System and method for automatically predicting the outcome of expert forecasts

Abstract

Systems and methods are provided that predict the accuracy of expert forecasts based on a corpus of prior expert forecasts. Expert forecasts (also referred to as opinions) are analyzed and processed to determine an opinion and related positions expressed by the expert. Each opinion is combined with relevant industry information to create a structured model for the opinion. A plurality of structured models for a given expert are then analyzed along with information related to the actual outcome of prior predictions to create a decision model for the particular expert. A new opinion issued by an expert is then analyzed using the decision model for that expert to predict the accuracy of the new opinion from the expert. Recommended actions in accordance with or against the new expert opinion may then be provided.

1 . A technical system for determining the accuracy of expert forecasts embodied in a target natural language document, comprising:
 a data storage area attached to a prediction server, wherein the data storage area stores a plurality of structured models and decision models;   a communication interface device that receives information from a data communication network, said information including natural language documents;   an opinion extraction module that receives a natural language document comprising an opinion, parses the words in said natural language document and identifies an expert who authored the natural language document, extracts an opinion, and creates a structured model of said natural language document, wherein the structured model identifies the expert and includes said extracted opinion, and stores the structured model in the data storage area;   a decision model module that obtains a plurality of structured models for an expert and processes the plurality of structured models to produce a decision model for said expert and store said decision model in the data storage area;   a prediction module configured to receive a target natural language document, identify an expert who authored the target natural language document, obtain a decision model for said expert from the data storage area, and process the target natural language document in accordance with the decision model; and   a display to present the results of the decision model processing of the target natural language document.