System for providing intelligent part of speech processing of complex natural language

• US 10,599,767 B1
• Filed: 05/31/2018
• Issued: 03/24/2020
• Est. Priority Date: 05/31/2018
• Status: Active Grant

• Current Assignee: UKG, Inc. (Seahawk Acquisition Corp.)
• Original Assignee: Ultimate Software Group Incorporated (Kronos (USA), Inc.)
• Inventors: Mattera, Michael L., Maliani, John A., Vanderwall, Robert L., Santiago, Dionny, Muras, Brian R., Briggs, Keith A., Adamo, David, King, Tariq
• Primary Examiner(s): Pullias, Jesse S
• Application Number:
  US15/994,869
• Time in Patent Office: 663 Days
• Field of Search: 704 1, 704 9, 704 10, 704257
• CPC Class Codes: G06F 40/242 Dictionaries
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates

Abstract

A system for providing intelligent part of speech processing of complex natural language is disclosed. The system identifies a multiword concept from an input and replaces the multiword concept with a token to be tagged as a desired part of speech. The system passes the modified text including the token to a part-of-speech tagger to tag each word in the text with the appropriate part-of-speech. The system may replace the token with the original text that the token was utilized to replace so that the original intent of the text is evident. The system may analyze the tagged text to generate analyses and interpretations associated with the input. When multiple multiword concepts are identified, the system may evaluate them by computing scores for each of the multiword concepts that may be replaced with tokens, for each of the modified texts including the tokens, or for any interpretations and analyses thereof.

Claims

1. A system, comprising:

a memory that stores instructions; and

a processor that executes the instructions to perform operations, the operations comprising;

identifying a set of words from among a plurality of words in an input, wherein the set of words describe a concept, idea, title, grouping, or a combination thereof;

replacing the set of words identified in the input with a token correlating to the set of words to generate a first modified input, wherein the token is constructed by the system to appear as a selected part of speech to a part of speech tagger;

tagging, by utilizing the part of speech tagger, parts of speech for each word and the token in the first modified input to generate tagged part of speech text for the first modified input, wherein the token is tagged by the part of speech tagger according to the selected part of speech; and

replacing, in the tagged part of speech text for the first modified input, the token with the set of words identified in the input to generate a second modified input.

2. The system of claim 1, wherein the operations further comprise selecting, by utilizing a segmenter, sentences, paragraphs, phrases, groupings of the words, a portion of the words, a document, or a combination thereof, from the input to separate the input into a plurality of portions of text.

3. The system of claim 2, wherein the operations further comprise analyzing the second modified input to generate analyses and interpretations associated with words in the plurality of words, the sentences, the paragraphs, the phrases, the groupings of the words, the portion of the words, the document, the plurality of the portions of the text, or a combination thereof.

4. The system of claim 1, wherein the operations further comprise identifying the set of words from among the plurality of words by detecting sets of capitalized words in the plurality of words, sets of nouns in the plurality of words, sets of compound nouns in the plurality of words, words adjacent to the set of capitalized words, words adjacent to the sets of nouns, words adjacent to the sets of compound nouns, any type of word, or any combination thereof.

5. The system of claim 1, wherein the operations further comprise identifying the set of words from the among the plurality of words by comparing the input to parameters of sources comprising documents, software manuals, hardware manuals, software requirements, hardware requirements, agile stories, software test cases, hardware test cases, release notes, dictionaries, domain specific data, glossaries, speaker tagged data, author tagged data, an application under evaluation by the system, remote internet software, cloud software, software in an immediate computing environment of the system, agglomerated models, or any combination thereof.

6. The system of claim 5, wherein the operations further comprise scoring the set of words identified in the input based on a number of words in the set of words, a descriptiveness of words in the set of words, a number of times the set of words have been replaced by one or more tokens, a genericity of the set of words, any scoring criteria, or a combination thereof.

7. The system of claim 1, wherein the operations further comprise identifying an emoji, a GIF, an emoticon, a symbol, video content, audio content, media content, or a combination thereof, in the input, and wherein the operations further comprise replacing the emoji, the GIF, the emoticon, the symbol, the video content, the audio content, the media content, or a combination thereof, with the token or a different token.

8. The system of claim 1, wherein the operations further comprise altering an arrangement of the words in the plurality of words or altering a word in the plurality of words prior to tagging the parts of speech for each word.

9. The system of claim 8, wherein the operations further comprise reverting, after tagging the parts of speech, the altered arrangement of the words in the plurality of words back to the arrangement of the words as arranged prior to the altering or the altered word in the plurality of words back to the word as the word existed prior to the alerting.

10. The system of claim 1, wherein the operations further comprise detecting a missing concept, word, idea, or a combination thereof, from the input, and wherein the operations further comprise replacing the missing concept, word, idea, or a combination thereof, from the input with a different token or the token.

11. The system of claim 1, wherein the operations further comprise detecting an error in the plurality of words, an error in a grouping of words in the plurality of words, or a combination thereof, and wherein the operations further comprise replacing the error in the plurality of words, the error in the grouping of words in the plurality of words, or a combination thereof, with the token or a different token.

12. The system of claim 1, wherein the operations further comprise recursively proceeding through the operations to identify portions of text in the input having a confidence below a threshold, wherein the identified portions represent words or phrases with missing content, a grammatical error, or a combination thereof, to be replaced with the token or a different token.

