Reply Content Scoring Using Candidate Assessment Point Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods struggle to accurately assess employee skills due to diverse and lengthy reply content, leading to increased scoring difficulty and reduced accuracy in customer service evaluations.

Innovation Solution

A scoring method that extracts M groups of candidate assessment points from reply content using a sliding window, removes characters based on smoothness criteria, and calculates scores considering similarity and polarity, ensuring accurate and efficient evaluation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If reply content is directly compared with standard answers, then scoring process is simple, but scoring accuracy deteriorates due to diverse and lengthy content forms

Engineering Contradiction:
Improvescoring accuracyVSAvoidscoring process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the reply content into multiple fixed-length character sequences (e.g., 5-character windows with step size of 1) to generate candidate assessment points. This segmentation transforms the complex task of evaluating diverse and lengthy reply content into a series of simpler comparisons between standardized segments and target assessment points, thereby improving scoring accuracy while managing process complexity.

Inventive Principle:
Principle #1Segmentation

2Productivity

If reply content is segmented into fixed-length sequences, then scoring efficiency is improved, but information loss may occur

Engineering Contradiction:
Improvescoring efficiencyVSAvoidcontent information loss
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent extracts candidate assessment points from segmented reply sequences by comparing them with target assessment points. The extraction process identifies and selects meaningful segments that match the evaluation criteria, thereby maintaining scoring efficiency while minimizing information loss by focusing on relevant content rather than processing all segments equally.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent adjusts parameters such as window size and step size to optimize the balance between scoring efficiency and information retention. By changing these parameters, the system can adapt to different content types and lengths, improving efficiency while preserving essential information through appropriate segment overlap and coverage.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4390795B1Scoring method and related apparatus
Publication Date: 2026.01.28 HUAWEI TECH CO LTD
  • EP4390795B1 patent drawingFigure 1~2
  • EP4390795B1 patent drawingFigure 3~4
  • EP4390795B1 patent drawingFigure 5~6

AI summary

Embodiments of this application disclose a scoring method and a related apparatus, to score reply content of a user, so as to accurately assess an employee. The method in embodiments of this application includes: obtaining reply content of a user to a target question; extracting M groups of candidate assessment points from the reply content based on M target assessment points, where the M target assessment points are obtained based on a target answer to the target question, each group of candidate assessment points includes N candidate assessment points and corresponds to one target assessment point, each candidate assessment point includes a plurality of consecutive characters in the reply content, and both M and N are positive integers; and obtaining a score of the reply content based on the M groups of candidate assessment points and the M target assessment points.