Systems and methods for analyzing entity profiles
Inventors
Skarin, Bruce • Duchon, Andrew • Allopenna, Paul • Dejordy, Rich
Assignees
Publication Number
US-9123022-B2
Publication Date
2015-09-01
Expiration Date
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Embodiments of the subject invention comprise a computer based system and methods to collect and compare the attributes of a group of entities using data representing topic data of the entity and interaction data between entities. Embodiments of the invention comprise using minimally invasive means to automatically collect and model both an entity's attributes such as their knowledge/work/interest as well as model the social interactions of the entity together with a means to identify opportunities to influence changes in the entity attributes. Minimally invasive means to collect and model attributes include semantic analysis and topic modeling techniques. Means to model social interactions include social network analysis techniques that can incorporate location data of the entity. Embodiments of the invention further provide a sharable index of the attributes of the entities and the group of entities.
Core Innovation
Embodiments of the subject invention comprise a computer based system and methods to collect and compare the attributes of a group of entities using data representing topic data of the entity and interaction data between entities. Embodiments of the invention comprise using minimally invasive means to automatically collect and model both an entity's attributes such as their knowledge/work/interest as well as model the social interactions of the entity together with a means to identify opportunities to influence changes in the entity attributes. Embodiments of the invention further provide a sharable index of the attributes of the entities and the group of entities.
As the numbers of people an individual interacts with increases, it is more and more difficult for a given individual to retain a 'running tally' of the interests of all of their colleagues. Relying on rigid bureaucratic hierarchies does not solve the problem of connecting people with overlapping interests. A tool is needed to facilitate rapid formation of effective human networks in a system that non-invasively monitors the rich content of digital media and conversation, builds knowledge of ad hoc and potential networks, and accurately recommends new opportunities for collaboration.
Minimally invasive means to collect and model attributes include semantic analysis and topic modeling techniques. Means to model social interactions include social network analysis techniques that can incorporate location data of the entity. The system integrates text analysis of digital documents and communications to construct an organizational model of members' areas of knowledge/work, interaction analysis of digital communication and sensor based interactions, and methods for identifying and initiating collaboration between members of the organization.
Claims Coverage
One independent claim is presented (claim 1) and it recites multiple main inventive features.
Automatic interaction model determination
automatically determining an interaction model of a set of interaction information from a group of entities;
Interaction model as pairwise representation
the interaction model comprises a representation of interactions between a first entity from the group of entities and a second entity from the group of entities;
Entities as individuals
the first entity is a first individual and the second entity is a second individual;
Automatic topic model determination
automatically determining a topic model for at least the first entity and the second entity;
Entity topic profile generation
determining a first entity topic profile for the first entity and a second entity topic profile for the second entity;
Topic profile comparison
comparing the first entity topic profile and the second entity topic profile to identify a degree of topic similarity of the first entity and the second entity;
Entity interaction profile generation
determining a first entity interaction profile for the first entity and a second entity interaction profile for the second entity from the interaction model;
Interaction profile comparison
comparing the first entity interaction profile and the second entity interaction profile to identify a degree of interaction similarity of the first entity and the second entity;
Cross-comparison for influence identification
comparing the degree of topic similarity of the first entity and the second entity and the degree of interaction similarity of the first entity and the second entity whereby a change in one of the first entity topic profile or the second entity topic profile can be identified to influence the degree of topic similarity of the first entity and the second entity.
Claim 1 covers automated creation of interaction and topic models for entity pairs, generation and comparison of entity topic and interaction profiles, and cross-comparison to identify how changes in profiles can influence topic similarity.
Stated Advantages
Increase the efficiency of organizations by profiling and optimizing the patterns of creative interaction in an organization.
Drastically reduce both information disparity and information overload by maintaining an accurate, global awareness of the organization's knowledge and ensuring that only value-adding opportunities for collaboration are detected and pushed to the organization's members in a timely fashion.
Facilitate rapid formation of effective human networks and accurately recommend new opportunities for collaboration.
Non-invasively monitor the rich content of digital media and social interaction and build knowledge of informal and potential networks.
Provide a sharable index of the attributes of the entities and the group of entities.
Documented Applications
Identifying opportunities for more efficient information sharing for an individual in a group based on their knowledge model as pulled from their desktop and compared to the knowledge model and social network model of other members of that group.
Identifying collaboration opportunities for an individual in a group based on their knowledge model as pulled from their desktop and compared to the knowledge model and social network model of other members of that group.
Identifying sales opportunities for a retail organization based on the knowledge model of an ideal purchaser and comparing that to the knowledge model and social network model of a group of individuals.
Identifying target individuals surfing the web based on the knowledge model of an ideal target and comparing that model to the knowledge model and social network model pulled from web browser logs.
Identifying opportunities for targeting individuals using network resources based on the knowledge model of an ideal target and compare that model to the knowledge model and social network model pulled from their network traffic and network traffic pattern.
Provide targeted advertising, discover new resources for groups and individuals, and aid other social knowledge discovery activities.
Interested in licensing this patent?