Recommendation Engine Filtering by Historical Feature Usage

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional recommendation engines fail to provide search results for devices that consider the user's historical feature usage of their current device, leading to unhelpful and time-consuming shopping experiences as they include features not important to the user.

Innovation Solution

A method to modify search results by monitoring and analyzing user interactions with their current device to determine frequently and infrequently used features, filtering out devices that do not include important features, and providing customized recommendations based on historical usage patterns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If conventional recommendation engines provide search results based on shopping patterns of other users, then the quantity of search results is increased, but the relevance of search results to the user's actual needs deteriorates

Engineering Contradiction:

Improvequantity of search resultsVSAvoidrelevance of search results to user needs

Core Design Contradiction:

Quantity of substanceVSLoss of information

Solution Approach 1:

The system performs preliminary action by monitoring and analyzing the user's historical interactions with their current device before the shopping process begins. Usage data is collected and stored in advance, allowing the recommendation engine to pre-determine which features are important to the user based on actual usage patterns rather than generic shopping patterns.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by continuously monitoring how the user interacts with the current device and using this information to refine future recommendations. The recommendation engine analyzes usage data, compares it against product features, and adjusts search results accordingly, creating a closed-loop system that learns from user behavior.

Inventive Principle:

Principle #23Feedback

2Device complexity

If conventional recommendation engines do not consider user's historical feature usage, then the device complexity is reduced, but the time required for user to identify suitable products increases

Engineering Contradiction:

Improvecomplexity of recommendation systemVSAvoidtime required for shopping process

Core Design Contradiction:

Device complexityVSLoss of time

Solution Approach 1:

The system performs preliminary action by collecting and storing usage data in advance before the user needs to shop for a replacement device. By pre-analyzing historical interactions and determining important features beforehand, the system eliminates the need for users to manually review extensive product listings, significantly reducing shopping time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements self-service by automatically monitoring, analyzing, and using the user's own usage data to generate personalized recommendations without requiring active input from the user during the shopping process. The system serves itself by leveraging the user's historical behavior patterns to autonomously determine relevant product features.

Inventive Principle:

Principle #25Self-service

3Adaptability or versatility

If search results include devices without features the user frequently uses, then the coverage of product options is improved, but the ease of operation for users deteriorates

Engineering Contradiction:

Improvecoverage of product optionsVSAvoidease of identifying suitable devices

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system extracts and removes from search results devices that do not include features the user frequently uses. By analyzing historical usage data to identify important features, the system selectively extracts and eliminates irrelevant options, presenting only devices that match the user's actual needs and preferences.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system applies local quality by tailoring search results to the specific user's preferences and usage patterns rather than providing a generic list. Each user receives customized recommendations based on their individual historical feature usage, making the search results locally optimized for their specific needs.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20230385346A1Filtering Results Based on Historic Feature Usage

Publication Date: 2023.11.30 CAPITAL ONE SERVICES LLC

AI summary

Aspects described herein may provide modification of recommendations from a recommendation engine for a product or a device. The recommendation engine may provide initial search results for a particular type of device. The initial search results may be modified to ensure inclusion of devices that include features that the user considers important. Features that the user considers important may be determined based on observing the user's interaction with another device of the same type. By observing the user's interaction with the other device over a period of time, features that the user commonly uses and features that the user sparingly uses may be determined. The initial search results may then be modified to remove devices that do not include the features frequently used by the user and therefore considered important to the user.