What are natural language processing and machine learning and how could they assist with contractual drafting in the future?
What is machine learning?
Machine learning is a field of computer science that allows computers to recognize patterns and learn from them. With machine learning algorithms are used to learn from and make predictions on data. Arthur Samuel, a pioneer of the tech industry with a background in electrical engineering, was the first one to label this field of computer science as machine learning and defined it as “Field of study that gives computers the ability to learn without being explicitly programmed” in 1959.
What is Natural Language Processing (NLP)?
Wikipedia defines NLP as a field of computer science, artificial intelligence, and computational linguistics concerned with the interactions between computers and human (natural) languages. This NLP is not to be confused with another concept often given the same acronym of NLP - Neuro-Linguistic Programming which is an approach to communication, personal development and psychotherapy.
The first experimentation in relation to NLP began in the 1950s with such tests as the Georgetown Experiment in 1954 where more than sixty Russian sentences were translated by computers into English. Those who were responsible for this experiment predicted that within three to five years machines would be able to fully translate text from any language. However, machine translation proved more difficult and progress was much slower. It has not been until much more recently that technologies such as Siri, Google Translate and Predictive Text have been successfully added to humanity’s technological repertoire.
This is probably because language is not simply a “code” but a complicated system for exchanging information that requires knowledge about context and the world itself in order to be understood. A sophisticated NLP system requires “knowledge” not only about complicated structures in language itself, which often do not follow logical rules. But also a manner in which it can comprehend and analyze context. Machine learning can assist with this. NLP together with machine learning can enable computers to recognise patterns within language and make predictions or suggestions based on this learning.
So, how could machine learning and NLP be applied to B2B contracting in the future?
Contractual language could be said to be its own language. There are particular patterns and vocabulary used that are not the same as those used in everyday conversations and written communications. Here are a few suggestions about how NLP and machine learning could possibly be used to transform the contracting space in the future.
- Large amounts of contracts could be reviewed and analyzed so as to predict which set of words for a particular clause or term in an agreement are likely to cause the least friction. NLP and machine learning could also be used to identify problematic terms and clauses which may need to be discarded;
- You may be able to talk to a contractual assistant (computer), describe the type of contract you want to create and have it draft the contract it believes is the best for your particular situation, based on past data. This means you could draft an entire contract without having to type or touch a button;
- You may also be able to converse with a computer contractual assistant to ask about particular problems you may be likely to have with a contract you are working on so you can anticipate and plan for these;
- Contractual software may be able to record conversations between yourself and a counterparty and analyze this to determine where you may be disagreeing and provide suggestions for how these issues could be resolved. It may also be able to analyze the tone of voice of your counterparty for “agreeability”, predict how long they will likely take to agree with you on a particular point or deal and make suggestions on how to interact with the counterparty to influence them to your advantage.
- NLP and machine learning could also be used in the translation of contracts into other languages. This technology may also be able to do this while also realizing that the clauses which work in other languages may not be exact translations of those which work in english. The same here could be said for contracts in the same language but for different jurisdictions - some clauses may work well in one jurisdiction but not in another.
So there are some ideas about how NLP and machine learning may affect contracting and deal making in the future.
What are your thoughts? How do you think these computer science fields will affect how we agree in the future?
ContractRoom is a contract management system that will use NLP and machine learning to predict how people will agree - Predictive AgreementTM. To find out more about ContractRoom or book a free demo, click here: