The present disclosure provides a question answering method and apparatus for query information. The method may include: receiving query information input by a user, and analyzing a query target comprised in the query information; recalling candidate answers from a pre-generated knowledge graph based on the query target, where the knowledge graph is constructed based on inherent data in a map database and dynamic data of historical users, and the dynamic data includes at least one of comment data, search data, or spatiotemporal big data; and returning, in response to that there is a target answer whose matching degree with the query target exceeds a preset threshold in the candidate answers, the target answer to the user.

A system and a method creating an intelligent memory and providing contextual intelligent recommendations is provided. The invention provides extracting electronic communications data associated with active user data. Further, the invention provides performing a keyword tagging operation on conversation data present in the extracted electronic communications data based on a pre-generated keywords map. The invention provides generating a multi-relational model representative of conversation data associated with the electronic communications data in the form of graph nodes based on the keywords stored as the first tag and the second tag. The invention provides transmitting one or more electronic Recommendation Action Communication (RAC) with embedded application program interface calls based on the multi-relational model, the embedded application program interface calls enabling actions to be taken on information units via a single click.

A central learning model is deployed as a user model and as an assistant model. Sensitive information utterances from a corpus of previously stored conversation language corresponding to user queries and chat agent responses thereto are used to train the user model to become an updated user model and to train the assistant model to become an updated assistant model, respectively. The user model provides user contexts corresponding to user queries to the assistant model and the assistant model provides assistant contexts corresponding to chat agent responses to the user model. During training, the user model does not provide plain-text queries to the assistant model and the assistant model does not provide plain-text responses to the user model. The updated assistant model may facilitate a federated training process produce an updated central model. An updated central model may be used to provide real-time chat agent responses to live user queries.

Techniques for request throttling in a provider network environment are described. A throttle handler controls whether requests will be processed through maintaining a token-based record, per type of request, having a token value indicative of a number of requests that can be processed over a time period. For a request, the token value of the token-based record corresponding to the request type is updated based on calculating an elapsed time between a last update time of the token-based record and the current time, calculating an intermediate token value as the existing token value plus a value of the elapsed time multiplied by a rate, and updating the token value to be the minimum between the intermediate token value and a burst value. The request is serviced when the updated token value is determined to be greater than or equal to a number of tokens needed to perform the request.

A method comprising receiving first data representative of a query. A representation of the query is generated using an encoder neural network and the first data. Words for a rephrased version of the query are selected from a set of words comprising a first subset of words comprising words of the query and a second subset of words comprising words absent from the query. Second data representative of the rephrased version of the query is generated.

Using a natural language analysis, it is determined that a compendium requires a natural language response to a natural language query, the compendium comprising a set of stored natural language responses to natural language queries. A relevance of a portion of narrative text to the natural language query is scored according to a query relevance measure, the portion extracted from a corpus of narrative text. The compendium is enhanced according to the query relevance score with information in the portion.

Systems and methods for determining a query for a data store are described. A natural language text may be analyzed using heuristic processing and one or more machine learning models. Query parameters may be determined from the heuristic processing and machine learning and combined to form a query in a query language. In the heuristic processing, parsing rules may be used to remove conditional terms to generate a base question. The base question may be input to the one or more machine learning model to generate a base query which may be combined with query parameters related to the conditional terms.

An information processing device acquires question information. The information processing device acquires vehicle state information representing a state of the vehicle. The information processing device acquires answer information in response to the question information, the answer information including an image for display. The information processing device, in a case in which the vehicle state information represents that the vehicle is traveling, stores the answer information in a storage. The information processing device, in a case in which the information processing device acquires vehicle state information representing that the vehicle is stopped, outputs the answer information stored in the storage.

Systems and methods for emotionally intelligent automated chatting are provided. The systems and method provide emotionally intelligent automated (or artificial intelligence) chatting by determining a context and an emotion of a conversation with a user. Based on these determinations, the systems and methods may select one or more responses from a database of responses to a reply to a user query. Further, the systems and methods are able update or train based on user feedback and/or world feedback.

Techniques for performing outputting additional content associated with but nonresponsive to an input command are described. A system receives input data from a device. The system determines an intent representing the input data and receives first output data responsive to the input data. The system determines, based on context data, that additional content associated with the first output data but nonresponsive to the input data should be output. The system receives second output data associated with but nonresponsive to the input data thereafter. The system then presents first content corresponding to the first output data and second content corresponding to the second output data.