Identity Verification Using Acquaintance Account Groups

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing social networking service (SNS) reliability analysis systems fail to accurately determine whether multiple accounts belong to the same person, especially when users post different content on friend accounts, leading to inaccuracies in similarity degree calculations.

Innovation Solution

An information processing apparatus and method that determines first and second acquaintance account groups associated with target accounts, and decides identity based on similarity between person groups using these account groups, incorporating SNS information and likelihood computations to assess identity matching.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If similarity degree is calculated based on posted images from friend accounts, then the system can leverage available social networking data, but the accuracy deteriorates when the same person posts different images on different friend accounts

Engineering Contradiction:

Improveamount of SNS data utilizedVSAvoidaccuracy of identity determination

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent segments the identity verification process into multiple independent similarity calculations: (1) similarity between target account's posted images, (2) similarity between acquaintance account groups, and (3) combination of these similarities with weighting. This segmentation allows the system to use diverse data sources while maintaining accuracy by not relying on a single potentially misleading metric.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent merges multiple similarity metrics (posted image similarity and acquaintance account group similarity) into a combined similarity degree calculation. By integrating these different types of evidence and applying appropriate weighting, the system overcomes the limitation of using only posted images from friend accounts, which may be inconsistent when the same person uses multiple accounts.

Inventive Principle:

Principle #5Merging (Combining)

2Productivity

If reliability analysis prioritizes friend account similarity when target account reliability is low, then the system can make decisions with limited direct evidence, but the determination accuracy worsens due to inconsistent posting behavior across friend accounts

Engineering Contradiction:

Improvespeed of identity decisionVSAvoidaccuracy of similarity degree

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent implements dynamic weighting of similarity metrics based on the reliability degrees of the target accounts. When target account reliability is low, the system dynamically adjusts the weighting to incorporate acquaintance account group similarity more heavily, rather than rigidly prioritizing friend account similarity. This dynamic adaptation maintains accuracy across different reliability scenarios.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the parameters of the similarity calculation by introducing reliability degrees as weighting factors. Instead of using a fixed formula, the system adjusts the contribution of different similarity metrics (posted image similarity vs. acquaintance account group similarity) based on the reliability parameters of the accounts involved, thereby maintaining accuracy across varying conditions.

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If the system uses posted images from friend accounts to determine similarity, then it can utilize existing SNS data structures, but the complexity of accurately representing person identity increases due to varied posting patterns

Engineering Contradiction:

Improvecompatibility with SNS data structureVSAvoidcomplexity of identity representation

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary structure called 'acquaintance account groups' that mediate between the raw SNS data (friend accounts and posted images) and the identity determination process. This intermediary organizes the data into meaningful groups based on relationships, simplifying the representation of person identity while maintaining compatibility with existing SNS data structures.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent adds a new dimension to the analysis by considering not just the content of posted images but also the relational structure of acquaintance accounts. This dimensional expansion transforms the problem from simple image matching to a multi-dimensional assessment that includes social relationship patterns, thereby simplifying the overall identity representation through structural insights.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS20240422231A1Information processing apparatus, information processing method, and storage medium

Publication Date: 2024.12.19 NEC CORP

AI summary

To accurately decide identity of a person using a plurality of target accounts, an information processing apparatus includes a determination unit and a decision unit. The determination unit determines a first acquaintance account group and a second acquaintance account group being associated with a first target account and a second target account, respectively. The decision unit decides identity of a person using the first target account and the second target account, based on similarity between a first person group and a second person group using the first acquaintance account group and the second acquaintance account group, respectively.