System and method for behavioural and contextual data analytics
View Patent ↗A server arrangement for managing observation data of wireless devices, including data input logic for obtaining observation data from wireless devices, the obtained data including behavioral and contextual raw data relative to the wireless devices, data mining logic for establishing a number of derived data elements, on the basis of processing and analyzing the obtained observation and optional supplementary data, the processing and analyzing incorporating aggregation procedures. At least one derived data element includes usage metrics with contextual dimension relative to applications or other features of wireless devices and users, data storage for storing the obtained data and the number of derived information elements, and a data distribution logic providing derived data. The distribution logic may serve a data query constructed by an external entity through provision of derived information from derived data elements according to the query parameters. A corresponding method for execution by the server arrangement is presented.
1. A computer system to process observational data received from wireless devices, comprising:
a memory including machine readable instructions; and
a processor to execute the instructions to:
obtain observational data from the wireless devices, the observational data including application usage data;
process the observational data to identify temporally adjacent applications to generate usage metric data; and
transmit advertisements to the wireless devices based on the usage metric data derived from the temporally adjacent applications, wherein the processor is to build a behavior model based on the identified temporally adjacent applications, the behavior model to describe user behaviors associated with the wireless devices, wherein the processor is to apply the behavior model to predict a usage duration of a second application in response to usage of a first application.
2. A computer system to process observational data received from wireless devices, comprising:
a memory including machine readable instructions; and
a processor to execute the instructions to:
obtain observational data from the wireless devices, the observational data including application usage data;
process the observational data to identify temporally adjacent applications to generate usage metric data; and
transmit advertisements to the wireless devices based on the usage metric data derived from the temporally adjacent applications, wherein the processor is to build a behavior model based on the identified temporally adjacent applications, the behavior model to describe user behaviors associated with the wireless devices, wherein the processor is to apply the behavior model to calculate a probability of using a second application based on detected usage of a first application.
3. A computer system as defined in claim 2 , wherein the first and second applications are temporally adjacent.
4. A computer system as defined in claim 2 , wherein the temporally adjacent applications include a first application executing during a first time period and a second application executing during a second time period, the second time period starting upon termination of the first time period.
5. A computer system as defined in claim 2 , wherein the processor is to receive supplementary data from an external data source.
6. A computer system as defined in claim 2 , wherein the processor is to:
serve a data query constructed by an external entity through usage metric data according to query parameters; and
push information relating to usage metric data to an external entity according to push logic.
7. A computer system as defined in claim 2 , wherein the processor is to generate data to adapt at least one of (i) a network service or (ii) an application for the wireless devices according to the usage metric data derived from the temporally adjacent applications.
8. A computer system as defined in claim 2 , wherein the processor is to generate data for personalized advertising according to the usage metric data.
9. A computer system as defined in claim 2 , wherein the processor is to determine an intensity variable including an activity indicator in a time domain relative to a selected time unit for the observational data.
10. A computer system as defined in claim 9 , wherein the processor is to use the intensity variable to determine a usage frequency.
11. A computer system as defined in claim 2 , wherein the processor is to process the observational data using at least one of correlation, additional clustering, or factoring.
12. A computer system as defined in claim 11 , wherein the processor is to apply a clustering algorithm on the usage metric data to at least one of behaviorally segment wireless device users to form behaviorally coherent user groups, apply factor analysis to categorize applications, apply factor analysis to categorize content, or apply pattern recognition to calculate relationships between at least one of actions, users, or applications.
13. A computer system as defined in claim 2 , wherein the observational data includes at least one of data relating to a voice communication action, data relating to a data communication action, microphone usage data, media reproduction data, camera usage data, user input data, user interface usage data, media recording data, location data, time data, identification data, device status data, cellular tower signal strength data, throughput rate data, signal-to-noise data, or data usage data.
14. A computer system as defined in claim 2 , wherein the processor is to determine a usage intensity for at least one of an application or an application category, and to calculate a derived factor from the usage intensity, the factor including generic multimedia usage indices based on at least one of music, imaging, or video application usage variables.
15. A computer system of claim 2 , wherein the processor is to calculate execution start and stop times for the observational data to determine usage session durations for at least one of user interface-level applications or background applications.
16. A computer system as defined in claim 2 , wherein the processor is to calculate a statistical model of usage behavior to provide an estimate for a status of each wireless device at a particular instant using the observational data.
17. A method to process observational data received from wireless devices, comprising:
obtaining, with a processor, observational data from the wireless devices, the observational data including application usage data;
processing the observational data to identify temporally adjacent applications to generate usage metric data;
transmitting advertisements to the wireless devices based on the usage metric data derived from the temporally adjacent applications;
building a behavior model based on the identified temporally adjacent applications, the behavior model to describe user behaviors associated with the wireless devices; and
applying, with the processor, the behavior model to predict a usage duration of a second application in response to usage of a first application.
18. A method to process data received from wireless devices, comprising:
obtaining, with a processor, observational data from the wireless devices, the observational data including application usage data;
processing the observational data to identify temporally adjacent applications to generate usage metric data;
transmitting advertisements to the wireless devices based on the usage metric data derived from the temporally adjacent applications;
building a behavior model based on the identified temporally adjacent applications, the behavior model to describe user behaviors associated with the wireless devices; and
applying, with the processor, the behavior model to calculate a probability of using a second application based on detected usage of a first application.
19. A method as defined in claim 18 , further including calculating a statistical model of usage behavior to provide a status estimate for each wireless device at a particular instant using the observational data.
20. A method as defined in claim 18 , wherein the first and second applications are temporally adjacent.
21. A method as defined in claim 18 , wherein the temporally adjacent applications include a first application executing during a first time period and a second application executing during a second time period starting upon termination of the first time period.
22. A tangible computer readable storage device or storage disk comprising instructions that, when executed, cause a machine to, at least:
process observational data to identify temporally adjacent applications to generate usage metric data, the observational data obtained from wireless devices, the observational data including application usage data;
transmit advertisements to the wireless devices based on the usage metric data derived from the temporally adjacent applications;
calculate a statistical model of usage behavior to provide a status estimate for the wireless devices at a particular instant using the observational data; and
apply the behavior model to calculate a usage duration of a second application in response to usage of a first application.
23. A computer readable storage device or storage disk comprising instructions that, when executed, cause a machine to, at least:
process observational data to identify temporally adjacent applications to generate usage metric data, the observational data obtained from wireless devices, the observational data including application usage data;
transmit advertisements to the wireless devices based on the usage metric data derived from the temporally adjacent applications; and
calculate a statistical model of usage behavior to provide a status estimate for the wireless devices at a particular instant using the observational data; and
apply the behavior model to calculate a probability of using a second application based on detected usage of a first application.
24. A computer readable storage device or storage disk as defined in claim 23 , wherein the instructions further cause the machine to calculate a statistical model of usage behavior to provide a status estimate for the wireless devices at a particular instant using the observational data.