Automated Predictive Model Generation for Lead Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The manual generation of business decision analytic models is cumbersome and prone to errors due to the exponential growth of data, requiring significant time and manpower, and lacks adaptability to changing data inflows, necessitating an automated system that can convert unstructured data into structured information and regenerate models based on new data.
Innovation Solution
A computer-implemented system and method that automates the generation of analytic models by processing data sets to derive elapsed-time variables, creating comparison models, and selecting operational models based on quality metrics, while converting unstructured data into structured information using natural language processing and model testing to ensure model efficiency and adaptability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual model generation is used, then model accuracy can be ensured through human judgment, but the process requires huge investments in time and manpower
Solution Approach 1:
The system enables automated model generation where the computational system performs model creation, variable selection, and optimization autonomously without requiring human experts to manually build each model. The system serves itself by automatically processing data, selecting variables, and generating predictive models based on predefined algorithms and quality metrics.
Solution Approach 2:
The system automatically adjusts model parameters, variable selections, and model configurations through computational algorithms rather than manual human judgment. By changing the approach from human-driven parameter selection to algorithm-driven parameter optimization, the system reduces time investment while maintaining model quality through systematic evaluation metrics.
2Reliability
If manual model generation is used, then models can be created with human expertise, but the process is cumbersome and prone to errors
Solution Approach 1:
The system replaces the manual mechanical process of model generation with an automated computational system. Instead of human experts manually selecting variables and building models (a mechanical human process), the system uses algorithms and computational methods to automatically perform these tasks, reducing human error and process complexity.
Solution Approach 2:
The system incorporates automated feedback mechanisms where model quality is continuously evaluated against predefined metrics, and the model generation process is iteratively improved based on this feedback. This systematic feedback loop ensures model reliability while reducing the complexity of manual quality assurance processes.
3Adaptability or versatility
If traditional model generation systems are used, then models can be created based on existing data, but they lack adaptability to changing data inflows
Solution Approach 1:
The system implements dynamic model generation where models are automatically updated and regenerated in response to changing data inflows. Rather than static models created once, the system continuously adapts models to new data conditions through automated processes, making the modeling system dynamic and responsive to changing business environments.
Solution Approach 2:
The system creates a universal automated framework that can handle multiple data types, model types, and business scenarios through a single automated platform. This multi-functional system can process structured and unstructured data, generate various types of predictive models, and adapt to different business domains, enhancing adaptability while maintaining high automation levels.
4Quantity of substance
If large volumes of data are used, then model comprehensiveness improves, but variable selection becomes a cumbersome task
Solution Approach 1:
The system automatically extracts and selects relevant variables from large volumes of available data using computational algorithms. Instead of human experts manually sifting through numerous variables, the system automatically identifies and extracts the most relevant variables based on their predictive power and relationships with target outcomes, making variable selection easier despite increased data volume.
Solution Approach 2:
The system uses automated copying and replication of successful variable selection patterns from historical data to identify relevant variables in new datasets. By copying proven variable selection methodologies and automatically applying them to large datasets, the system maintains ease of operation while effectively handling increased data volumes.
Data Source
AI summary
A computer implemented system for automating the generation of an analytic model includes a processor configured to process a plurality of data sets. Each data set includes values for a plurality of variables. A time-stamping module is configured to derive values for a plurality of elapsed-time variables for each data set, and the plurality of variables and plurality of elapsed-time variables are included in a plurality of model variables. A model generator is configured to create a plurality of comparison analytic models each based on a different subset of model variables. Each comparison analytic model is configured to operate on new data sets associated with current leads, and to output a likelihood of successfully closing an associated transaction. A model testing module is configured to select an operational analytic model from among the comparison analytic models based on a quality metric.