Paralel haber kaynaklarından eşanlamlı ifadeler ayıklama

Different expressions of the same statement is said to be paraphrases of each other. Anexample is the phrases 'solved' and 'found a solution to' in 'Alice solved the problem' and'Alice found a solution to the problem'. Paraphrase Extraction is the method of finding andgrouping such paraphrases from free text. Finding equivalent paraphrases and structures canbe very beneficial in a number of NLP applications, such as Question Answering, MachineTranslation, and Multi-text Summarization, e.g. in Question Answering, alternative questionscan be created using alternative paraphrases. We attack the problem by first grouping newsarticles that describe the same event and then collecting sentence pairs from these articles thatare semantically close to each other, and then finally extracting paraphrases out of thesesentence pairs to learn paraphrase structures. The precision of finding two equivalentdocuments turned out to be 0.56 and 0.70 on average, when matching criterion was strict andflexible, respectively. We tried 9 different evaluation techniques for sentence-level matching.Although, exact word match count approach had a better precision value than the n-gramprecision count approaches, paraphrase extraction phase shows that the latter approachescatch sentence pairs with higher quality pairs for paraphrase extraction. Our system canextract paraphrases with 0.66 precision when only equivalent document pairs are used as atest set.