Apparatus and methods are disclosed for processing messages from agents of a network environment including the use of collectors. Collectors can use configurable pipelines to improve processing of messages received from the agents. In one example of the disclosed technology, a number of networked agents are configured to gather data describing operational aspects of an agent's computing host. A collector is configured to receive reports from the agent and send the gathered data to one or more destination agent data consumers designated by augmentation information in the reports. In some examples, the collector transforms data using one or more stage selector rules.

A system and method for upgrading an executable chatbot is disclosed. The system may include a processor including a fallout utterance analyzer, a response identifier, a deviation identifier, a flow generator and enhancer. The fallout utterance analyzer may receive chats logs comprising a plurality of utterances and corresponding bot responses. The fallout utterance analyzer may classify the plurality of utterances into multiple buckets pertaining to at least one of an out-of-scope intent, a newly identified intent, and a new variation of an existing intent. The response identifier may generate auto-generated responses corresponding to new intents for upgrading the executable chatbot. The deviation identifier may overlay corresponding intent in the chat logs with the prestored flow dialog network to designate an extent of deviation with respect to flow prediction performance by the executable chatbot. The flow generator and enhancer may generate an auto-generated conversational dialog flow for upgrading the executable chatbot.

A chat bot computing system includes a bot controller and a natural language processor. The natural language processor receives a first textual input and identifies concepts represented by the first textual input. An indication of the concepts is output to the bot controller which generates a response to the first textual input. The concepts output by the natural language processor are also fed back into the input to the natural language processor, as context information, along with an expiration indicator when a second textual input is received. The natural language processor then identifies concepts represented in the second textual input, based on the second natural language, textual input and unexpired context information.

A chat bot computing system includes a bot controller and a natural language processor. The natural language processor receives a first textual input and identifies concepts represented by the first textual input. An indication of the concepts is output to the bot controller which generates a response to the first textual input. The concepts output by the natural language processor are also fed back into the input to the natural language processor, as context information, along with an expiration indicator when a second textual input is received. The natural language processor then identifies concepts represented in the second textual input, based on the second natural language, textual input and unexpired context information.

Certain aspects of the present disclosure provide techniques for placing targeted messages in communications within a software application using machine learning models. An example method generally includes retrieving, from a repository, a data set of targeted messages. For each respective targeted message in the data set of targeted messages, an effectiveness score for a party associated with the respective targeted message, a distance score between the party associated with the respective targeted message and a host party, and a match score between the party associated with the respective targeted message and the host party. Based on the effectiveness score, the distance score, and the match score for each respective targeted message in the data set of targeted messages, a message is selected to be included in one or more communications by the host party. The one or more communications including the selected message are generated and output for transmission.

Certain aspects of the present disclosure provide techniques for placing targeted messages in communications within a software application using machine learning models. An example method generally includes retrieving, from a repository, a data set of targeted messages. For each respective targeted message in the data set of targeted messages, an effectiveness score for a party associated with the respective targeted message, a distance score between the party associated with the respective targeted message and a host party, and a match score between the party associated with the respective targeted message and the host party. Based on the effectiveness score, the distance score, and the match score for each respective targeted message in the data set of targeted messages, a message is selected to be included in one or more communications by the host party. The one or more communications including the selected message are generated and output for transmission.

To enable provision of appropriate information for a user query even in a case there are multiple information provision modules which are different in answer generation processing. A query sending unit 212 sends a user query to each one of a plurality of information provision module units 220 that are different in the answer generation processing and that each generate an answer candidate for the user query. An output control unit 214 performs control such that the answer candidate acquired from each one of the plurality of information provision module units 220 is displayed on a display unit 300 on a per-agent basis with information on an agent associated with that information provision module unit 220.

The present disclosure is a method and apparatus for generating a suggested message. In one embodiment, a method for generating a suggested message includes monitoring, by an application server, a first device operated by a first user, detecting, by the application server, a triggering event relevant to the first user, and automatically generating the suggested message in response to the triggering event, where the suggested message is addressed to a second user.

Systems, methods, and computer-readable media for providing grouped data of interest. In some configurations, a computing device can include a user interface with one or more groups configured to receive messages. The groups can be computer-defined groups, such as by the computing device or other system, or user-defined groups. Techniques and technologies described herein receive a message in the one or more groups, and determine a message relevance (e.g., utility) based on the content and/or context (e.g., time of day, sender, geo-location of recipient, message urgency, etc.) of the message. In some configurations, the message relevance can be based, at least in part, on the particular group or a cluster of groups to which the message is associated.

Techniques for implementing a chatbot that utilizes context embeddings are described. An exemplary method includes determining a next turn by: applying a language model to the utterance to determine a probability of a sequence of words, generating a context embedding for the utterance based at least on one or more of: a dialog act as defined by a chatbot definition of the chatbot, a topic vector identifying a domain of the chatbot, a previous chatbot response, and one or more slot options; performing neural language model rescoring using the determined probability of a sequence of words as a word embedding and the generated context embedding to predict an hypothesis; determining at least a name of a slot and type to be fulfilled based at least in part on the hypothesis and the chatbot definition; and determining a next turn based at least in part on the chatbot definition, any previous state, and the name of the slot and type to be fulfilled.