Development of an Enhanced Knowledge Retrieval System Using Web 2.0 Technology and Vector Space Model

Abstract

There is an increasing pool of information on the web and a major contributor is web 2.0 technology on which social media is based. Searching for specific information in this pool is always tasking, therefore, the need to harness this information as a means of enhancing retrieval and reuse of relevant ones. Some researches and development have been carried out in the field of Knowledge Retrieval using Vector Space Model (VSM) and Latent Semantic Indexing (LSI), but the approach used is based on large pool of information available online, which makes getting most relevant information relatively difficult at the point of retrieval, this is a major setback.

Collaborations on Facebook and Twitter (web 2.0 technology) were harvested using APIs and stored in the Knowledge Repository, The collaboration on social media served as the source of information in the Knowledge Repository. An Enhanced Knowledge Retrieval System (EKRS) applying VSM was developed and implemented. The use of VSM was to calculate the Cosine Similarity and Term Frequency to aid effective retrieval of relevant documents from the repository based on user’s needs. In this project, we were able to achieve the aim of retrieving relevant documents. EKRS was able to employ both web 2.0 and VSM to meet specific user’s information needs.

Keywords: web 2.0, Knowledge retrieval, Vector Space Model, Latent Semantic Indexing, Knowledge Repository, Cosine Similarity and Term Frequency.