Law firm’s client claiming software plagiarism engages Baker Tilly to provide data analysis
A law firm claiming software plagiarism on their client’s intellectual property engaged Baker Tilly to measure the amount of intellectual property contained within multiple online software web applications. Read this case study to lean how Baker Tilly provided the company the data needed to support their damage claims.
Client background
This company is a law firm representing clients with multiple intellectual property disputes related to the usage of licenses software source code and data structures within multiple online web applications.
The business challenge
The company believed software plagiarism had occurred with their intellectual property and wanted to prove that their code was unlicensed and being used by the defendant. Due to the nature of this dispute, the company needed someone to measure the amount of intellectual property that was contained and utilized within the online software web applications that were used to send and receive electronic data interchange and online retail sales and deliveries from the online platform. These applications consisted of source code files, configuration files, databases and sample data transmissions.
This ask requited developing an objective measurement process to measure similarity over large repositories consisting of over 50 million terms, codes, text statements and data structures, before providing an expert analysis and opinion on if and how much intellectual property was contained and utilized within the online web applications.
Strategy and solution
Baker Tilly used a natural language processing (NPL) approach to identify and measure the similarity of items from both online software web application repositories. We started by running an optical character recognition (OCR) conversion of all the PDF documents containing business transactions, producing searchable data.
Using Amazon Web Services (AWS) products such as AWS Sage Maker, AWS Textrack and AWS Open Search, we were able to conduct a similarity search to detect any of the company’s intellectual property on unstructured text such as code, configuration files, PDFs, Word documents and images being used by the defendant. This allowed us to observe how much code from the company was being used, unlicensed, by the defendant.
As a result, Baker Tilly’s analysis provided the company the data needed to support their multiple intellectual property cases damage claim calculation of $35 million.
