Multi-source, multi-dimensional, cross-entity, multimedia analytical model sharing database platform apparatuses, methods and systems
First Claim
1. An analytical model sharing processor-implemented method for privacy enhanced analytical model sharing through the use of contextual privacy dataset modifications, comprising:
- obtaining, using one or more data processors, a request to publish an analytical model operating on user data retrieved from a distributed linking node mesh, wherein the distributed linking node mesh includes a node representing an observable entity, a node representing a deduced entity derived through aggregating information associated with the user, and a node representing a meta concept derived through identifying sentiment words in the associated information;
determining, using the one or more data processors, whether the analytical model requested for publishing includes a model dataset used to develop the analytical model;
when the one or more data processors determine that the analytical model includes the model dataset, determining, using the one or more data processors, whether the model dataset used to develop the analytical model contains private consumer data that should not be shared;
when the one or more data processors determine that the model dataset contains private consumer data, determining, using the one or more data processors, whether the model dataset is excludable from the analytical model while still allowing the analytical model to be properly used;
when the one or more data processors determine that the model dataset is excludable from the analytical model, creating a modified analytical model that does not include the model dataset by deleting the model dataset from the analytical model;
wherein the request to publish the analytical model contains a user generated analytical model;
converting the user generated analytical model into a model containing contextless meta-data;
wherein converting the user generated analytical model comprises;
determining a language definition corresponding to a language of the user generated analytical model;
creating a mapping between the language definition and commands contained within the user generated analytical model;
applying a transformation, using the mapping, to the user generated analytical model; and
publishing, using the one or more data processors, the modified analytical model to an analytical model sharing service in response to the request to publish.
1 Assignment
0 Petitions
Accused Products
Abstract
The MULTI-SOURCE, MULTI-DIMENSIONAL, CROSS-ENTITY, MULTIMEDIA ANALYTICAL MODEL SHARING DATABASE PLATFORM APPARATUSES, METHODS AND SYSTEMS (“AMSDB”) transform data aggregated from various computer resources using AMSDB components into updated entity profiles and/or social graphs. In one implementation, the AMSDB aggregates data records including search results, purchase transaction data, service usage data, service enrollment data, and social data. The AMSDB identifies data field types within the data records and their associated data values. From the data field types and their associated data values, the AMSDB identifies an entity. The AMSDB generates correlations of the entity to other entities identifiable from the data field types and their associated data values. The AMSDB also associates attributes to the entity by drawing inferences related to the entity from the data field types and their associated data values. Using the generated correlations and associated attributes, the AMSDB generates an updated profile and social graph of the entity. The AMSDB provides the updated profile and social graph for an automated web form filling request.
-
Citations
16 Claims
-
1. An analytical model sharing processor-implemented method for privacy enhanced analytical model sharing through the use of contextual privacy dataset modifications, comprising:
-
obtaining, using one or more data processors, a request to publish an analytical model operating on user data retrieved from a distributed linking node mesh, wherein the distributed linking node mesh includes a node representing an observable entity, a node representing a deduced entity derived through aggregating information associated with the user, and a node representing a meta concept derived through identifying sentiment words in the associated information; determining, using the one or more data processors, whether the analytical model requested for publishing includes a model dataset used to develop the analytical model; when the one or more data processors determine that the analytical model includes the model dataset, determining, using the one or more data processors, whether the model dataset used to develop the analytical model contains private consumer data that should not be shared; when the one or more data processors determine that the model dataset contains private consumer data, determining, using the one or more data processors, whether the model dataset is excludable from the analytical model while still allowing the analytical model to be properly used; when the one or more data processors determine that the model dataset is excludable from the analytical model, creating a modified analytical model that does not include the model dataset by deleting the model dataset from the analytical model; wherein the request to publish the analytical model contains a user generated analytical model; converting the user generated analytical model into a model containing contextless meta-data; wherein converting the user generated analytical model comprises; determining a language definition corresponding to a language of the user generated analytical model; creating a mapping between the language definition and commands contained within the user generated analytical model; applying a transformation, using the mapping, to the user generated analytical model; and publishing, using the one or more data processors, the modified analytical model to an analytical model sharing service in response to the request to publish. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
Specification