Custom Model Creation Wizard for Analytics Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The complexity of creating computer models for analyzing machine data requires advanced data science, modeling, and programming skills, limiting users to only pre-defined models and preventing the customization of models for specific use cases.

Innovation Solution

A computer system provides a layer of abstraction through a model creation wizard interface, allowing users to input custom model parameters, which are then transformed into executable models, enabling the creation and execution of custom models.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a fixed set of pre-defined models are provided with the enterprise system, then the system complexity is reduced and ease of operation is improved, but the adaptability to specific use cases deteriorates

Engineering Contradiction:
Improveease of model creationVSAvoidcustomization capability
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

A model creation wizard interface is introduced as an intermediary between the user and the complex model execution environment. The wizard accepts high-level user inputs (time series data, parameter specifications) and automatically generates executable computer models through automated model generation techniques, shielding users from underlying complexity while enabling custom model creation.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system enables users to create their own custom models through the wizard interface without requiring expert data science knowledge. The automated model generation process allows users to self-serve by specifying their own parameters and data sources, and the system automatically constructs and executes appropriate models.

Inventive Principle:
Principle #25Self-service

2Manufacturing precision

If advanced data science and programming skills are required to create models, then the manufacturing precision of models is improved, but the ease of manufacture deteriorates

Engineering Contradiction:
Improvemodel accuracyVSAvoidease of model creation
Core Design Contradiction:
Manufacturing precisionVSEase of manufacture

Solution Approach 1:

Manual model creation processes requiring expert data science skills are replaced with automated model generation techniques. The system uses algorithmic approaches to automatically construct models from user-specified parameters and time series data, substituting the need for manual expert intervention with automated computational processes that maintain model accuracy.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system allows users to specify high-level parameters through the wizard interface (data sources, time ranges, model types) rather than requiring detailed programming specifications. The automated generation process translates these parameter specifications into accurate executable models, maintaining precision while simplifying the creation process.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If custom models are allowed for specific use cases, then the adaptability is improved, but the device complexity increases

Engineering Contradiction:
Improvecustom model supportVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The model creation wizard serves as an intermediary layer that manages system complexity. It provides a simplified interface for users to request custom models while handling the complex automated model generation processes in the background, thus enabling adaptability without exposing users to system complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The model creation process is segmented into distinct steps within the wizard interface (data selection, parameter specification, model generation, execution). This segmentation allows the complex process of creating custom models to be broken down into manageable stages, reducing the perceived complexity for users while maintaining full adaptability.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11875275B1Custom time series models in computer analytics systems
Publication Date: 2024.01.16 CISCO TECHNOLOGY INC
  • US11875275B1 patent drawing
  • US11875275B1 patent drawing
  • US11875275B1 patent drawing

AI summary

A custom use case framework in a computer analytics system is shown and described. The custom use case framework includes a custom model creation wizard interface that guides a user through submitting custom model parameters of a custom model definition. The computing system transforms custom model parameters of the custom model definition into a custom model. The custom model is executed in an analytics system. Thus, one or more embodiments provide a simplified method for a user to generate a custom model that is executable by a computer system.