1
$\begingroup$

Okay lets say i have two data structures . two phone data for example containing their Name and spec ( cpu , ram , display etc ) . I want to check if these two phones are the same or not . Their names can be different as their other features . For example we have a :

Name : Galaxy s3, CPU : Dual core , Display : 4 inches

and on the other hand i have :

Name : galaxy SIII , CPU : dualcore 1.2 , Display : 4'

as a human can understand these two phones are the same . but how can a machine say this ? On top of my head i have a solution . I can set weight for each field and calculate the similarity between each two fields by simple text comparison and then multiple to its weight . define a threshold and thats it . BUT i want to know if there is any scientific field about this or not . Are there any deep learning solutions or anything else . Thanks

$\endgroup$
  • 1
    $\begingroup$ I don't think you need any deep learning for this. A modest number of simple techniques from natural language processing and information retrieval (e.g. case folding, stemming) will do it. How much work it takes largely depends on the number of variants that you have. I suggest that you look up "shallow parsing", which may give you some ideas. Another hint is to look up geocoding. It's a completely different field (mapping street addresses to coordinates), but understanding street address information is a very similar problem to the one you have. $\endgroup$ – Pseudonym Dec 13 '15 at 23:53
  • 1
    $\begingroup$ Oh, one more comment: Be prepared for the fact that the best solution may be to have a database of all known phone models to compare with. $\endgroup$ – Pseudonym Dec 13 '15 at 23:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.