Vector Space Modeling: An Intuitive Approach to Information Retrieval

The purpose of this independent study is to introduce readers with a strong enough background in mathematics to the field of information retrieval. The project is laid out in three chapters. In the first chapter, a variety of perspectives are introduced that explore the motivation and background of information retrieval. In the second chapter, the focus becomes devising a model using some techniques from linear algebra. As this model is introduced, reasons for other models are introduced. In the final chapter, more advanced techniques derived from discrete mathematics are employed to generalize this model and make a model that is theoretically consistent and still intuitively appealing. This paper is not written as a guide for understanding information retrieval in its many aspects and forms, but is meant to introduce the topic to an undergraduate audience. The literature referenced throughout this document serves a variety of purposes, providing general overviews of the models and as more specific instances of their development. By tracing a development of one of the simpler models in this paper, the reader should find the study of IR more accessible