Toward Automatically Identifying Legally Relevant Factors

Abstract

In making legal decisions, courts apply relevant law to facts. While the law typically changes slowly over time, facts vary from case to case. Nevertheless, underlying patterns of fact may emerge. This research focuses on underlying fact patterns commonly present in cases where motorists are stopped for a traffic violation and subsequently detained while a police officer conducts a canine sniff of the vehicle for drugs. We present a set of underlying patterns of fact, that is, factors of suspicion, that police and courts apply in determining reasonable suspicion. We demonstrate how these fact patterns can be identified and annotated in legal cases and how these annotations can be employed to fine-tune a transformer model to identify the factors in previously unseen legal opinions.

Discovering significant topics from legal decisions with selective inference

“…After the ECtHR found this to be a violation in 1996, 22 several similar and ultimately successful cases were raised in which the ECtHR expressly 'recalls that it has already examined' such complaints and found violations. 23…”

“…Other researchers have prioritized more explainable methods over end-toend prediction. Typically, algorithms are first developed to automatically extract case attributes and other legally relevant variables before using these variables to model outcomes [8,[21][22][23][24]. The goal is not necessarily predictive accuracy alone, but also to identify and explain what motivates legal decisions.…”

Empirical legal analysis simplified: reducing complexity through automatic identification and evaluation of legally relevant factors

“…In the pilot study [3,4], we hypothesized that a program can employ ML/NLP methods to learn automatically to identify the essential elements in DIAS cases that courts consider in assessing the legality of the stop; that is, the factors which courts use to determine if a police officer had reasonable suspicion that the automobile contained drugs.…”

“…From the cases returned by our searches, we then selected the first 1 211 relevant cases that addressed a motion to suppress evidence. Of these, the courts found reasonable suspicion in 63% of cases and found no reasonable suspicion in 37% [3,4].…”

“…One can use a similarity metric to select the cases that maximize factors shared with the focal case and minimize unshared ones. 3 Comparing the current fact situation with the most similar cases may current fact situation:…”