System and method for determining the characteristics of human personality and providing real-time recommendations
The present disclosure relates to systems, methods, and non-transitory computer-readable media for identifying a personality of a human subject based on correlations between personality traits obtained from the subject's physical features, which may include a movement pattern of the subject, such as the subject's gait. Embodiments in accordance with the present disclosure are further capable of providing a recommendation to the subject for a product or service based on the identified personality of the subject.
1. A system for identifying a personality of a human subject comprising:
one or more hardware processors; and
a computer-readable medium storing instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
receiving visual data of the human subject from one or more hardware sensors;
validating the visual data using a facial recognition algorithm, wherein the validation of the visual data comprises: after determining that the visual data comprises an image of a distorted anatomical feature, correcting the distorted anatomical feature in the visual data;
detecting at least one anatomical feature of the human subject from the validated visual data, wherein detecting the at least one anatomical feature of the human subject comprises:
locating a physical region corresponding to each of the at least one anatomical feature in the validated visual data,
extracting a geometrical representation of each of the at least one anatomical feature from the physical region, and
associating a first personality factor with each of the at least one anatomical feature, wherein the first personality factor is determined based on the geometrical representation of the anatomical feature; and
determining the personality of the human subject based on the first personality factor associated with each of the detected at least one anatomical feature using a hidden Markov model algorithm, and a plurality of hidden Markov models each of which corresponding to a plurality of personalities by:
determining a hidden Markov model from the plurality of hidden Markov models that maximizes a probability of observing a sequence of first personality factors associated with the detected anatomical features, and
determining the personality from a plurality of personalities corresponding to the determined hidden Markov model by determining a centroid of a N-dimension region corresponding to the plurality of personalities corresponding to the determined hidden Markov model.
2. The system according to claim 1 , wherein the operation of validating the visual data further comprises determining a gender of the human subject based on a support vector machine algorithm.
3. The system according to claim 2 , wherein the medium stores further instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
determining the plurality of hidden Markov models based on the gender of the human subject.
4. The system according to claim 1 , wherein the medium stores further instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
extracting at least one movement pattern of the human subject from the visual data, and
associating a second personality factor with each of the extracted at least one movement pattern,
wherein determining the personality of the human subject comprises determining the personality of the human subject based on the first personality factor associated with each of the detected at least one anatomical feature and the second personality factor associated with each of the extracted at least one movement pattern using a hidden Markov model algorithm and a plurality of hidden Markov models corresponding to a plurality of personalities by:
determining a hidden Markov model from the plurality of hidden Markov models that maximizes a probability of observing a sequence of first personality factors associated with the detected anatomical features and of second personality factors associated with the extracted movement patterns.
5. The system accordingly to claim 4 , wherein the at least one movement pattern includes a gait of the human subject.
6. The system according to claim 4 , wherein the second personality factor associated with each of the extracted at least one movement pattern is determined based on a hidden Markov model algorithm.
7. The system according to claim 4 , wherein the first personality factor and the second personality factor comprise one or more weighted personality traits, and wherein the personality of the human subject is determined based on correlations between the personality factors associated with each of the detected at least one anatomical feature and each of the extracted at least one movement pattern, the correlations being obtained from the one or more weighted personality traits.
8. The system according to claim 1 , wherein the medium stores further instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
providing a real-time recommendation for a product to the human subject based on the determined personality of the human subject.
9. The system according to claim 8 , wherein the medium stores further instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
receiving feedback via a computing device associated with the human subject; and
improving the real-time recommendation based on at least a machine-learning algorithm and the received feedback.
10. The system according to claim 9 , wherein the received feedback comprises an amount of time the human subject observed the recommendation for the product.
11. The system according to claim 1 , wherein the correction of the distorted anatomical feature in the visual data comprises:
extracting a first field corresponding to one or more peaks in image intensity of the distorted anatomical feature;
extracting a second field corresponding to one or more valleys in image intensity of the distorted anatomical feature;
extracting a third field corresponding to one or more edges of image intensity of the distorted anatomical feature; and
correcting the distorted anatomical feature based on an image potential function and the first, second, and third fields.
12. A non-transitory computer-readable medium storing instructions for identifying a personality of a human subject that, when executed by one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
receiving visual data of the human subject from one or more hardware sensors;
validating the visual data using a facial recognition algorithm, wherein the validation of the visual data comprises: after determining that the visual data comprises an image of a distorted anatomical feature, correcting the distorted anatomical feature in the visual data;
detecting at least one anatomical feature of the human subject from the validated visual data, wherein detecting the at least one anatomical feature of the human subject comprises:
locating a physical region corresponding to each of the at least one anatomical feature in the validated visual data,
extracting a geometrical representation of each of the at least one anatomical feature from the physical region, and
associating a first personality factor with each of the at least one anatomical feature, wherein the first personality factor is determined using the geometrical representation of the anatomical feature; and
determining the personality of the human subject based on the first personality factor associated with each of the detected at least one anatomical feature using a hidden Markov model algorithm, and a plurality of hidden Markov models each of which corresponding to a plurality of personalities by:
determining a hidden Markov model from the plurality of hidden Markov models that maximizes a probability of observing a sequence of first personality factors associated with the detected anatomical features, and
determining the personality from a plurality of personalities corresponding to the determined hidden Markov model by determining a centroid of a N-dimension region corresponding to the plurality of personalities corresponding to the determined hidden Markov model.
