AI based Multi Template Data Extraction Engine for Logistic
This project was developed for handling Logistics Invoices of different companies. The goal of the project was to extract: 1) Invoice Number 2) Invoice Total 3) Pickup Address(These could be multiple in one invoice)- Address, Zip code and States were further classified 4) Pickup Date&Time(These could be multiple in one invoice) 5) Delivery Address(These could be multiple in one invoice)- Address, Zip code and States were further classified 6) Delivery Date&Time(These could be multiple in one invoice) There was no point of hard coding every template as the variation of the templates was infinite in terms of content. Therefor, state of the art Machine Learning, NLP and OpenCV techniques were used to develop the Extraction engine. 94% Extraction Accuracy was achieved.
About Me
Computer Engineer with over three years of research and development experience in Image Processing and Machine Learning/AI. I'm passionate about building machine learning and computer vision technologies that have impact on the "real world". Expertise: - Machine Learning in production, with a focus on Object Detection, Facial Recognition, Pose Estimation and Human Feature Analysis with the State of the Art Deep Learning Algorithms and Techniques. - Level 1 Seller on Fiverr, developing Machine Learning solutions in variety of different Domains. - Algorithm Research and Optimization in Machine Learning and Operations Research. - Data-driven decision making practice in Development Life Cycle of a Machine Learning project. - Ability of self-learning and strong power of execution. - Experience with Product Problem Understanding with clients, proposing and delivering scalable Machine Learning Solutions.
$ 20 USD/hr
NEW FREELANCER!
