Hi! I’m working on a project that involves extracting data from scanned shop receipts in Brazil. We can already get the text from a receipt image, and now we have to actually get the relevant info from this text. The pieces of information we need to extract are:
- Place
- Date/Time
- Products bought
- Total price
- Amount of taxes
Even though there isn’t a standardized format for receipts, the categories above are guaranteed to be in our input text. With that being said, I would like to know if Rasa’s entity extraction features would be a suitable tool for this job. It’s a similar question to the one found here, although we have a few more categories to extract, and the added caveat of having the text in Portuguese. Thanks in advance!