Skripsi
PEMODELAN TOPIK MENGGUNAKAN BERTOPIC DENGAN KEYBERT UNTUK EKSTRAKSI KATA KUNCI SEBAGAI TOPIC REPRESENTATION TUNING
In the era of technological advancement, the use of social media such as Twitter has become commonplace as a medium for online interaction. Every day, a vast number of tweets are generated by Twitter users from around the world. To determine which topics are trending, reading all the tweets on Twitter would take an extremely long time due to the sheer volume of tweets. One method used to efficiently extract information from Twitter tweets is topic modeling. Topic modeling is a method for discovering topics from various texts. This research aims to perform topic modeling on Indonesian-language tweets using BERTopic with KeyBERT for keyword extraction in each topic. KeyBERT will generate keywords for each topic cluster and will be used by BERTopic to enrich the results of the topic modeling. The dataset used consists of 10,000 Indonesian language tweets taken from the Twitter account @detikcom. The data is divided into two parts: 8,000 tweets are used for training data and 2,000 tweets are used for testing. Based on the topic modeling results with BERTopic, a total of 50 topics were obtained. Topic Modeling evaluation was conducted using coherence score, yielding an average of 0.765 on the training data and 0.675 on the testing data.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2407003793 | T147741 | T1477412024 | Central Library (REFERENCES) | Available but not for loan - Not for Loan |
No other version available