# US 10,873,545 B2 · Automatic action responses > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,873,545 B2 - Document ID: US10873545B2 - Status: Granted - Kind: B2 - Application number: 15/620,671 - Filed: 2017-06-12 - Published: 2018-12-13 - Granted: 2020-12-22 - Earliest claimed: 2017-06-12 - Assignees: Microsoft Technology Licensing, LLC - Applicants: Microsoft Technology Licensing, LLC - Inventors: Amy Huyen Phuoc Nguyen; Chia-Jung Lee; Ivan Valeryevich Zhiboedov; Philipp Cannons; Rachel Imogen Solimeno; Dong Hwi Yoo; Yamin Wang; Milad Shokouhi - Web page: https://patentgrep.com/patent/US10873545B2 ## CPC - H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages - G06F 16/2365: {Ensuring data consistency and integrity} - G06F 16/3329: Natural language query formulation - G06N 20/00: Machine learning - G06Q 10/107: Computer-aided management of electronic mailing [e-mailing] - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR ## Abstract Systems and methods are provided that automatically process message input and provide action responses according to the processing results. The automatic action response system may leverage at least one machine-learning algorithm that is trained using a dataset. The provided action responses may comprise of default action responses and/or intelligent action responses that are based at least in part on prior conversational data between a user and a sender. Some intelligent action responses may include text-based replies, which eliminate the need for a user to type a reply on a device screen, thereby saving previous time, conserving device battery life, and preserving the integrity of the device hardware. A portion of a message may be highlighted manually by a user or automatically by the action response system to initiate the automatic action response system. In this way, a more efficient and productive user experience across various devices and applications is achieved. ## Claims ### Claim 1 A processor-implemented method of providing action responses, comprising: detecting at least one portion of a message input from a computing device; generating a semantic representation for at least one part of the at least one portion of the message input; determining a set of action responses corresponding to the at least one portion of the message input, wherein the set of action responses comprises an intelligent action response that corresponds to the at least one portion of the message input and a default action response that corresponds to the at least one portion of the message input, and wherein the intelligent action response is based at least in part on the semantic representation; determining whether to provide the intelligent action response or the default action response based at least in part on a comparison of the intelligent action response and the default action response with data from a database containing historical message data and corresponding action responses; and automatically providing at least one action response of the set of action responses corresponding to the at least one portion of the message input, wherein the at least one action response is the intelligent action response or the default action response. ### Claim 2 (depends on 1) The method of claim 1 further comprising tokenizing the at least one portion of the message input based at least in part on a preset database of tokens. ### Claim 3 (depends on 1) The method of claim 1 further comprising determining at least one domain classifier to apply to the at least one portion of the message input based at least in part on a preset database of domain classifiers. ### Claim 4 (depends on 3) The method of claim 3, wherein determining at least one domain classifier to apply to the at least one portion of the message input is further based at least in part on past conversational data. ### Claim 5 (depends on 1) The method of claim 1, wherein determining the set of action responses is based at least in part on a preset database of action responses. ### Claim 6 (depends on 1) The method of claim 1, wherein determining the set of action responses is based at least in part on identifying an action response corresponding to past conversational data. ### Claim 7 (depends on 1) The method of claim 1, wherein determining the set of action responses is based at least in part on a public profile. ### Claim 8 (depends on 6) The method of claim 6, wherein the action response corresponding to past conversational data was retrieved from the database containing historical message data and corresponding action responses. ### Claim 9 (depends on 1) The method of claim 1, wherein the at least on portion of the message input is text-based. ### Claim 10 (depends on 1) The method of claim 1, wherein the at least one portion of the message input is speech-based. ### Claim 11 (depends on 1) The method of claim 1, wherein the database is associated with a machine-learning algorithm. ### Claim 12 (depends on 1) The method of claim 1, wherein determining the set of action responses corresponding to the at least one portion of the message input is based at least in part on identifying at least one third-party application. ### Claim 13 A computer device comprising: at least one processing unit; at least one memory storing processor-executable instructions that when executed by the at least one processing unit cause the computing device to: detect at least one portion of a message input from a computing device; generate a semantic representation for at least one part of the at least one portion of the message input; determine a set of action responses corresponding to the at least one portion of the message input, wherein the set of action responses includes an intelligent action response corresponding to the at least one portion of the message input and a default action response that corresponds to the at least one portion of the message input, wherein the intelligent action response is based at least in part on the semantic representation; determine whether to provide the intelligent action response or the default action response based at least in part on a comparison of the intelligent action response and the default action response with data from a database containing historical message data and corresponding action responses; and automatically provide at least one action response of the set of action responses corresponding to the at least one portion of the message input, wherein the at least one action response is the intelligent action response or the default action response. ### Claim 14 (depends on 13) The computing device of claim 13, wherein determining the set of action responses is based at least in part on a preset database of action responses. ### Claim 15 (depends on 14) The computing device of claim 14, wherein determining the set of action responses is based at least in part on identifying an action response corresponding to prior conversational data. ### Claim 16 (depends on 13) The computing device of claim 13, wherein the database is stored locally on a device not connected to a network. ### Claim 17 (depends on 13) The computing device of claim 13, wherein the at least one portion of the message input corresponds to a discussion forum. ### Claim 18 (depends on 13) The computing device of claim 13, wherein the at least one portion of the message input corresponds to at least one portion of a voicemail message. ### Claim 19 A computer storage medium storing instructions for executing, on one or more processors of a computing device, a method of providing action responses, the method comprising: detecting at least one portion of a message input from a computing device; generating a semantic representation for at least one part of the at least one portion of the message input; determining a set of action responses for the at least one portion of the message input, the set of action responses comprising an intelligent action response corresponding to the at least one portion of the message input and a default action response corresponding to the at least one portion of the message input, wherein the intelligent action response is based at least in part on the semantic representation; determining whether to provide the intelligent action response or the default action response based at least in part on a comparison of the intelligent action response and the default action response with data from a database containing historical message data and corresponding action responses; and automatically providing at least one action response of the set of action responses corresponding to the at least one portion of the message input, wherein the at least one action response is the intelligent action response or the default action response. ### Claim 20 (depends on 19) The computer storage medium of claim 19, wherein the method executed further comprises: tokenizing the at least one portion of the message input; extracting at least one feature from the at least one portion of the message input; and determining at least one domain classifier to apply to the at least one portion of the message input, wherein the semantic representation for the at least one part of the at least one portion of the message input generated based on the determination of the at least one domain classifier.