| CPC G06Q 10/04 (2013.01) [G06Q 10/06375 (2013.01); G06Q 50/18 (2013.01); G06F 40/30 (2020.01)] | 18 Claims |
|
1. A legal information processing system comprising a processor that executes computer-executable instructions that cause the processor to act as:
a key person setting section configured to set one or more key people that are related people related to a law, regulation, or standard;
an information gathering section configured to gather at least public information including a history of revisions to the law, regulations, or standard, the key person set by the key person setting section being involved in creating or publishing the public information; and
a revision predicting section configured to predict a revision trend of the law, regulation, or standard with a neural network based inference engine that has undergone a learning process based on learning data and a parameter updating section, the learning data is content of the public information including the history of revisions, the public information being gathered by the information gathering section and the parameter updating section updates each value of a parameter group in a manner to decrease an error generated by a comparison of an actual output value of the inference engine with an idealized output value of the inference engine, wherein
the legal information processing system handles the public information or private information with one or more computers, wherein general key people out of the one or more key people, or a key person related to product development in his or her own company is involved in creating or publishing the public information or the private information,
the legal information processing system further comprising:
an information classifying unit that classifies information in prescribed syntax units obtained by parsing the public information or the private information, into each category of the law, regulation, or standard; and
a matching information extracting unit that extracts matching information indicating matching points or differing points in interpretation between enactment content of the law, regulation, or standard and a development category of a product, and
a client terminal apparatus that provides notification about content of a current law, regulation, or standard requiring revision corresponding to a development category of the product, the content being the matching information extracted in the matching information extracting step, in a case where the development category of the product is input as the public information or the private information,
the revision predicting section includes
a feature amount generating module that generates one or more feature amounts to be provided in the learning process of the inference engine, and
a learning process module that controls the learning process of the inference engine and includes a learning data generating section and a parameter updating section that reference a history of revisions that have actually occurred and generate the learning data, and
wherein a determination that a revision of the law, regulation, or standard has occurred automatically triggers performance of the learning process with the learning process module.
|