The Institute of Information Technology of ANAS hosted a scientific seminar dedicated to the theme "Analysis of methods of search engine efficiency: problems and approaches".

Suleyman Suleymanzada, the Ph.D. student of the institute, presented the main ideas of the Information Retrieval (IR). Describing the key components of IR systems, he said that finding the relevant documents across large text collections is the main goal of traditional IR systems.

"Search engines form the basis of IR systems. Depending on the application areas, there are different types of search programs such as vertical and desktop, search engine analytics, and common search engines", he noted.

S.Suleymanzada noted that the information search systems fulfill two important functions, such as indexing and searching. He said that a large number of benchmarking architectures were offered for information search engines. According to him, the first issue of the information search system is the initial processing of documents.

He said there are many classic models in the field of information search and provided extensive information about Boolean IR Model, Vector Space, Okapi BM25, statistical language, probable IR models, neural networks models.

The effectiveness of the information search methods in the search engine was tested on popular bases such as Cranfield, Reuters Corpus Volume I (RCV1), "OHSUMED", and "TREC", he noted.

Views on presentation were exchanged and questions were answered.

© All rights reserved. Citing to www.ict.az is necessary upon using news