Critical Sentence Identification in Legal Cases Using Multi-Class Classification

Jayasinghe, Sahan; Rambukkanage, Lakith; Silva, Ashan; Silva, Nisansa de; Perera, Amal Shehan

doi:10.1109/iciis53135.2021.9660657

Cited by 5 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The top cases for these topics all involved similar fact patterns. 13 BTO L 21 captures cases involving rejected asylum seekers who argued that they faced real risks of being subjected to treatment violating Article 3 if they were sent home. This allegedly because of their previous membership in military organizations that had clashed with their countries' current governments.…”

Section: (B) Echr Resultsmentioning

confidence: 99%

“…001-140017. 13 HUDOC Case nos. 001-140017, 001-112097, 001-95882, 001-95457, 001-92119, 001-93121, 001-150311, 001-146390 and 001-70853.…”

Section: (B) Echr Resultsunclassified

“…topic models [7][8][9][10][11], word embeddings [12,13] and language models [14][15][16]. Given their centrality in legal analysis, court decisions in particular have attracted significant scholarly attention.…”

Section: Introductionmentioning

confidence: 99%

“…Researchers have increasingly sought to develop automated processes for converting unstructured legal texts to structured variables [ 3 , 4 ]. Depending on the texts involved and variables required, these have included term frequency counts [ 5 ], regular expressions [ 6 ], topic models [ 7 – 11 ], word embeddings [ 12 , 13 ] and language models [ 14 – 16 ]. Given their centrality in legal analysis, court decisions in particular have attracted significant scholarly attention.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Discovering significant topics from legal decisions with selective inference

Soh Tsin Howe

2024

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

We propose and evaluate an automated pipeline for discovering significant topics from legal decision texts by passing features synthesized with topic models through penalized regressions and post-selection significance tests. The method identifies case topics significantly correlated with outcomes, topic-word distributions which can be manually interpreted to gain insights about significant topics, and case-topic weights which can be used to identify representative cases for each topic. We demonstrate the method on a new dataset of domain name disputes and a canonical dataset of European Court of Human Rights violation cases. Topic models based on latent semantic analysis as well as language model embeddings are evaluated. We show that topics derived by the pipeline are consistent with legal doctrines in both areas and can be useful in other related legal analysis tasks. This article is part of the theme issue ‘A complexity science approach to law and governance’.

show abstract

Section: (B) Echr Resultsmentioning

confidence: 99%

“…001-140017. 13 HUDOC Case nos. 001-140017, 001-112097, 001-95882, 001-95457, 001-92119, 001-93121, 001-150311, 001-146390 and 001-70853.…”

Section: (B) Echr Resultsunclassified

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Discovering significant topics from legal decisions with selective inference

Soh Tsin Howe

2024

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

show abstract

Legal Natural Language Processing From 2015 to 2022: A Comprehensive Systematic Mapping Study of Advances and Applications

Quevedo,

Cerny,

Rodriguez

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The surge in legal text production has amplified the workload for legal professionals, making many tasks repetitive and time-consuming. Furthermore, the complexity and specialized language of legal documents pose challenges not just for those in the legal domain but also for the general public. This emphasizes the potential role and impact of Legal Natural Language Processing (Legal NLP). Although advancements have been made in this domain, particularly after 2015 with the advent of Deep Learning and Large Language Models (LLMs), a systematic exploration of this progress until 2022 is nonexistent. In this research, we perform a Systematic Mapping Study (SMS) to bridge this gap. We aim to provide a descriptive statistical analysis of the Legal NLP research between 2015 and 2022. Categorize and sub-categorize primary publications based on their research problems. Identify limitations and areas of improvement in current research. Using a robust search methodology across four reputable indexers, we filtered 536 papers down to 75 pivotal articles. Our findings reveal the diverse methods employed for tasks such as Multiclass Classification, Summarization, and Question Answering in the Legal NLP field. We also highlight resources, challenges, and gaps in current methodologies and emphasize the need for curated datasets, ontologies, and a focus on inherent difficulties like data accessibility. As the legal sector gradually embraces Natural Language Processing (NLP), understanding the capabilities and limitations of Legal NLP becomes vital for ensuring efficient and ethical application. The research offers insights for both Legal NLP researchers and the broader legal community, advocating for continued advancements in automation while also addressing ethical concerns.

show abstract