Client Device Deal Filtering with Local User Profiles

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Merchants face challenges in targeting deals to specific consumers while maintaining consumer privacy, as existing systems often require user profiles to be stored on servers accessible via the internet, compromising privacy.

Innovation Solution

A system where user profiles are stored on client devices, allowing consumers to control their data, and deals are filtered and displayed based on relevance using a user profile within the client device, using metrics like latent Dirichlet allocation to determine deal relevance and render deals in a visual hierarchy on a mobile application.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If user profiles are stored on servers accessible via the internet, then deal targeting and personalization can be achieved, but consumer privacy is compromised

Engineering Contradiction:

Improvedeal targeting capabilityVSAvoidprivacy compromise

Core Design Contradiction:

Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The patent extracts the user profile data from centralized servers and stores it locally on consumer devices. This allows the profile information to remain accessible for deal targeting while removing it from internet-accessible servers, thereby protecting consumer privacy while maintaining personalization capabilities

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an intermediary mechanism where deal relevance is determined by computing similarity between deal attributes and user profile attributes locally on the device. This intermediary process enables personalized deal selection without requiring direct access to user profile data on external servers

Inventive Principle:

Principle #24Intermediary (Mediator)

2Object-affected harmful factors

If user profiles are stored locally on client devices, then consumer privacy is maintained, but system complexity increases

Engineering Contradiction:

Improveprivacy protectionVSAvoidlocal processing complexity

Core Design Contradiction:

Object-affected harmful factorsVSDevice complexity

Solution Approach 1:

The patent transforms the user profile into a structured format with defined attributes (e.g., category preferences, price sensitivity) that can be systematically compared with deal attributes. This parameterization enables automated relevance computation using similarity metrics, reducing the complexity of local profile processing while maintaining privacy

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If deals are filtered and displayed based on relevance using local user profiles, then deal personalization is achieved, but computational resources on client devices are consumed

Engineering Contradiction:

Improvedeal personalizationVSAvoidclient device computational energy

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent computes relevance scores for multiple deals against the user profile and selects only the top-N most relevant deals for display. This partial action approach processes a larger set of deals computationally but presents only a subset to the user, balancing personalization quality with energy consumption by limiting the scope of displayed results

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS10762535B2Client-based deal filtering and display

Publication Date: 2020.09.01 BYTEDANCE INC

AI summary

Various implementations relate to a system that provides a collection of deals to a mobile application on a client device. The mobile application then filters the deals based on relevance using a user profile within the client device, and displays the deals to a user. The display of the deals may be governed by the relevance of the respective deals as determined by matching of the deals with the user profile within the client device.