I am a newbie in machine learning but I have a coursework to create program that can extract some concrete features from the given text.

For example: If I want to extract number of red apples and green apples, I will extract 3, 5 from "On that tree 3 red apples and 5 green apples grow" and 10, 1000 from "In the box there are 1000 green apples and 10 of red". Actually the real examples can be more complicated like a extracting a car specification from the ad.

The concrete features are known before the data set processing.

Could you suggest what kind of algorithms I should use? Previously, I used only linear regression.

  • $\begingroup$ I guess neural networks would help you in this. If you create a network which could distinguish between numbers and characters i guess it will help you. Split the sentence according to spaces ,pass each word to network for identification ,and if it is a no just extract it. $\endgroup$ – Rohit Jere May 3 '18 at 5:18

I guess if you know what you want to extract you can just find it using regular expressions for integers or car specifications.

  • $\begingroup$ No, it does not work. The samples are in various formats and may contain typos. $\endgroup$ – Victor Khovanskiy Sep 25 '17 at 13:31

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.