Applications

Content

• pre-trained data baked into the model
• Context provided content as Text
• Context provided content as Embeddings
• Fine tuning
• pre-filtering and pre-analysis of local Embeddings
• Scale up all local and referenced data into a Data-Lake

Tasks

Summarization

• consolidate a long text into its main components
• rephrase the content in an abstracted way

Sentiment analysis and classification

• analyzing a list of posts to identify which ones were answered and which not
• analyze a list of features or requirements to identify the type and cluster them in groups
• Named Entity Recognition by identifying certain categories or tagging a special type

Content consolidation and linking

• identify similarities and duplicates
• identify contradictions
• identify relationships between text elements
• identify references between text elements

Question answering

• search for relevant info
• provide an answer to question with a reference

Example applications

• Test on Astro web framework