Semantic recommender system for Malaysian tourism industry / Tirad Mohammed Aref Almalahmeh

Tirad, Mohammed Aref Almalahmeh (2014) Semantic recommender system for Malaysian tourism industry / Tirad Mohammed Aref Almalahmeh. PhD thesis, University Malaya.

Preview

PDF (Full Text) Download (4Mb) | Preview

Abstract

Semantic Malaysian Tourism Recommender System (SMTRS) adopts the natural language interface, recommender system and semantic technology to analyse users’ query and provide answers from the Malaysian tourism domain based on the tourists’ preferences. Tourists usually search for information through different search engines. However, as found by various researchers the retrieved answers have two main problems: overloaded and not-related answers. A Recommender System (RS) is one application that can provide personalized information, with the optimal goal of providing personalized information recommendation in order to customize the World Wide Web (WWW). Regular RS users query the system by choosing from a fixed set of attributes represented by option sets or dropdown lists. Menu-driven navigation and keyword search currently provided by most commercial sites have considerable limitations because they tend to overwhelm and frustrate users with lengthy, rigid, and ineffective interactions. This research proposes incorporating semantic technology with a recommender system to deliver information that is more related to the tourists’ interests. At the same time a User-friendly Natural Language Interface is also included to assure convenient query access to the Semantic Web data, where the Natural Language Interfaces are perceived as the most acceptable by end-users. The approach results in a prototype with an architecture consisting of a Content-based Recommender System, Semantic Technology, ontology engineering in the Malaysian Tourism domain, and Natural Language Interface. This research found, users are satisfied with the proposed services giving it an excellent rating based on the System Usability Scale (SUS) acceptability score.

Actions (For repository staff only : Login required)