# US 12,002,058 B2 · Customer service ticket processing using cluster-based data driven guidebooks > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,002,058 B2 - Document ID: US12002058B2 - Status: Granted - Kind: B2 - Application number: 16/864,396 - Filed: 2020-05-01 - Published: 2021-11-04 - Granted: 2024-06-04 - Earliest claimed: 2020-05-01 - Assignees: Dell Products L.P. - Applicants: Dell Products L.P. - Inventors: Anat Parush Tzur; Amihai Savir; Avitan Gefen - Web page: https://patentgrep.com/patent/US12002058B2 ## CPC - G06Q 30/016: After-sales - G06F 16/338: Presentation of query results - G06F 16/35: Clustering; Classification - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N 20/00: Machine learning - 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 - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Techniques are provided for customer service ticket processing using cluster-based data driven guidebooks. One method comprises obtaining a customer service ticket; extracting features related to the customer service ticket, wherein the features comprise a representation of a problem associated with the customer service ticket; assigning the customer service ticket to a given cluster of multiple customer service ticket clusters based on the features; obtaining a customer service ticket processing guidebook associated with the given cluster that identifies independent actions to perform to address the problem; and processing the customer service ticket based on the customer service ticket processing guidebook. A customer service ticket processing guidebook may be generated for each customer service ticket cluster using historical customer service tickets from the respective cluster. The customer service ticket processing guidebooks can be generated by clustering (i) possible independent actions and (ii) possible solutions identified in the historical customer service tickets of the given cluster. ## Claims ### Claim 1 A method, comprising: obtaining a plurality of clusters of historical customer service data records, wherein a clustering of the historical customer service data records is based at least in part on a plurality of features extracted from the historical customer service data records; obtaining a plurality of processor-readable data record processing guidebooks, wherein each of the plurality of processor-readable data record processing guidebooks is associated with a different cluster of the plurality of clusters, and wherein each of the processor-readable data record processing guidebooks is generated by performing the following steps for each of one or more of the plurality of clusters: (i) extracting a plurality of activity features and a plurality of solution features from the historical customer service data records in the respective cluster; (ii) clustering, within the respective cluster, a first set of possible independent computer-implemented actions, using the extracted plurality of activity features, identified in the plurality of historical customer service data records from the respective cluster, and (iii) clustering, within the respective cluster, a second set of possible solutions, using the extracted plurality of solution features, identified in the plurality of historical customer service data records from the respective cluster; performing the following steps, in response to obtaining a customer service data record to be processed, wherein the plurality of clusters and the plurality of processor-readable data record processing guidebooks are generated prior to the obtaining of the customer service data record to be processed: processing the customer service data record to extract a plurality of features related to the customer service data record, wherein the plurality of features comprises a representation of a problem associated with the customer service data record; processing the customer service data record to assign the customer service data record to a given cluster of the plurality of clusters of customer service data records based at least in part on the extracted plurality of features; selecting, using the assigned given cluster, a given processor-readable data record processing guidebook of the plurality of processor-readable data record processing guidebooks, wherein the given processor-readable data record processing guidebook is associated with the assigned given cluster, wherein the given processor-readable data record processing guidebook identifies one or more independent computer-implemented actions to perform to address the problem; and processing the customer service data record using at least one of the independent computer-implemented actions of the selected processor-readable data record processing guidebook; wherein the method is performed by at least one processing device comprising a processor coupled to a memory. ### Claim 2 (depends on 1) The method of claim 1, wherein the processing the customer service data record to extract the plurality of features related to the customer service data record further comprises assigning a numerical representation to one or more issues associated with the customer service data record. ### Claim 3 (depends on 1) The method of claim 1, wherein the processing the customer service data record to assign the customer service data record to the given cluster of the plurality of clusters employs an unsupervised clustering technique. ### Claim 4 (depends on 1) The method of claim 1, further comprising generating the processor-readable data record processing guidebook for each of the plurality of clusters of customer service data records using a plurality of historical customer service data records from the respective cluster. ### Claim 5 (depends on 1) The method of claim 1, further comprising grouping a plurality of dependent computer-implemented actions, obtained from the plurality of historical customer service data records from the given cluster, performed in a given order, into a single independent computer-implemented action, such that the dependent computer-implemented actions are performed in the given order. ### Claim 6 (depends on 1) The method of claim 1, further comprising generating a classification model using binary features and a fixed set of labels obtained from the clustered first set of possible independent computer-implemented actions and the clustered second set of possible solutions. ### Claim 7 (depends on 6) The method of claim 6, wherein the classification model comprises one or more decision tree classifiers and wherein a root of each decision tree comprises a computer-implemented action providing a substantially highest information gain. ### Claim 8 (depends on 7) The method of claim 7, wherein the computer-implemented action providing the substantially highest information gain comprises the computer-implemented action that most discriminates the historical data in the plurality of historical customer service data records from the respective cluster. ### Claim 9 A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device to perform the following steps: obtaining a plurality of clusters of historical customer service data records, wherein a clustering of the historical customer service data records is based at least in part on a plurality of features extracted from the historical customer service data records; obtaining a plurality of processor-readable data record processing guidebooks, wherein each of the plurality of processor-readable data record processing guidebooks is associated with a different cluster of the plurality of clusters, and wherein each of the processor-readable data record processing guidebooks is generated by performing the following steps for each of one or more of the plurality of clusters: (i) extracting a plurality of activity features and a plurality of solution features from the historical customer service data records in the respective cluster; (ii) clustering, within the respective cluster, a first set of possible independent computer-implemented actions, using