Topic Modeling using Transformer Based Embeddings | DataHour by Bharath Kumar Bolla

Customers' reviews and comments are important for businesses to understand users' sentiment about the products and services. However, this data needs to be analyzed to assess the sentiment associated with topics/aspects to provide efficient customer assistance. LDA and LSA fail to capture the semantic relationship and are not specific to any domain. BERTopic, is a novel method that generates topics using sentence embeddings and is applied to Consumer Financial Protection Bureau (CFPB) data.

In this DataHour, Bharath will show how BERTopic is flexible and yet provides meaningful and diverse topics compared to LDA and LSA. Furthermore, he will explain how domain-specific pre-trained embeddings (FinBERT) yield even better topics.


For more amazing datahour session, visit: https://datahack.analyticsvidhya.com/...

Stay on top of your industry by interacting with us on our social channels:

Follow us on Instagram: Instagram: analytics_vidhya
Like us on Facebook: Facebook: AnalyticsVidhya
Follow us on Twitter: Twitter: AnalyticsVidhya
Follow us on LinkedIn:LinkedIn: analytics-vidhya

• #education
• #machine_learning
• #natural_language_processing
• #deep_learning