|
I have to work on a problem of extracting company description from "About us" page that is present in most of the websites. I did some reading, and it seems like this is query based summarization problem. Query can be something like "What this company manufactures" or "What services are provided by this company". How should I approach this problem. |
|
Yes I think it is focused summarisation problem. You have to define rules for better/worser sentences like contains company_name, contains definition word etc and then do summarisation to extract most frequent and non duplication sentences. |
This looks more like question-answering than summarization itself.
If I approach it as a question-answering problem, how can I extract relevant content.