Cognitive modeling apparatus for assessing values qualitatively across a multiple dimension terrain
The present design is directed to a system for measuring values qualitatively across a terrain including multiple dimensions using cognitive computing techniques, comprising a plurality of event reception components configured to operate on each event in a stream relevant to the terrain, the plurality of event reception components including a threshold application component configured to apply a threshold to each element in the stream, a terrain updater configured to update the terrain based on at least one event, an outlier analysis module configured to determine any outlier in the stream of events, a threshold violation predictor configured to predict threshold violations based on the stream of events, a time-ordered behavior evaluator configured to evaluate behavior based on the stream of events, and a graph updater to update a graph based on the stream of events.
1. A system for assessing values qualitatively across a terrain comprising multiple dimensions using cognitive computing techniques, comprising:
a plurality of event reception components configured to operate on each event in a stream relevant to the terrain, the plurality of event reception components comprising:
a threshold application component configured to apply a threshold to each element in the stream;
a terrain updater configured to update the terrain based on at least one event;
an outlier analysis module configured to determine any outlier in the stream of events, wherein the outlier analysis module determines whether at least one outlier in the stream of events represents a beginning of a different data pattern or a change to an existing data pattern;
a threshold violation predictor configured to predict threshold violations based on the stream of events;
a time-ordered behavior evaluator configured to evaluate behavior based on the stream of events; and
a graph updater to update a graph based on the stream of events;
wherein the system changes a rate of change threshold and an outlier boundary threshold upon the outlier analysis module determining the beginning of the different data pattern or the change to the existing data pattern.
2. The system of claim 1 , further comprising an event acquirer configured to acquire an event comprising an associated date and set of data fields, and pass event related data to the plurality of event reception components.
3. The system of claim 2 , wherein time-ordered behavior evaluator employs a Markov Graph, to learn a general sequence of events performed by one actor during a time period.
4. The system of claim 1 , further comprising a periodic set of components configured to operate periodically on demand, the periodic set of components configured to perform peer to peer analysis, actor correlation analysis, actor behavior analysis, semantic rule analysis, and predict rates of change.
5. The system of claim 4 , further comprising a threat detector configured to detect threats.
6. The system of claim 5 wherein the threat detector employs an extreme vigilance application programming interface.
7. The system of claim 4 , wherein the periodic set of components comprises a peer to peer analyzer, an actor correlation analyzer, an actor behavior analyzer, a rate of change predictor, and a semantic rule analyzer.
8. The system of claim 7 , wherein the actor behavior analyzer is configured to examine any changes in actor behavior over time by comparing similarity of past behavior with current behavior.
9. The system of claim 7 , wherein the semantic rule analyzer is configured to encode conditional, provisional, cognitive, operational, and functional knowledge.
10. The system of claim 7 , wherein the rate of change predictor is configured to store similarity and correlation results over time and rates of change over time.
11. The system of claim 1 , further comprising a signal manager configured to exclude signals based on content properties of data transmitted.
12. The system of claim 1 , further comprising an additional signal manager connected to a signal filter configured to receive signal weights from a signal weight repository.
13. A system for measuring values qualitatively across a terrain comprising multiple dimensions using cognitive computing techniques, comprising:
a plurality of event reception components configured to operate on each event in a stream relevant to the terrain;
a periodic set of components configured to operate periodically on demand to analyze and predict based on information received from the plurality of event reception components, the periodic set of components comprising an outlier analysis module configured to determine at least one outlier in the stream, wherein the outlier analysis module determines whether at least one outlier in the stream represents a beginning of a different data pattern or a change to an existing data pattern; and
a plurality of signal managers interfacing with the plurality of event reception components and the periodic set of components, wherein the plurality of signal managers is configured to exclude signals based on content properties of data transmitted;
wherein the system changes a rate of change threshold and an outlier boundary threshold upon the outlier analysis module determining the beginning of the different data pattern or the change to the existing data pattern.
14. The system of claim 13 , wherein the plurality of event reception components comprises:
a threshold application component configured to apply a threshold to each element in the stream;
a terrain updater configured to update the terrain based on at least one event;
a threshold violation predictor configured to predict threshold violations based on the stream of events;
a time-ordered behavior evaluator configured to evaluate behavior based on the stream of events; and
a graph updater to update a graph based on the stream of events.
15. The system of claim 13 , further comprising a signal manager configured to exclude signals based on content properties of data transmitted.
16. The system of claim 13 , further comprising a threat detector configured to detect threats.
17. A method for measuring values qualitatively across a terrain comprising multiple dimensions using cognitive computing techniques, comprising:
operating on each event in a stream relevant to the terrain using a plurality of event reception components;
operating on demand to analyze and predict based on information received from the plurality of event reception components using a periodic set of components comprising:
determining at least one outlier in the stream; and
determining whether the at least one outlier in the stream represents a beginning of a different data pattern or a change to an existing data pattern;
excluding selected signals based on content properties of data transmitted;
changing a rate of change threshold and an outlier boundary threshold upon determining the beginning of the different data pattern or the change to the existing data pattern.
18. The method of claim 17 , wherein operating on each event in the stream comprises:
applying a threshold to each element in the stream;
updating the terrain based on at least one event;
predicting threshold violations based on the stream of events;
evaluating behavior based on the stream of events; and
updating a graph based on the stream of events.
19. The method of claim 17 , further comprising excluding signals based on content properties of data transmitted using a signal manager.
20. The method of claim 17 , further comprising detecting threats using a threat detector.