# US 11,443,113 B2 · Sourcing object creation from natural language inputs > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,443,113 B2 - Document ID: US11443113B2 - Status: Granted - Kind: B2 - Application number: 17/207,521 - Filed: 2021-03-19 - Published: 2021-07-08 - Granted: 2022-09-13 - Earliest claimed: 2019-10-17 - Assignees: Sap SE - Applicants: Sap SE - Inventors: Nishant Kumar; Panish Ramakrishna; Kumaraswamy Gowda; Rajendra Vuppala; Vidhya Neelakantan; Erica Vandenhoek; Nithya Rajagopalan - Web page: https://patentgrep.com/patent/US11443113B2 ## CPC - G06F 40/205: Parsing - G06F 3/0481: based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance - G06F 40/174: Form filling; Merging - G06F 40/284: Lexical analysis, e.g. tokenisation or collocates - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/08: Learning methods - G06N 3/09: Supervised learning - G06N 3/096: Transfer learning - G06N 5/04: Inference or reasoning models - G06N 20/00: Machine learning - G06F 40/186: Templates - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract User-generated input is received that includes a sequence of words associated with initiation of a computer-implemented event. Thereafter, such input is parsed using at least one natural language processing (NLP) model. This parsed input is then used by a machine learning model to determine a suggested template having a plurality of fields for initiating the event. The template can then be presented in a graphical user interface. Related apparatus, systems, techniques and articles are also described. ## Claims ### Claim 1 A method for implementation by one or more computing devices comprising: receiving user-generated input comprising a sequence of words associated with initiation of a computer-implemented event; parsing the user-generated input using at least one pre-trained language model that classifies each word to one of a plurality of pre-defined classes, a first of the pre-defined classes being unknown and a second of the pre-defined classes being event type; determining, by a machine learning model using the parsed user-generated input, a suggested template comprising a plurality of fields for initiating the event, the machine learning model being different than the pre-trained language model, wherein the words classified as being unknown are ignored by the machine learning model; and providing the suggested template for presentation. ### Claim 2 (depends on 1) The method of claim 1 wherein the suggested template is provided for presentation in a graphical user interface, the method further comprising: receiving, via the graphical user interface, further user-generated input populating at least a portion of the fields. ### Claim 3 (depends on 2) The method of claim 2 further comprising: initiating the event once all of the fields have been completed. ### Claim 4 (depends on 3) The method of claim 3, wherein the event comprises a computer-implemented procurement process. ### Claim 5 (depends on 1) The method of claim 1 further comprising: populating at least a portion of the fields based on an output of the machine learning model. ### Claim 6 (depends on 1) The method of claim 1, wherein the suggested template is provided for presentation in a graphical user interface, and wherein the graphical user interface forms part of a web browser. ### Claim 7 (depends on 1) The method of claim 1, wherein the suggested template is provided for presentation in a graphical user interface, and wherein the graphical user interface forms part of a chatbot interface. ### Claim 8 (depends on 1) The method of claim 1, wherein the user input is received via e-mail. ### Claim 9 (depends on 1) The method of claim 1, wherein the machine learning model is an extreme gradient boosting model. ### Claim 10 (depends on 1) The method of claim 1, wherein the machine learning model was trained using historical user selections of events. ### Claim 11 A system comprising: at least one data processor; and memory storing instructions which, when executed by the at least one data processor, result in operations comprising: receiving user-generated input comprising a sequence of words associated with initiation of a computer-implemented event; parsing the user-generated input using at least one pre-trained language model that classifies each word to one of a plurality of pre-defined classes, a first of the pre-defined classes being unknown and a second of the pre-defined classes being event type; determining, by a machine learning model using the parsed user-generated input, a suggested template comprising a plurality of fields for initiating the event, the machine learning model being different than the pre-trained language model, wherein the words classified as being unknown are ignored by the machine learning model; and providing the suggested template for presentation. ### Claim 12 (depends on 11) The system of claim 11, wherein the suggested template is provided for presentation in a graphical user interface, and wherein the operations further comprise: receiving, via the graphical user interface, further user-generated input populating at least a portion of the fields. ### Claim 13 (depends on 12) The system of claim 12, wherein the operations further comprise: initiating the event once all of the fields have been completed. ### Claim 14 (depends on 13) The system of claim 13, wherein the event comprises a computer-implemented procurement process. ### Claim 15 (depends on 11) The system of claim 11, wherein the operations further comprise: populating at least a portion of the fields based on an output of the machine learning model. ### Claim 16 (depends on 11) The system of claim 11, wherein the suggested template is provided for presentation in a graphical user interface, and wherein the graphical user interface forms part of a web browser. ### Claim 17 (depends on 11) The system of claim 11, wherein the suggested template is provided for presentation in a graphical user interface, and wherein the graphical user interface forms part of a chatbot interface. ### Claim 18 (depends on 11) The system of claim 11, wherein the user input is received via e-mail. ### Claim 19 (depends on 11) The system of claim 11, wherein the machine learning model was trained using historical user selections of event, and wherein the machine learning model is an extreme gradient boosting model. ### Claim 20 A non-transitory computer program product storing instructions which, when executed by at least one computing device, result in operations comprising: receiving user-generated input comprising a sequence of words associated with initiation of a computer-implemented event; parsing the user-generated input using at least one pre-trained language model that classifies each word to one of a plurality of pre-defined classes, a first of the pre-defined classes being unknown and a second of the pre-defined classes being event type; determining, by a machine learning model using the parsed user-generated input, a suggested template comprising a plurality of fields for initiating the event, the machine learning model being different than the pre-trained language model, wherein the words classified as being unknown are ignored by the machine learning model; and providing the suggested template for presentation.