Evaluating the Effectiveness of the Vector Space Retrieval Model Indexing

초록

Modern information retrieval activities are supported with software systems that facilitate the users' information searching. Information retrieval systems are significantly improved in the past few decades. Now days, there are three types of retrieval models: Boolean, Vector Space and Probabilistic. In this study, we examined the vector space model where documents and queries are represented as vectors. We conducted a number of experiments on the indexing technique of the vector space model to quantitatively describe the effectiveness of the techniques using Lemur Toolkit. The result indicates that stop word removal and steaming techniques improve the quality of the index terms.

키워드