Topic modeling with BERTopic
BERTopic is a topic modeling algorithm that is based on the BERT word embeddings. In Chapter 14, LDA and BERTopic, we will learn the key components of BERTopic and build our own model. In addition, the BERTopic modeling has its own visualization functions that are similar to pyLDAvis, as seen in Figure 1.4. We will learn to use all the visualization functions as well.
Figure 1.5 shows the top words for eight topics:
ght topics:
Figure 1.5 – An overview of topic modeling results by BERTopic (also presented in Chapter 14)
I trust these introductions have given you a strong appetite to dive into each chapter and apply the models discussed in your future work. Now, let's get familiar with the terminology commonly used in NLP.
