# US 12,004,256 B2 · Systems and methods for inferring intent to opt-out of communications > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,004,256 B2 - Document ID: US12004256B2 - Status: Granted - Kind: B2 - Application number: 18/075,359 - Filed: 2022-12-05 - Published: 2024-01-25 - Granted: 2024-06-04 - Earliest claimed: 2021-12-03 - Assignees: Augusta AI LLC - Applicants: Augusta AI LLC - Inventors: David Phillips; Eric Kobbe; Joshua Bates - Web page: https://patentgrep.com/patent/US12004256B2 ## CPC - H04L 51/212: using filtering or selective blocking - H04W 76/34: Selective release of ongoing connections - H04W 72/0453: Resources in frequency domain, e.g. a carrier in FDMA - H04W 76/20: Manipulation of established connections - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) ## Abstract A processing server assists in selectively removing or opting-out a client device from further communications. The processing server includes a hardware processor, and memory storing computer instructions that when executed perform transmitting a message to an entity, indicating directions to opt-out of receiving further communications, receiving, in response to the transmitting of the message, a reply, from the entity, inferring an intent of the reply pertaining to opting-out of receiving communications, and selectively terminating the communications based on the inferred intent. ## Claims ### Claim 1 A server system configured to infer opt-out instructions in a reply message from a user, the server system comprising: one or more hardware processors; and memory storing computer instructions, the computer instructions when executed by the one or more hardware processors configured to perform: receiving a reply from the user of a client device, in response to a communication sent to the client device; evaluating sentiment and context of the reply to infer intent whether to opt-out of receiving future communications and to generate a confidence score associated with the intent, the evaluating using a computer model; selectively instructing to terminate the future communications based on the inferred intent and the confidence score; receiving feedback, the feedback including an instruction to opt back in after instructing to terminate the future communications or including an adjustment of the confidence score; and training the computer model based on the feedback. ### Claim 2 (depends on 1) The server system of claim 1, wherein the sentiment is based on an existence or an absence of profanity within the reply. ### Claim 3 (depends on 1) The server system of claim 1, wherein the sentiment is based on historical behavior associated with the user of the client device. ### Claim 4 (depends on 3) The server system of claim 3, wherein the historical behavior is based on a frequency of opting back in following the instructing to terminate the future communications. ### Claim 5 (depends on 1) The server system of claim 1, wherein the context is based on the communication sent to the client device. ### Claim 6 (depends on 1) The server system of claim 1, wherein the computer model is generated by a trained machine learning component. ### Claim 7 (depends on 1) The server system of claim 1, wherein the evaluating the sentiment and the context of the reply includes evaluating media components within the reply. ### Claim 8 (depends on 1) The server system of claim 1, wherein the computer instructions, when executed by the one or more hardware processors, are configured to perform adjusting a frequency of the future communications based on an ambivalent confidence score. ### Claim 9 A processor-based method of inferring opt-out instructions in a reply message from a user, the method comprising: receiving a reply from the user of a client device, in response to a communication sent to the client device; evaluating sentiment and context of the reply to infer intent whether to opt-out of receiving future communications and to generate a confidence score associated with the intent, the evaluating using a computer model; selectively instructing to terminate the future communications based on the inferred intent and the confidence score; receiving feedback, the feedback including an instruction to opt back in after instructing to terminate the future communications or including an adjustment of the confidence score; and training the computer model based on the feedback. ### Claim 10 (depends on 9) The processor-based method of claim 9, wherein the sentiment is based on an existence or an absence of profanity within the reply. ### Claim 11 (depends on 9) The processor-based method of claim 9, wherein the sentiment is based on historical behavior associated with the user of the client device. ### Claim 12 (depends on 11) The processor-based method of claim 11, wherein the historical behavior is based on a frequency of opting back in following the instructing to terminate the future communications. ### Claim 13 (depends on 9) The processor-based method of claim 9, wherein the context is based on the communication sent to the client device. ### Claim 14 (depends on 9) The processor-based method of claim 9, wherein the computer model is generated by a trained machine learning component. ### Claim 15 (depends on 9) The processor-based method of claim 9, wherein the evaluating the sentiment and the context of the reply includes evaluating media components within the reply. ### Claim 16 (depends on 9) The processor-based method of claim 9, further comprising adjusting a frequency of the future communications based on an ambivalent confidence score.