Scalable and explainable legal prediction

Branting, L. Karl; Pfeifer, Craig; Brown, Bradford; Ferro, Lisa; Aberdeen, John S.; Weiss, Brandy; Pfaff, Mark S.; Liao, Bill

doi:10.1007/s10506-020-09273-1

Cited by 67 publications

(60 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Such cases may involve analogy [4], or some kind of common sense ontology. This suggests that the key role for ML is not the prediction of outcomes, but the identification of the factors as in [3], [19] and [9].…”

Section: Discussion Of Challengesmentioning

confidence: 99%

“…A major weaknesses of these approaches is, however, that they are unable to explain their reasoning in an acceptable manner. Traditional explanations of ML such as listing or highlighting the most influential words in the texts have been shown to be unhelpful [9] because they are difficult to relate to the relevant law. Moreover there are good reasons why any such explanation would be inappropriate in a legal context [8].…”

Section: Introductionmentioning

confidence: 99%

“…In order to explain the predicted outcomes in appropriate terms, researchers have turned to the extensive body of work on explanation developed in AI and Law [5]. In particular the type of explanation advocated has been based on the use of factors, as developed in CATO [2], e.g [9], [16] and [6]. Specifically, using the set of argumentation schemes designed to capture the reasoning of CATO from [17] is advocated in [16].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Explaining Factor Ascription

Mumford

Atkinson

Bench‐Capon

2021

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Explanation and justification of legal decisions has become a highly relevant topic in light of the explosion of interest in the use of machine learning (ML) approaches to predict legal decisions. Current suggestions are to use the established factor based explanations developed in AI and Law as the basis for explaining such programs. We, however, identify factor ascription as an important aspect of explanation of case outcomes not currently covered, and argue that explanations must also include this aspect. Finally, we outline our proposal for a hybrid system approach that combines ML and Abstract Dialectical Framework (ADF) layers to engender an explainable process.

show abstract

Section: Discussion Of Challengesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Explaining Factor Ascription

Mumford

Atkinson

Bench‐Capon

2021

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

show abstract

“…To verify the assumption, IBP infers large numbers of past legal judgments [6]. A new model, CNN-BiGRU, is suggested to have better prediction than a single CNN or RNN model [7], and we propose the generalized Gini-PLS algorithm, which is based on the simple Gini-PLS model, to develop a judicial prediction system [8].…”

Section: Prediction Of Legal Judgmentsmentioning

confidence: 99%

AI Model for Predicting Legal Judgments to Improve Accuracy and Explainability of Online Privacy Invasion Cases

Park

Chai

2021

Applied Sciences

View full text Add to dashboard Cite

Since there are growing concerns regarding online privacy, firms may have the risk of being involved in various privacy infringement cases resulting in legal causations. If firms are aware of consequences from possible cases of invasion of online privacy, they can more actively prevent future online privacy infringements. Thus, this study attempts to predict the probability of judgment types caused by various invasions within US judicial cases that are related to online privacy invasions. Since legal judgment results are significantly influenced by societal factors and technological development, this study tries to identify a model that can accurately predict legal judgment with explainability. To archive the study objective, it compares the prediction performance by applying five types of classification algorithms (LDA, NNET, CART, SVM, and random forest) of machine learning. We also examined the relationship between privacy infringement factors and adjudications by applying network text analysis. The results indicate that firms could have a high possibility of both civil and criminal law responsibilities if they distributed malware or spyware, intentionally or non-intentionally, to collect unauthorized data. It addresses the needs of reflecting both quantitative and qualitative approach for establishing automatic legal systems for improving its accuracy based on the socio-technical perspective.

show abstract

“…Only one allegation code could be predicted with an MCC greater than 0.15, meaning that predictive accuracy for most allegation codes was only slightly higher than chance. We attempted to develop an annotation scheme for complaint texts so that we could apply the methodology described in [6] to the corpus, but the complaints' extreme variability and disorganization frustrated these annotation efforts.…”

Section: Predicting Decisions From Complainant Textsmentioning

confidence: 99%

Judges Are from Mars, Pro Se Litigants Are from Venus: Predicting Decisions from Lay Text

Branting

Balhana

Pfeifer

et al. 2020

Frontiers in Artificial Intelligence and Applications

Self Cite

View full text Add to dashboard Cite

Access to justice could be significantly expanded if decision support systems were able to accurately interpret statements of fact by pro se (self-represented) litigants. Prior research, which has demonstrated that case decisions can often be predicted by machine-learning models trained on judges’ statements of facts, suggests the hypothesis that these same learning algorithms could be effectively applied to pro se litigants’ fact statements. However, there has been a dearth of corpora on which to test this hypothesis. This paper describes an experiment testing the ability to predict the outcome of pro se litigants’ complaints on a corpus of 5,842 cases initiated by citizen complaints. The results of this experiment were strikingly negative, suggesting that fact statements by unguided pro se litigants are far less amenable to simple machine-learning techniques than judges’ texts and appearing to disconfirm the hypothesis above.

show abstract

Scalable and explainable legal prediction

Cited by 67 publications

References 39 publications

Explaining Factor Ascription

Explaining Factor Ascription

AI Model for Predicting Legal Judgments to Improve Accuracy and Explainability of Online Privacy Invasion Cases

Judges Are from Mars, Pro Se Litigants Are from Venus: Predicting Decisions from Lay Text

Contact Info

Product

Resources

About