Acronym:

Knowledge Discovery in Textual Data 

From: 2007-01-01 To: 2013-06-30 (Completed)

MZOS 2007 

- 

Bojana, Dalbelo-Bašić

6129-871 

The development of the www has revolutionized the availability of information in digital form. However, while the quantity of information grows, human capacity to understand and process it remains unchanged. Knowledge discovery methods in textual data have the objective to relieve people as much as possible of repetitious processing of saved and previously processed data and enable them to concentrate on making decisions based on results of knowledge discovery processes in textual data. Knowledge discovery in textual data incorporates artificial intelligence (machine learning, natural language processing) and computer linguistics, and includes modelling with the fundamental mathematical and statistical methods. This project’s research results will primarily relate to language independent methods and models, but will also incorporate systems developed by integrating this project’s results with research at language specific level, which is to be part of other projects of the program. This project’s aim is to show by which approach and which models a new level of intelligent knowledge discovery systems for textual data may be achieved, particularly for texts in Croatian language. Proposed knowledge discovery models and new methods will be experimentally verified, specifically on document collection in Croatian language.