Information Retrieval(Q2)

junio 2nd, 2009 · No hay Comentarios

The information Retrieval,or IR, is the science of the search of information out of the documents, the search of the documents themselves, the search of metadata that describe documents, or the search of data base through the internet, intranet, texts, images, sound, or other characteristic´s data.

IR is an interdisciplinary study. it covers so many disciplines that we usually have just a parcial knowledge of it. Some of the ativities that take partin these studies are the cognitive psycology, information´s architecture informations7´s design, the behavior of humans towards the information, linguistic, semiotic, computer science, documentation, and librarianship.

Searching programmes like Google, Lycos, and Copernic are some of the most popular of the information retrieval. basically we just need to creat a vocabulay, a list of terms in natural language, an algorithm that includes the logical rules of the search table and a true assessment of the amount of results or information obtained or possible.

Sometimes, some issues may occur such as the silence(because of the synonymy), the noise(because of the polysemy)homography, ambiguity,…

The documents are transformed into strog representations so the information retrieval works. there are devided in two dimentions:
-First dimention mathematical basis:
a)Set-theoretic models
Standard Boolean model
Extended Boolean model
Fuzzy retrieval
b)Algebraic models
Vector space model
Generalized vector space model
Topic-based vector space model
Extended Boolean model
Enhanced topic-based vector space model
Latent semantic indexing aka latent semantic analysis
c)Probabilistic models
Binary independence retrieval
Probabilistic relevance model (BM25)
Uncertain inference
Language models
Divergence-from-randomness model
Latent Dirichlet allocation