A system, method and computer program product, which given in input a question in natural language format, delivers personalized insights related to the answer. Personalized insights are selected among candidate insights mined from the data and ranked based on closeness to (mined) user-preference, relevance to the question, and surprise factor. Two core components include: Question analysis and meaningful insight look up and Multi-dimensional insight ranking. The Question analysis and meaningful insights lookup module performs a semantic analysis of the questions and, uses techniques including “templates” to build new questions which could uncover insights from the data. The Multi-dimensional insight ranking module takes in input a list of insights returned from Question analysis and meaningful insights lookup and rank such insights based on such factors as: relevance to the query, surprise factor, and user preferences.

Embodiments provide search answer extensibility by using either one or more primary applications that is executing or residing on, or is accessible to, a first computing device that receives a natural language (NL) query and/or one or more assigned applications that is executing or residing on the first computing device or on a different second computing device to process the NL query and provide a response to the NL query.

A computing system uses neural networks to translate natural language queries to database queries. The computing system uses a plurality of machine learning based models, each machine learning model for generating a portion of the database query. The machine learning models use an input representation generated based on terms of the input natural language query, a set of columns of the database schema, and the vocabulary of a database query language, for example, structured query language SQL. The plurality of machine learning based models may include an aggregation classifier model for determining an aggregation operator in the database query, a result column predictor model for determining the result columns of the database query, and a condition clause predictor model for determining the condition clause of the database query. The condition clause predictor is based on reinforcement learning.

In some examples, a word embedding is received. A word embedding includes a plurality of vectors in vector space. Each vector of the plurality of vectors represents a natural language word or other character sequence. Each vector is oriented in vector space based on a semantic similarity between each of the natural language word or other character sequence. A first distance is determined between a first vector and a second vector. A second distance is determined between a third vector and the second vector. Based at least in part on the first distance between the first vector and the second vector and the second distance between the third vector and the second vector, the third vector is moved closer or further away from the second vector. The moving is indicative of introducing a bias or removing a bias between the third vector and the second vector. This introducing or removing of bias supports more accurate and inclusive results from an application such as search of healthcare data, among other things.

Systems and methods are described for mitigating errors introduced during processing of user input such as voice input. A query may be derived from processed user input. A performance predictor analyzes the query and uses historical data to predict whether the query will return relevant results if executed. If the query's predicted performance is below a threshold, a query rewriter may identify potential alternatives to the query from a library of “known good” queries. Different analyzers may be applied to identify different sets of alternatives, and machine learning models may be applied to rank the outputs of the analyzers. The best-matching alternatives from each analyzer may then be provided as inputs to a further machine learning model, which assesses the probability that each of the identified alternatives reflects the intent of the user. A most likely alternative may then be selected to execute in place of the original query.

A user device transmit a natural language query with a request for a description of videos stored in a media repository. A query system receives the query and determines a command associated with obtaining the requested description of the videos stored in the media repository requested by the query. The determined command corresponds to an image analysis to perform on at least a portion of the stored videos. The query system determines, based at least in part on the determined command, an artificial intelligence model to execute on at least the portion of the stored videos. The query system determines, by executing the determined artificial intelligence model, a model output that includes the requested description of the videos stored in the media repository. The query system provides a response to the query. The response includes the requested description of the videos stored in the media repository.

A system, program product, and method for use with an artificial intelligence (AI) platform to dynamically amend a knowledge base responsive to query evaluating and processing. A received or detected query is subject to natural language processing to identify, annotate, and map one or more query tokens against a knowledge base. The query tokens are evaluated against the knowledge base to identify one or more query tokens absent from the knowledge base and leverage a neural network to predict a probability relationship between the query tokens absent from the knowledge base and one or more tokens populated in the knowledge base. The natural language (NL) query is translated to a structured query language (SQL) and the SQL query is executed and evaluated, and the knowledge base is selectively and dynamically amended subject to the SQL evaluation.

Systems and methods are provided for self-learning natural language predictive searching including receiving a first input, the first input being related to the desired outcome; retrieving a first information related to the first input; determining a first output based on at least the first input and the first information; outputting the first output; receiving a second input based on the outputted first output in response to the first output being different from the desired outcome, the second input being related to the desired outcome; retrieving, by the processor, a second information related to the second input; determining a second output based on at least the second input, the second information, the first input and the first information; and outputting the second output.

Methods, systems, and apparatus for receiving a request that includes an entity identifier of an entity referenced by a search query and one or more context terms that are included in the search query; determining that the one or more context terms describe a relationship connecting the entity referenced by the search query with a plurality of other entities; and in response to determining that the one or more context. terms describe the relationship, generating user interface elements that provide facts related to the entity referenced. Generating the user interface may include identifying relationship knowledge elements that include facts relating to at least some of the other entities connected to the entity by the relationship, and identifying one or more additional knowledge for the entity referenced by the search query. The ranking the relationship knowledge elements may be highest ranked knowledge elements in the user interface elements.

An information output receives a conversational sentence, parameterizes one or more entities included in one or more received conversational sentences, searches a database using the to acquire a search result, and outputs the search result.