ABSTRACT
For our financial statement analysis project which we are planning to develop there are many free websites available which basically analyse your statement and give you a response and the response is also divided into various categories, but they are not able to provide accuracy. What paid versions are giving is basically they are giving accuracy but the issue with that is that they are very much costly. What we are planning is basically to make a system which is providing accuracy and that is to be free for all. We would like to achieve the maximum accuracy and the service should be free for all. This is a huge problem for all of those who are basically working in any organisation, the people who are managing the finances of the organisations they have to do all these things manually. We are trying to automate this thing. Secondly it is very much helpful for those who are doing too many online transactions and those individuals who are in their last year their financial year are very much confused or very much concerned about how and why they are spending their money. Since this project looks very simple and there is very little user interaction but tricky and interesting machine learning models involved for building the system. We have to train on a model for future use. The categories are not so simple for dividing them individually since the uploaded file will be in PDF format and we have to convert that into text format and then we have to make the data consistent and finally we have to provide the data to the machine learning model. After that machine learning model will be trained in such a way that it will give maximum possible accuracy.
|
DOI:
10.17148/IJARCCE.2021.10638
