| CPC G06Q 30/018 (2013.01) [G06Q 50/26 (2013.01)] | 11 Claims |
|
1. A prediction system comprising:
a first device that collects relevant information relating to revision of legislation, regulations, or standards;
a second device that performs predetermined conversion processing for converting the relevant information into an index relating to a degree of influence on the revision; and
a third device that predicts based on the index, a stage of the revision or properness of negotiation timing relating to the revision,
wherein:
the relevant information includes prescribed information determined as motiving the revision,
the prescribed information includes first information which is information on media reports or research reports about atmospheric pollution,
the second device includes a first trained model trained through machine learning,
the first trained model is trained through training processing using a first training data set to provide information representing a degree of influence on the revision in response to an input,
the first training data set includes a plurality of pieces of training data obtained by labeling information on a predetermined item with information representing the degree of influence, and
the second device:
performs prescribed processing to extract the information on the predetermined item from the first information,
sets the information on the predetermined item as an input to the first trained model,
converts the information representing the degree of influence provided from the first trained model into a point by using a predetermined first function, and
sets an average value of points for a first prescribed period as the index.
|