13. The system of claim 1, wherein the operations further comprise identifying a plurality of candidate sets of words of the plurality of words to potentially be replaced by tokens.

14. The system of claim 13, wherein the operations further comprise scoring the plurality of candidate sets of words of the plurality of words, and wherein the operations further comprise scoring analyses, interpretations, outputs, or a combination thereof, generated based on analyzing the plurality of candidate set of words as normalized numbers, confidences, or a combination thereof.

15. The system of claim 14, wherein the operations further comprise selecting a highest scoring candidate set of words of the plurality of candidate sets of words to be replaced by a different token.

16. The system of claim 1, wherein the operations further comprise analyzing the second modified input to generate analyses and interpretations associated with words in the plurality of words by identifying concepts associated with the words in the plurality of words and relationships between the words in the plurality of words.

17. The system of claim 1, wherein the operations further comprise adding a character or a plurality of characters to a token name of the token to indicate that the token is a verb, noun, adverb, adjective, or other part of speech, to the part of speech tagger, or wherein the operations further comprise adjusting a position of the token in the first modified input to indicate that the token is a verb, noun, adverb, adjective, or other part of speech.

18. The system of claim 1, wherein the operations further comprise scoring or weighing an accuracy of the tagging or analyses and interpretations generated based on analyzing the second modified input, and wherein the scoring or weighing of the accuracy of the tagging or the analyses and interpretations is based on feedback from a hypothesis tester, a test generator, a discovery module, a human evaluator, any criteria, or a combination thereof.

19. The system of claim 18, wherein the scoring or weighing is utilized to adjust a learning engine involved in identifying the set of words from among the plurality of words in the input.

20. The system of claim 1, wherein the operations further comprise receiving the input from a source, wherein the source comprises a collection or compendium of documents, a document, a paragraph, a sentence, a phrase, an image, a video, audio, or a combination thereof.

21. A method, comprising:

receiving, at a system, an input from a source;

identifying a set of words from among a plurality of words in the input that correspond to a parameter, product, component, a method name, a user profile name, a role profile name, a window name, a field name, a label name, a widget name, or any combination thereof;

replacing the set of words identified in the input with a token correlating to the set of words to generate a first modified input, wherein the token is constructed to be characterized as a selected part of speech by a part of speech tagger;

tagging, by utilizing the part of speech tagger and by utilizing instructions from a memory that are executed by a processor of the system, parts of speech for each word and the token in the first modified input to generate tagged part of speech text for the first modified input;

replacing, in the tagged part of speech text for the first modified input, the token with the set of words identified in the input to generate a second modified input; and

analyzing the second modified input to generate analyses and interpretations associated with words in the plurality of words.

22. The method of claim 21, further comprising validating whether the token tagged by the part of speech tagger is tagged according to the selected part of speech.

23. The method of claim 22, further comprising modifying, if the token tagged by the part of speech tagger is not tagged according to the selected part of speech, the first modified input to adjust a location of the token within the first modified input until the token is tagged by the part of speech tagger according to the selected part of speech.

24. The method of claim 21, further comprising updating an agglomerated model based on the analyses and interpretations generated by the analyzing.

25. The method of claim 21, further comprising facilitating understanding of an intent of the input without requiring training for the system.

26. The method of claim 21, further comprising evaluating a plurality of sets of words identified from among the plurality of words in the input by computing confidence values for each of the plurality of sets of words identified, computing confidence values of the first and second modified inputs, computing confidence values of the analyses and interpretations, or a combination thereof.

27. The method of claim 26, further comprising selecting a set of words from the plurality of sets of words having a highest confidence value.

28. The method of claim 21, further comprising classifying the first modified input based on a sentiment of sentences in the first modified input, an intent of sentences in the first modified input, or a combination thereof.

29. The method of claim 21, further comprising analyzing the first modified input including the token to generate supplementary analyzes and interpretations.

30. A non-transitory computer-readable device comprising instructions, which when loaded and executed by a processor, cause the processor to perform operations comprising:

receiving, at a system, an input from a source;

identifying a set of words from among a plurality of words in the input, wherein the set of words correspond to a parameter;

replacing the set of words identified in the input with a token correlating to the set of words to generate a first modified input, wherein the token is constructed to be characterized as a selected part of speech by a part of speech tagger;

tagging, by utilizing the part of speech tagger and by utilizing instructions from a memory that are executed by a processor of the system, parts of speech for each word and the token in the first modified input to generate tagged part of speech text for the first modified input;

replacing, in the tagged part of speech text for the first modified input, the token with the set of words identified in the input to generate a second modified input;

analyzing the second modified input to generate analyses and interpretations associated with words in the plurality of words; and

computing scores for the analyses and interpretations associated with the words in the plurality of words.

31. The non-transitory computer-readable device of claim 30, wherein the operations further comprise training or improving a model, a parameter, weight, a dictionary, a threshold, a confidence, or a filter associated with the generating of a future token, wherein the training or improving is based on feedback, test results, an internal source, an external source, the application under evaluation, hypothesis testing, hypothesis testing results, an analyses, a confirmation or rejection of the token, the token, another token related to the token, another token, any other source, or combination thereof.