Abstract
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.
Advisor
Moynihan, Matthew
Department
Mathematics
Recommended Citation
Yerkes, Gregory R.P., "Vector Space Modeling: An Intuitive Approach to Information Retrieval" (2015). Senior Independent Study Theses. Paper 6518.
https://openworks.wooster.edu/independentstudy/6518
Disciplines
Discrete Mathematics and Combinatorics | Logic and Foundations | Theory and Algorithms
Keywords
Vector Space Models, Information Retrieval, Natural Language Processing, Human Language Technology, Search Engines
Publication Date
2015
Degree Granted
Bachelor of Arts
Document Type
Senior Independent Study Thesis
© Copyright 2015 Gregory R.P. Yerkes