Web Usage Mining Issues in Big Data: Challenges and Opportunities

Introduction

World Wide Web is an internet server which hosts the abundant amount of information in the form of web pages. Search Engines like Google, Yahoo, Bing, DuckDuckGo use data mining techniques to extract the web pages from the web and the whole process is known as web mining (WM) (Furukawa et al., 2015). According to analysis targets, Web Mining is organized into three main areas: web content mining (WCM), web structure mining (WSM) and web usage mining (WUM). Content Mining is process of analyzing the contents of web pages. WSM is the process of mining the knowledge about the web pages like the ranking of web pages, how these web pages are interlinked with one another (Kumar & Singh, 2017). In WUM, web log data related to a user is analyzed while the user surfs the web (Neelima & Rodda, 2015). This usage information is further used in order to predict the future needs of the user and for the neighboring users. Nevertheless, the abundant amount of information available on the web creates a challenge to both the customers and the companies. The customer is presented with multiple choices of products for a specific need which leads to product overload. Consequently, the need for computing based advertising strategies like one to one marketing and client Relationship Management (CRM) has been stressed upon by both researchers and companies. An effective strategy to overcome this product overload is by providing personalized web recommendations in which the user is interested in. Malik and Fyfe (2012) describe web personalization in three different phases: Learning, Matching and Recommendation. The Learning phase is further subdivided into two types: Implicit Learning and Explicit Learning. The second phase is matching. Matching phase includes filtration processes which are Collaborative Filtering (CF), Content Based Filtering (CBF), and Hybrid Filtering. The last phase that is recommendation phase is responsible for providing the set of personalized results to the users (Al-Shargabi et al, 2020; Aljawarneh, 2012; Aljawarneh et al, 2017; Chehbi-Gamoura et al, 2018; Esposito et al, 2018; Jaswal et al, 2019; Kalpana et al, 2018; Lizcano et al, 2020; Malhotra et al, 2019; Mohammed et al, 2019; Mouchili et al, 2018; Singh,2011).

The speed and ease with which electronic transactions can be carried out on the web have been the key force behind the instant growth of electronic business. Eirinaki and Vazirgiannis (2003) describe the need for combining the e-commerce and semantic web and evaluated the benefit of this grouping. Their work further insists on the importance of semantic web technique and emphasized that the modern web technology has the ability to extremely encourage the prospect improvement on the internet. The beginning of big data and data technologies instigated novel research issues to the personalization community. Big data is a novel field that deals with large collection of data- both structured and unstructured- which is increasing exponentially with time. The development of big data has occurred only recently. The act of collecting and storing enormous amount of data started in the early 1950s when mainframe computers were came to the market. In the period from 1950s to 1990s, the data generated slowly because of high cost of computers, storage devices, network; hence expensive access to computers. The data during this period was exceptionally structured, essentially to support operationally information systems. The advent of World Wide Web in the 1990s generated enormous amount of data hence development of big data analytics. The hugeness, robustness, and versatility of huge data shift Web personalization into a new research situation.