Artificial Intelligence (AI) refers to systems or machines that mimic human intelligence in its ability to reason, understand and perform tasks, and learn iteratively over time, by using patterns in training data. AI technology is improving human performance and productivity by automating processes or tasks that once required human energy and were often tedious or repetitive, and by making sense of data on a much larger scale than humanly possible.
Machine Learning is the learning process for AI. It’s a method of data analysis based on the idea that systems can learn from data, identify patterns, and make decisions with minimal human intervention. Some examples of familiar ways machine learning is used today: online recommendations from Amazon or video recommendation from Netflix; fraud detection; and Google Gmail suggestions when replying to emails.
A neural network is a computing system made up of several simple, highly interconnected processing elements [neurons], which process information and provide data outputs. Made up of multiple layers, such as the input layer, where data is inputted; the hidden layers, where the data is processed; and the output layer, where the final data is released, neural networks solve a wide variety of problems in natural language processing, computer vision, forecasting, data validation, risk management, customer research, and more. In contracting, neural networks are used in natural language processing or computer vision to understand how to read a document and associate terms and phrases.
Computer vision is another subset of AI that enables machines to perceive as if they had human vision. It allows the AI to read and understand images. Relevant to contract management, it allows a system to parse text out of scanned images and documents using OCR (Optical Character Recognition), as well as read logos, signatures, and tables.
Deep learning is a subset of ML-based on neural networks and is another method to train AI similar to how the brain processes information. It trains the AI to learn to detect objects, perform named entity recognition, translate languages, and multiple other tasks.
Natural language processing (a form of ML and DL) is a type of AI that enables it to program, process, and analyze language data. In contract management, NLP plays an important role in metadata extraction, and provision identification. It can help identify standard clauses, contextualize terms to locate clauses that are non-standard but acceptable, and pinpoint non-standard, non-acceptable clauses. It can also contextualize terms and dates to provide renewal and expiration dates when they aren’t exactly spelled out. Here are the components of NLP that help identify clause, term, and metadata:
In Contracting AI there are several components of NLP that identify Clause, Term, and Metadata
Structure of words and sentences, contextual spelling corrections trained on a legal lexicon
Understanding the syntax of words or phrases (e.g., n-grams) for parsing and extraction purposes
Identifying the polarity (positive/negative) of a statement in a legal context
Understanding the meaning of words, sentences, and complete paragraphs using custom-built embedding representations, context, and document-level characteristics