Abstract:Mobile devices become more and more prevalent in recent years, especially in young groups. The rapid progress of mobile devices promotes the development of M-Commerce business. The purchase on mobile terminals accounts for a considerable percentage in the total trading volume of E-Commerce and begins to draw the attention of E-Commerce corporation. Alibaba held a Mobile Recommendation Algorithm Competition aiming to recommend appropriate items for mobile users at the right time and place. The dataset provided by Alibaba consists of about 6 billion operation logs made by 5 million Taobao users towards over 150 million items spanning a period of one month. Compared with traditional scenarios in purchase predicting, the competition raised three challenges: (1)The dataset is too large to be processed in personal computers, (2)Some days with great discounts provided by Taobao Marketplace are within the period of dataset, (3)Positive samples are too few compared to the dimension of features. In this paper we study the problem of predicting the purchase behaviour of M-Commerce users, by exploring the solution for Alibaba's Mobile Recommendation Algorithm Competition. We first deeply study the habit of customers and filter many outliers. After that we adopt the method of "sliding window" to supply positive samples of training dataset and smooth the burst of sales near Dec 12th. We design a feature engineering framework to extract 6 categories of features that aim to capture the buying potential of user-item pairs. Our features exploit the interaction of user-item pair, user's shopping habit and item' attraction for users. Then we apply Gradient Boost Decision Trees (GBDT) as the training model. In the end, we combine outputs of individual GBDT together by Logistic Regression to get the final predictions. Our solution achieves 8.66% F1 score, and ranks the third place in the final round.

E-commerce Purchase Prediction Approach by User Behavior Data.

Purchase Prediction Using Tmall-specific Features

Predicting Users' Purchasing Behaviors Using Their Browsing History

Browsing Behavioral Intent Prediction on Product Recommendation Pages of E-commerce Platform

Research on Statistics-Based Model for E-Commerce User Purchase Prediction

A Method of Purchase Prediction Based on User Behavior Log

Purchase Behavior Prediction in M-Commerce with an Optimized Sampling Methods

Research on Consumer Behavior Prediction Based on E-commerce Data Analysis

Purchase Prediction Via Machine Learning in Mobile Commerce

Personalized Recommendation Model for Mobile E-commerce Users

Online Consumptions Prediction Via Modeling User Behaviors and Choices

User Behavior Analysis and Prediction Algorithm Design of E-commerce Platform Based on Big Data Analysis

Impact of recommender systems on unplanned purchase behaviours in e-commerce

Making Recommendations Better: The Role of User Online Purchase Intention Identification

Enhancing customer behavior prediction in e-commerce: A comparative analysis of machine learning and deep learning models

Can Online User Behavior Improve the Performance of Sales Prediction in E-commerce?

Predicting Purchase Behaviors from Social Media

Multi-Classes Feature Engineering with Sliding Window for Purchase Prediction in Mobile Commerce

A Preliminary Study Of E-Commerce User Behavior Based On Mobile Big Data

Click Prediction for Product Search on C2c Web Sites

Personalized Recommendation with User Categories of Different Behavior Modes: A Case in an E-Commerce Platform in China