The Challenge
This project was for an invoice processing business. The company was facing a lot of issues in their working efficiency due to a large number of invoices being received and the inability to organize them.
Due to such issues, the company was seeking a portal. A tool that could help them upload and organize a large number of invoices, as most of their data were difficult to organize due to its raw forms, like pdf, image, and document formats.
The Solution
To this problem of our client, the solution provided by the genius heads of CloudZenia was to have a portal that can analyze the text directly from the raw forms of data, regardless of the format being PDF, image, or document. The complete solution was built on Serverless Stack, which enabled customers to work with a very large number of invoices processing.
The solution we generated was to organize the data on a dashboard that would not only be easy to understand but also show all the statistics efficiently. This organization would then help the business study its business to grow and also provide a statistical overview of the entire business and its working data, like the number of sales, the total number of invoices revived from one particular source, etc.
This entire solution would enable the projection of raw data into an organized dashboard with all the easily accessible details.
Our Process
To bring up this solution in action, this is what we did:
- Dashboard: We used React JS to build the dashboard.
- Frontend: Coding for frontend purposes was done in React JS but hosted on S3.
- Backend: The backend was made Serverless using the technology AWS Lambda, API Gateway, SNS, S3, DynamoDB, AWS ElasticSearch, and AWS Textract.
Working
After bringing the entire process into action and finalizing the OCR tool; the OCR tool started its work, as such when an invoice was uploaded to the portal, regardless of its form, the invoice would get directed to S3, after which the backend Lambda would take over.
This backend Lambda would then be triggered and get on to its next step of sending the invoice to the AWS Textract. The AWS Textraxt analyses the particular file form and extracts the data from it. The extracted data in the form of the text is sent to Lambda for processing purposes.
After this step, it is finally sent to the DynamoDB for the saving of data.