Figure - uploaded by Marcello Ceci
Content may be subject to copyright.
Source publication
Semantic legal metadata provides information that helps with understanding and interpreting legal provisions. Such metadata is therefore important for the systematic analysis of legal requirements. However, manually enhancing a large legal corpus with semantic metadata is prohibitively expensive. Our work is motivated by two observations: (1) the e...
Contexts in source publication
Context 1
... total, the ground truth has 1202 annotations covering 1177 phrases (25 phrases have double annotations). A detailed breakdown is provided in the ground truth column of Table 8. Similar to the qualitative study, we observed no occurrences of result and a very low number of occurrences of constraint. ...Context 2
... evaluation results are presented in columns 3 through 8 of Table 8. For each legal concept (metadata type), we provide the number of matches, the number of misclassified annotations, the number of missed annotations, and scores for precision and recall. ...Context 3
... determine the root causes for the automation inaccuracies observed, we analyzed the misclassified and missed annotations. Of the 31 misclassifications in Table 8, 20 are related to polysemous concept markers. For example, the term "seizure" is a marker for sanction, since the term may refer to the confiscation of a possession. ...Context 4
... total, the ground truth has 1202 annotations covering 1177 phrases (25 phrases have double annotations). A detailed breakdown is provided in the ground truth column of Table 8. Similar to the qualitative study, we observed no occurrences of result and a very low number of occurrences of constraint. ...Context 5
... evaluation results are presented in columns 3 through 8 of Table 8. For each legal concept (metadata type), we provide the number of matches, the number of misclassified annotations, the number of missed annotations, and scores for precision and recall. ...Context 6
... determine the root causes for the automation inaccuracies observed, we analyzed the misclassified and missed annotations. Of the 31 misclassifications in Table 8, 20 are related to polysemous concept markers. For example, the term "seizure" is a marker for sanction, since the term may refer to the confiscation of a possession. ...Similar publications
Semantic legal metadata provides information that helps with understanding and interpreting legal provisions. Such metadata is therefore important for the systematic analysis of legal requirements. However, manually enhancing a large legal corpus with semantic metadata is prohibitively expensive. Our work is motivated by two observations: (1) the e...
