Development of an Automatic Document to Digital Record Association Feature for a Cloud-Based Accounting Information System
Documents such as contracts; receipts; and sales invoices are proofs of transactions generated by various functions of business organizations. Though some organizations have initiatives to digitize paper-based proof of transactions; their business processes do not remove paper trails entirely. Organ...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | text |
| Published: |
Archīum Ateneo
2021
|
| Subjects: | |
| Online Access: | https://archium.ateneo.edu/discs-faculty-pubs/213 https://link.springer.com/chapter/10.1007/978-3-030-80119-9_59 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Ateneo De Manila University |
| Summary: | Documents such as contracts; receipts; and sales invoices are proofs of transactions generated by various functions of business organizations. Though some organizations have initiatives to digitize paper-based proof of transactions; their business processes do not remove paper trails entirely. Organizations normally scan business document transactions; manually classify digitized documents; and associate digitized documents to digital records in a database management system. Hence; the digitization process introduced more work rather than efficiency. This study seeks to eliminate the additional work brought about by the document digitization process. It specifically looks at the application of image enhancing techniques and open-source Optical Character Recognition (OCR) technology to automatically classify and associate business documents to digital records in a database management system. The study presents how an alternative document digitizer and image enhancing feature is integrated into an accounting information system to facilitate the automatic classification and association of digitized documents to specific database records. The application of image cropping and grayscale color processing image enhancing techniques contributed to achieving an average of 90% level of confidence in extracting field labels while 91.5% level of confidence in extracting field values in business documents. |
|---|