Named Entity Recognition and Resolution in Legal Text

Fine-Grained Named Entity Recognition in Legal Documents

“…To adapt categories for the legal domain, the set of NE classes was redefined in the approaches described above. Thus, Dozier et al [13] focused on legal NEs (e.g., judge, lawyer, court). Cardellino et al [8] extended NEs on NERC level to document, abstraction, and act.…”

Developing an innovative entity extraction method for unstructured data

“…The diversity of domains which rely on NLP (e.g., news media, law, biomedicine, pharmaceutical/pharmacogenomics, chemistry, etc. [7][8][9][10][11][12]) is growing and so is the variety of languages (other than English). The long-term goal of NLP is to have algorithms capable of automatically reading and obtaining knowledge from the text [13].…”

Temporal Dependence in Legal Documents

“…Dozier and colleagues have applied a hybrid method of controlled vocabulary lookup, contextual rules and statistical models to a test set of 600 US trial documents to identify named entities such as jurisdiction, court, document title, document type and name of the judge [7]. They report high precision for identifying judges (98%) and high recall for capturing jurisdictions (87%).…”

“…The consistent identification of this category is largely an open problem. While Dozier et al mention the need to identify such references 5 they do not attempt it in their analysis [7]. Quaresma and Gonçalves on the other hand emphasize the importance of this category: "we can state that these documents have a high number of references to other documents and articles [...] Table 1 gives an overview of the four categories, their definition and how often they occurred in our exploratory text sample.…”