# US 12,182,183 B2 · Robot response method, apparatus, device and storage medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,182,183 B2 - Document ID: US12182183B2 - Status: Granted - Kind: B2 - Application number: 17/785,369 - Filed: 2020-04-20 - Published: 2023-01-26 - Granted: 2024-12-31 - Earliest claimed: 2019-12-27 - Assignees: Jingdong Technology Holding Co., Ltd. - Applicants: Jingdong Technology Holding Co., Ltd. - Inventors: Yuyu Zheng - Web page: https://patentgrep.com/patent/US12182183B2 ## CPC - G06F 16/3329: Natural language query formulation - G06F 16/3344: {using natural language analysis} - G06F 16/35: Clustering; Classification - G06N 3/006: based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] - G06N 3/045: Combinations of networks - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/088: Non-supervised learning, e.g. competitive learning - G06N 5/02: Knowledge representation; Symbolic representation - H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract The present application provides a robot response method, apparatus, device and storage medium. The method includes: obtaining, by a robot, current query voice; extracting semantic information of the current query voice; matching the semantic information of the current query voice with multiple semantic information clusters stored in advance to get a matched target semantic information cluster, where each semantic information cluster includes: at least one Q&A instance, and each Q&A instance includes: semantic information corresponding to a historical query voice and a query question selected in a query list corresponding to the historical query voice; and obtaining, by the robot, the number of times each query question was selected in the target semantic information cluster, determining, according to the number of times each query question was selected, a target query question corresponding to the current query voice, and outputting a query response corresponding to the target query question. ## Claims ### Claim 1 A robot response method, comprising: obtaining, by a server, multiple semantic information clusters, wherein each semantic information cluster comprises: at least one question and answer (Q&A) instance, and each Q&A instance comprises: semantic information corresponding to a historical query voice and a query question selected in a query list corresponding to the historical query voice; sending, by the server, the multiple semantic information clusters to a robot; obtaining, by the robot, a current query voice; extracting, by the robot, semantic information of the current query voice; matching, by the robot, the semantic information of the current query voice with multiple semantic information clusters stored in advance to get a matched target semantic information cluster; and obtaining, by the robot, a number of times each query question was selected in the target semantic information cluster, determining, by the robot, according to the number of times each query question was selected, a target query question corresponding to the current query voice, and outputting a query response corresponding to the target query question. ### Claim 2 (depends on 1) The method according to claim 1, wherein the matching the semantic information of the current query voice with the multiple semantic information clusters stored in advance, comprises: determining, by the robot, a similarity between the semantic information of the current query voice and the semantic information corresponding to each historical query voice in the multiple semantic information clusters; and if the similarity between the semantic information of the current query voice and the semantic information corresponding to the historical query voice in one semantic information cluster is greater than a first preset similarity, using, by the robot, the semantic information cluster as the target semantic information cluster. ### Claim 3 (depends on 2) The method according to claim 2, wherein the determining, according to the number of times each query question was selected, a target query question corresponding to the current query voice, comprises: determining, by the robot, among the query questions, the query question whose number of times of being selected is greater than a preset number of times as the target query question. ### Claim 4 (depends on 2) The method according to claim 2, further comprising: receiving, by the robot, the multiple semantic information clusters sent by a server. ### Claim 5 (depends on 1) The method according to claim 1, wherein the determining, according to the number of times each query question was selected, a target query question corresponding to the current query voice, comprises: determining, by the robot, among the query questions, the query question whose number of times of being selected is greater than a preset number of times as the target query question. ### Claim 6 (depends on 1) The method according to claim 1, further comprising: receiving, by the robot, the multiple semantic information clusters sent by a server. ### Claim 7 (depends on 1) A non-transitory computer-readable storage medium, wherein the computer-readable storage medium has stored therein computer-executable instructions, which, when executed by a processor, implement the robot response method according to claim 1. ### Claim 8 A robot response method, comprising: obtaining, by a server, multiple semantic information clusters, wherein each semantic information cluster comprises: at least one question and answer (Q&A) instance, and each Q&A instance comprises: semantic information corresponding to a historical query voice and a query question selected in a query list corresponding to the historical query voice; and sending, by the server, the multiple semantic information clusters to a robot; wherein before the obtaining multiple semantic information clusters, the method further comprises: obtaining, by the server, at least one basic semantic information cluster and at least one Q&A instance, wherein the at least one Q&A instance is periodically obtained; correspondingly, obtaining multiple semantic information clusters comprises: updating, by the server, the at least one basic semantic information cluster according to the at least one Q&A instance to obtain the multiple semantic information clusters; wherein updating the at least one basic semantic information cluster according to the at least one Q&A instance