System and Method for Identifying Text in Legal documents for Preparation of Headnotes
First Claim
1. A method for generating feature graphs employed for creation of a headnote in a legal document, the method comprising:
- identifying one or more predetermined features in a plurality of legal documents, wherein the one or more predetermined features are based on grammatical constituents of text in the legal document, the plurality of legal documents being manually identified as headnote and non headnote;
obtaining data related to the availability of the one or more identified predetermined features in the sentences manually identified as headnote and non-headnote in the plurality of legal documents;
computing likelihood of a sentence being a headnote based on the obtained data;
generating feature graphs corresponding to each predetermined feature based on the computed likelihood and obtained data; and
storing the generated feature graphs in a repository.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for generating feature graphs employed for creation of a head note in a legal document is provided. The method enables identifying one or more predetermined features in a plurality of legal documents. The one or more predetermined features are based on grammatical constituents of text in the legal document. The plurality of legal documents is manually identified as headnote and non headnote. The method further enables obtaining data related to the availability of the one or more identified predetermined features in the sentences manually identified as headnote and non headnote in the plurality of legal documents. Furthermore, the method enables computing likelihood of a sentence being a headnote based on the obtained data. The method further enables generating feature graphs corresponding to each predetermined feature based on the computed likelihood and obtained data and storing the generated feature graphs in a repository.
24 Citations
37 Claims
-
1. A method for generating feature graphs employed for creation of a headnote in a legal document, the method comprising:
-
identifying one or more predetermined features in a plurality of legal documents, wherein the one or more predetermined features are based on grammatical constituents of text in the legal document, the plurality of legal documents being manually identified as headnote and non headnote; obtaining data related to the availability of the one or more identified predetermined features in the sentences manually identified as headnote and non-headnote in the plurality of legal documents; computing likelihood of a sentence being a headnote based on the obtained data; generating feature graphs corresponding to each predetermined feature based on the computed likelihood and obtained data; and storing the generated feature graphs in a repository. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A method for extracting text in a legal document for preparation of headnotes, the method comprising:
-
detecting one or more predetermined features in each sentence of the legal document, wherein the one or more predetermined features are based on grammatical constituents of text in the legal document; computing occurrence of the detected one or more predetermined features in each sentence of the legal document; retrieving feature graphs corresponding to the one or more detected predetermined features from a repository, wherein the feature graphs represent likelihood of a sentence being a headnote for a specific number of occurrences of the predetermined features in the sentences; analyzing the detected one or more predetermined features based on the computed occurrence of the one or more detected predetermined features and the retrieved feature graphs; and rendering the legal document with text tagged as headnote on a user interface based on the analysis. - View Dependent Claims (10)
-
-
11. A method for extracting text in a legal document for preparation of headnotes, the method comprising:
-
detecting one or more predetermined features in each sentence of the legal document, wherein the one or more predetermined features are based on grammatical constituents of text in the legal document; computing occurrence of the detected one or more predetermined features in each sentence of the legal document; assigning a score to each predetermined feature detected in each sentence by referring to feature graphs corresponding to each predetermined feature, wherein the feature graphs are retrieved from a repository; combining the assigned score in each sentence to obtain a final headnote score for each sentence; comparing the final headnote score with a predetermined threshold; tagging text in each sentence as headnote and non headnote based on the comparison; and rendering the legal document with text tagged as headnote on a user interface. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25)
-
-
26. A system for generating feature graphs employed for creation of a headnote in a legal document, the system comprising:
-
a feature identification module configured to identify one or more predetermined features in a plurality of legal documents, wherein the one or more predetermined features are based on grammatical constituents of text in the legal document and further wherein the feature identification module receives the plurality of legal documents manually identified as headnote and non headnote from a user interface; a feature availability computing module configured to; obtain data related to the availability of the one or more identified predetermined features in the sentences manually identified as headnote and non headnote in the plurality of legal documents; and store the obtained data in a repository; and a feature graph generating module configured to; compute likelihood of a sentence being a headnote based on the data stored in the repository; generate feature graphs corresponding to each predetermined feature based on the computed likelihood and stored data; and store the generated feature graphs in the repository. - View Dependent Claims (27, 28, 29, 30, 31)
-
-
32. A headnote preparation system for extracting text in a legal document for preparation of headnotes, the system comprising:
-
a repository configured to store feature graphs corresponding to one or more predetermined features; a feature detecting module configured to detect one or more predetermined features in each sentence of the legal document, wherein the one or more predetermined features are based on grammatical constituents of text in the legal document; a feature score assigning module configured to; compute occurrence of the detected one or more predetermined features in each sentence of the legal document; assign a score to each predetermined feature detected in each sentence by referring to feature graphs corresponding to each predetermined feature stored in the repository based on the computed occurrence of the detected one or more features; and a feature integrating module configured to combine the assigned score in each sentence to obtain a final headnote score for each sentence; a headnote generating module configured to; compare the final headnote score with a predetermined threshold; tag text in each sentence as headnote and non headnote based on the comparison; and render the legal document with text tagged as headnote on a user interface, wherein the text tagged as headnote facilitates a user to prepare a headnote for the legal document. - View Dependent Claims (33)
-
-
34. A method for extracting text in a legal document for preparation of headnotes, the method comprising:
-
retrieving feature graphs corresponding to one or more of predetermined features, wherein the feature graphs represent probability of a sentence being a headnote for a specific number of occurrence of the predetermined features in the sentence; assigning a score to each predetermined feature in each sentence by referring to the retrieved feature graphs; combining each assigned score in each sentence to obtain a final headnote score for each sentence; tagging text in the legal document as headnote and non headnote by comparing the final headnote score in each sentence with a predetermined threshold; and rendering text in the legal document that are tagged as headnotes on a user interface.
-
-
35. A headnote preparation system for extracting text in a legal document for preparation of headnotes, the system comprising:
-
a repository for storing feature graphs corresponding to each predetermined feature, wherein the feature graphs represent probability of a sentence being a headnote for a specific number of occurrence of the predetermined features in the sentence; a feature score assigning module configured to assign a score to each predetermined feature in each sentence of the legal document by referring to the feature graphs retrieved from the repository; a feature integration module configured to combine each assigned score in each sentence to obtain a final headnote score for each sentence; a headnote generating module configured to tag text in the legal document as headnote and non headnote by comparing the final headnote score in each sentence with a predetermined threshold; and a user interface configured to receive the legal document with text tagged as headnotes.
-
-
36. A computer program product comprising:
-
a non-transitory computer-readable medium having computer-readable program code stored thereon, the computer-readable program code comprising instructions that when executed by a processor, cause the processor to; detect one or more predetermined features in each sentence of the legal document, wherein the one or more predetermined features are based on grammatical constituents of text in the legal document; compute occurrence of the detected one or more predetermined features in each sentence of the legal document; retrieve feature graphs corresponding to the one or more detected predetermined features from a repository, wherein the feature graphs represent likelihood of a sentence being a headnote for a specific number of occurrences of the predetermined features in the sentences; analyze the detected one or more predetermined features based on the computed occurrence of the one or more detected predetermined features and the retrieved feature graphs; and render the legal document with text tagged as headnote on a user interface based on the analysis. - View Dependent Claims (37)
-
Specification