Tools and methods for determining semantic relationship indexes

• US 8,478,702 B1
• Filed: 09/17/2012
• Issued: 07/02/2013
• Est. Priority Date: 02/08/2012
• Status: Active Grant

• Current Assignee: Gatsby Technologies, LLC
• Original Assignee: Adam Treiser
• Inventors: Treiser, Adam
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): COUGHLAN, PETER D
• Application Number:
  US13/621,717
• Time in Patent Office: 288 Days
• Field of Search: None
• US Class Current: 706/12
• CPC Class Codes: G06F 16/288 Entity relationship models
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Systems, apparatuses, and methods for determining an individual'"'"'s sentimental baseline, based on a plurality of data items and characteristics. The data items may include objective and quantitative data, as well as subjective and qualitative data. The system, apparatus, or method may obtain a number of relationships between information atoms, identify sentiments associated with the relationships, and calculate sentimental baselines for those relationships. Differences from any baseline may also be calculated, to determine true changes in sentiment. Relationships between those differences and other data or relationships may also be calculated, to determine how a change in sentiment is related to other changes in behavior. For example, relationships between a particular difference and changes in any metric, sub-metric, group of characteristics, data item, data source, characteristic, sentiment, or groups thereof may be determined. These relationships may also be used to predict future behavior or sentiment.

Claims

1. A method of identifying baselines for relationships, comprising:

obtaining, at a computer, a first set of relationships between information atoms, wherein at least one of the information atoms comprises a descriptor of an individual;

identifying baselines for the first set of relationships;

obtaining a second set of relationships between the information atoms;

determining a set of differences between the baselines and the relationships in the second set of relationships;

identifying a third set of relationships between the set of differences and the descriptor; and

associating the differences with the descriptor, based on the third set of relationships.

2. The method of claim 1, wherein the baselines comprise first baselines and the set of differences comprise comprises a first set of differences, the method comprising:

identifying a first set of sentiments for the first set of relationships;

identifying second baselines for the first set of sentiments;

identifying a second set of sentiments for the second set of relationships; and

determining a second set of differences between the second baselines and the second set of sentiments.

3. The method of claim 2, wherein the information atoms further comprise statements made by an individual, or topics to which the statements pertain.

4. The method of claim 2, wherein the information atoms comprise a first set of information atoms, the method comprising:

predicting a second set of information atoms related to the descriptor, based on the differences and third set of relationships; and

outputting an information atom from the second set of information atoms.

5. The method of claim 4, wherein the set of differences comprises a first set of differences, the method comprising:

obtaining a third set of information atoms;

computing a fourth set of relationships between the second and third sets of information atoms;

predicting a second set of differences related to the descriptor, based on the third and fourth sets of relationships; and

outputting a difference from the second set of differences.

6. The method of claim 5, comprising:

computing a fifth set of relationships between the first set of differences and the second set of differences;

predicting a fourth set of information atoms related to the descriptor, based on the fifth set of relationships; and

outputting an information atom from the fourth set of information atoms.

7. The method of claim 5, wherein outputting a difference from the second set of differences comprises:

identifying a third set of sentiments, based on the second set of differences; and

outputting a sentiment from the third set of sentiments.

8. The method of claim 5, wherein the descriptor comprises a first descriptor, and the information atoms further comprise a group of one or more second descriptors, the method further comprising:

obtaining fifth relationships between the first descriptor and the group of second descriptors; and

wherein predicting a second set of differences comprises predicting a second set of differences for the first descriptor, based on the fifth set of relationships and the third and fourth sets of relationships associated with the group of second descriptors.

9. The method of claim 1, comprising:

storing the baselines;

recognizing a pattern in the differences; and

identifying a new baseline based on the pattern.

10. The method of claim 1, wherein at least one of the information atoms comprises a metric, the method comprising;

determining a fourth set of relationships between the metric and information atoms other than the metric;

identifying the differences related to the metric and the descriptor, based on the fourth set of relationships;

determining a result based on the identified differences related to the metric; and

outputting the result.

11. The method of claim 1, comprising;

calculating a score for the descriptor, based on the differences and the third set of relationships; and

outputting the score.

12. The method of claim 1, comprising;

outputting one or more of the differences associated with the descriptor, based on the third set of relationships.

