Comparison of information retrieval models for question answering

Question Answering Systems (QAS) are an important research topic triggered and at the same time stimulated by the immense amount of texts available in digital form. As the quantity of natural language information increases, the necessity of new methods to precisely retrieve the exact information from massive textual databases becomes inevitable. Although QAS have already been well explored, there are still many aspects to be solved, particularly those which are language specific. The main goal of the research presented in this paper was to compare three proven information retrieval (IR) models in order to accurately determine the relevant documents which contain the correct answer to questions posed in Macedonian language. It was accomplished using a real-life corpus of lectures and related questions existing in our e-testing system. In order to compare the results, we designed a small system capable of learning the correct answer. We revealed that the modified vector space model is the most suitable for our collection. The results we obtained are promising and they encouraged us for further improvement adopting some of the existing IR models, or even proposing a new one.