I am at the start of a project where I would like to map/match/link external product names to the respective internal product names. The goal should be to ingest related external information (e.g. stock number) of the external products into our system by joining the same products based on their product names. Short, the external product name should be matched to the internal representation of the product name.

Problem and Question

I'm now doing some research about potential solutions and I'm having difficulties finding out if the nature of the problem can be allocated to Entity Resolution or Entity Linkage or if it even includes both of them. I'm asking this because I'm afraid to go down the wrong path when researching for a potential way to tackle the problem. I have seen the post about key differences about entity linking and entity matching, but it's still hard for me to allocate the nature of my problem to one of them. Can please someone tell me if the problem can be allocated to Entity Resolution, Entity Linkage or both, and why this is the case?

Thanks a lot!

  • 1
    $\begingroup$ I think that the answers to the question you linked should be useful. A slide from one of the links mentions the irony that the subject itself has many names which might require disambiguation. My suggestion is to research all these phrases in papers, for example in arxiv Learning subject and see which one is the most popular / relevant to you. $\endgroup$
    – Valentas
    Commented Oct 5, 2023 at 9:47
  • 1
    $\begingroup$ I also posted your question verbatim to ChatGPT. Here is one paragraph from the answer: "In summary, your problem involves both entity resolution (within your internal dataset) and entity linkage (between external and internal datasets). You may need to perform entity resolution first to ensure the accuracy and consistency of your internal data. Then, you can use entity linkage to map external product names to their corresponding internal representations." $\endgroup$
    – Valentas
    Commented Oct 5, 2023 at 10:11


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.