Partner Article

Solving Contract Analytics Using Machine Learning

SEAL Software’s promise is to accurately and quickly extract the most important provisions of a contract. Whilst this doesn’t involve apocalyptic androids scanning through mountains of paper copies, it does imply creating software smart enough to analyse and comprehend legal language.

Computers find contract extractions hard. The task at hand isn’t as straightforward as being able to recognise the legal provisions - it is about both identification and localisation. For a human brain reading through a page these two might seem the same; however, for a machine, detecting which of the many different combinations of words best fits a provision type is much more difficult than answering a simple yes or no question. Image recognition technology has a similar challenge: it is harder to single out a cat in an image depicting many animals than to correctly distinguish a cat from a dog.

To make matters worse, the SEAL system’s main function is to perform these extractions on data never seen before. The machine is expected to inherently deal with novelties and uncertainties. These predictive properties may happen naturally and unconsciously in the brain of an expert lawyer, but are far trickier to attain in a logical computer. In fact the legal language itself is continually evolving, and contract analysis can sometimes feel a lot like trying to hit a constantly moving target.

Machine Learning provides very effective tools to deal with these challenges. These algorithms mimic the human approach to recognition, and apply it to finding contract provisions. Just like a toddler learns what a cat looks like from real-life examples, not by being told to look out for fur, whiskers or tails. Their inner workings aren’t manually designed, but are automatically deduced from examples. Over time, more examples do not add complexity but generally improve accuracy, which makes these methods flexible and powerful. These algorithms also make use of the underlying uncertainties that are inherent to predictive extractions, and are an exceptional match for the dynamic and elaborate task at the heart of contract analytics.

Natural Language Processing (generally defined as “understanding spoken and written language”), is a fast-paced field at the centre of a multidisciplinary crossroad, and relies heavily on Machine Learning. Our team comprises of Computer Scientists, Data Scientists, Mathematicians and Linguists sharing their knowledge and collaborating continuously: From designing an efficient computational framework to support the processing, the mastering the intricate mathematical concepts that underpin the algorithms, to the knowledge of advanced syntactic and semantic linguistic tools. A cohesive team with varied backgrounds is pivotal to exploit all the aspects of language processing.

The accuracy of our systems depends as much on the quality of their design and inner-workings as on the material they use to learn. In other words, a machine is only as smart as its teachers! To get the finest quality and required quantity of data to teach our systems, it is crucial to team up with experienced legal experts. Their insights on the meaning and content of the extractions proved to be incredibly useful for our analytical designs. Perhaps the most exciting part of this collaboration however, is exploring the ways to directly transfer the experts’ knowledge and understanding to our systems. By trying to create a symbiotic relationship between the lawyers and the machines based on feedback and iteration, we leverage both human and computer analytical powers in what has become a rewarding and game-changing experience.

We push the boundaries of contract analytics performance with the use of a combination of industry state-of-the-art algorithms, as well as our own. For analyzing the complex structures of legal documents, one class of algorithms is king: Deep Learning.

Apart from being particularly trendy, Deep Learning is a “disruptive” branch of Machine Learning which harnesses the analytical power of vast networks of algorithms. This technology was born over 40 years ago when computer scientists tried to imitate the brain’s neural structure. Despite being promising, these computations were too slow and costly to be effective. The tremendous advances in computer hardware of the last few years have sent neural networks back to the top of the A.I pyramid, as deep learning systems are now capable of crunching large amounts of data and handling very sophisticated tasks. This explains both why it is at the forefront of natural language processing, and why it is a perfect candidate for our contractual extractions. The deep architectures allows the systems to really grasp the meaning of sentences rather than recognise sequences of words. Using Deep Learning puts us ever closer to understanding the intent of the contract’s author.

Using cutting edge technology and developing our own, the SEAL Research and Development team is dedicated to making sure you get the most accurate analysis on your contracts. Our upmost priority is to make sure that we find all the cats you’re looking for!

This was posted in Bdaily's Members' News section by Seal Software .