Towards a documents processing tool using traceability information retrieval and content recognition through machine learning in a big data context

Abstract

In 1980, an application was developed to track, manage and store documents in electronic format. Scan technology has enabled organizations to digitize papers for easier document storage and tracking. Document management tools have since developed by introducing new functionalities, related to security, users services, workflow and audit. Our research is part of the context of improving the efficiency of document processing by proposing an approach using information traceability retrieval and content recognition techniques through machine learning. In this sense, we started by proposing the exploitation and extraction of relationships between documents based on the traceability links and the calculation of similarity using information retrieval. Then, in order to improve the processing of documents, we proposed a contribution of the use of recognition of content techniques by machine learning approaches. Thus, the visualization of the results, according to user profiles, motivated us to offer recommendations dedicated to the document management system. A Big Data environment is proposed because of the exponential growth of data and also to our needs concerning the analysis and the distributed calculation of the voluminous masses of data.