The goal of our project is to create a window of opportunity for the E-commerce sites to increase their sell value by understanding the customers and their needs.The more we know about the customers,the easier it is to identify opportunities to sell them new products and target them with appropriate offers.
Our system will take a sequence of click events performed by some user during a typical session in an E-commerce website as input.The aim is to predict whether the user is going to buy something or not, and if he is buying, what would be the items he is going to buy. The task could therefore be divided into two sub goals:
- Is the user going to buy items in this session?
- If yes, what are the items that are going to be bought?
The German company Yoochoose, which provides recommendations for ecommerce platforms, news and media, sponsors the competition. They give us a dataset of completely anonymized (unless otherwise proven) implicit feedback data coming from an ecommerce business located in Europe. The business sells all kind of stuff such as garden tools, toys, clothes, electronics and much more. The dataset has: 33 MILLION CLICKS 1.1 MILLION BUYS 9 MILLION SESSIONS (USERS) 53K ITEMS
The data represents six months worth of clicks grouped in sessions, from April 1st to September 30th. For some of the sessions (around 5%), there are also buying events,which is the interesting part, as the goal of this challenge is to predict whether the user(a session) is going to buy something or not, and if she is actually buying, what are the items bought. At this point it is important to note that user and sessions are equivalent in this problem. Since the data is anonymized, the terms session and users are interchangeable.
Item clicks + Buys/click + popular items + Hour of Day + Month Of Year = 45821