USING CLASSIFIED TEXT AND DEEP LEARNING ALGORITHMS TO ASSESS RISK AND PROVIDE EARLY WARNING
Deep learning is used to identify specific risks to an enterprise of a pending litigation and identify documents of interest for the litigation. The system involves mining and using existing classifications of data (e.g., from a litigation database) to train one or more deep learning algorithms, and then examining the electronically stored information with the trained algorithm, to generate a scored output that will enable enterprise personnel to review risks to the enterprise, e.g. to enable enterprise personnel to assess the nature and extent of the potential damage from the litigation, and to identify relevant documents that would be saved to prevent spoliation.
1 . A method of using classified text and deep learning algorithms to assess risk and identify relevant documents comprising:
creating one or more training datasets for textual data corresponding to a specific risk classification, wherein said risk classification comprises a nature of a recently filed lawsuit;
training one or more deep learning algorithms using said one or more training datasets;
collecting and extracting a corpus of documents comprising electronically stored information stored by an enterprise;
applying said one or more deep learning algorithms to said corpus of documents to identify and report one or more documents of interest in the said corpus of documents for an early assessment of the potential harm to the enterprise of said lawsuit;
determining if said identified one or more documents of interest is a false positive or a true positive; and
re-training said one or more deep learning algorithms if said identified one or more documents of interest is a false positive.
2 . The method of claim 1 , wherein said one or more deep learning algorithms is a framework for natural language processing of text.
3 . The method of claim 1 , wherein said one or more deep learning algorithms is a recurrent neural network with a multiplicity of layers and various features, including but not limited to long short-term memory or gated recurrent units.
4 . The method of claim 1 , wherein the one or more deep learning algorithms have been trained with different classifiers using previously classified data sourced and provided by a subject matter expert to become models for specific threats or risks of interest.
5 . The method of claim 1 , wherein each one of said one or more deep learning algorithms has also been trained with one or more datasets unrelated to the threats or risks of interest.
6 . The method of claim 1 , wherein said one or more training datasets is obtained by mining one or more litigation databases.
7 . A method of using classified text and deep learning algorithms to identify risk and provide early warning comprising:
creating one or more training datasets by mining one or more litigation databases for textual data corresponding to a specific threat or risk of interest;
training one or more deep learning algorithms using said one or more training datasets;
collecting and extracting a corpus of documents comprising electronically stored information stored by an enterprise;
applying said one or more deep learning algorithms to said corpus of documents to identify and report one or more documents of interest in the said corpus of documents for an early assessment of the potential harm to the enterprise of said lawsuit;
determining if said identified one or more documents of interest is a false positive or a true positive; and
re-training said one or more deep learning algorithms if said identified one or more documents of interest is a false positive.
8 . The method of claim 7 , wherein each one of said one or more deep learning algorithms scores the data for accuracy with the deep learning model classification of the data.
9 . The method of claim 8 , wherein said report comprises providing the scores and related data to one or more designated users.
10 . The method of claim 8 , wherein the report may be limited to scores which surpass a specified threshold associated with each of said one or more deep learning algorithms.
11 . The method of claim 8 , wherein the report and the documents of interest are exported to an existing case management system for investigation and review and possible further action.