Clustering of the Web Search Results in Educational Recommender Systems

Introduction

Learners are asking for intelligent services in order to discover and access the content they need. The mechanism for discovering Web documents are powerful search engines, with specialized discovery services, indexes, and databases. But, a simple query can produce hundreds or thousands of results, making it practically impossible for the trainee to check the relevance of each of them. .This chapter argues that grouping the results into relevant clusters might make the learner to use them more efficiently. The chapter describes a meta-search engine of an education recommender system, attached to an e-assessment application for project management knowledge. The meta-search engine is meant to deliver bibliography (web pages) from the internet according to trainees’ results obtained at an e-assessment task. Meta-search means that, for a specific query (or mistake made by the trainee), several search mechanisms for suitable bibliography (further reading) could be applied. The lists of results delivered by the standard search mechanisms are used to build thematically homogenous groups using an ontology-based clustering algorithm. The clustering process uses an educational ontology and WordNet lexical database to create its categories. Organizing search results in clusters is not meant to replace the classical way of presenting results in ranked lists. Its purpose is to provide supplementary organization for those results. The clustering method will provide a series of search results clusters, with the properties that the pages inside one cluster are similar to each other, and the pages belonging to different clusters differ from one another. Inside each cluster, the initial ranking order provided will be preserved.

The chapter demonstrates that a meta-search approach using an ontology-based clustering algorithm for results’ presentation produces high quality recommendations for the students who want to use the e-assessment tool in project management and, thus, sharpen their knowledge. Exploiting the clustering technologies improves the performance issues of the educational recommender system attached to the e-assessment application, which gains a formative value and becomes a learning tool. The issue of educational recommender systems, closely related to web search applications, is a common concern in Technology Enhanced Learning (TEL) domain: “Since information retrieval (in terms of searching for relevant learning resources to support teachers or learners) is a pivotal activity in TEL, the deployment of recommender systems has attracted increased interest.” (Manouselis, Drachsler, Vuorikari, Hummel, & Koper, 2010) An insight of other researches regarding ontology-based clustering of web-meta search in recommender systemsis also presented, before describing the original solution. Experimentation results are presented and the efficiency of the proposed solution is discussed in the context of similar applications.