Cognitive neuro-linguistic behavior recognition system for multi-sensor data fusion
View Patent ↗Embodiments presented herein describe techniques for generating a linguistic model of input data obtained from a data source (e.g., a video camera). According to one embodiment of the present disclosure, a sequence of symbols is generated based on an ordered stream of normalized vectors generated from the input data. A dictionary of words is generated from combinations of the ordered sequence of symbols based on a frequency at which combinations of symbols appear in the ordered sequence of symbols. A plurality of phrases is generated based an ordered sequence of words from the dictionary observed in the ordered sequence of symbols based on a frequency by which combinations of words in ordered sequence of words appear relative to one another.
1. A method for generating a neuro-linguistic model of input data obtained from one or more data sources, the method comprising:
generating, via a processor, a sequence of symbols based on an ordered stream of normalized vectors generated from the input data;
generating a dictionary of words from combinations of the ordered sequence of symbols based on a frequency at which combinations of symbols appear in the ordered sequence of symbols; and
generating a plurality of phrases based an ordered sequence of words from the dictionary observed in the ordered sequence of symbols based on a frequency by which combinations of words in ordered sequence of words appear relative to one another.
2. The method of claim 1 , further comprising, determining, for at least one symbol, word, or phrase, an unusualness score based on a frequency of the at least one symbol, word, or phrase, as generated from the ordered stream of normalized vectors generated from the input data.
3. The method of claim 1 , further comprising, identifying anomalous behavior present in subsequent symbols, words, and phrases generated from the ordered stream of normalized vectors generated from the input data, based on a comparison with the sequence of symbols, generated dictionary and generated plurality of words.
4. The method of claim 1 , wherein words included in the dictionary include a set of words having a most statistical relevance relative to a complete set of words generated from combinations of the ordered sequence of symbols.
5. The method of claim 1 , wherein each normalized vector represents an element of the input data in a range of 0 to 1, inclusive.
6. The method of claim 1 , wherein one of the data sources is a sequence of video frames.
7. The method of claim 1 , wherein generating the dictionary comprises building a dictionary of words up to a maximum length of symbols.
8. A non-transitory computer-readable storage medium storing instructions, which, when executed on a processor, performs an operation for generating a linguistic model of input data obtained from a data source, the operation comprising:
generating, via a processor, a sequence of symbols based on an ordered stream of normalized vectors generated from the input data;
generating a dictionary of words from combinations of the ordered sequence of symbols based on a frequency at which combinations of symbols appear in the ordered sequence of symbols; and
generating a plurality of phrases based an ordered sequence of words from the dictionary observed in the ordered sequence of symbols based on a frequency by which combinations of words in ordered sequence of words appear relative to one another.
9. The non-transitory computer-readable storage medium of claim 8 , wherein the operation further comprises, determining, for at least one symbol, word, or phrase, an unusualness score based on a frequency of the at least one symbol, word, or phrase, as generated from the ordered stream of normalized vectors generated from the input data.
10. The non-transitory computer-readable storage medium of claim 8 , wherein the operation further comprises, identifying anomalous behavior present in subsequent symbols, words, and phrases generated from the ordered stream of normalized vectors generated from the input data, based on a comparison with the sequence of symbols, generated dictionary and generated plurality of words.
11. The non-transitory computer-readable storage medium of claim 8 , wherein the operation further comprises, wherein words included in the dictionary include a set of words having a most statistical relevance relative to a complete set of words generated from combinations of the ordered sequence of symbols.
12. The non-transitory computer-readable storage medium of claim 8 , wherein each normalized vector represents an element of the input data in a range of 0 to 1, inclusive.
13. The non-transitory computer-readable storage medium of claim 8 , wherein one of the data sources is a sequence of video frames.
14. The non-transitory computer-readable storage medium of claim 8 , wherein generating the dictionary comprises building a dictionary of words up to a maximum length of symbols.
15. A system, comprising:
a processor; and
a memory storing one or more application programs configured to perform an operation for generating a linguistic model of input data obtained from a data source, the operation comprising:
generating a sequence of symbols based on an ordered stream of normalized vectors generated from the input data,
generating a dictionary of words from combinations of the ordered sequence of symbols based on a frequency at which combinations of symbols appear in the ordered sequence of symbols, and
generating a plurality of phrases based an ordered sequence of words from the dictionary observed in the ordered sequence of symbols based on a frequency by which combinations of words in ordered sequence of words appear relative to one another.
16. The system of claim 15 , wherein the operation further comprises, determining, for at least one symbol, word, or phrase, an unusualness score based on a frequency of the at least one symbol, word, or phrase, as generated from the ordered stream of normalized vectors generated from the input data.
17. The system of claim 15 , wherein the operation further comprises, identifying anomalous behavior present in subsequent symbols, words, and phrases generated from the ordered stream of normalized vectors generated from the input data, based on a comparison with the sequence of symbols, generated dictionary and generated plurality of words.
18. The system of claim 15 , wherein the operation further comprises, wherein words included in the dictionary include a set of words having a most statistical relevance relative to a complete set of words generated from combinations of the ordered sequence of symbols.
19. The system of claim 15 , wherein each normalized vector represents an element of the input data in a range of 0 to 1, inclusive.
20. The system of claim 15 , wherein one of the data sources is a sequence of video frames.
21. The system of claim 15 , wherein generating the dictionary comprises building a dictionary of words up to a maximum length of symbols.