Facebook, Inc.

Stepan Parunashvili

Laurent Nicolas Landowski

Kemal El Moujahid

Willy Blandin

Andreea-Loredana Crisan

Martin Jean Raison

Techniques to response to respond to user requests using natural-language machine learning based on branching example conversations are described. In one embodiment, an apparatus may comprise a bot application interface component operative to receive an example-interaction repository, the example-interaction repository comprising a plurality of example user-to-bot interactions, including one or more branching example user-to-bot interactions; and an interaction processing component operative to generate a linearized example-interaction repository by replacing the one or more branching example user-to-bot interactions with the plurality of linearized example user-to-bot interactions; submit the example-interaction repository to a natural-language machine learning component; and receive a sequence model from the natural-language machine learning component in response to submitting the example-interaction repository; and a client communication component operative to perform a user-to-bot conversation based on the sequence model. Other embodiments are described and claimed.

Techniques to respond to user requests using natural-language machine learning based on branching example conversations

A messaging server that concurrently provides a business or other recipient with a message sent to the recipient and the set of entities detected within the message using natural language processing (NLP). A business interacts with the messaging server to specify the types of NLP to perform on messages received by the business. The messaging server creates a configuration token describing the specified NLP configuration. The messaging server receives a message sent from a client to the business and generates a payload describing the message and including the configuration token. The messaging server then sends the payload to a NLP server. The NLP server performs NLP on the message in the payload according to the configuration specified by the token to detect a set of entities. The messaging server forms an enriched payload including the message and detected entities and forwards the enriched payload to the business.

Entity detection in messages

Techniques to customize bot messaging behavior are described. In one embodiment, an apparatus may comprise bot setup component operative to receive a bot customization control directive at a page bot associated with a web page, the web page comprising a plurality of page fields; present a bot setting user interface in response to the control directive; receive a first control directive in the bot setting user interface to enable a response assistant component of the page bot; receive a second control directive in the bot setting user interface to enable a response to an intent; receive a customization setting in the bot setting user interface to modify the enabled response to the intent; and save the customization setting for the response assistant component. Other embodiments are described and claimed.

Techniques to customize bot messaging behavior

Techniques to configure a web-based application for bot configuration are described. In one embodiment, an apparatus may comprise a web application hosting component operative to receive a bot-configuration request from a web client, the bot-configuration request corresponding to a developer account for a bot system; and provide a web-based application to the web client; and a web-application preparation component operative to prepare the web-based application for the web client, the web-based application comprising a simulated user environment interface, a code development interface, and a natural-language-processing log interface, wherein the simulated user environment interface corresponds to a user interface for the bot system. Other embodiments are described and claimed.

Techniques to configure a web-based application for bot configuration

A method may comprise receiving a bot creation control directive from a web page; instantiating a page bot from a template; authorizing the page bot to access the contents of the web page; associating the page bot with the web page; and enabling the page bot to respond to messages sent to the web page. The method may further comprise receiving a message at the page bot from a client device; requesting a meaning of the message from a natural language processing component; identifying a field in the web page associated with the page bot according to the meaning of the message; retrieving data from the identified field; composing a message response including the retrieved data; and sending the message response to the client device. Other embodiments are described and claimed.

Growing community of inventors

San Francisco, CA, United States of America

Stepan Parunashvili