A computer implemented method for analyzing feedback with respect to a virtual assistant includes identifying a technical support problem and a corresponding resolution, wherein the technical support problem corresponds to a query, and wherein the corresponding resolution corresponds to the virtual assistant's response, collecting user feedback provided by one or more users corresponding to the technical support problem and the corresponding resolution, creating a set of user profiles corresponding to the one or more users, generating weighted user feedback according to the set of user profiles, identifying contradictory feedback patterns corresponding to the one or more users, adjusting the set of user profiles according to the identified contradictory feedback patterns, and recommending improvements to the identified corresponding resolution.

A computer-implemented method and system for enrichment of responses in a multimodal conversation environment are disclosed. A Question Answer (QA) engine, such as a reinforcement document transformer exploits a document template structure or layout, adapts the information extraction using a domain ontology, stores the enriched contents in a hierarchical form, and learns context and query patterns based on the intent and utterances of one or more queries. The region of enriched content for preparing a response to a given query is expanded or collapsed by navigating upwards or downwards in the hierarchy. The QA engine returns the most relevant answer with the proper context for one or more questions. The responses are provided to the user in one or more modalities.

A method may include receiving, from a first client, a first message. The first message may be matched to a second user based on a similarity between a first keyword included in the first message and a second keyword included in a profile of a second user. The first keyword may be determined to be similar to the second keyword based on a distance between a first vector representation of the first keyword and a second vector representation of the second keyword not exceeding a threshold value. In response to the first message being matched with the second user, the first message may be sent to a second client associated with the second user. In response to receiving, from the second client, a second message responsive to the first message, the second message may be sent to the first client. Related systems and articles of manufacture are also provided.

Systems and methods for computational shelf forecasting are described. This may include receiving, by a computational action system, a network document associated with a corporate entity; initiating a data parsing process on the network document to determine official corporate filing data associated with the corporate entity; using the official corporate filing data associated with the corporate entity, determining a target stock price associated with a subset of shelf availability and a corresponding date of the target stock price; receiving, by the computational action system via a user interaction at a user interface, future data associated with the corporate entity; adjusting one or more shelf statuses using the future data; and providing a forecast action based on the one or more shelf statuses.

A computer-implemented method, computer system, and computer program product for contextualizing searches in a collaborative session having two or more users. The method may include generating, by a processor, one or more keywords from user context sources of the collaborative session. Users engaged in the collaborative session may use computing devices interconnected with each other via a collaborative tool. The user context source may comprise a document, a file, a webpage, a search history, or an application. Context of the collaborative session, having a start and stop, may be established using a natural language processing system. The method may include adding one of the one or more keywords to the search string of one of the users participating in the collaborative session. In some embodiments, one user may be an expert user whose user context source may be the only user context source collected and analyzed during the collaborative session.

Systems and methods for generating and annotating cell documents include extracting tables from a document using a table extraction engine. Headers are extracted for each of the tables using a header detection engine. Cells are extracted from each of the tables using a cell extraction engine. A cell document is generated for each of the cells which are each correlated to corresponding portions of the headers, each cell document recording the correlation between the cells and the headers. Each cell document is annotated to generate annotated cell documents with a cell recognition model trained to perform natural language processing on the cell documents by classifying each term in each of the cell documents and extracting relationships between the terms of each of the cell documents.

Various embodiments of the teachings herein include a fault processing method comprising: receiving two historical faults similar to a target fault; searching keywords in a description of the target fault and each historical fault, wherein the keywords are classified into N grades, and for each system component in a grade, the grade comprises at least one keyword for describing the component, wherein N is an integer no less than 2; for each of the N grades, counting a quantity of identical system components represented by the keywords in the text description of each historical fault and the target fault; and comparing a degree of similarity of each historical fault to the target fault according to the quantity of identical system components counted in each grade of the N different grades, wherein a historical fault relating to a larger number of high-grade identical system components has a higher degree of similarity to the target fault.

An approach and system incorporating obtaining a semantic model for a device, identifying properties from the semantic model, processing the properties, obtaining a property role, identifying a property role filter that correlates to the property role, and generating parameter values associated with the property role filter.

A method, system and computer-usable medium are disclosed for improving search engine ranking of a landing page using automated analysis of landing pages of third-party entities. Certain embodiments include receiving, at a user interface, a primary keyword associated with a targeted landing page of a primary entity; transmitting the primary keyword to a search engine; and receiving a search engine results page from the search engine. The search engine results page may be used to identify landing pages of third-party entities having a higher rank than the targeted landing page. Secondary keywords occurring on the third-party landing pages may be identified and analyzed to determine whether inclusion of the secondary keyword in the targeted landing page will increase ranking of the targeted landing page in the search engine.

A method for issuing control signals in response to sentiment. In some embodiments, the method includes: assigning, to each of a plurality of comments, a respective topic vector of length k; determining whether a largest element of the topic vector of a first comment of the plurality of comments exceeds a weight threshold; in response to the determining that the largest element exceeds the weight threshold, classifying the first comment into a first topic, of k topics, the first topic corresponding to the position, in the topic vector, of the largest element of the topic vector; calculating a first sentiment score; calculating an average sentiment score, based in part on the first sentiment score; determining whether the average sentiment score meets a criterion; and in response to the determining that the criterion is met, generating a control signal, the control signal including a message related to the first topic.