Artificial intelligence explaining for natural language processing
AssigneeInternational Business Machines Corporation
InventorsTakumi Yanagawa, Fumihiko Terui, Kensuke Matsuoka, Sayaka Furukawa
Life of this patent
Estimating the term…
Abstract
In an approach to AI explaining for natural language processing, responsive to receiving an input text for a machine learning model, an output is generated from the machine learning model. A plurality of alteration techniques are applied to the input text to generate one or more alternate outputs, where each alternate output corresponds to an alteration technique. A variation rate of the alternate output is calculated for each alteration technique. A preferred technique of generating neighboring data of the input text is generated based on a comparison of the variation rate of the alternate output for each alteration technique.
Claims17
A computer-implemented method for artificial intelligence explaining for natural language processing, the computer-implemented method comprising:
- responsive to receiving an input text for a machine learning model, generating, by one or more computer processors, an output from the machine learning model;
- applying, by the one or more computer processors, a plurality of alteration techniques to the input text to generate one or more alternate outputs, wherein each alternate output corresponds to an alteration technique of the plurality of alteration techniques;
- calculating, by the one or more computer processors, a plurality of variation rates of the alternate output for each respective alteration technique of the plurality of alteration techniques, wherein the plurality of variation rates represent a difference between the output and the alternate output divided by the output;
- selecting, by the one or more processors, a neighbor data generation technique based on a largest variation rate value of the plurality of variation rates;
- and
- generating, by the one or more computer processors, one or more neighboring data of the input text based on the selected neighbor data generation technique.
A computer program product for artificial intelligence explaining for natural language processing, the computer program product comprising one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the program instructions including instructions to:
- responsive to receiving an input text for a machine learning model, generate an output from the machine learning model;
- apply a plurality of alteration techniques to the input text to generate one or more alternate outputs, wherein each alternate output corresponds to an alteration technique of the plurality of alteration techniques;
- calculate a plurality of variation rates of the alternate output for each respective alteration technique of the plurality of alteration techniques, wherein the plurality of variation rates represent a difference between the output and the alternate output divided by the output;
- select a neighbor data generation technique based on a largest variation rate value of the plurality of variation rates;
- and
- generate one or more neighboring data of the input text based on the selected neighbor data generation technique.
A computer system for artificial intelligence explaining for natural language processing, the computer system comprising:
- one or more computer processors;
- one or more computer readable storage media;
- and
- program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more computer processors, the stored program instructions including instructions to:
- responsive to receiving an input text for a machine learning model, generate an output from the machine learning model;
- apply a plurality of alteration techniques to the input text to generate one or more alternate outputs, wherein each alternate output corresponds to an alteration technique of the plurality of alteration techniques;
- calculate a plurality of variation rates of the alternate output for each respective alteration technique of the plurality of alteration techniques, wherein the plurality of variation rates represent a difference between the output and the alternate output divided by the output;
- select a neighbor data generation technique based on a largest variation rate value of the plurality of variation rates;
- and
- generate one or more neighboring data of the input text based on the selected neighbor data generation technique.
Dependencies are inferred from the claim text.