Our product i.e. AutoFeedbacker has been developed with an intent of eliminating an inherent flaw in the feedback generation and usage system currently in place. An inherent flaw in feedback generation exists in the disparity in its importance between the consumer and the organization. While, organic and genuine feedback is a huge asset to the growth of the company, filling out feedback forms every now and then is an unnecessary and cumbersome process for the customers. What we are gunning for is an automated system which replaces this dysfunctional system by an automated feedback and recommendation generation system.

While, filing out separate feedback forms is tedious and only a few do it, customer care calls happen in magnitude of thousands or greater every single day and are recorded for future analysis. That's where we come in. We'll analyse these customer care calls in real time providing the executive with a real time emotional and sentimental analysis of the ongoing call represented graphically and further generating keywords out of it leading to accurate recommendations. Not only this would lift the quality of the currently ongoing call and improve consumer relations, it has multiple benefits –

1. Generation of genuine feedback using the call trends
2. Reduction in storage requirements
3. Graphical representation and analysis of the call
4. Increase in the Daily Active Users for government generated applications and schemes including Bhamashah, REAP, E-Mitra etc. which are great but lack consumer trust
5. Employee rating and call management can be managed automatically

While fulfilling all of these facets, we have been able to develop a prototype that reaches the same with 85-90% accuracy which would further go up with time. Thus, we are providing a one stop coherent solution to building consumer relations and managing them in the best way possible.

1. reactjs
2. react-graphs
3. tensorflow
4. NLTK
5. flask