Product list optimization project, part 1.3. GA tuning: Pagetype dimension

This post is a little diversion from the main course of this project.

Earlier I’ve been working on another UX research project which goal was to discover behavioral patterns in users’ sessions, in order to improve information architecture of the website. My approach was to apply hierarchical clustering to a number of sessions. But it’s hard to find any patterns when you parametrize session via URLs of each page in this session. Simply, it’s too much of diversity.

So I came up with idea to make a new dimension, Pagetype. Roughly speaking, it’s a place (coordinate) of this page in the website’s information architecture. It looks like this:

“<Senior section of the website> <level X> <particular type of the page>”

For example, “catalog level 4 product list”, “catalog level 5 product page”, “news level 2”, etc.

Performing some magic on distance measure and using Ward's minimum variance method while clustering, it gives very interesting outcomes for the information architecture analysis.

But let’s get back to our project!

I’ve mentioned Pagetype dimension because it helped here in this product list project too. Website I’ve been working on has product lists on different levels of catalog, from 2 to 5. So I used Pagetype custom dimension to filter product list pages from pages of other types.

This website happened to have a little bit weird CMS, and only javascript after page is loaded can decide whether it is a product list or not. So I’ve made a dedicated “Window loaded” tag in Google Tag Manager to track the Pagetype custom dimension.

I recommend to have such custom dimension if you work with behavioral patterns.

Kaggle Walmart competition

Just submitted my code on GitHub, which gave me the first Kaggle badge (top 25%).
https://github.com/Oleg-Davydov/kaggle_walmart_competition
I used Gradient Boosting with R package caret. But before that, there was a lot of preprocessing.
This was my second Kaggle competition. I finished 86th out of 485 participants, 1% of score difference with the winner.
Probably not bad for the start :)

Wine quality assessment app

Developing Data Products course, ninth and last in Coursera's Data Science Specialization, finished. As a course project, we had to build web-application with elements of machine learning. And then make a pitch presentation for that app. We were expected to use Shiny package and Shinyapps.io platform to publish our projects on the web.

I found an interesting dataset of Portugal wines' chemical characteristics and assessed quality in UCI repository. It gave me an idea that one can predict wine quality after chemical analysis.
Random Forest was chosen as a modelling tool. I hosted the model right on my github account:
https://github.com/Oleg-Davydov/winequality

And there you are! :)
https://laborant.shinyapps.io/winequality

Pitch presentation was restricted by five pages including the cover. I added a bit of humor in it :) Use right and left arrows on your keyboard to turn pages.
http://oleg-davydov.github.io/winequality/Rpresenter.html

Statistical Inference and Regression Models

That was different! Two new courses in Data Science Specialization on Coursera. Two courses of almost pure mathemathics. Pure pleasure! :) Many thanks to my instructor Brian Caffo, professor of Johns Hopkins University!
Even though these concepts are mostly familiar to me thanks to my alma mater, Moscow State University, it's very useful to see them again in combination with R language.