An Automated Approach for an Online Grievance System for Categorization, Tagging and Analysis of Sentiments of Grievances Through a Web and Mobile Portal Using Deep Neural Network

Abstract: A grievance is a discontent or dispute that may arise at any level in any organization. In many circumstances, people of an organization fail to state their issues and cannot seek support for the issues they are facing in an organization. Surveying the already implemented grievance redressal systems, it can be said that the people want to remain anonymous, however there is no provision as such for now. The management of the institutions and organizations are constantly on the lookout for various issues its people are facing, track them down, work on them and update their status, however no current systems address this. Also, these systems are not autonomous and need human intervention for classifying the issues into various categories. Thus, we have come up with a novel approach that keeps all these factors in mind and developed a grievances tracker and redressal system. The end users can post their issues on the portal completely anonymously and the authorities at their various levels can provide solutions to those. Others facing the same issue can upvote the issue for higher visibility. The management can easily track the issues that are currently ongoing, completed, or invalid and can have statistics and reports on one click. The system having natural language processing capabilities can easily and autonomously categorize the various issues based on the content. Also, NSFW content that might be uploaded by the rogue end users can be hampered by using a deep neural network for classifying images before uploading.d. You can use this document as both an instruction set and as a template into which you can type your own text.

Keywords: Grievances Tracking, Issues Tracking, Neural Network, Portal, Management