A data mining application for extracting semantic information from Turkish texts

Today, with spreading of internet, information word has a new problem about to reach the right information from the large amounts of data. Existing search engines can only fetch the data even no relation with the search subject. Semantic web technology is created to reach the related information from the web. For this technology, Information Extraction is extracting related and structured information from natural language raw texts. ACE determines whether this processes as a result of conferences and workshops. With ACE the problem deals with three main tasks; Entity Detection and Tracking , Relation Detection and Characterization, and Event Detection and Characterization.The purpose of this study was to examine Turkish texts can be possible to identify relationships in the process of information extraction. In this study we used Automated Content Extraction(ACE) as a guide to detect relation of entities from Turkish texts. In this context Turkish and English studies examined and assuming that if an entity is equal to a name of a system designed. With this system, news page contents collected and some sentences ,words and verb of sentence abstracted and every couple words in a sentece with the verb of the sentence recorded with some other specifications about the locations in the phrese. After building this frequency data base various views created to determine if the word couple is related. With this views classification made by word couple and verb rate and tested over support vector machine alghoritm. A web page content prepared in Turkish can be brought into line with the standards specified with this system for the submission of a preliminary study.