Films were originally live images or moving photos forcing the viewer to see continuous motion between different objects quickly and successively. The world of cinema is currently increasingly being embellished with technological advances, so that public interest in watching films is also increasing. Before launching a new film, the film industry always provides a brief description first in the form of writing how the story of the film is called a synopsis. Searching a synopsis of a film that is done manually takes a long time because of the large amount of data available on social networks. This makes it difficult for users to find the most relevant synopsis to what they are looking for. The information retrieval system for the film synopsis is used to meet the needs of users in searching for information from a document. One of the document ranking methods that can be used for information retrieval is the BM25F method. Testing was performed on BM25F independent parameters. The best value in the precision@k and r-precision tests on 300 film synopsis documents against 6 queries using the free parameter BM25F, the best combination of boost values obtained in the boost variable test is in the title field = 5 and content = 1, the best bc value from bc variable testing is 0.75, and the best k1 value from the results of k1 variable testing is 1.2. The best average value for precision@k is 0.93 and for the best results in the r-precision test, the value is 1.
Copyrights © 2022