ChatGPT: Information Retrieval from Image using Robotic Process Automation and OCR

摘要

ChatGPT hit 1M users in just 5 days once it went public, the leading companies such as Facebook, Instagram, Spotify, Netflix took several months to years to achieve this mark. As a result of the conversational AI Chabot being hailed as a technological revolution, ChatGPT has emerged as one of the year’s most popular buzzwords. This research study intends to initiate interaction with ChatGPT more convenient to users by using user sent WhatsApp images containing text as input by converting it into text-based prompt which is accepted by ChatGPT and provide the user with the response generated by ChatGPT. This is achieved by integrating RPA to automate an in-depth examination of images for information discovery using OCR to extract text and ChatGpt to retrieve information. This research study also focuses on providing a quick response to the user by comparing images with previous users’ inputs and sending the same response in case there is repetition of images which increases efficiency in terms of computation resources too. The proposed solution is designed to provide a quick suitable response to the user inscribed text in the image.