A Framework for Multi-Objective and Comprehensive Analysis of Web User Behavioural Patterns

Abstract

In the contemporary era, businesses are driven by web applications. Customers or users of web applications play vital role in the growth of business. In this context, for organizations, it became indispensable to understand user behaviour towards the web application and its services or products. Unless user behaviour is analysed, it is not possible to improve the tool through which organizations drive business round the clock. Web user behaviour analysis is the process of collecting, quantifying and analysing users’ interaction with the web application and its components associated with different services. In order to improve web user experience, the organizations need to analyse users’ behaviour and improve the application and services from time to time towards improvement in business. Many researchers contributed towards the techniques used to analyse web user behaviour. However, a comprehensive framework that looks into multiple objectives and aspects of user behaviour is still desired. To fill this gap, in this paper, we proposed a framework for multi-objective and comprehensive analysis of web user behavioural patterns. An algorithm by name Multi-Objective Web Log Analysis (MO-WLA) is proposed. Since the major source of user behaviour is web logs, the proposed algorithm analyses multiple aspects of user behavioural patterns to gain insights into the behaviour of web users. It analyses user activities and interaction with different web components. Python data science platform is used to develop a prototype application. The empirical study revealed that the proposed algorithm outperforms state of the art.