I'm working on a project (Python) to enforce the company naming convention of products on product lists provided by clients/suppliers. I'm having a list of company names (Standardised names) and those of external. I'm considering typos too - generating this list using GPT.

Here's are the models I'm considering:
Sequence-to-Sequence (Seq2Seq): LTSM over RNN
Transformer-Based Models: BERT on custom data

Additionally, I'm looking up fuzzy string matching.

Could anyone recommend other approaches, or if I'm missing something? Greatly appreciated :)

This a sample of the dataset I'm dealing with. This is the correct names. I'm creating my own dataset with around 20 incorrect names alongside the correct ones. There are about 20k+ unique names.


New contributor
Secret Ambush is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
  • $\begingroup$ Could you provide some sample data. This will help others understand your problem well and give relevant answers. $\endgroup$ Commented 2 days ago
  • $\begingroup$ Hi @VidyadharRao, thanks for the suggestion. I've added a snip of the dataset. $\endgroup$ Commented 2 days ago
  • $\begingroup$ "Additionally, I'm looking up fuzzy string matching." Did you already try fuzzy string matching algorithms instead of some models. I could imagine they suit your needs better. $\endgroup$
    – Broele
    Commented 19 hours ago


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.