SYSTEM AND METHODS FOR DETERMINING SENTIMENT BASED ON CONTEXT

• US 20140278365A1
• Filed: 03/17/2013
• Published: 09/18/2014
• Est. Priority Date: 03/12/2013
• Status: Active Grant

• Current Assignee: Linfo IP LLC (Pueblo Nuevo LLC)
• Original Assignee: Chizhong Zhang, Guangsheng Zhang
• Inventors: Zhang, Guangsheng, Zhang, Chizhong
• Granted Patent: US 9,697,196 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

System and methods are disclosed for determining the connotation or sentiment type of a text unit comprising multiple terms and with a grammatical structure, such as subject+verb, verb+object, adjective+noun, noun+noun, noun+preposition+noun. The connotation or sentiment type of the text unit is determined by applying context rules where the context of the grammatical structure may change the inherent or default connotations of individual terms in the text unit. The methods provide a solution to the challenge of correctly or accurately determining the sentiment type of various linguistic structures under different context, and to the simplistic approach of using the inherent or default connotation of individual terms for the linguistic structure containing such terms.

Claims

1. A method for determining the connotation of a multi-term text unit, comprising:

identifying a first term and a second term in a text unit comprising two or more terms;

identifying a grammatical attribute associated with the terms in the text unit, where in the grammatical attributes includes at least the grammatical roles of a subject, an object, a head, a modifier, and parts of speech of a noun, a verb, a preposition, an adjective, and adverb;

defining a group of terms comprising a subset of terms in a language, or attaching a label to a subset of terms in a language contained in a dictionary, based on the behavior pattern of the terms in determining the connotation of a multi-term text unit containing one of the terms; and

determining the connotation of the text unit based on the group membership or the attached label of at least one of the terms in the text unit, and the grammatical attribute associated with at least one of the terms in the text unit, wherein the connotation of the text unit can be positive, negative, or neutral.

2. The method of claim 1, wherein the behavior pattern of a term includes the default connotation of the term and the semantic characteristics of the term.

3. The method of claim 2, wherein the semantic characteristics of the term includes a meaning of the term that generally signifies a process of increasing or decreasing, or a process of creating or removing, or the alike.

4. The method of claim 1, wherein the first term is a verb, and the second term is a noun, the method further comprising:

identifying a first group of terms comprising a subset of verbs in a language; and

determining the connotation or the sentiment type of the text unit based on whether the verb is a member of the first group, and based on the connotation of the verb or the connotation of the noun.

5. The method of claim 4, wherein the first term is a verb, and the second term is a noun that is an object of the verb.

6. The method of claim 4, wherein the first term is a verb, and the second term is a noun that is a subject of the verb.

7. The method of claim 4, further comprising:

identifying a first group of verbs containing the verb in the text unit, assigning a positive connotation to the text unit if the verb has a positive connotation and the noun also has a positive connotation, or if the verb has a negative connotation and the noun also has a negative connotation;

or

assigning a negative connotation to the text unit if the verb has a positive connotation and the noun has a negative connotation, or if the verb has a negative connotation and the noun has a positive connotation.

8. The method of claim 4, further comprising:

identifying a second group of verbs containing the verb in the text unit, assigning a positive connotation to the text unit if the verb has a positive connotation and the noun has a negative connotation, or if the verb has a negative connotation and the noun has a positive connotation;

or

assigning a negative connotation to the text unit if the verb has a positive connotation and the noun also has a positive connotation, or if the verb has a negative connotation and the noun also has a negative connotation.

9. The method of claim 4, further comprising:

identifying a third group of verbs containing the verb in the text unit, assigning a positive connotation to the text unit if the verb has a positive connotation;

or assigning a negative connotation to the text unit if the verb has a negative connotation.

10. The method of claim 1, wherein the first term is a noun and the second term is also a noun, the method further comprising:

identifying a group of terms comprising a subset of nouns in a language; and

determining the connotation or the sentiment type of the text unit based on whether the first noun is a member of the group or not, and based on the connotation of the first noun or the second noun.

11. The method of claim 10, wherein the first noun is derived from a verb, or is semantically related to an action.

12. The method of claim 10, further comprising:

identifying a first group of nouns containing the first noun in the text unit, assigning a positive connotation to the text unit if the first noun has a positive connotation and the second noun also has a positive connotation, or if the first noun has a negative connotation and the second noun also has a negative connotation;

or assigning a negative connotation to the text unit if the first noun has a positive connotation and the second noun has a negative connotation, or if the first noun has a negative connotation and the second noun has a positive connotation.

13. The method of claim 10, further comprising:

identifying a second group of nouns containing the first noun in the text unit, assigning a positive connotation to the text unit if the first noun has a positive connotation and the second noun has a negative connotation, or if the first noun has a negative connotation and the second noun has a positive connotation;

or assigning a negative connotation to the text unit if the first noun has a positive connotation and the second noun also has a positive connotation, or if the first noun has a negative connotation and the second noun also has a negative connotation.

14. The method of claim 10, further comprising:

identifying a third group of nouns containing the first noun in the text unit, assigning a positive connotation to the text unit if the first noun has a positive connotation;

or assigning a negative connotation to the text unit if the first noun has a negative connotation.

15. The method of claim 10, wherein the text unit further comprises a third term associated with the parts of speech of a preposition, wherein the preposition connects the first noun and the second noun, forming a noun+preposition+noun structure.

16. The method of claim 1, wherein the first term is an adjective, wherein the second term is a noun, wherein the text unit has an adjective+noun structure, the method further comprising:

identifying an adjective group comprising a subset of adjectives in a language, wherein each adjective in the group is associated with a positive or negative connotation; and

a noun group comprising a subset of nouns in a language; and

determining the connotation or the sentiment type of the text unit based on whether the adjective in the text unit is a member of the adjective group or not, and whether the noun is a member of the noun group or not.

17. The method of claim 16, further comprising:

identifying a first group of adjectives containing the adjective in the text unit;

identifying a first group of nouns, or a second group of nouns, or a third group of nouns;

assigning a positive connotation to the text unit if the adjective has a positive connotation and the noun is a member of the first group of nouns, or if the adjective has a negative connotation and the noun is a member of the second group of nouns;

or

assigning a negative connotation to the text unit if the adjective has a negative connotation and the noun is a member of the first group of nouns, or if the adjective has a positive connotation and the noun is a member of the second group of nouns;

or

assigning a neutral connotation to the text unit if the noun is a member of the third group of nouns.

18. The method of claim 16, wherein the text unit further comprises a third term having the parts of speech of a preposition, wherein the adjective and the noun in the text unit is connected by the preposition in the form of adjective+preposition+noun.

19. The method of claim 18, wherein the second term in the text unit is a verb, wherein the adjective and the verb in the text unit is connected by the preposition in the form of adjective+preposition+verb, the method further comprising:

identifying different verb groups that include subsets of verbs in a language, and determining the connotation of the text unit based on whether the adjective is a member of a particular adjective group, and whether the verb is a member of a particular verb group.

20. The method of claim 1, wherein the first term is a subject, and the second term is an adjective predicative, wherein the subject and the adjective is connected by a linking verb in the form of subject+be+adjective, the method further comprising:

identifying different noun groups that include subsets of nouns in a language, and different adjective groups that include subsets of adjectives in a language, and determining the connotation of the text unit based on whether the adjective is a member of a particular adjective group, and whether the subject noun is a member of a particular noun group.