Network server arrangement for processing non-parametric, multi-dimensional, spatial and temporal human behavior or technical observations measured pervasively, and related method for the same
View Patent ↗This invention generally discusses wireless devices, servers and communications networks. In particular the invention pertains to performing observations in one or more mobile terminals and processing and distributing the related data in a server side system through layered data processing activities, and conversion of non-parametric data into parameterized form through the utilization of statistical filtering and semantic data structures. It is further explained how such multi-layer, parametrized data can be utilized for predictive purposes, and how feedback loops can be built with the physical world to improve future predictions. The invention is applicable in various applications, for example in systems where precise digital profiles of users need to be built on a continuous basis, and such profiles need to be dynamically linked to one or several actions triggered by emerging characteristics in the data. The multi-layer approaches makes it possible to structure output statistics into continuous and standardized, periodic datasets, even though input data is highly unorganized, non-chronological, and sporadic. Similarly, the invention describes how the multi-layer data storage structure and chosen statistical operations make it possible to build virtually an infinite number of further aggregations and averages based on the output data streams.
1. A system, comprising:
a data input to receive multi-dimensional, non-parametric data obtained from mobile devices;
a processor to parametrize the multi-dimensional, non-parametric data into a plurality of different abstraction layers as multi-layered data;
a memory to store the parametrized data;
an aggregation apparatus configured to aggregate the parametrized data in batches using at least one of time-series, averaging, or sum operations, the aggregation is performed relative to at least one of (1) a time period, (2) a location, (3) a mobile application, (4) a mobile application category, (5) a mobile user, or (6) a mobile user group, the aggregation apparatus is to determine, from at least one of the data batches, at least one of:
descriptive higher-level behavioral indicators for the mobile devices, or
descriptive higher-level technical indicators for the mobile devices,
the aggregation apparatus is configured to be activated based on at least one of a sufficient amount of data becoming available or a trigger; and
a data export entity to provide at least one of the behavioral indicators or the technical indicators to an external entity for determining at least one of personalized content or a network performance optimization for at least one of the mobile devices.
2. The system of claim 1 , further including a prediction entity to host a prediction model, the prediction model to provide a likelihood of at least one of (1) a future state, (2) a future pattern, or (3) a future location for at least one of the mobile devices based on previously obtained data.
3. The system of claim 2 , wherein the prediction model includes a Markov state machine.
4. The system of claim 2 , further including a feedback entity to provide information to the prediction entity on whether a prediction was successful or not to enable adaptation of the prediction model.
5. The system of claim 1 , wherein the aggregation apparatus is configured to determine a number of multi-dimensional behavioral vectors for behavioral indicator values, the multi-dimensional behavioral vectors including a vector indicative of at least one of traveling activity, movement activity, music consumption activity, or sleeping activity.
6. The system of claim 5 , wherein the aggregation apparatus is configured to aggregate at least one of previously determined behavioral indicators or vectors to construct further statistics.
7. The system of claim 5 , wherein the aggregation apparatus is configured to process behavioral indicators to determine at least one of:
dynamic behavior of a mobile device, or
differences between at least two mobile devices using at least one of Pearson correlation coefficient or regression analysis, the aggregation apparatus to use at least two behavioral vectors of behavioral indicator values to generate a behavioral similarity index or time trend data including an average slope thereof.
8. The system of claim 5 , wherein the aggregation apparatus is configured to trigger an alarm on a basis of a triggering condition relative to a comparison result between two behavioral vectors of behavioral indicator values or relative to a calculation of a new behavioral indicator.
9. The system of claim 5 , wherein the aggregation apparatus is configured to abstract behavioral indicators by determining a number of multi-dimensional behavioral vectors on a basis of already available behavioral vectors of behavioral indicator values to describe a behavioral pattern.
10. The system of claim 5 , wherein the aggregation apparatus is configured to determine a substantially continuous behavioral vector by automatically augmenting one or more missing portions using one or more of preceding data or following data.
11. The system of claim 1 , wherein the processor is to parametrize the multi-dimensional, non-parametric data according to one or more of (1) categorization, (2) structuring, or (3) structuring based on content, the processor is to add one or more of location tags, web domain data, or mobile application data, the processor to determine an application or mobile content entity identification.
12. The system of claim 11 , wherein the processor is to categorize using at least one of: application usage data, web browsing usage data, network performance data, access network scan data, cellular network scan data, WiFi scan data, memory usage data, device feature usage data, device system data, alarm clock data, calendar data, media usage data, content usage data, phone book content, message logs, voice call logs, or positioning data.
13. The system of claim 1 , further including a support engine entity to add one or more of remotely received parameters or locally generated parameters to the received non-parametric data, the one or more of the remotely received parameters or the locally generated parameters including one or more location stamps.
14. The system of claim 1 , wherein the aggregation apparatus is configured to process a behavioral and statistic indicator to determine at least one of: average browsing time in predetermined time units, average sleeping time during a predetermined period, average span of daily movements, average entropy of location dynamics for a predetermined period, application usage activity, or application usage frequency.
15. The system of claim 1 , wherein the aggregation apparatus includes a semantic model entity to associate semantic data including at least one of movement, location names, nature of locations, application consumed, or data type consumed with received non-parametric data for enabling natural language oriented semantic data queries.
16. A processor-based method, comprising:
accessing, with a processor, multi-dimensional, non-parametric data obtained from mobile devices;
parameterizing, with the processor, the multi-dimensional, non-parametric data into a plurality of different abstraction layers as multi-layered data;
storing the parametrized data in a memory;
aggregating, with the processor, the parametrized data in batches using at least one of time-series, averaging or sum operations, the aggregation is preformed relative to at least one of (1) a time period, (2) a location, (3) a mobile application, (4) a mobile application category, (5) a mobile user, or (6) a mobile user group, the aggregating of the parameterized data to determine at least one of:
descriptive higher-level behavioral indicators for the mobile devices, or descriptive higher-level technical indicators for the mobile devices,
wherein the aggregating of the parameterized data to determine to be activated based on at least one of a sufficient amount of data becoming available or a trigger; and
providing at least one of the behavioral indicators or technical indicators to an external entity for determining at least one of personalized content or a network performance optimization for at least one of the mobile devices.
17. The processor-based method of claim 16 , further including determining a number of multi-dimensional behavioral vectors for behavioral indicator values the multi-dimensional behavioral vector including a vector indicative of at least one of traveling activity, movement activity, music consumption activity, or sleeping activity.
18. The processor-based method of claim 17 , further including aggregating previously determined behavioral indicators or vectors to construct further statistics.
19. The processor-based method of claim 17 , further including processing behavioral indicators to determine at least one of:
dynamic behavior of at least one of the mobile devices, or
differences between at least two mobile devices; and
using at least one of Pearson correlation coefficient or regression analysis, using at least two behavioral vectors of behavioral indicator values to generate a behavioral similarity index or time trend data including an average slope thereof.
20. The processor-based method of claim 17 , further including triggering an alarm on the basis of a triggering condition relative to a comparison result between two behavioral vectors of behavioral indicator values or relative to a calculation of a new behavioral indicator.