EKSTRAKTIVNA SUMARIZACIJA RECENZIJA HOTELA UPOTREBOM WORD2VEC, DOC2VEC I GPT-3.5 MODELA
Ključne reči:
Ekstraktivna sumarizacija, NLP, Word2Vec, Doc2Vec, GPT-3.5, TextRank
Apstrakt
U radu je prikazana ekstraktivna sumarizacija recenzija hotela upotrebom Word2Vec, Doc2Vec i GPT-3.5 modela. Objašnjene su teorijske osnove NLP-a, pomenutih modela i TextRank algoritma. Prikazan je proces sumarizacije teksta korišćen u implementaciji ovog rešenja. Skup podataka za treniranje i evaluaciju modela je preuzet sa sajta booking.com. Podaci su javno dostupni i nalaze se u vlasništvu sajta. U radu su prikazani eksperimenti izvršeni upotrebom različitih varijanti modela, kao i upotrebom različitih skupova podataka. Rezultati eksperimenata su poređeni međusobno, kao i sa rezultatima jednog od srodnih radova.
Reference
[1] LI, JUN; HUANG, GUIMIN; FAN, CHUNLI; SUN, ZHENGLIN; and ZHU, HONGTAO (2019) "Key word extraction for short text via word2vec, doc2vec, and textrank," Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 27: No. 3, Article 17.
[2] Milad Moradi, Maedeh Dashti, Matthias Samwald, "Summarization of biomediacal articles using domain-specific word embeddings and graph ranking", Journal of Biomedical Informatics: Vol. 107, July 2020, 103452.
[3] https://www.geeksforgeeks.org/natural-language-processing-overview/ (pristupljeno u januaru 2024.)
[4] https://medium.com/sciforce/towards-automatic-text-summarization-extractive-methods-e8439cd54715 (pristupljeno u februaru 2024.)
[5] https://www.geeksforgeeks.org/python-word-embedding-using-word2vec/ (pristupljeno u februaru 2024.)
[6]https://wiki.app.uib.no/info216/images/c/c4/IntroToWordEmbeddings.pdf (pristupljeno u februaru 2024.)
[7] https://medium.com/@manansuri/a-dummys-guide-to-word2vec-456444f3c673 (pristupljeno u februaru 2024.)
[8] https://www.geeksforgeeks.org/doc2vec-in-nlp/ (pristupljeno u februaru 2024.)
[9] A. Hernández-Castñeda, R. A. García-Hernández, Y. Ledeneva and C. E. Millán-Hernández, "Extractive automatic text summarization based on lexical-semantic keywords", IEEE Access, vol. 8, pp. 49896-49907, 2020.
[10] R. Mihalcea and P. Tarau, "Textrank: Bringing order into text", Proceedings of the 2004 conference on empirical methods in natural language processing, pp. 404-411, July 2004.
[11] https://www.linkedin.com/pulse/chatgpts-guide-understanding-gpt-35-architecture-heena-koshti#:~:text=The%20GPT%2D3.5%20architecture%20is%20a%20neural%20network%2Dbased%20language,text%20completion%2C%20and%20question%20answering (pristupljeno u februaru 2024.)
[12] https://medium.com/nerd-for-tech/gpt3-and-chat-gpt-detailed-architecture-study-deep-nlp-horse-db3af9de8a5d (pristupljeno u februaru 2024.)
[2] Milad Moradi, Maedeh Dashti, Matthias Samwald, "Summarization of biomediacal articles using domain-specific word embeddings and graph ranking", Journal of Biomedical Informatics: Vol. 107, July 2020, 103452.
[3] https://www.geeksforgeeks.org/natural-language-processing-overview/ (pristupljeno u januaru 2024.)
[4] https://medium.com/sciforce/towards-automatic-text-summarization-extractive-methods-e8439cd54715 (pristupljeno u februaru 2024.)
[5] https://www.geeksforgeeks.org/python-word-embedding-using-word2vec/ (pristupljeno u februaru 2024.)
[6]https://wiki.app.uib.no/info216/images/c/c4/IntroToWordEmbeddings.pdf (pristupljeno u februaru 2024.)
[7] https://medium.com/@manansuri/a-dummys-guide-to-word2vec-456444f3c673 (pristupljeno u februaru 2024.)
[8] https://www.geeksforgeeks.org/doc2vec-in-nlp/ (pristupljeno u februaru 2024.)
[9] A. Hernández-Castñeda, R. A. García-Hernández, Y. Ledeneva and C. E. Millán-Hernández, "Extractive automatic text summarization based on lexical-semantic keywords", IEEE Access, vol. 8, pp. 49896-49907, 2020.
[10] R. Mihalcea and P. Tarau, "Textrank: Bringing order into text", Proceedings of the 2004 conference on empirical methods in natural language processing, pp. 404-411, July 2004.
[11] https://www.linkedin.com/pulse/chatgpts-guide-understanding-gpt-35-architecture-heena-koshti#:~:text=The%20GPT%2D3.5%20architecture%20is%20a%20neural%20network%2Dbased%20language,text%20completion%2C%20and%20question%20answering (pristupljeno u februaru 2024.)
[12] https://medium.com/nerd-for-tech/gpt3-and-chat-gpt-detailed-architecture-study-deep-nlp-horse-db3af9de8a5d (pristupljeno u februaru 2024.)
Objavljeno
2024-09-06
Sekcija
Elektrotehničko i računarsko inženjerstvo
