Online e-commerce applications are becoming a primary vehicle for people to find, compare, and ultimately purchase products. One of the fundamental questions that arises in e-commerce is to characterize, understand, and model user long-term purchasing intent, which is important as it allows for personalized and context relevant e-commerce services.
Understanding online purchase intent and its buildup over time is important because individuals spend large amounts of time and resources on online shopping—in the U.S. alone, e-commerce sales have reached over 350 billion USD per year and are expected to grow at around 15% annually.
Visit Business Understanding Notebook for more detail.
In this project we aim to answer the following analytical questions:
- How accurately are we able to predict that a customer is going to make an online purchase on a website?
- What and what is the range of driving factors that lead to a purchase?
- Can the probability of online purchase produced through data mining represent the "real" probability of customer online purchase?
- Do those who purchase products tend to be of a certain region? returning or new visitor? buying near a special day, particular month or weekend? or spending more time on a specific page or click-thorugh (bounce) freuqently? What pages do they visit and time spent on them before exit?
The origial dataset can be found here: UCI Data
- Our project is divided into 3 main parts
- Steps and important findings will be shared in this report. If you want to see how the code was executed links to full Notebooks are provided at each step.
See 1_Data_Report.ipynb for the final report notebook with explained steps and reccommendations.