US 12,032,564 B1
Transforming natural language request into enterprise analytics query using fine-tuned machine learning model
Suchin Chouta, Udupi (IN); and Shashank Mohan Jain, Karnataka (IN)
Assigned to SAP SE, Walldorf (DE)
Filed by SAP SE, Walldorf (DE)
Filed on Jan. 3, 2023, as Appl. No. 18/149,263.
Int. Cl. G06F 16/24 (2019.01); G06F 16/242 (2019.01); G06F 16/2455 (2019.01); G06N 5/02 (2023.01)
CPC G06F 16/243 (2019.01) [G06F 16/2455 (2019.01); G06N 5/027 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A system associated with enterprise business analytics, comprising:
a Machine Learning (“ML”) model data store containing at least one generic ML model;
a fine-tuning data store containing prior natural language user requests and associated enterprise business database queries that were manually generated by business intelligence engineers;
an enterprise business data store containing enterprise business data; and
a transformation framework coupled to the ML model data store, the fine-tuning data store, and the enterprise data store, including:
a computer processor, and
a computer memory coupled to the computer processor and storing instructions that, when executed by the computer processor, cause the transformation framework to:
(i) retrieve the generic ML model from the ML model data store,
(ii) fine-tune the generic ML model using the prior user requests and associated enterprise business database queries to create a fine-tuned ML model,
(iii) receive a new natural language request from a user associated with the enterprise via an interactive chatbot and a remote user device,
(iv) use the fine-tuned ML model and new natural language request to automatically create a new enterprise business analytics query that contains both multiple dimension parameters identifying a business application and measure parameters identifying response information, and
(v) execute the new enterprise business analytics query to fetch enterprise business analytics data from the enterprise business data store.