Abstract: A method of automating incoming message prioritization. The method including training a global classifier of a computer system using training data. Dynamically training a user-specific classifier of the computer system based on a plurality of feedback instances. Inferring a topic of the incoming message received by the computer system based on a topic-based user model. Computing a plurality of contextual features of the incoming message. Determining a priority classification strategy for assigning a priority level to the incoming message based on the computed contextual features of the incoming message and a weighted combination of the global classifier and the user specific classifier. Classifying the incoming message based on the priority classification strategy.

Abstract: Visualizing content change of a data collection over time. A topic may be split into multiple linear, non-overlapping sub-topics along a timeline by satisfying a diverse set of semantic, temporal, and visualization constraints simultaneously. For each derived sub-topic, a set of representative keywords may be automatically selected to summarize the main content of the sub-topic.

Abstract: An apparatus for automating a prioritization of an incoming message, including a batch learning module that generates a global classifier based on training data that is input to the batch learning module. A feedback learning module that generates a user-specific classifier based on a plurality of feedback instances. A feature extraction module that receives the incoming message and a topic-based user model, infers a topic of the incoming message based on the topic-based user model, and computes a plurality of contextual features of the incoming message. A classification module that dynamically determines a priority classification strategy for assigning a priority level to the incoming message based on the plurality of contextual features of the incoming message and a weighted combination of the global classifier and the user-specific classifier, and classifies the incoming message based on the priority classification strategy.

Abstract: A method of modeling a user includes performing a role-based classification of tangible interactions involving the user performed via a computer system of an organization, creating a collection of role-specific interactions, creating, from the collection of role-specific interactions, a plurality of role-specific models of the user, wherein the plurality of role-specific models constitute a user model of the user, outputting one or more of the role-specific models to a different user model associated with a different user, and consolidating the output one or more of the role-specific models with a second plurality of role-specific models of the different user model within the different user model.

Abstract: Detecting propensity profile for a person may comprise receiving artifacts associated with the person; detecting profile characteristics for the person based on the artifacts; receiving a plurality of predefined profiles comprising a plurality of characteristics and relationships between the characteristics over time, each of the plurality of predefined profiles specifying an indication of propensity; matching the profile characteristics for the person with one or more of the plurality of predefined profiles; and outputting one or more propensity indicators based on the matching, the propensity indicators comprising at least an expressed strength of a given propensity in the person at a given time.

Abstract: Automatically generating a meeting digest of a set of meetings is provided. A set of topics of interest to the parties to the set of meetings is detected utilizing a user model associated with a user that is based on at least one of communications, relationships, and roles of the parties to the set of meetings. Topic-related content associated with the set of topics of interest to the parties is extracted from meeting data corresponding to the set of meetings. Then, the meeting digest of the set of meetings is generated using the topic-related content associated with the set of topics of interest to the parties extracted from the meeting data corresponding to the set of meetings.

Abstract: One embodiment of the present method and apparatus for robust input interpretation by conversation systems includes receiving a user request containing at least one un-interpretable term. The present invention conveys the conversation system's interpretation capabilities to the user, for example by suggesting at least one alternative request in context of the un-interpretable user request, where the alternative request is phrased in a manner that is substantially fully interpretable. A user may then select a suggested alternative request or formulate a new request in accordance with knowledge of the conversation system capabilities learned from the conveyed information. In another embodiment, a method may independently (e.g., without user prompting) adapt the un-interpretable user request in accordance with learned information regarding the meaning of any un-interpretable terms in the user request.

Abstract: A method including generating a global topic model based on a set of data that is updated according to an activity of each user of a plurality of users, the global topic model including a topic representation for a topic, generating a plurality of user models, each user model being generated based on the activity of a respective user, generating an expertise model for the topic based on the activity of at least one user of the plurality of users, the expertise model for the topic setting a target level of knowledge for a first user of the plurality of users, comparing a user model of the first user with the expertise model for the topic, the comparing being performed by a processor of a computer system, and recommending an activity associated with the set of data to the first user based on the comparison.

Abstract: A system including a global topic model building module that builds a global topic model based on a set of data that is updated according to an activity of each user of a plurality of users, a user model building module that builds a plurality of user models, each user model being built based on an activity of a respective user, an expertise model building module that builds an expertise model for the topic based on the activity of at least one user of the plurality of users, the expertise model for the topic setting a target level of knowledge for a first user of the plurality of users, a processor to compare a user model of the first user with the expertise model for the topic; and an expertise assessment and learning recommendation module that recommends an activity associated with the set of data to the first user based on the comparison.

Abstract: Methods and apparatus are provided for business outcome-based personalized ranking of information objects. Information objects in outcome-based business processes are ranked by recording information objects generated during a creation of the outcome-based business processes. At least one business process comprises an outcome attribute, and at least one business process comprises in-process information objects. A composite graph is generated of the information objects and business processes. Each node in the graph corresponds to an information object or a business process. Links between two of the information object nodes have a strength based on a content similarity and a social network distance. The in-process information object node connects to a corresponding business process node, and two business process nodes have a link if they are indicated as related in the business process information system.

