Sentiment Analysis System Using Emoji Overlay for Call Center Monitoring

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Call center agents and customers' sentiment trends are difficult to detect and analyze, making it challenging to determine how agents handle communications, especially negative or aggressive sentiments, in existing electronic communication systems.

Innovation Solution

An electronic communication system that includes a sentiment determination system with a call recorder, transcription module, sentiment extraction engine, database, highlight production module, overlay production module, and score calculation module, which records, transcribes, and analyzes communications to generate sentiment information and display it using emojis, allowing for sentiment trend analysis and performance reviews.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Difficulty of detecting and measuring

If sentiment analysis is implemented in existing electronic communication systems, then sentiment detection capability is improved, but system complexity increases

Engineering Contradiction:
Improvesentiment detection capabilityVSAvoidsystem complexity
Core Design Contradiction:
Difficulty of detecting and measuringVSDevice complexity

Solution Approach 1:

The patent implements sentiment analysis by nesting multiple processing modules within the electronic communication system: a recording module that captures communications, a transcription module that converts speech to text, a sentiment analysis module that processes the text, and a display module that presents results. Each module is embedded within the system architecture, allowing sentiment detection functionality to be integrated without requiring a complete system redesign, thus improving detection capability while managing complexity through modular nesting

Inventive Principle:
Principle #7Nested doll (Nesting)

2Measurement precision

If comprehensive sentiment analysis is performed on all communications, then sentiment analysis accuracy is improved, but processing time increases

Engineering Contradiction:
Improvesentiment analysis accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary transcription of communications into text format before conducting sentiment analysis. This preliminary action prepares the data in advance, allowing the sentiment analysis module to process structured text rather than raw audio, thereby improving analysis accuracy while reducing the time required during the actual sentiment evaluation phase

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent divides the communication analysis process into distinct segments: recording, transcription, sentiment analysis, and display. By segmenting the overall process, each module can be optimized independently - the transcription module handles audio-to-text conversion efficiently, while the sentiment analysis module focuses specifically on text processing, thereby improving overall accuracy without proportionally increasing total processing time

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3905176B1Electronic communication system and method with sentiment analysis
Publication Date: 2024.08.21 MITEL CORP
  • EP3905176B1 patent drawingFigure 1
  • EP3905176B1 patent drawingFigure 2
  • EP3905176B1 patent drawingFigure 3

AI summary

Electronic communication methods and systems for determining sentiment associated with a communication, conveniently displaying indicia of the sentiment information, and scoring the sentiment information are disclosed. The methods and systems can include associating an emoji with a sentiment and annotating information, for example, a highlight reel or a waveform, with the emoji.