Binary Search Scoring for Lending Query Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complicated data searches involving multiple negotiable parameters often require computationally intensive brute-force processing, making real-time or near real-time results difficult to obtain.
Innovation Solution
The use of binary searches and analytical models to identify minimum and maximum entries within a data set, generating scores for entries between these boundaries, and returning responsive entries based on their scores, thereby reducing the need for extensive brute-force processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If brute-force iteration is used to search data with multiple negotiable parameters, then search accuracy is improved, but processing time and computational complexity increase significantly
Solution Approach 1:
The patent segments the search space by identifying minimum and maximum boundary entries first, then dividing the remaining entries into a set between these boundaries. This segmentation allows the system to focus computational effort on relevant portions of the data rather than iterating through all entries brute-force, thereby maintaining search accuracy while reducing processing time.
Solution Approach 2:
The patent performs preliminary actions by first conducting binary searches to identify minimum and maximum boundary entries before processing the actual search query. This preliminary boundary identification creates a focused search space, eliminating the need for exhaustive brute-force iteration across the entire dataset and significantly reducing computational time while preserving accuracy.
2Reliability
If brute-force iteration is used to search data with multiple negotiable parameters, then comprehensive result coverage is improved, but computational efficiency deteriorates
Solution Approach 1:
The patent segments the dataset into boundary entries (minimum and maximum) and a set of intermediate entries between these boundaries. This segmentation enables comprehensive coverage of relevant results by systematically processing entries within the identified range while excluding irrelevant entries outside the range, thereby maintaining result coverage without requiring exhaustive brute-force computation across the entire dataset.
Solution Approach 2:
The patent extracts and identifies minimum and maximum boundary entries through binary search before processing the main search query. By taking out these boundary definitions first, the system establishes clear limits for the search space, allowing comprehensive coverage of relevant results within these boundaries while improving computational efficiency by excluding entries outside the range.
3Speed
If analytical models are used to generate scores for entries, then processing speed is improved, but model complexity increases
Solution Approach 1:
The patent applies analytical models to generate scores for boundary entries (minimum and maximum) and intermediate entries in advance, storing these scores for later retrieval. This preliminary scoring action eliminates the need to recompute scores during query processing, significantly improving processing speed. The model complexity is justified by the elimination of repeated computations across multiple queries.
Solution Approach 2:
The patent applies analytical models selectively to only those entries that fall within the identified minimum and maximum boundaries, rather than computing scores for all entries in the dataset. This partial application of the analytical model reduces unnecessary computational overhead while maintaining processing speed for relevant entries, balancing model complexity with actual processing needs.
Data Source
AI summary
A search request may define a minimum value and a maximum value for a numerical parameter of the search. A plurality of binary searches may identify a minimum entry and a maximum entry, each binary search comprising selecting a single entry and using at least one analytical model to generate a score for the single entry, wherein the score of the minimum entry corresponds with the minimum value and the score of the maximum entry corresponds with the maximum value. The at least one analytical model may be used to generate a score for each of a set of entries between the minimum entry and the maximum entry while ignoring entries outside of the set of entries between the minimum entry and the maximum entry. Entries responsive to the request may be identified within the scored set of entries.