13. The method of claim 1, wherein at least one relationship in the first, second or third set of relationships comprises a magnitude.

14. The method of claim 1, wherein at least one relationship in the first, second, or third set of relationships comprises a direction.

15. The method of claim 1, comprising:

recognizing a pattern in the differences; and

adjusting a relationship in the second set of relationships based on the pattern and the baselines.

16. The method of claim 1, comprising:

recognizing a pattern in the differences; and

adjusting an information atom based on the pattern and the baselines.

17. A non-transitory computer-readable storage medium encoded with instructions that, when executed on a processor, perform a method of identifying baselines for relationships, the method comprising:

obtaining, at a computer, a first set of relationships between information atoms, wherein at least one of the information atoms comprises a descriptor of an individual;

identifying baselines for the first set of relationships;

obtaining a second set of relationships between the information atoms;

determining a set of differences between the baselines and the relationships in the second set of relationships;

identifying a third set of relationships between the set differences and the descriptor; and

associating the differences with the descriptor, based on the third set of relationships.

18. The storage medium of claim 17, wherein the baselines comprise first baselines and the set of differences comprises a first set of differences, the method comprising:

identifying a first set of sentiments for the first set of relationships;

identifying second baselines for the first set of sentiments;

identifying a second se sentiments for the second set of relationships; and

determining a second set of differences between the second baselines and the second set of sentiments.

19. The storage medium of claim 17, wherein the method comprises:

storing the baselines;

recognizing a pattern in the differences; and

identifying a new baseline based on the pattern.

20. The storage medium of claim 17, wherein at least one of the information atoms comprises a metric, the method comprising:

determining a fourth set of relationships between the metric and information atoms other than the metric;

identifying the differences related to metric and the descriptor, based on the fourth set of relationships;

determining a result based on the identified differences related to the metric; and

outputting the result.

21. The storage medium of claim 17, wherein the wherein the method comprises:

recognizing a pattern in the differences; and

adjusting a relationship in the second set of relationships based on the pattern and the baselines.

22. The storage medium of claim 17, wherein the method comprises:

recognizing a pattern in the differences; and

adjusting an information atom based on the pattern and the baselines.

23. A characteristic-based server comprising:

a processor; and

memory containing instructions executed by the processor to perform a method comprising;

obtaining, at a computer, a first set of relationships between information atoms, wherein at least one of the information atoms comprises a descriptor of an individual;

identifying baselines for the first set of relationships;

obtaining a second set of relationships between the information atoms;

determining a set of differences between the baselines and the relationships in the second set of relationships;

identifying a third set of relationships between the set of differences and the descriptor; and

associating the differences with the descriptor, based on the third set of relationships.

24. The server of claim 23, wherein the baselines comprise first baselines, the set of differences comprises a first set of differences, and the memory contains instructions executed by the processor to perform a method further comprising:

identifying a first set of sentiments for the first set of relationships;

identifying second baselines for the first set of sentiments;

identifying a second set of sentiments for the second set of relationships; and

determining a second set of differences between the second baselines and the second set of sentiments.

25. The server of claim 23, the memory containing instructions executed by the processor to perform a method further comprising:

storing the baselines;

recognizing a pattern in the differences; and

identifying a new based on the pattern.

26. The server of claim 23, wherein at least one information atom comprises a metric, and the memory contains instructions executed by the processor to perform a method further comprising:

determining a fourth set of relationships between the metric and information atoms other than the metric;

identifying the differences related to the metric and the descriptor, based on the fourth set of relationships;

determining a result based on the identified differences related to the metric; and

outputting the result.

27. The server of claim 23, wherein the memory further comprises instructions operable to:

calculate a score for the descriptor, based on the differences and the third set of relationships; and

output the score for the descriptor.

28. The server of claim 23, wherein the memory further comprises instructions operable to output a difference for the descriptor, based on the third set of relationships.

29. The system of claim 23, wherein the memory further comprises instructions operable to:

recognize a pattern in the differences; and

adjust a relationship in the second set of relationships based on the pattern and the baselines.

30. The system of claim 23, wherein the memory further comprises instructions operable to:

recognize a pattern in the differences; and

adjust an information atom based on the pattern and the baselines.