A Movie E-Shop Recommendation Model Based on Web Usage and Ontological Data

Introduction

Nowadays, e-commerce features as an extremely dynamic economic sector and at the same time, as one of the most appealing ways of beginning or expanding a business activity. Statistics for its future are quite optimistic indicating that the business-to-consumer (B2C) e-commerce will grow rapidly in the years to come (Browne et al., 2003). In fact, despite the recent talking about businesses cases that went online but did not manage to profit, looking at the big picture reveals that such cases comprise just a small portion of the market and most “inhabitants” of the e-commerce world are not only surviving, but also doing very good business. This is a strong indication for assuming that there are certain distinctive features that make the difference and, which result in either success or failure. Moreover, surveys and studies have shown that consumers are increasingly giving online commerce a chance (E-commerce Alert, 2006). Taking also into account the rapidly growing number of those that have access to the Internet and those that gain familiarity with the medium, we have a quite promising picture (Internet World Stats, 2006).

In this paper we argue that one of the factors that may block or urge online sales concerns the individuality of Internet shoppers. There is no dispute about whether or not e-commerce has a future; it is a new philosophy in conducting business and it is here to stay. The question is how easily Internet users become e-consumers and which are the internal “mechanisms” and external factors that participate in an e-purchase. The problem arises from the fact that shoppers with varying needs, preferences and background navigate through large and complex web structures and are confronted with too many options, missing in many cases the goal of their inquiry. Generally, search engines are used for filtering pages according to explicit users’ queries. However, their results are often poor since the produced lists are long, unmanageable, and contain irrelevant pages (Middleton et al., 2004a).

Web personalization is one of the most promising approaches to alleviate this information overload problem and to provide users with tailored experiences. Eirinaki and Vazirgiannis (2003) define personalization as “any action that adapts the information or services provided by a web site to the knowledge gained from the users’ navigational behavior and individual interests, in combination with the content and the structure of the site”. In this direction, recent web technological advances help online companies to acquire individual customer’s information in real-time and with low cost. Based on this information, they construct detailed profiles and provide personalized services. Thus, e-shops have now the opportunity to improve their performance by addressing individual user preferences and needs, increasing satisfaction, promoting loyalty, and establishing one-to-one relationships.

Recommendations systems or recommender systems (RSs) that comprise the most popular forms of personalization are becoming significant business tools. They emerged in the middle of 1990’s (Resnick & Varian, 1997) and from novelties used by a few web sites have changed to important tools incorporated to many e-commerce applications (e.g. Amazon.com, eBay.com, CDNow.com). Specifically, these systems take advantage of users’ and/or communities’ opinions in order to help individuals identify information or products of interest (i.e., relevant to their needs and/or preferences). The recommendations may have various forms e.g. personalized offers, prices, products or services, inserting or removing paragraphs, sections or units, sorting, hiding, adding, removing, or highlighting links, explanations or detailed information, etc. (Brusilovsky, 2001).