Explainable passage classification
An embodiment includes tokenizing an input passage into an n-gram sequence of tokens. The embodiment also includes evaluating the input passage using a trained classification model that generates an output indicative of a classification of the input passage. The embodiment also includes generating a first token vector for a first token of the sequence of tokens and projecting the first token vector to a higher dimensional space, resulting in a first projected token vector. The embodiment also includes generating a first similarity score for the first projected token vector based on comparisons of the first projected token vector to each of a plurality of class representations. The embodiment also includes generating a ranked list of the tokens, wherein the generating of the ranked list includes ranking the first token among others of the tokens based on the first similarity score.
1. A computer implemented method comprising:
receiving an input passage and a class selection, the input passage being a part of a service request received over an application programming interface (API) from a user device, and the input passage comprising Natural Language data that is tokenizable in an n-gram sequence of tokens;
tokenizing the input passage into the n-gram sequence of tokens;
evaluating the input passage using a trained classification model within a natural language processing neural network that generates an output indicative of a classification of the input passage in one of a plurality of predefined classes, the predefined classes including a class corresponding to the class selection;
generating a first token vector for a first token of the n-gram sequence of tokens;
projecting the first token vector to a higher dimensional space, resulting in a first projected token vector, the higher dimensional space being a multiple of a dimensional space associated with the trained classification model;
generating a first similarity score for the first projected token vector based on comparisons of the first projected token vector to each of a plurality of class representations, each class representation corresponding to one of the plurality of predefined classes, and the first projected token vector being associated with each of the plurality of predefined classes for which a similarity distance value meets a predetermined threshold;
generating a ranked list of the tokens, wherein the generating of the ranked list includes ranking the first token among others of the tokens based on the first similarity score; and
displaying the ranked list of the tokens.
2. The method of claim 1 , wherein the trained classification model is a trained deep semantic classification model.
3. The method of claim 2 , wherein the output of the trained classification model includes a vector of probability values corresponding to respective classes of the plurality of predefined classes.
4. The method of claim 1 , further comprising generating a second token vector for a second token of the n-gram sequence of tokens.
5. The method of claim 4 , further comprising projecting the second token vector to the higher dimensional space, resulting in a second projected token vector.
6. The method of claim 5 , wherein the trained classification model is an n-dimensional model, and wherein the higher dimensional space is a 2n-dimensional space.
7. The method of claim 5 , further comprising generating a second similarity score for the second projected token vector based on comparisons of the second projected token vector to each of the plurality of class representations.
8. The method of claim 7 , wherein the generating of the ranked list of tokens further comprises ranking the second token among others of the tokens based on the second similarity score.
9. The method of claim 1 , wherein the plurality of class representations includes a first reference passage in a designated class from among the plurality of predefined classes and a second reference passage in one of the plurality of predefined classes other than the designated class, and wherein the comparisons of the first projected token vector to each of the plurality of class representations includes a first comparison of the first projected token vector to the first reference passage and a second comparison of the first projected token vector to the second reference passage.
10. The method of claim 9 , further comprising automatically designating, as the designated class, the one of the plurality of predefined classes in which the trained classification model classified the input passage.
11. The method of claim 9 , further comprising:
receiving a user input indicative of a class selected by a user; and
designating, as the designated class, the class selected by the user.
12. The method of claim 9 , wherein the generating of the first similarity score further comprises adding a first value resulting from the first comparison and subtracting a second value resulting from the second comparison.
13. The method of claim 1 , further comprising rendering a display of the ranked list of the tokens.
14. The method of claim 1 , wherein the plurality of class representations includes a first reference feature in the one of the plurality of predefined classes in which the trained classification model classified the input passage, and a second reference feature in one of the plurality of predefined classes other than the one of the plurality of predefined classes.
15. A computer program product, the computer program product comprising one or more non-transitory computer readable storage media, and program instructions collectively stored on the one or more non-transitory computer readable storage media, the program instructions executable by one or more processors to cause the one or more processors to perform operations comprising:
receiving an input passage and a class selection, the input passage being a part of a service request received over an application programming interface (API) from a user device, and the input passage comprising Natural Language data that is tokenizable in an n-gram sequence of tokens;
tokenizing the input passage into the n-gram sequence of tokens;
evaluating the input passage using a trained classification model within a natural language processing neural network that generates an output indicative of a classification of the input passage in one of a plurality of predefined classes, the predefined classes including a class corresponding to the class selection;
generating a first token vector for a first token of the n-gram sequence of tokens;
projecting the first token vector to a higher dimensional space, resulting in a first projected token vector, the higher dimensional space being a multiple of a dimensional space associated with the trained classification model;
generating a first similarity score for the first projected token vector based on comparisons of the first projected token vector to each of a plurality of class representations, each class representation corresponding to one of the plurality of predefined classes, and the first projected token vector being associated with each of the plurality of predefined classes for which a similarity distance value meets a predetermined threshold;
generating a ranked list of the tokens, wherein the generating of the ranked list includes ranking the first token among others of the tokens based on the first similarity score; and displaying the ranked list of the tokens.
16. The computer program product of claim 15 , wherein the stored program instructions are stored in a computer readable storage device in a data processing system, and wherein the stored program instructions are transferred over a network from a remote data processing system.
17. The computer program product of claim 15 , wherein the stored program instructions are stored in a computer readable storage device in a server data processing system, and wherein the stored program instructions are downloaded in response to a request over a network to a remote data processing system for use in a computer readable storage device associated with the remote data processing system, further comprising:
program instructions to meter use of the program instructions associated with the request; and
program instructions to generate an invoice based on the metered use.
18. The computer program product of claim 15 , wherein the plurality of class representations includes a first reference passage in a designated class from among the plurality of predefined classes and a second reference passage in one of the plurality of predefined classes other than the designated class, and wherein the comparisons of the first projected token vector to each of the plurality of class representations includes a first comparison of the first projected token vector to the first reference passage and a second comparison of the first projected token vector to the second reference passage.
19. The computer program product of claim 18 , further comprising automatically designating, as the designated class, the one of the plurality of predefined classes in which the trained classification model classified the input passage.
20. A computer system comprising one or more processors and one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable by the one or more processors to cause the one or more processors to perform operations comprising:
receiving an input passage and a class selection, the input passage being a part of a service request received over an application programming interface (API) from a user device, and the input passage comprising Natural Language data that is tokenizable in an n-gram sequence of tokens;
tokenizing the input passage into the n-gram sequence of tokens;
evaluating the input passage using a trained classification model within a natural language processing neural network that generates an output indicative of a classification of the input passage in one of a plurality of predefined classes, the predefined classes including a class corresponding to the class selection;
generating a first token vector for a first token of the n-gram sequence of tokens;
projecting the first token vector to a higher dimensional space, resulting in a first projected token vector, the higher dimensional space being a multiple of a dimensional space associated with the trained classification model;
generating a first similarity score for the first projected token vector based on comparisons of the first projected token vector to each of a plurality of class representations, each class representation corresponding to one of the plurality of predefined classes, and the first projected token vector being associated with each of the plurality of predefined classes for which a similarity distance value meets a predetermined threshold;
generating a ranked list of the tokens, wherein the generating of the ranked list includes ranking the first token among others of the tokens based on the first similarity score; and displaying the ranked list of the tokens.