Using Linguistic Information and Machine Learning Techniques to Identify Entities from Juridical Documents

Quaresma, Paulo; Gonçalves, Teresa

doi:10.1007/978-3-642-12837-0_3

Cited by 22 publications

(12 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The consensus verdict from Table II has been subsequently used to compute the system's search and retrieval performance using the system-based evaluation metrics [45]- [48] given in Equations…”

Section: System Evaluation and Resultsmentioning

confidence: 99%

A Semantic Retrieval System for Case Law

Ebietomere

Ekuobase

2019

Applied Computer Systems

View full text Add to dashboard Cite

Legal reasoning, the core of legal practice in many countries, is “stare decisis” and its soundness is usually strengthened by relevant case law consulted. However, the task of relevant case law access and retrieval is tiring to legal practitioners and constitutes a serious drain on their productivity. Existing efforts at addressing this problem are conceptional, restrictive or unreliable. Specifically, existing semantic retrieval (SR) systems for case law are desirous of exceptional retrieval precision. Ontology promises to meet this desire, if introduced to the SR system. As a consequence, an ontology-based SR system for case law has been built using the systems analysis and design methodology. In particular, the component-based software engineering and the agile methodologies are employed to implement the system. Finally, the search and retrieval performance of the resultant SR system has been evaluated using the heuristics evaluation method. The retrieval system has shown to have a search and retrieval performance of about 94 % precision, 80 % recall and 84 % F-measure. Overall, the paper implements the SR system for case law with excellent precision and affirms the superiority of ontology approach over other semantic approaches to SR systems for document retrieval in the legal domain.

show abstract

Section: System Evaluation and Resultsmentioning

confidence: 99%

A Semantic Retrieval System for Case Law

Ebietomere

Ekuobase

2019

Applied Computer Systems

View full text Add to dashboard Cite

show abstract

“…Although these approaches give better results compared to other approaches, the drawback of using handcrafted rules remains the same. Another approach by Quaresma and Gonçalves combined linguistic and ML techniques for identifying entities from a corpus of legal documents. Here, NEs such as location names, organization names, dates, references to other documents, and document articles are extracted using the semantic information from the output of a natural language parser.…”

Section: Related Work On Ner Approachesmentioning

confidence: 99%

“…Srihari et al used a combination of the maximum entropy model (MaxEnt), hidden Markov model (HMM), and handcrafted patterns for extracting NEs from natural language text. Quaresma and Gonçalves proposed a mixed approach, using linguistic and ML techniques to identify entities from a corpus of legal documents. Rule‐based approaches have several advantages: first, rule‐based techniques can work without depending on the availability of annotated training data, which can be expensive to obtain.…”

Section: Introductionmentioning

confidence: 99%

An innovative hybrid approach for extracting named entities from unstructured text data

Thomas

Sangeetha

2019

Computational Intelligence

View full text Add to dashboard Cite

Named entity recognition (NER) is the core part of information extraction that facilitates the automatic detection and classification of entities in natural language text into predefined categories, such as the names of persons, organizations, locations, and so on. The output of the NER task is crucial for many applications, including relation extraction, textual entailment, machine translation, information retrieval, etc. Literature shows that machine learning and deep learning approaches are the most widely used techniques for NER. However, for entity extraction, the abovementioned approaches demand the availability of a domain‐specific annotated data set. Our goal is to develop a hybrid NER system composed of rule‐based deep learning as well as clustering‐based approaches, which facilitates the extraction of generic entities (such as person, location, and organization) out of natural language texts of domains that lack generic named entities labeled domain data sets. The proposed approach takes the advantages of both deep learning and clustering approaches but separately, in combination with a knowledge‐based approach by using a postprocessing module. We evaluated the proposed methodology on court cases (judgments) as a use case since it contains generic named entities of different forms that are poorly or not present in open‐source NER data sets. We also evaluated our hybrid models on two benchmark data sets, namely, Computational Natural Language Learning (CoNLL) 2003 and Open Knowledge Extraction (OKE) 2016. The experimental results obtained from benchmark data sets show that our hybrid models achieved substantially better performance in terms of the F‐score in comparison to other competitive systems.

show abstract

“…They report high precision for identifying judges (98%) and high recall for capturing jurisdictions (87%). Quaresma and Gonçalves use linguistic information to identify named entities in a corpus of legal documents from the International Agreements/External Relations section of the EUR-Lex portal 1 and classify documents based on some of the named entities discovered [8]. Unlike Dozier et al, Quaresma and Gonçalves found only very few person names in their corpus (presumably due to nature of international agreements).…”

Section: Named Entity and Date Extraction In Legal Textmentioning

confidence: 99%

Temporal Dependence in Legal Documents

Isemann

Ahmad

Fernando

et al. 2013

Intelligent Data Engineering and Automated Learning – IDEAL 2013

View full text Add to dashboard Cite

Abstract. Tasks and difficulties inherent in the largely open problem of temporal information extraction from legal text are outlined. We demonstrate the efficacy of tools and concepts available "off-the-shelf" and suggest refinements for such applications. In particular, the frequent references between regulatory texts have to be addressed as a separate named entity recognition task that bears relevance to an analysis of the temporal ordering of legislation. A regular expression-based approach as a robust first step towards addressing this problem is tested.

show abstract

Using Linguistic Information and Machine Learning Techniques to Identify Entities from Juridical Documents

Cited by 22 publications

References 25 publications

A Semantic Retrieval System for Case Law

A Semantic Retrieval System for Case Law

An innovative hybrid approach for extracting named entities from unstructured text data

Temporal Dependence in Legal Documents

Contact Info

Product

Resources

About