Information retrieval - P500003
|
|
|
|
|
||
|
A number of electronic documents grows much faster than a human is able to deal with. The information retrieval methods help to identify documents likely containing a given information. The selection of documents is based on keywords, that are assigned to characterize document content and used to specify the aims of user search. To achieve this aim, information retrieval utilizes the methods of linear algebra that work with the vector model, statistical and probability methods, methods of computational linguistics or classification and clustering methods of artificial intelligence.
Last update: Svozil Daniel (25.05.2018)
|
|
||
|
Students will know:
Last update: Svozil Daniel (23.05.2018)
|
|
||
|
R: Baeza-Yates, R., Ribeiro-Neto, B.: Modern Information Retrieval. Second edition, Addison-Wesley, 2011. R: Weiss, S.M. et all: Text Mining? Predictive Methods for Analyzing Unstructured Information. Springer, 2005. Last update: Svozil Daniel (23.05.2018)
|
|
||
|
Lecturer materials Last update: Svozil Daniel (23.05.2018)
|
|
||
|
Introduction to information retrieval, uncertainty, relevance, text document normalization, Zipf's law Text documents indexing, querying and searching - metrics, vector model - dimensionality reduction, latent semantic indexing Document and keyword clustering, distance, similarity metrics, centroid, clustering algorithms Document classification, Bayesian classification, k nearest neighbors, decision trees, metoda support vector machines The aims and capabilities of text mining, linguistic methods in text mining, tokenization, part-of-speech tagging, named entity recognition, parsing, coreferences Text mining in information retrieval: document content extraction, automatic document summarization, automatic question answering Last update: Svozil Daniel (25.05.2018)
|
|
||
|
oral exam Last update: Svozil Daniel (23.05.2018)
|