Abstract: Business outcome-based personalized ranking of information objects is provided. Information objects in outcome-based business processes are ranked by recording information objects generated during a creation of the outcome-based business processes. At least one business process comprises an outcome attribute, and at least one business process comprises in-process information objects. A composite graph is generated of the information objects and business processes. Each node in the graph corresponds to an information object or a business process. Links between two information object nodes have a strength based on a content similarity and a social network distance. The in-process information object node connects to a corresponding business process node, and two business process nodes have a link if they are indicated as related in the business process information system. The information objects are ranked based on the link strengths.

Abstract: Visualizing content change of a data collection over time. A topic may be split into multiple linear, non-overlapping sub-topics along a timeline by satisfying a diverse set of semantic, temporal, and visualization constraints simultaneously. For each derived sub-topic, a set of representative keywords may be automatically selected to summarize the main content of the sub-topic.

Abstract: A method of automating incoming message prioritization. The method including training a global classifier of a computer system using training data. Dynamically training a user-specific classifier of the computer system based on a plurality of feedback instances. Inferring a topic of the incoming message received by the computer system based on a topic-based user model. Computing a plurality of contextual features of the incoming message. Determining a priority classification strategy for assigning a priority level to the incoming message based on the computed contextual features of the incoming message and a weighted combination of the global classifier and the user specific classifier. Classifying the incoming message based on the priority classification strategy.

Abstract: Automatically generating a meeting digest of a set of meetings is provided. A set of topics of interest to the parties to the set of meetings is detected utilizing a user model associated with a user that is based on at least one of communications, relationships, and roles of the parties to the set of meetings. Topic-related content associated with the set of topics of interest to the parties is extracted from meeting data corresponding to the set of meetings. Then, the meeting digest of the set of meetings is generated using the topic-related content associated with the set of topics of interest to the parties extracted from the meeting data corresponding to the set of meetings.

Abstract: An apparatus for automating a prioritization of an incoming message, including a batch learning module that generates a global classifier based on training data that is input to the batch learning module. A feedback learning module that generates a user-specific classifier based on a plurality of feedback instances. A feature extraction module that receives the incoming message and a topic-based user model, infers a topic of the incoming message based on the topic-based user model, and computes a plurality of contextual features of the incoming message. A classification module that dynamically determines a priority classification strategy for assigning a priority level to the incoming message based on the plurality of contextual features of the incoming message and a weighted combination of the global classifier and the user-specific classifier, and classifies the incoming message based on the priority classification strategy.

Abstract: Automated contextual information retrieval techniques are provided based on multi-tiered user modeling and a dynamic retrieval strategy. Content relevant to a current message is presented by initially obtaining a multi-tiered user model containing a multi-tiered representation of interactions of a first user with each contact, wherein the multi-tiered representation includes a plurality of topic models each corresponding to interactions between the first user and one contact. The topic models contain a set of topics, each containing topic keywords. Context information is extracted based on content of the current message, a sender and/or a recipient of the current message, and the multi-tiered user model. A retrieval strategy is determined based on the extracted context information. Contextual queries are generated to search the information repositories selected based on the determined retrieval strategy.

Abstract: A method and system of producing a visual text summarization. According to the method, a plurality of topics may be extracted from a collection of text documents, where each of the topics comprises a distribution of topic keywords. An importance ranking for each of the topics may be determined, and an importance ranking for each of the topic keywords of a topic may also be determined. A graph may be displayed having a plurality of stacked layers representing a group of topics selected from the plurality of topics based on the importance ranking of the topics. A keyword cloud within each layer of the graph may also be displayed, where the one keyword cloud is a group of topic keywords selected from the extracted topic keywords based on the importance ranking of the extracted topic keywords.

Abstract: A method and system of producing a visual text summarization. According to the method, a plurality of topics may be extracted from a collection of text documents, where each of the topics comprises a distribution of topic keywords. An importance ranking for each of the topics may be determined, and an importance ranking for each of the topic keywords of a topic may also be determined. A graph may be displayed having a plurality of stacked layers representing a group of topics selected from the plurality of topics based on the importance ranking of the topics. A keyword cloud within each layer of the graph may also be displayed, where the one keyword cloud is a group of topic keywords selected from the extracted topic keywords based on the importance ranking of the extracted topic keywords.

Abstract: A method for instance-based sentence boundary determination optimizes a set of criteria based on examples in a corpus, and provides a general domain-independent framework for the task by balancing a comprehensive set of sentence complexity and quality constraints. The characteristics and style of naturally occurring sentences are simulated through the use of semantic grouping and sentence length distribution. The method is parameterized so that it is easily adapts to suit a Natural Language Generation (NLG) system's generation.

Abstract: Methods and apparatus are provided for integrating a visual query interface and a natural language interface. The disclosed interface embeds a first of the visual query interface and the natural language interface into a second of the visual query interface and the natural language interface. The disclosed interface can also receives one or more natural language expressions in the natural language interface and one or more visual query expressions in the visual query interface in a same search turn. The disclosed interface can also receive a first query comprised of a first of a natural language query and a visual query; process the first query; and creates a substantially equivalent query to the first query in a second of the natural language query and the visual query The disclosed interface can process a natural language query to determine if a portion of the natural language query is not understood; and convert at least a portion of the natural language query to a visual query.