In the law, there unfortunately exists no easy way for a lawyer to determine whether the judicial opinions they cite in support of their arguments retain their validity. There exist services that provide this information, but they cost enough that many small legal practices can’t afford them. Their clients are thus left at a steep disadvantage by their lawyers’ inability to verify the validity of their arguments. Ashkon identified the opportunity to apply his academic background to help rectify this problem. To reduce the need for human input into the process of aggregating negative treatments, decrease the cost of producing that information and increase the accessibility of relevant services, Ashkon is building a machine learning system that automates the process of extracting negative treatments of judicial opinions from legal corpora.

This system models the task of identifying negative treatments articulated in judicial opinions as a classification problem. It builds a training data set by using regular expressions to extract the small subset of negative treatments expressed in a predictably structured manner from a dataset of judicial opinions. It subsequently uses this limited training set to train a generalized classifier that is able to extract a much larger number of negative treatments expressed in more diverse ways from the same set of opinions.

Project Leader: Ashkon Farhangi, former CodeX fellow

The descriptions of current and past projects of CodeX non-residential fellows are provided to illustrate the kind of work our non-residential fellows are carrying out. These projects are listed here for informational purposes only and are not endorsed by CodeX, Stanford Law School, or Stanford University.