Click Stream Analysis

7

Click here to load reader

Transcript of Click Stream Analysis

Page 1: Click Stream Analysis

Bisen Vikrantsingh

Kodamasimham Pridhvi

Vaibhav Singh Rajput

1

Page 2: Click Stream Analysis

Intro

Dataset

Analysis

Approach

2

Page 3: Click Stream Analysis

The goal To design models

To support web-site personalization and

To improve the profitability of the site by increasing customer response.

3

Page 4: Click Stream Analysis

SET - I

Session, cookies ID, date, time

Customer ID, visit count, gender, age, martial

status,Income, occupation, home market value , howDidYouHearAboutUs, HowDidYouFindUs, U.S. state

Page View View count of 80+ different

pages/product, last page view, assortment path with level

4

SET - II

Session ID, date, time

Customer ID, visit count, gender, age, martial

status,Income, occupation, home market value , howDidYouHearAboutUs, HowDidYouFindUs, U.S. state

Order Date,time, amount, tax, discount,

shipping amount, promotion code

Page 5: Click Stream Analysis

Questions - When given a set of page views,will the visitor view another page on the site or leave?

which product brand will the visitor view in the remainder of the session?

characterize heavy spenders

characterize killer pages

5

Page 6: Click Stream Analysis

Stage-I Data cleaning

Filtering attributes

Load into RDBMS

Stage-II Classification of users

RFV analysis

Classify users into potential or not

Decision tree algorithm

Clustering of products

Dimension to be consider Product, age group, location, purchase amount

Fuzzy clustering

Find correlation

Correlation between {advertise,gender,income,brand} and {product view/purchase}

Stage-III Answer question mention in previous slides 6

Technology Stack:• Python• MYSQL

Page 7: Click Stream Analysis

7