An analysis device includes a storage unit that stores input sentences in association with information for distinguishing users, an extraction unit that extracts the input sentences stored in the storage unit on a per-user basis for respective corresponding functions, a classification unit that classifies the input sentences into intra-user similarity groups on the per-user basis for respective corresponding functions so that the input sentences extracted by the extraction unit form the intra-user similarity group consisting of input sentences similar to each other, an aggregation unit that aggregates the intra-user similarity groups among users on the per-function basis so that the intra-user similarity groups form an inter-user similarity group consisting of intra-user similarity groups similar to each other, and an output unit that outputs an aggregation result of the aggregation unit.

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.

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately and flexibly generating topic divergence classifications for digital videos based on words from the digital videos and further based on a digital text corpus representing a target topic. Particularly, the disclosed systems utilize a topic-specific knowledge encoder neural network to generate a topic divergence classification for a digital video to indicate whether or not the digital video diverges from a target topic. In some embodiments, the disclosed systems determine topic divergence classifications contemporaneously in real time for livestream digital videos or for stored digital videos (e.g., digital video tutorials). For instance, to generate a topic divergence classification, the disclosed systems generate and compare contextualized feature vectors from digital videos with corpus embeddings from a digital text corpus representing a target topic utilizing a topic-specific knowledge encoder neural network.

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately and flexibly generating topic divergence classifications for digital videos based on words from the digital videos and further based on a digital text corpus representing a target topic. Particularly, the disclosed systems utilize a topic-specific knowledge encoder neural network to generate a topic divergence classification for a digital video to indicate whether or not the digital video diverges from a target topic. In some embodiments, the disclosed systems determine topic divergence classifications contemporaneously in real time for livestream digital videos or for stored digital videos (e.g., digital video tutorials). For instance, to generate a topic divergence classification, the disclosed systems generate and compare contextualized feature vectors from digital videos with corpus embeddings from a digital text corpus representing a target topic utilizing a topic-specific knowledge encoder neural network.

Disambiguating question answering responses by receiving voice command data associated with a first user, determining a first user identity according to the first user voice command data, determining a first user activity context according to the first user voice command data, determining a first response for the first user, receiving voice command data associated with a second user, determining a second user identity according to the second user voice command data, determining a second user activity context according to the second user voice command data, determining a second response for the second user, determining a predicted ambiguity between the first response and the second response, altering the first response according to the predicted ambiguity, and providing the first response and the second response.

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.

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 are described for a contextual natural language understanding (cNLU) framework that is able to incorporate contextual signals of variable history length to perform joint intent classification (IC) and slot labeling (SL) tasks. A user utterance provided by a user within a multi-turn chat dialog between the user and a conversational agent is received. The user utterance and contextual information associated with one or more previous turns of the multi-turn chat dialog is provided to a machine learning (ML) model. An intent classification and one or more slot labels for the user utterance are then obtained from the ML model. The cNLU framework described herein thus uses, in addition to a current utterance itself, various contextual signals as input to a model to generate IC and SL predictions for each utterance of a multi-turn chat dialog.

Methods, apparatuses, and computer program products are described herein that are configured to express a time in an output text. In some example embodiments, a method is provided that comprises identifying a time period to be described linguistically in an output text. The method of this embodiment may also include identifying a communicative context for the output text. The method of this embodiment may also include determining one or more temporal reference frames that are applicable to the time period and a domain defined by the communicative context. The method of this embodiment may also include generating a phrase specification that linguistically describes the time period based on the descriptor that is defined by a temporal reference frame of the one or more temporal reference frames. In some examples, the descriptor specifies a time window that is inclusive of at least a portion of the time period to be described linguistically.

Methods, apparatuses, and computer program products are described herein that are configured to express a time in an output text. In some example embodiments, a method is provided that comprises identifying a time period to be described linguistically in an output text. The method of this embodiment may also include identifying a communicative context for the output text. The method of this embodiment may also include determining one or more temporal reference frames that are applicable to the time period and a domain defined by the communicative context. The method of this embodiment may also include generating a phrase specification that linguistically describes the time period based on the descriptor that is defined by a temporal reference frame of the one or more temporal reference frames. In some examples, the descriptor specifies a time window that is inclusive of at least a portion of the time period to be described linguistically.