Interaction Party Recommendation Using Historical Pattern Matching

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users face challenges in finding interaction parties that resemble their regular interaction patterns efficiently, requiring substantial time and computing resources due to limited information and uncertainty about review reliability, especially in unfamiliar geographic locations.

Innovation Solution

A recommendation system that utilizes machine learning to analyze user interaction history and other users' histories to suggest local entities matching user preferences based on factors like time, day, and geographic location, using supervised learning to enhance accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If users manually search for interaction parties in unfamiliar geographic locations, then they can find potential interaction parties, but it requires substantial time and computing resources

Engineering Contradiction:

Improvecertainty of finding matching interaction partiesVSAvoidtime for manual searches

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system pre-processes and stores interaction history data in structured formats before queries are made. Historical interaction data is organized by interaction types, locations, and patterns, enabling rapid retrieval and matching without requiring real-time manual analysis during user searches

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

A machine learning model acts as an intermediary between user queries and interaction party databases. The model receives query parameters, automatically matches them against historical data patterns, and returns recommended interaction parties, eliminating the need for users to manually search and compare options

Inventive Principle:

Principle #24Intermediary (Mediator)

2Loss of information

If users manually search for interaction parties with limited information, then they can find potential matches, but computing resources are substantially consumed

Engineering Contradiction:

Improveinformation about interaction patternsVSAvoidcomputing resources

Core Design Contradiction:

Loss of informationVSUse of energy by moving object

Solution Approach 1:

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Manual information gathering and analysis is replaced with automated machine learning models that process structured historical data. The system substitutes computational algorithms for human information processing, efficiently handling large datasets without proportional increases in computing resource consumption

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Device complexity

If the system recommends interaction parties without using machine learning, then it is simpler, but accuracy and certainty of recommendations are reduced

Engineering Contradiction:

Improvesystem complexityVSAvoidaccuracy of recommendations

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The recommendation system is segmented into distinct functional modules: data collection module that gathers interaction history, processing module that structures the data, machine learning model module that analyzes patterns, and output module that generates recommendations. This modular architecture manages complexity while enabling sophisticated analysis

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The machine learning model serves multiple functions: it analyzes historical interaction patterns, predicts user preferences, ranks interaction parties by relevance, and adapts to different user contexts. This multi-functionality justifies the added complexity by delivering superior recommendation accuracy across diverse scenarios

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12481664B2Machine learning model for recommending interaction parties

Publication Date: 2025.11.25 CAPITAL ONE SERVICES LLC

AI summary

In some implementations, a system may receive interaction data associated with interactions between a user and subsets of a plurality of interaction parties. The system may store the interaction data and the as historical interaction data associated with historical interactions of the user. The system may provide the historical interaction data as input to a machine learning model, which may be trained using supervised learning and the historical interactions of the user or historical interactions of one or more other users with one or more of the plurality of interaction parties. The system may receive an output, based on applying the machine learning model to the historical interaction data, that may indicate one or more recommended interaction parties based at least in part on one or more factors, wherein the one or more recommended parties may be local entities local to a geographic location associated with the user.