The task is to extract semantically and syntactically similar sentences. Focus needs to be more on semantics.
Retrieve at max top 10 semantically and syntactically similar sentences for each sentence.
The input needs to be in an excel format with 2 columns (column A ‘Risk Description’ and column B ‘Control Description’)
The similarity comparison will be made on column B ‘Control Description’.
There need to be two scores associated with similar sentences.
1. Similarity %
2. Similarity rationalization (low, medium and high). If similarity % < 50 then its low, if similarity % is between 50 and 75 then its medium and if its > 75 then its high.
Output will be in an excel format with the following columns ’Control Description’, Similar Sentences, Risk Description, % similarity for each similar sentence,Similarity rationalization, reasoning why the controls are similar. )