发明名称 |
Artificial intelligence expert system for screening |
摘要 |
An artificial intelligence expert system for screening provides characteristic profiles to candidates to perform a particular task. The profiles have individual screening items within them that are expected to be related to whether or not a person is suitable for the task. The responses from the persons to the items are received by a computer implemented expert system. The expert system applies a combined model to the responses to generate a forecasted performance of the person to the task. The combined model is a linear combination of two or more path dependent regressions performed on data from a set of N training persons with known abilities to do the task. The number of parameters in each path dependent model is limited to a fraction of the number N so that the path dependent models are not over fit to the data. A suitable fraction is ⅕. |
申请公布号 |
US9280745(B1) |
申请公布日期 |
2016.03.08 |
申请号 |
US201514793841 |
申请日期 |
2015.07.08 |
申请人 |
Applied Underwriters, Inc. |
发明人 |
Clark David Alan;Smith Justin N. |
分类号 |
G06F15/18;G06N5/04;G06N99/00 |
主分类号 |
G06F15/18 |
代理机构 |
|
代理人 |
Nowotarski Mark |
主权项 |
1. An artificial intelligence expert learning system for screening comprising:
a) a computer implemented screening item measuring instrument comprising a measuring instrument output device and a measuring instrument input device; b) a task performance database comprising task performance metric data for a set of N training persons wherein said N training persons all have a same task function characterized by said task performance metric; c) a computer implemented modeling engine comprising a modeling engine output device and a modeling engine input device; d) a computer implemented screening engine comprising a screening engine output device and a screening engine input device; and e) a permanent memory comprising computer executable instructions to physically cause:
i) said measuring instrument to provide a characteristic profile comprising one or more screening items and one or more non-screening items to said set of N training persons through said measuring instrument output device;ii) said measuring instrument to read in responses to said items from said set of N training persons through said measuring instrument input device;iii) said modeling engine to:
1) read in said responses from said N training persons from said measuring instrument; and2) read in said task performance metric data for said N training persons from said task performance database;iv) said modeling engine to fit a first path dependent model of said task performance metric data using responses to a first subset of said screening items, said first path dependent model comprising not more than M parameters where M is less than or equal to N/E wherein E has a value of 5 or greater;v) said modeling engine to fit a second path dependent model of said task performance metric using responses to a second subset of said screening items, said second path dependent model comprising not more than M parameters, and wherein said first subset of said screening items is different than said second subset of said screening items by at least one screening item;vi) said modeling engine to form a linear combination of said first and second path dependent models to form a combined model;vii) said screening engine to read in said combined model;viii) said screening engine to provide said characteristic profile to a candidate for said task function through said screening engine output device;ix) said screening engine to receive responses to the screening items in said characteristic profile from said candidate through said screening engine input device;x) said screening engine to execute said combined model using said candidate responses to produce an forecasted task performance metric for said candidate; andxi) said screening engine to reject said candidate for said task when said projected task performance metric is less than a minimum threshold task performance metric. |
地址 |
Omaha NE US |