Systems for crowd typing by hierarchy of influence

• US 9,916,370 B1
• Filed: 01/23/2015
• Issued: 03/13/2018
• Est. Priority Date: 01/23/2014
• Status: Active Grant

• Current Assignee: Element Data, Inc.
• Original Assignee: Element Data, Inc.
• Inventors: Cooper, Joshua N., Harry, Keith A.
• Primary Examiner(s): Le, Hung
• Application Number:
  US14/604,163
• Time in Patent Office: 1,145 Days
• Field of Search: None
• CPC Class Codes: G06F 16/26 Visual data mining; Browsing structured data
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

Systems and methods for providing hierarchy scores are described. Generally, influence scores for authors of a crowd may be determined based on emotional scores and echoing of time series data strings. One or more regression lines may be determined based on the influence scores to provide a raw hierarchy score and/or a central hierarchy score. Analysis and/or comparisons of the hierarchy scores may be used to classify the crowd type and output an influential score report.

Claims

1. A computerized method implemented by at least one processor executing computer executable code having instructions that when executed by the at least one processor cause the at least one processor to perform steps comprising:

analyzing a plurality of time series data strings provided by two or more authors, each time series data string having at least one pre-determined word to generate a first data set including a plurality of emotional scores with respect to the pre-determined word within the data strings;

determining influence scores for each author based on the emotional scores and echoing of the time series data strings, the authors forming a crowd;

determining a first regression line based on the influence scores to provide a raw hierarchy score using slope of the first regression line;

determining a second regression line to provide a central hierarchy score using slope of the second regression line;

comparing the raw hierarchy score and the central hierarchy score to classify crowd type of the two or more authors; and

,

outputting an influential score report indicating the crowd type based on comparison of the raw hierarchy score and the central hierarchy score.

2. The method of claim 1, wherein the raw hierarchy score is provided using a method of least squares.

3. The method of claim 1, wherein the central hierarchy score is provided using a Theil-Sen method.

4. The method of claim 1, wherein the time series data strings are provided to the processor via an external source.

5. The method of claim 4, wherein the external source is a social media server.

6. The method of claim 1, wherein a transfer entropy technique is used to determine echoing of the data strings.

7. The method of claim 1, wherein the influential score report includes a ranking of the authors by influential scores.

8. The method of claim 1, further comprising mapping of the influential scores on a log-linear plot.

9. The method of claim 1, further comprising the step of determining at least one of heterogeneity and homogeneity of the crowd.

10. The method of claim 1, wherein the influential score report includes predefined parameters for a marketing plan based on the crowd type.

11. The method of claim 1, wherein each time series data string includes at least one pre-determined word, and wherein the method further comprises generating data sets including a plurality of emotional scores with respect to the at least one pre-determined word within the time series data strings, and the plurality of emotional scores are used to determine types of emotions each author is exhibiting in each time series data string.

12. A system, comprising:

a host system having a microprocessor; and

,

a computer readable medium storing a set of instructions that when executed by the microprocessor, cause the microprocessor to;

obtain and record in a first database, scores indicative of emotional scoring;

obtain at least two textual sources of information having time information indicative of the times when the textual sources of information were created;

analyze and determine an emotional score for the textual sources of information using the first database;

analyze and determine an influential score for authors of the textual sources of information, the authors forming a crowd;

map the influential scores for authors of the textual sources of information;

analyze and determine crowd type of the authors of the textual source of information using the map of the influential scores; and

,

generate and pass digital data indicative of an influential score report containing the crowd type to at least one user device in communication with the host system.

13. The system of claim 12, wherein the map is a log-linear mapping of the influential scores.

14. The system of claim 13, wherein determining crowd type of the authors of the textual source of information using the map of the influential scores includes fitting at least one regression line to the influential scores in the map.

15. The system of claim 14, wherein the set of instructions that when executed by the microprocessor, cause the microprocessor to determine a first slope of a first regression line and to interpret the first slope to provide a raw hierarchy score for the crowd type.

16. The system of claim 15, wherein the set of instructions that when executed by the microprocessor, cause the microprocessor to determine a second slope of a second regression line and interpret the second slope to provide a central hierarchy score for the crowd type.

17. The system of claim 16, wherein the raw hierarchy score and the central hierarchy score are compared to determine heterogeneity and homogeneity of crowd.