to obtain the multiple semantic information clusters, comprises: determining, by the server, a similarity between each semantic information in the at least one Q&A instance and each semantic information in the at least one basic semantic information cluster; and for each Q&A instance in the at least one Q&A instance, dividing, by the server, the Q&A instance into a basic semantic information cluster in the at least one basic semantic information cluster whose similarity with the Q&A instance is greater than a third preset similarity. ### Claim 9 (depends on 8) The method according to claim 8, wherein before the obtaining multiple semantic information clusters, the method further comprises: obtaining multiple Q&A instances; correspondingly, the obtaining multiple semantic information clusters, comprises: determining a similarity of the semantic information in every two Q&A instances among the multiple Q&A instances; and classifying Q&A instances whose similarity is greater than a second preset similarity into a same one semantic information cluster. ### Claim 10 (depends on 9) The method according to claim 9, further comprising: obtaining, by the server, the number of times each query question was selected in the multiple semantic information clusters; and sending the number of times each query question was selected in the multiple semantic information clusters to the robot. ### Claim 11 (depends on 8) The method according to claim 8, further comprising: obtaining, by the server, the number of times each query question was selected in the multiple semantic information clusters; and sending the number of times each query question was selected in the multiple semantic information clusters to the robot. ### Claim 12 (depends on 8) The method according to claim 8, further comprising: obtaining, by the server, the number of times each query question was selected in the multiple semantic information clusters; and sending, by the server the number of times each query question was selected in the multiple semantic information clusters to the robot. ### Claim 13 (depends on 8) The method according to claim 8, further comprising: obtaining, by the server, the number of times each query question was selected in the multiple semantic information clusters; and sending, by the server, the number of times each query question was selected in the multiple semantic information clusters to the robot. ### Claim 14 A robot response system, comprising: a transceiver of a server, a memory of the server, a processor of the server, and a computer program stored on the memory of the server and operable on the processor of the server, wherein the processor of the server, when running the computer program, is configured to: obtain multiple semantic information clusters, wherein each semantic information cluster comprises: at least one question and answer (Q&A) instance, and each Q&A instance comprises: semantic information corresponding to a historical query voice and a query question selected in a query list corresponding to the historical query voice; and send the multiple semantic information clusters to a robot; wherein the robot response system further comprises a transceiver of the robot, a memory of the robot, a processor of the robot, and a computer program of the robot stored on the memory of the robot and operable on the processor of the robot, wherein the processor of the robot, when running the computer program, is configured to: obtain a current query voice; extract semantic information of the current query voice; match the semantic information of the current query voice with the multiple semantic information clusters stored in advance to get a matched target semantic information cluster; obtain the number of times each query question was selected in the target semantic information cluster; determine, according to the number of times each query question was selected, a target query question corresponding to the current query voice; and control the transceiver of the robot to output a query response corresponding to the target query question; wherein before the processor of the server obtaining multiple semantic information clusters, the processor of the server is further configured to: obtain at least one basic semantic information cluster and at least one Q&A instance, wherein the at least one Q&A instance is periodically obtained; update the at least one basic semantic information cluster according to the at least one Q&A instance to obtain the multiple semantic information clusters; wherein the processor of the server is further configured to: determine a similarity between each semantic information in the at least one Q&A instance and each semantic information in the at least one basic semantic information cluster; and for each Q&A instance in the at least one Q&A instance, divide the Q&A instance into a basic semantic information cluster in the at least one basic semantic information cluster whose similarity with the Q&A instance is greater than a third preset similarity. ### Claim 15 (depends on 14) The system according to claim 14, wherein the processor of the robot is configured to: determine a similarity between the semantic information of the current query voice and the semantic information corresponding to each historical query voice in the multiple semantic information clusters; and if the similarity between the semantic information of the current query voice and the semantic information corresponding to the historical query voice in one semantic information cluster is greater than a first preset similarity, use the semantic information cluster as the target semantic information cluster. ### Claim 16 (depends on 15) The system according to claim 15, wherein the processor of the robot is configured to: determine, among the query questions, the query question whose number of times of being selected is greater than a preset number of times as the target query question. ### Claim 17 (depends on 14) The system according to claim 14, wherein the processor of the robot is configured to: determine, among the query questions, the query question whose number of times of being selected is greater than a preset number of times as the target query question. ### Claim 18 (depends on 14) The system according to claim 14, wherein the processor of the robot is configured to: control the transceiver of the robot to receive the multiple semantic information clusters sent by the server.