Abstract: Large batches of social media communications may be automatically annotated. This provides techniques to create large labeled datasets without the assistance of human labelers. For instance, social media communications may be fetched and annotated as actionable or noise for a given account (e.g., a brand handle on Twitter®) without human review. Social media communications from users who are attempting to engage with a brand or entity on a social media platform may be annotated as actionable, whereas other communications may be labeled as noise.

Abstract: A computer-implement process generally pertains to ephemeral chat messaging, and in particular, maintaining context of a conversation between a BOT and a user from one webpage of a web browser to another webpage of the web browser. The computer-implemented process includes adding a message to a chat widget. The chat widget is displayed within the webpage of the web browser and the message being added is from the other webpage that the user responded to. The computer-implemented process also includes discarding one or more previous messages in the chat widget that are not responded to by the user, allowing the conversation between the BOT and the user to persist while retaining the context of the conversation from the other webpage.

Abstract: A noise reduction and smart ticketing application for social media-based communication systems may identify social media-based communications from users who are attempting to engage with a brand or entity on a social media platform as actionable, and distinguish other communications as noise. The noise reduction and smart ticketing system may use machine learning to determine which social media communications are actionable for a given company or other organization, and generates tickets for actionable communications. Actionable communications may include, but are not limited to, technical support issues, inquiries about a product release date, grievances, incidents, suggestions to improve service, critiques of company policies, etc. Non-actionable communications (i.e., “noise”) may include, but are not limited to, suggestions to other users, promotions, coupons, offers, marketing campaigns, affiliate marketing, statements that a user is attending an event, etc.

Abstract: A noise reduction and smart ticketing application for social media-based communication systems may identify social media-based communications from users who are attempting to engage with a brand or entity on a social media platform as actionable, and distinguish other communications as noise. The noise reduction and smart ticketing system may use machine learning to determine which social media communications are actionable for a given company or other organization, and generates tickets for actionable communications. Actionable communications may include, but are not limited to, technical support issues, inquiries about a product release date, grievances, incidents, suggestions to improve service, critiques of company policies, etc. Non-actionable communications (i.e., “noise”) may include, but are not limited to, suggestions to other users, promotions, coupons, offers, marketing campaigns, affiliate marketing, statements that a user is attending an event, etc.

Abstract: A method for notifying a subset of users of a messaging channel may include publishing one or more shadow channels for the messaging channel. Each shadow channel is configured to prevent transmission of an event notification to one or more unregistered user devices. The method may also include broadcasting one or more event notifications to one or more registered user devices when an event associated with the one or more shadow channels is triggered. The one or more registered user devices are subscribers to the one or more shadow channels.