Chatbot Event Generation Using LLM-Derived Runtime Parameters

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Generating user-defined events in computing systems is a highly technical task often unavailable to run-time users, leading to undetected events that adversely impact throughput and accuracy.

Innovation Solution

A chatbot event generation system (EGS) leveraging large language models (LLM) allows users to input user parameters through a chatbot interface, with the LLM generating derived parameters to create events, simplifying the process and minimizing human error.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If event generation is made accessible to run-time users, then ease of operation improves, but device complexity increases due to the need for automated parameter derivation

Engineering Contradiction:

Improveease of event generationVSAvoidsystem complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent introduces an automated event generation system comprising an event specification repository, parameter derivation module, and event generator as intermediary components between the user and the event detection mechanism. The user provides simple event names through the chatbot interface, while the automated system handles the complex parameter derivation and event creation processes, resolving the contradiction by shielding users from complexity while maintaining accessibility

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The event generation system performs self-service by automatically deriving parameters from available data sources and generating events without requiring user expertise. The parameter derivation module autonomously queries data sources, extracts relevant parameters, and configures events based on predefined schemas, enabling the system to serve itself rather than requiring complex user intervention

Inventive Principle:

Principle #25Self-service

2Productivity

If automated parameter derivation is implemented, then productivity improves, but use of energy increases due to LLM processing requirements

Engineering Contradiction:

Improveevent generation throughputVSAvoidcomputational energy consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent implements partial action by using the LLM only for generating event names from user inputs, while relying on deterministic algorithms for parameter derivation and event creation. This selective application of expensive LLM processing only where necessary (name generation) while using lighter computational methods elsewhere resolves the energy-productivity contradiction

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system performs preliminary action by pre-defining event schemas and parameter relationships in the event specification repository before runtime. This allows the parameter derivation module to efficiently generate events using cached schemas and predefined logic, reducing the need for expensive real-time LLM processing and lowering energy consumption while maintaining high productivity

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If simplified user input is accepted, then ease of operation improves, but measurement precision deteriorates due to fewer user-specified parameters

Engineering Contradiction:

Improveuser input simplicityVSAvoidparameter specification accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent replaces the mechanical system of manual parameter specification with an automated information retrieval and derivation system. Instead of users manually specifying parameters (mechanical input), the system automatically queries data sources, extracts parameters using derivation logic, and populates event configurations, maintaining precision while simplifying user input to simple event names

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system implements feedback loops where the parameter derivation module queries available data sources, validates derived parameters against event schemas, and iteratively refines parameter values. This feedback mechanism ensures that even though users provide minimal input, the system verifies and corrects parameters to maintain specification accuracy through automated validation and error handling

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20260065012A1Chatbot event generation system leveraging LLM capabilities

Publication Date: 2026.03.05 SAP SE

AI summary

System, method, and various embodiments for a chatbot event generation system are described herein. An embodiment operates by receiving a request to create an event, and identifying a set of parameters associated with generating the event. A first subset of parameters as provided with the request are identified. A prompt for a large language model (LLM) is generated, and a second subset of parameters is received from the LLM. The event is generated based on the first subset of parameters provided with the request and the second subset of parameters provided by the LLM. The event is detected and an additional action is performed responsive to the detecting the event.