We believe that the world is sure to adopt a form of natively-digital contracting. But there are several problems with contract data.
Problem 1: Contracts are unstructured, un standardized, and use nuanced legal language.
Problem 2. Contracts exist to guard against rare and potentially catastrophic occurrences, so tolerance for false negatives and false positives are pretty close to zero. Almost as soon as the pandemic began, our customers started asking for more information about their contracts In order to solve the above problems, we can use AI and Cloud Auto ML Natural Language. Here are the steps of how we should do that.
First, we need to upload a small, curated set of contracts and label three properties: entity name, signature date, and signer name. Secondly, after a few hours of training, the signature date had precision and recall rates surpassing 90%. Therefore, this is the best result we’d ever achieved over three years of on-and-off experiments — and, incredibly, Google needs a relatively tiny data set to achieve it. Finally, the model was immediately live on Google Cloud AI Platform for predictions, so we could start testing the user experience that every day.In summary, with the development of data and AI, we are exploring new ways to apply Google Cloud AI to make contracting faster and smarter for the customers.
—— written by Xun_cui