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.

An audio stream is detected during a communication session with a user. Natural language processing on the audio stream is performed to update a set of attributes by supplementing the set of attributes based on attributes derived from the audio stream. A set of filter values is updated based on the updated set of attributes. The updated set of filter values is used to query a set of databases to obtain datasets. A probabilistic program is executed during the communication session by determining a set of probability parameters characterizing a probability of an anomaly occurring based on the datasets and the set of attributes. A determination is made if whether the probability satisfies a threshold. In response to a determination that the probability satisfies the threshold, a record is updated to identify the communication session to indicate that the threshold is satisfied.

Examples of the present disclosure describe systems and methods for automatically assisting conversations using a graph database. In order to minimize misunderstanding of words and phrases used by participants during a conversation, phrases from the conversation may be received by conversation assistance application as the conversation takes place. Entities may be extracted from the phrase based on natural language recognition according to a domain context of the participant being assisted. One or more tags may be looked up from a graph database, and may be provided to the participant as a list of hashtags related to the conversation. Links to documents may be extracted based on the tags for the participant for viewing during the conversation.

Examples of the present disclosure describe systems and methods for automatically assisting conversations using a graph database. In order to minimize misunderstanding of words and phrases used by participants during a conversation, phrases from the conversation may be received by conversation assistance application as the conversation takes place. Entities may be extracted from the phrase based on natural language recognition according to a domain context of the participant being assisted. One or more tags may be looked up from a graph database, and may be provided to the participant as a list of hashtags related to the conversation. Links to documents may be extracted based on the tags for the participant for viewing during the conversation.

In certain embodiments, intent prediction and dialogue generation may be facilitated. In some embodiments, a chat initiation request may be obtained from a user. The latest activity information associated with the user may be provided to a prediction model to obtain a first set of predicted intents of the user. For each intent of the first set of predicted intents, a candidate question may be selected from a question set based on the candidate question matching the intent. In some embodiments, the candidate questions may be simultaneously presented on the chat interface.

An embodiment may involve a database containing a first user profile that specifies a first preferred language of a first user and a second user profile that specifies a second preferred language of a second user. The embodiment may also involve one or more processors configured to: receive, from the first user and within a chat session, a first set of messages in the first preferred language; cause the first set of messages to be translated into the second preferred language; provide, to the second user and within the chat session, the first set of messages as translated; receive, from the second user and within the chat session, a second set of messages in the second preferred language; cause the second set of messages to be translated into the first preferred language; and provide, to the first user and within the chat session, the second set of messages as translated.

Embodiments described herein provide a dynamic topic tracking mechanism that tracks how the conversation topics change from one utterance to another and use the tracking information to rank candidate responses. A pre-trained language model may be used for response selection in the multi-party conversations, which consists of two steps: (1) a topic-based pre-training to embed topic information into the language model with self-supervised learning, and (2) a multi-task learning on the pretrained model by jointly training response selection and dynamic topic prediction and disentanglement tasks.

A service identifies a level of specificity of one or more identified entities in a user input comprising a query, within one of multiple levels of a hierarchy of a hierarchical coding system. Responsive to determining that additional levels of specificity beyond the identified level of specificity are recommended to return a minimum answer set to the query, the service returns one or more answers requesting one or more additional inputs refining the query based on one or more values identified in a next level. Responsive to determining that no additional levels of specificity beyond the identified level of specificity are recommended to return the minimum answer set to the query, the service returns an answer set comprising a selection of information for the current level of specificity from an ingested corpus of knowledge mapped to the hierarchical coding system.

Systems, devices, and methods of the present invention create a searchable index that includes informative portions of text. In an example, a computer-implemented method creates a discourse tree from a body of text. For each non-terminal node in the discourse tree, the method identifies a rhetorical relationship associated with the non-terminal node. The method labels each terminal node associated with the non-terminal node as either a nucleus or a satellite. The method further accesses a rule associated with the rhetorical relationship, and selects, based on the rule, selects the fragment associated with the nucleus. The method creates a searchable index including the selected fragments.

Systems, devices, and methods of the present invention create a searchable index that includes informative portions of text. In an example, a computer-implemented method creates a discourse tree from a body of text. For each non-terminal node in the discourse tree, the method identifies a rhetorical relationship associated with the non-terminal node. The method labels each terminal node associated with the non-terminal node as either a nucleus or a satellite. The method further accesses a rule associated with the rhetorical relationship, and selects, based on the rule, selects the fragment associated with the nucleus. The method creates a searchable index including the selected fragments.