the extracted plurality of activity features, identified in the plurality of historical customer service data records from the respective cluster, and (iii) clustering, within the respective cluster, a second set of possible solutions, using the extracted plurality of solution features, identified in the plurality of historical customer service data records from the respective cluster; performing the following steps, in response to obtaining a customer service data record to be processed, wherein the plurality of clusters and the plurality of processor-readable data record processing guidebooks are generated prior to the obtaining of the customer service data record to be processed: processing the customer service data record to extract a plurality of features related to the customer service data record, wherein the plurality of features comprises a representation of a problem associated with the customer service data record; processing the customer service data record to assign the customer service data record to a given cluster of the plurality of clusters of customer service data records based at least in part on the extracted plurality of features; selecting, using the assigned given cluster, a given processor-readable data record processing guidebook of the plurality of processor-readable data record processing guidebooks, wherein the given processor-readable data record processing guidebook is associated with the assigned given cluster, wherein the given processor-readable data record processing guidebook identifies one or more independent computer-implemented actions to perform to address the problem; and processing the customer service data record using at least one of the independent computer-implemented actions of the selected processor-readable data record processing guidebook. ### Claim 10 (depends on 9) The non-transitory processor-readable storage medium of claim 9, wherein the processing the customer service data record to extract the plurality of features related to the customer service data record further comprises assigning a numerical representation to one or more issues associated with the customer service data record. ### Claim 11 (depends on 9) The non-transitory processor-readable storage medium of claim 9, wherein the processing the customer service data record to assign the customer service data record to the given cluster of the plurality of clusters employs an unsupervised clustering technique. ### Claim 12 (depends on 9) The non-transitory processor-readable storage medium of claim 9, further comprising generating the processor-readable data record processing guidebook for each of the plurality of clusters of customer service data records using a plurality of historical customer service data records from the respective cluster. ### Claim 13 (depends on 9) The non-transitory processor-readable storage medium of claim 9, further comprising grouping a plurality of dependent computer-implemented actions, obtained from the plurality of historical customer service data records from the given cluster, performed in a given order into a single independent computer-implemented action, such that the dependent computer-implemented actions are performed in the given order. ### Claim 14 (depends on 9) The non-transitory processor-readable storage medium of claim 9, further comprising generating a classification model using binary features and a fixed set of labels obtained from the clustered first set of possible independent computer-implemented actions and the clustered second set of possible solutions. ### Claim 15 (depends on 14) The non-transitory processor-readable storage medium of claim 14, wherein the classification model comprises one or more decision tree classifiers and wherein a root of each decision tree comprises a computer-implemented action providing a substantially highest information gain. ### Claim 16 An apparatus comprising: at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured to implement the following steps: obtaining a plurality of clusters of historical customer service data records, wherein a clustering of the historical customer service data records is based at least in part on a plurality of features extracted from the historical customer service data records; obtaining a plurality of processor-readable data record processing guidebooks, wherein each of the plurality of processor-readable data record processing guidebooks is associated with a different cluster of the plurality of clusters, and wherein each of the processor-readable data record processing guidebooks is generated by performing the following steps for each of one or more of the plurality of clusters: (i) extracting a plurality of activity features and a plurality of solution features from the historical customer service data records in the respective cluster; (ii) clustering, within the respective cluster, a first set of possible independent computer-implemented actions, using the extracted plurality of activity features, identified in the plurality of historical customer service data records from the respective cluster, and (iii) clustering, within the respective cluster, a second set of possible solutions, using the extracted plurality of solution features, identified in the plurality of historical customer service data records from the respective cluster; performing the following steps, in response to obtaining a customer service data record to be processed, wherein the plurality of clusters and the plurality of processor-readable data record processing guidebooks are generated prior to the obtaining of the customer service data record to be processed: processing the customer service data record to extract a plurality of features related to the customer service data record, wherein the plurality of features comprises a representation of a problem associated with the customer service data record; processing the customer service data record to assign the customer service data record to a given cluster of the plurality of clusters of customer service data records based at least in part on the extracted plurality of features; selecting, using the assigned given cluster, a given processor-readable data record processing guidebook of the plurality of processor-readable data record processing guidebooks, wherein the given processor-readable data record processing guidebook is associated with the assigned given cluster, wherein the given processor-readable data record processing guidebook identifies one or more independent computer-implemented actions to perform to address the problem; and processing the customer service data record using at least one of the independent computer-implemented actions of the selected processor-readable data record processing guidebook. ### Claim 17 (depends on 16) The apparatus of claim 16, further comprising generating the processor-readable data record processing guidebook for each of the plurality of clusters of customer service data records using a plurality of historical customer service data records from the respective cluster. ### Claim 18 (depends on 16) The apparatus of claim 16, further comprising grouping a plurality of dependent computer-implemented actions, obtained from the plurality of historical customer service data records from the given cluster, performed in a given order into a single independent computer-implemented action, such that the dependent computer-implemented actions are performed in the given order. ### Claim 19 (depends on 16) The apparatus of claim 16, further comprising generating a classification model using binary features and a fixed set of labels obtained from the clustered first set of possible independent computer-implemented actions and clustered second set of possible solutions. ### Claim 20 (depends on 19) The apparatus of claim 19, wherein the classification model comprises one or more decision tree classifiers and wherein a root of each decision tree comprises a computer-implemented action providing a substantially highest information gain.