<![CDATA[Current technological developments make it easier for humans to explore a lot of information using the internet such as review information or opinions about films. Public opinion about the film can be found on the IMDB website. By doing a sentiment analysis on public opinion about the film, we can conclude whether a film gets more positive or negative opinions. To perform this sentiment analysis, one of the deep learning methods is used, namely Long Short-Term Memory (LSTM) with FastText as a vector representation of words in the IMDB movie reviews dataset of 50,000 data. Performance using the Long Short-Term Memory and FastText methods produces an accuracy of 0.863; precision of 0.865; recalls of 0.861; and f1-score of 0.863. This LSTM and FastText method produces better performance than using LSTM alone with a difference of 0.053 on the f1-score value with details of accuracy reaching 0.808; precision reaches 0.804; recalls reached 0.816; f1-score reaches 0.810 for the LSTM method only.]]>
Copyrights © 2022
