and Markov Chains. Followed by 3 practical applications with the R language:
- Frequent path analysis
- Future click prediction
- Transition probabilities mapping

Published in: Data & Analytics
  • If we are speaking about saving time and money this site ⇒ ⇐ is going to be the best option!! I personally used lots of times and remain highly satisfied.
    Are you sure you want to  Yes  No
    Your message goes here
  • Check the source ⇒ ⇐ This site is really helped me out gave me relief from headaches. Good luck!
    Are you sure you want to  Yes  No
    Your message goes here

Clickstream Analytics with Markov Chains

  1. 1. Clickstream Analytics with Markov Chains MeasureCamp #12 Alexandros Papageorgiou
  2. 2. Next 30 mins ● What is the clickstream ● Markov Chains overview ● Practical examples and techniques ● Discussion
  3. 3. About me Account Strategist @ Google Ireland Consumer behaviour analyst @ Analyst @
  4. 4. Why Clickstream ● Enable advanced types of analysis eg. Data mining/Machine Learning ● Get closer to the voices of our customers ● Go beyond standard segmentation analysis
  5. 5. Alternatives ?
  6. 6. What’s clickstream exactly ? A tidy example
  7. 7. The Weblog
  8. 8. 3 useful applications ● Frequent Path analysis ● Future Click prediction w/ Markov Chains ● Transition Probabilities mapping w/ Markov Chains
  9. 9. Markov Chains ● It’s a 100+ year old theory. ● Hidden Markov Models, Markov Chain Monte Carlo, higher order Markov Chains ● Used widely in science from physics to finance information science ● Models the evolution of dynamic systems in time
  10. 10. Markov Chains ● Page Rank ● Attribution Models ● Clickstream
  11. 11. Markov Chains key concepts Media Exposure through the Funnel: A Model of Multi-Stage Attribution Abhishek, Vibhanshu & Fader, Peter & Hosanagar, Kartik. (2012)
  12. 12. The clickstream R package. Package Author: Michael Scholz - Cluster your clickstream - Model the clickstream clusters as a markov chain - Visualise and calculate transition probabilities - Predict next click given a submitted click sequence. - Convert the clickstream to an object that is ready for association rules
  13. 13. R notebook
  14. 14. Final thoughts Clickstream analysis has been an academic research topic for quite some time Development of practical applications is now becoming more accessible thanks to open source tools and libraries There are several other methods that play nicely with the clickstream such as clustering, network analysis and association rules.
  15. 15. Useful References Clickstream package article on the Journal of Statistical Software Supercharging websites with a real-time R API Wikipedia clickstream rabbit holes MeaureCamp session Clickstream notebook
  16. 16. Questions ? Thank you!