Machine Learning Prospect Scoring for Hard-to-Borrow Securities Lending
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems are inefficient and time-consuming in identifying and prioritizing households with clients interested in participating in securities lending programs for hard-to-borrow securities, leading to delays in fulfilling time-sensitive requests.
Innovation Solution
A machine learning-based framework that utilizes ML models to generate model scores, filter households based on exclusion rules, and prioritize households for securities lending programs, generating user interfaces with prioritized households for timely communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual identification and prioritization methods are used, then system complexity is low, but productivity and speed are insufficient
Solution Approach 1:
The patent replaces manual mechanical identification processes with machine learning models that automatically analyze household data, generate model scores, and identify potential customers. This substitution dramatically increases identification speed while the system complexity is managed through modular architecture comprising data processing modules, scoring modules, and prioritization modules.
Solution Approach 2:
The system enables self-service through automated data processing where the machine learning models independently evaluate household data, generate scores, and prioritize candidates without requiring manual intervention at each step. The automated communication initiation further extends this self-service capability, allowing the system to autonomously reach out to identified prospects.
2Measurement precision
If comprehensive data analysis is performed, then measurement precision improves, but loss of time increases
Solution Approach 1:
The patent implements preliminary action by pre-processing household data into standardized formats, pre-training machine learning models on historical data, and pre-establishing scoring criteria before actual identification begins. This preparation enables rapid, accurate analysis when actual identification is needed, as the heavy computational lifting has already been done in advance.
Solution Approach 2:
The comprehensive data analysis is segmented into distinct processing stages: data collection, data cleaning, feature extraction, model scoring, and prioritization. Each segment can be processed independently and in parallel, reducing overall processing time while maintaining comprehensive analysis through the systematic progression through all segments.
3Productivity
If automated communication is implemented, then productivity increases, but ease of operation decreases
Solution Approach 1:
The automated communication system is designed with dynamic control capabilities, allowing the level of automation to be adjusted based on needs. Users can configure which communication steps are fully automated and which require human review, enabling the system to adapt between high productivity mode and high ease-of-operation mode depending on the situation.
Data Source
AI summary
A system for identifying a targeted set of prospect clients from a pool of existing clients is caused to receive a dataset having securities data, household data, and forecasted data, determine, with a machine learning model, a model score for each household of the households based on the received dataset, filter the households based on the model score for each household and a threshold score to generate remaining households, determine a prioritization score for each remaining household of the remaining households, prioritize the remaining households based on the prioritization score for each remaining household of the remaining households, and generate and display a user interface including a list of the prioritized remaining households having prospect clients from the pool of existing clients. Other example systems, methods, and non-transitory computer readable medium for identifying a targeted set of prospect clients from a pool of existing clients are also disclosed.