Information Retrieval Models

Many applications that handle information on the internet would be completely inadequate without the support of information retrieval technology. How would we find information on the world wide web if there were no web search engines? How would we manage our email without spam filtering? Much of the development of information retrieval technology, such as web search engines and spam filters, requires a combination of experimentation and theory. Experimentation and rigorous empirical testing are needed to keep up with increasing volumes of web pages and emails. Furthermore, experimentation and constant adaptation of technology is needed in practice to counteract the effects of people that deliberately try to manipulate the technology, such as email spammers. However, if experimentation is not guided by theory, engineering becomes trial and error. New problems and challenges for information retrieval come up constantly. They cannot possibly be solved by trial and error alone. So, what is the theory of information retrieval? There is not one convincing answer to this question. There are many theories, here called formal models, and each model is helpful for the development of some information retrieval tools, but not so helpful for the development others. In order to understand information retrieval, it is essential to learn about these retrieval models. In this chapter, some of the most important retrieval models are gathered and explained in a tutorial style.