: Vocabulaire du TAL: BTM

NLP methods and tools > NLP software > BTM

BTM

Biterm Topic Models find topics in collections of short texts. It is a word co-occurrence based topic model that learns topics by modeling word-word co-occurrences patterns which are called biterms. (Loterre)

BTM eliminates a system as soon as there is another system with a significantly higher score. (Mohankumar & Khapra, 2022)
First we find the BTM identifies the word co-occurrence by considering the bi-term frequency in the corpus-level. (Chen & Kao, 2015)
For example the relatively recent Amplayo and Song (2017) combine LDA with Biterm Topic Models. (Huber & Spiliopoulou, 2019)
However BTM will make the common words be performed excessively because BTM identifies the word co-occurrence information by the bi-term (Chen & Kao, 2015)
The BTM consists of a linear projection layer forward and backward long-short term memory (LSTM) layers and an additional LSTM to combine them. (Alkhouli, Rietig & Ney, 2015)

http://data.loterre.fr/ark:/67375/8LP-RVV5X596-4

RDF/XML TURTLE JSON-LD Dernière modification le 03/05/2024

Loterre