Sentic neurons: expanding intention awareness

Patent metadata

Field	Value
Publication number	US-10846601-B1
Application number	US-201615272995-A
Country	US
Kind code	B1
Filing date	Sep 22, 2016
Priority date	Sep 22, 2015
Publication date	Nov 24, 2020
Grant date	Nov 24, 2020

Abstract

Embodiments of the present invention may provide techniques to create a new framework by this invention applying the theory of Sentic Computing. For example, in an embodiment of the present invention, a computer-implemented method for data analysis may comprise receiving input data representing circumstantial semantics, processing the received input data representing circumstantial semantics with Intention Awareness processing, receiving input data representing conceptual and affective information associated with objects and actors of the operating environment, processing the received input data representing conceptual and affective information associated with objects and actors of the operating environment with Sentic Computing, generating a mapping of the Intention Awareness processing data to a first multi-dimensional coordinate vector, generating a mapping of the Sentic Computing processed data to a second multi-dimensional coordinate vector, and generating output data by fusing the first vector and the second vector over time.

First claim

What is claimed is:

1. A method for predicting and understanding behavior based on data relating to objects and actors in an operating environment, the method implemented in a computer system comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, the method comprising: receiving input data including natural language, visual, auditory and textual data, and circumstantial semantics of the operating environment, wherein at least some of the data is received from behavior monitoring instrumentation; processing the received input data with Intention Awareness processing to find conceptual and affective information associated with the objects and actors of the operating environment, and to combine the conceptual and affective information with information relating to prior events to construct an event space from which human intentions can be inferred; processing the received input data with Sentic Computing processing by parsing sets of personality and sentiment data to derive mind state from the input data, determine mean sentiment value, and identify personality types based on metrics including extroversion and conscientiousness; generating a mapping of the Intention Awareness processed data to a first multi-dimensional coordinate vector; generating a mapping of the Sentic Computing processed data to a second multi-dimensional coordinate vector; and generating output data comprising a global multidimensional dynamic stream by fusing the first vector and the second vector over time, wherein the output data include a prediction of behavior of the objects and actors in the operating environment.

2. The method of claim 1 , wherein components of each coordinate's vector are calculated as a barycenter of each coordinate's vector's weighted points.

3. The method of claim 1 , wherein the visual data comprises facial expressions that are used to determine sentiments, and wherein tendencies over time are used to enhance personality analysis.

4. The method of claim 1 , wherein the textual data is processed by:
pre-processing the textual data;
extracting concepts from the pre-processed data;
inferring semantics associated with the extracted concepts; and
extracting Sentic data from the inferred semantics.

5. A computer program product for predicting and understanding behavior based on data relating to objects and actors in an operating environment, the computer program product comprising a non-transitory computer readable storage having program instructions embodied therewith, the program instructions executable by a computer, to cause the computer to perform a method comprising: receiving input data including natural language, visual, auditory and textual data, and circumstantial semantics of the operating environment, wherein at least some of the data is received from behavior monitoring instrumentation; processing the received input data with Intention Awareness processing to find conceptual and affective information associated with the objects and actors of the operating environment, and to combine the conceptual and affective information with information relating to prior events to construct an event space from which human intentions can be inferred; processing the received input data with Sentic Computing processing by parsing sets of personality and sentiment data to derive mind state from the input data, determine mean sentiment value, and identify personality types based on metrics including extroversion and conscientiousness; generating a mapping of the Intention Awareness processed data to a first multi-dimensional coordinate vector; generating a mapping of the Sentic Computing processed data to a second multi-dimensional coordinate vector; and generating output data comprising a global multidimensional dynamic stream by fusing the first vector and the second vector over time, wherein the output data include a prediction of behavior of the objects and actors in the operating environment.

6. The computer program product of claim 5 , wherein components of each coordinate's vector are calculated as a barycenter of each coordinate's vector's weighted points.

7. The computer program product of claim 5 , wherein the visual data comprises facial expressions that are used to determine sentiments, and wherein tendencies over time are used to enhance personality analysis.

8. The computer program product of claim 5 , wherein the textual data is processed by: pre-processing the textual data;
extracting concepts from the pre-processed data;
inferring semantics associated with the extracted concepts; and
extracting Sentic data from the inferred semantics.

9. A system for predicting and understanding behavior based on data relating to objects and actors in an operating environment, the system comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor to perform: receiving input data including natural language, visual, auditory and textual data, and circumstantial semantics of the operating environment, wherein at least some of the data is received from behavior monitoring instrumentation; processing the received input data with Intention Awareness processing to find conceptual and affective information associated with the objects and actors of the operating environment, and to combine the conceptual and affective information with information relating to prior events to construct an event space from which human intentions can be inferred; processing the received input data with Sentic Computing processing by parsing sets of personality and sentiment data to derive mind state from the input data, determine mean sentiment value, and identify personality types based on metrics including extroversion and conscientiousness: generating a mapping of the Intention Awareness processed data to a first multi-dimensional coordinate vector; generating a mapping of the Sentic Computing processed data to a second multi-dimensional coordinate vector; and generating output data comprising a global multidimensional dynamic stream by fusing the first vector and the second vector over time, wherein the output data include a prediction of behavior of the objects and actors in the operating environment.

10. The system of claim 9 , wherein components of each coordinate's vector are calculated as a barycenter of each coordinate's vector's weighted points.

11. The system of claim 9 , wherein the visual data comprises facial expressions that are used to determine sentiments, and wherein tendencies over time are used to enhance personality analysis.

12. The system of claim 9 , wherein the textual data is processed by:
pre-processing the textual data;
extracting concepts from the pre-processed data;
inferring semantics associated with the extracted concepts; and
extracting Sentic data from the inferred semantics.

Assignees

• Howard Newton
• Cambria Erik

Inventors

• Howard Newton
• Cambria Erik

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N5/04 Primary

  Inference or reasoning models · CPC title

• G06F8/33 Primary

  Intelligent editors · CPC title

• G06N5/02

  Knowledge representation; Symbolic representation · CPC title

• G06N3/08 Primary

  Learning methods · CPC title