Machine Learning Model for Social Media Complaint Categorization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Businesses face challenges in monitoring and collecting customer complaints lodged on social media platforms, as these complaints are decentralized and difficult to track, posing risks for regulatory enforcement and compliance.

Innovation Solution

A system utilizing machine learning models to categorize and identify customer complaints from social media posts by training on existing customer interaction data, allowing for automated tagging and cataloging of risks, and enabling notifications for potential regulatory violations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If businesses monitor customer complaints on social media platforms, then compliance with regulatory provisions is improved, but the complexity of monitoring and collecting decentralized complaints increases

Engineering Contradiction:

Improvecompliance with regulatory provisionsVSAvoidcomplexity of monitoring and collecting complaints

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary system comprising a scraping service, database, and machine learning model that mediates between decentralized social media platforms and the business. The scraping service collects complaints from various social media platforms, the database stores and organizes them, and the machine learning model categorizes and identifies relevant complaints, thereby simplifying the monitoring process while maintaining compliance reliability

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces manual mechanical monitoring processes with automated machine learning-based systems. Instead of human analysts manually reviewing social media platforms, the system uses trained machine learning models to automatically detect, categorize, and prioritize complaints, significantly reducing operational complexity while improving compliance monitoring effectiveness

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If businesses manually monitor customer complaints on social media, then identification of relevant complaints is improved, but the time required to process vast amounts of data increases

Engineering Contradiction:

Improveidentification of relevant complaintsVSAvoidtime required to process data

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual data processing with automated machine learning models that rapidly analyze vast amounts of social media data. The trained models automatically detect patterns, categorize complaints, and identify relevant issues in real-time, eliminating the time-consuming manual review process while maintaining high measurement precision in complaint identification

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent implements preliminary action through pre-trained machine learning models that are ready to immediately process and categorize complaints upon receipt. The system performs preliminary classification and prioritization automatically, so that when new complaints arrive, they are instantly processed without requiring manual preparation or review, significantly reducing time loss

Inventive Principle:

Principle #10Preliminary action

3Reliability

If businesses collect all customer complaints from social media, then comprehensive monitoring is improved, but the volume of data to process increases

Engineering Contradiction:

Improvecomprehensive monitoringVSAvoidvolume of data to process

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent applies the extraction principle by using machine learning models to identify and extract only the relevant and actionable complaints from the vast volume of social media data. The trained models filter out irrelevant information and focus on extracting complaints that meet specific criteria (such as regulatory relevance, severity, and impact), thereby maintaining comprehensive monitoring while reducing the effective data volume that requires processing

Inventive Principle:

Principle #2Taking out (Extraction)

4Productivity

If businesses use automated systems to process social media data, then processing speed is improved, but the complexity of the system increases

Engineering Contradiction:

Improveprocessing speedVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent uses an intermediary machine learning model layer that simplifies the interaction between raw social media data and business decision-making. The model acts as a mediator that automatically processes data, categorizes complaints, and provides actionable insights, thereby achieving high processing speed while the model's trained nature reduces the operational complexity of managing and interpreting the system

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11935522B2Cognitive analysis of public communications

Publication Date: 2024.03.19 CAPITAL ONE SERVICES LLC

AI summary

Disclosed herein are system, method, and computer program product embodiments for categorizing customer complaints on social media using a model trained on customer voice calls or chats with agents. Additionally, users interested in monitoring regulatory compliance issues based on customer complaints can receive notifications regarding complaints that are linked to regulatory topic areas, without the need to manually scan vast numbers of social media postings.