0
$\begingroup$

I am trying to solve a problem where I need to map multiple variations of a company name to a single name. For example: say I have a company named Super Idea Corporation Limited.

I need to resolve the following to Super Idea Corporation Limited

  • SICL
  • Super Idea Corp Ltd
  • SIC Ltd
  • SIC Limited

Is there a non regex way of doing this? The reason I am averse to using regex is that there are a lot of business names that can be represented in many different ways. I want something that is more flexible and adaptive.

$\endgroup$
2
  • $\begingroup$ Can you explain how this question is related to Artificial Intelligence, as defined in our on-topic page (ai.stackexchange.com/help/on-topic)? $\endgroup$
    – nbro
    Commented Sep 15, 2021 at 1:13
  • $\begingroup$ @nbro, I am not very well researched on NLP. Asked here to see if there is any work done NLP-wise on this particular kind of problem. Hence the tag. $\endgroup$ Commented Sep 15, 2021 at 2:48

1 Answer 1

1
$\begingroup$

Have a look at Named Entity Recognition (NER); these algorithms are mainly concerned with recognising that there is an entity, but often also include normalising the name to a canonical form for information retrieval -- this is what you would need.

In a previous job I actually implemented this, using a fuzzy match with variable word order. You would still need to map Corp and Corporation onto each other as exchangable, and deal with acronyms, but that should be tractable.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .