Big Data Classification of Users Navigation and Behavior Using Web Server Logs

Abstract

Digital marketing is growing faster to increase business and sale products. All Ecommerce sites send recommendations of products or newly updated product list to the user. They also send exciting offers to users. We need to study and understand behavior and interest of user to recommend products, this is important to adapt to requirements of customer in ecommerce website. The information about users behavior is stored into server log file. We extract log file to get behavior and do pre-processing to get session action details. To analyze log files, proposed system implements a linear-temporal logic to analyze web server logs of ecommerce website. Web server logs are mapped to event logs by process of mapping log records, to capture behavior of customer. To find different behavioral patterns that refers to different action performed by users in session, different predefined queries are performed. We proposed the use of Temporal Logic and model checking approach as an alternative to traditional data mining techniques using and to use it in structured e-commerce websites. The goal of system is to analyze the usage of e-commerce websites and to extract customers behavioral patterns with the use of temporal logic formulas and to describe user behavior against the log model. The system can determine user behavior and interests, it will associate users with each other for better recommendation system.