13. The non-transitory computer-readable medium according to claim 12 , wherein the operation of validating the visual data further comprises determining a gender of the human subject based on a support vector machine algorithm.
14. The non-transitory computer-readable medium according to claim 12 , wherein the medium stores further instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
extracting at least one movement pattern of the human subject from the visual data, and
associating a second personality factor with each of the extracted at least one movement pattern, and
wherein determining the personality of the human subject comprises determining the personality of the human subject based on the first personality factor associated with each detected at least one anatomical feature and the second personality factor associated with each of the extracted at least one movement pattern using a hidden Markov model algorithm and a plurality of hidden Markov models corresponding to a plurality of personalities by:
determining a hidden Markov model from the plurality of hidden Markov models that maximizes a probability of observing a sequence of first personality factors associated with the detected anatomical features and of second personality factors associated with the extracted movement patterns.
15. The non-transitory computer-readable medium according to claim 14 , wherein the at least one movement pattern includes a gait of the human subject.
16. The non-transitory computer-readable medium according to claim 14 , wherein the second personality factor associated with each of the extracted at least one movement pattern is determined based on a hidden Markov model algorithm.
17. The non-transitory computer-readable medium according to claim 14 , wherein the first personality factor and the second personality factor comprise one or more weighted personality traits, and wherein the personality of the human subject is determined based on correlations between the personality factors associated with each of the detected at least one anatomical feature and each of the extracted at least one movement pattern, the correlations being obtained from the one or more weighted personality traits.
18. The non-transitory computer-readable medium according to claim 12 , wherein the medium stores further instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
providing a real-time recommendation for a product to the human subject based on the determined personality of the human subject.
19. The non-transitory computer-readable medium according to claim 18 , wherein the medium stores further instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
receiving feedback via a computing device associated with the human subject; and
improving the real-time recommendation based on at least a machine-learning algorithm and the received feedback.
20. A method for identifying a personality of a human subject comprising:
receiving, using one or more hardware processors, visual data of the human subject from one or more hardware sensors;
validating, using one or more hardware processors, the visual data using a facial recognition algorithm, wherein the validation of the visual data comprises: after determining that the visual data comprises an image of a distorted anatomical feature, correcting the distorted anatomical feature in the visual data;
detecting, using one or more hardware processors, at least one anatomical feature of the human subject from the validated visual data, wherein detecting the at least one anatomical feature of the human subject comprises:
locating a physical region corresponding to each of the at least one anatomical feature in the validated visual data,
extracting a geometrical representation of each of the at least one anatomical feature from the physical region, and
associating a first personality factor with each of the at least one anatomical feature, wherein the first personality factor is determined using the geometrical representation of the anatomical feature; and
determining, using one or more hardware processors, the personality of the human subject based on the first personality factor associated with each of the detected at least one anatomical feature using a hidden Markov model algorithm, and a plurality of hidden Markov models each of which corresponding to a plurality of personalities by:
determining a hidden Markov model from the plurality of hidden Markov models that maximizes a probability of observing a sequence of first personality factors associated with the detected anatomical features, and
determining the personality from a plurality of personalities corresponding to the determined hidden Markov model by determining a centroid of a N-dimension region corresponding to the plurality of personalities corresponding to the determined hidden Markov model.
21. The method according to claim 20 , wherein validating, using one or more hardware processors, the visual data further comprises determining a gender of the human subject based on a support vector machine algorithm.
22. The method according to claim 20 , further comprising:
extracting, using one or more hardware processors, at least one movement pattern of the human subject from the visual data, and
associating, using one or more hardware processors, a second personality factor with each of the extracted at least one movement pattern, and
wherein determining the personality of the human subject comprises determining the personality of the human subject based on the first personality factor associated with each of the detected at least one anatomical feature and the second personality factor associated with each of the extracted at least one movement pattern using a hidden Markov model algorithm and a plurality of hidden Markov models corresponding to a plurality of personalities by:
determining a hidden Markov model from the plurality of hidden Markov models that maximizes a probability of observing a sequence of first personality factors associated with the detected anatomical features and of second personality factors associated with the extracted movement patterns.
23. The method according to claim 22 , wherein the at least one movement pattern includes a gait of the human subject.
24. The method according to claim 22 , wherein the second personality factor associated with each of the extracted at least one movement pattern is determined based on a hidden Markov model algorithm.
25. The method according to claim 22 , wherein the first personality factor and the second personality factor comprise one or more weighted personality traits, and wherein the personality of the human subject is determined based on correlations between the personality factors associated with each of the detected at least one anatomical feature and each of the extracted at least one movement pattern, the correlations being obtained from the one or more weighted personality traits.
26. The method according to claim 20 , wherein the operations further comprise providing, using one or more hardware processors, a real-time recommendation for a product to the human subject based on the determined personality of the human subject.
27. The method according to claim 26 , further comprising:
receiving, using one or more hardware processors, feedback via a computing device associated with the human subject; and
improving, using one or more hardware processors, the real-time recommendation based on at least a machine-learning algorithm and the received feedback.