I would like to code a script that could locate a specific word or number in a financial statement. Financial statements roughly contain the same information, they are however not identical and organized in the same way. My thought is that by using Tensorflow I could train a neural network to locate the specific words or numbers for me. I am thinking that if I label different text and numbers in 1000 financial statements and use them to train the neural network, it will then be able to identify these numbers or words in all financial statements. For example, tell it in all 1000 training statements which number that is the profit of the company. Then when I give it an unseen financial statement, it should be able to identify which number that is the profit.

Is this doable? I have been working with coding in python for a couple of months and so far I've built some web scrapers and integrated them with twitter, slack and google sheets. However, this would be my first AI related project. I would be very grateful for all your thoughts on this project and if anyone could steer me in the right direction by sharing relevant tutorials.

Thanks a lot!

  • 1
    $\begingroup$ Hi and welcome to this community! To clarify, you say "For example, tell it in all 1000 training statements which number that is the profit of the company". Is this the only task you have, that is, you want to predict the profit of a certain company given financial statements, or you want to solve other tasks too? $\endgroup$ – nbro Nov 7 '19 at 16:04
  • 1
    $\begingroup$ Thanks a lot! I don't want it to predict anything. I just want it to identify specific numbers or words (names) in the financial statements. The goal is to have the key information in the financial statements automatically organized in a spreadsheet without having to type it in manually. $\endgroup$ – Lohant00 Nov 7 '19 at 17:02
  • 1
    $\begingroup$ Well, in that case, you don't really need neural networks or deep learning. You can just write a program that searches for the keywords you're looking for. However, which key information are you talking about? Do you know what these keywords are? $\endgroup$ – nbro Nov 7 '19 at 17:07
  • $\begingroup$ With the help of dictionaries it's possible to categorize individual words into groups. The next step is to label sentences with a sentiment. An LSTM neural network is trained on an example data-set. $\endgroup$ – Manuel Rodriguez Nov 7 '19 at 17:13
  • $\begingroup$ Well I do think that I need neural networks because the name of the ceo or the amount of profit for example varies every time. Their place in the financial statement also varies. I need the neural network to figure out the right name or the right number by looking at the context, just like a human does. $\endgroup$ – Lohant00 Nov 7 '19 at 17:30

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.