# US 2022/0277019 A1 · Displaying query results using machine learning model-determined query results visualizations > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2022/0277019 A1 - Document ID: US20220277019A1 - Status: Published application - Kind: A1 - Application number: 17/186,769 - Filed: 2021-02-26 - Published: 2022-09-01 - Earliest claimed: 2021-02-26 - Assignees: Micro Focus LLC - Applicants: Micro Focus LLC - Inventors: Tamir Mitelman - Web page: https://patentgrep.com/patent/US20220277019A1 ## CPC - G06F 16/26: Visual data mining; Browsing structured data - G06F 16/90335: {Query processing} - G06F 16/24535: {of sub-queries or views} - G06F 16/24578: {using ranking} - G06F 18/2148: {characterised by the process organisation or structure, e.g. boosting cascade} - G06N 3/09: Supervised learning - G06N 20/00: Machine learning - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A query is converted into a structured query tokenization. One or more query results visualizations for the query are determined by applying a supervised machine learning model to the structured query tokenization into which the query has been converted. Query results of the query are displayed using the determined query results visualizations. ## Claims ### Claim 1 A non-transitory computer-readable data storage medium storing program code executable by a processor to perform processing comprising: converting a query into a structured query tokenization; determining one or more query results visualizations for the query by applying a supervised machine learning model to the structured query tokenization; and displaying query results of the query using the determined query results visualizations. ### Claim 2 (depends on 1) The non-transitory computer-readable data storage medium of claim 1, wherein the structured query tokenization cannot be used for executing the query. ### Claim 3 (depends on 1) The non-transitory computer-readable data storage medium of claim 1, wherein data values within the query are replaced with data types of corresponding data fields within the structured query tokenization. ### Claim 4 (depends on 3) The non-transitory computer-readable data storage medium of claim 3, wherein field names within the query are retained within the structured query tokenization. ### Claim 5 (depends on 4) The non-transitory computer-readable data storage medium of claim 4, wherein relational operators relating the data values to the field names within the query are not retained within the structured query tokenization. ### Claim 6 (depends on 5) The non-transitory computer-readable data storage medium of claim 5, wherein logical operators separating sub-queries of the query are retained within the structured query tokenization, and wherein the structured query tokenization includes functions and/or calculated expressions not used for query evaluation. ### Claim 7 (depends on 1) The non-transitory computer-readable data storage medium of claim 1, wherein converting the query into the structured query tokenization comprises: dividing the query into one or more sub-queries, each sub-query including a field name, a data value, and a logical operator relating the field name to the data value; retaining the field name of each sub-query; replacing the data value of each sub-query with a data type of a corresponding data field; and removing the relational operator of each sub-query. ### Claim 8 (depends on 7) The non-transitory computer-readable data storage medium of claim 7, wherein the sub-queries are separated by logical operators within the query, and wherein converting the query into the structured query tokenization further comprises: retaining the logical operators. ### Claim 9 (depends on 1) The non-transitory computer-readable data storage medium of claim 1, wherein the processing further comprises: calculating a prediction score for the query as to each query results visualization with respect to the structured query tokenization; and culling the query results visualizations based on the calculated prediction scores of the query results visualizations. ### Claim 10 (depends on 9) The non-transitory computer-readable data storage medium of claim 9, wherein culling the query results visualizations based on the calculated prediction scores of the query results visualizations comprises one or more of: retaining only the query results visualizations having the calculated prediction scores that are greater than a threshold; retaining only a threshold number of the query results visualizations having the calculated prediction scores that are highest. ### Claim 11 (depends on 1) The non-transitory computer-readable data storage medium of claim 1, wherein the processing further comprises: receiving the query; and generating the query results of the query by executing the query. ### Claim 12 (depends on 1) The non-transitory computer-readable data storage medium of claim 1, wherein the processing further comprises: displaying graphical representations of the determined query results visualizations; and receiving a selected graphical representation of the displayed graphical representations of the determined query results visualizations, wherein displaying the query results of the query using the determined query results visualizations comprises displaying the query results of the query using the determined query results visualization having the selected graphical representation. ### Claim 13 A method comprising: training a supervised machine learning model using a training data set comprising a plurality of training data pairs that each comprise a training structured query tokenization of a training query and a training query results visualization for the training query; and displaying query results of an query using a query results visualization determined for the query by the trained supervised machine learning model. ### Claim 14 (depends on 13) The method of claim 13, further comprising: testing the trained supervised machine learning model using a testing data set comprising a plurality of testing data pairs that each comprise a testing structured query tokenization of a testing query and a testing query results visualization for the testing query. ### Claim 15 (depends on 14) The method of claim 14, further comprising: collecting data pairs that each comprise a query and a query results visualization for the query; converting the query of each data pair into a structured query tokenization; and dividing the data pairs into the training data set and the testing data set, wherein the data pairs of the testing data set are different than the data pairs of the training data set. ### Claim 16 (depends on 13) The method of claim 13, further comprising: continuously retraining the supervised machine learning model. ### Claim 17 (depends on 13) The method of claim 13, wherein for each training data pair, the training structured query tokenization is not used for executing the training query. ### Claim 18 (depends on 13) The method of claim 13, wherein for each training data pair, data values within the training query are replaced with data types of the data values within the training structured query tokenization, field names within the training query are retained within the training structured query tokenization, and relational operators relating the data values to the field names within the training query are not retained within the training structured query tokenization. ### Claim 19 A computing system comprising: a processor; and a memory storing instructions executable by the processor to: determine a query results visualization for a query by applying a supervised machine learning model to a structured query tokenization corresponding to the query; and display query results of the query using the determined query results visualization. ### Claim 20 (depends on 19) The computing system of claim 19, wherein the instructions are executable by the processor to further convert the query into the structured query tokenization by: dividing the query into one or more sub-queries, each sub-query including a field name, a data value, and a logical operator relating the field name to the data value; retaining the field name of each sub-query; replacing the data value of each sub-query with a data type of the data value; and removing the relational operator of each sub-query.