JACKSONVILLE, Fla., Feb. 16, 2021 /PRNewswire-PRWeb/ — Just about every organization in the world has large volumes of documents stored in a physical folder and/or in digital format in a data base in their server rooms or in the cloud. These documents represent years of hard work by the employees who produced them, as well as a large financial investment made by the organization. The challenge for organizations has been how to automate classifying and extracting pertinent information from all the different document types. The most current optical character recognition (OCR) technologies fall well short of being able to do this effectively.
To solve this challenge, the award-winning data science team at NLP Logix repurposed a neural network algorithm it used to accurately identify breast cancer tumors in pathology slides. The team then trained the algorithm to identify and extract characters and other features from document images, such as a logo. The end-to-end process to accomplish this complex task generally follows this workflow: collect large amounts of samples of the types of documents you are processing; artificially “clean” the documents by removing blemishes; run them through a neural network to train it to classify the type of document; identify the information you want; finally, extract the information you want and place it in a readable file format.
“We have invested years in perfecting the techniques,” said Robert Marsh, Chief Technology Officer, NLP Logix. “To have this fourth and latest patent awarded is a great achievement and testament to the team here at NLP Logix.”
Wilson Dutra, a Jacksonville-based intellectual property and patent law firm, successfully assisted and guided NLP Logix through the process of filing all four awarded patents.
“We really appreciate the work Wilson Dutra did for us in securing these patents,” said Ted Willich, Chief Executive Officer, NLP Logix. “It is great to have a high quality IP firm right down the street from our offices in Jacksonville, Florida.”
The patented data extraction technology is currently being used in the workers compensation, auto casualty insurance and healthcare industries. The Company is working to introduce it to the non-profit and government sectors in the first quarter of 2021.
Ted Willich, NLP Logix, 904 208 5065, email@example.com
SOURCE NLP Logix