Automated Answer Evaluation System for Test Paper Scoring

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional marking methods in education are time-consuming and subjective, leading to errors and high workload for teachers, especially when evaluating objective and subjective questions, which require precise answering methods.

Innovation Solution

An answer evaluation method using a pre-trained test question classification model to differentiate between objective and subjective question areas in an image of a test paper, followed by optical character recognition to identify and score questions, and a system that determines a total score value based on these classifications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional manual marking is used, then teachers can evaluate student answers, but the workload increases and time is consumed

Engineering Contradiction:

Improveevaluation accuracyVSAvoidmarking time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces the mechanical manual marking system with an automated image recognition and processing system. The system captures images of student test papers, uses OCR to recognize text, classifies questions into objective and subjective types, and automatically scores them, thereby eliminating the time-consuming manual marking process while maintaining evaluation accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent enables the test paper evaluation system to perform self-service by automatically completing the entire marking process without human intervention. The system independently captures paper images, recognizes text content, classifies question types, scores answers, and generates results, freeing teachers from the manual marking workload

Inventive Principle:

Principle #25Self-service

2Ease of operation

If manual review of subjective questions is performed, then evaluation can be conducted, but subjectivity causes large errors

Engineering Contradiction:

Improvemarking convenienceVSAvoidscoring accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent segments the evaluation process into distinct components: objective question scoring and subjective question scoring. For subjective questions, it further segments the analysis into keyword extraction, matching against standard answers, and automated scoring. This segmentation enables systematic processing that reduces subjective bias while maintaining operational simplicity

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary computational layer between the student's subjective answer and the final score. This intermediary process includes text recognition, keyword extraction, comparison with standard answers, and algorithmic scoring, which mediates the evaluation to reduce human subjectivity while keeping the process easy to operate

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If electronic marking of objective questions is implemented, then efficiency is improved, but requirements for answering methods increase

Engineering Contradiction:

Improvemarking efficiencyVSAvoidanswering convenience

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent changes the recognition parameters of the electronic marking system to accommodate various student answering methods. The OCR and image recognition systems are configured to recognize multiple formats including handwritten answers, printed text, and different answer positions, thereby maintaining high marking efficiency while allowing students flexibility in how they answer

Inventive Principle:

Principle #35Parameter changes

4Speed

If automated classification is used, then processing speed increases, but system complexity increases

Engineering Contradiction:

Improveevaluation speedVSAvoidsystem complexity

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

The patent segments the automated classification system into distinct functional modules: image capture module, OCR text recognition module, question type classification module, and scoring module. Each module performs a specific function with clear input-output interfaces, which enables fast processing while managing system complexity through modular design

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by pre-training the classification model with labeled data and pre-configuring the scoring criteria for different question types. This preliminary preparation enables the system to process new test papers quickly without complex real-time decision-making, balancing speed with manageable system complexity

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11790641B2Answer evaluation method, answer evaluation system, electronic device, and medium

Publication Date: 2023.10.17 BOE TECHNOLOGY GROUP CO LTD

AI summary

The present disclosure provides an answer evaluation method, an answer evaluation system, an electronic device, and a medium. The method comprises: acquiring an answer image, for a test paper answered by a use; classifying the answer image based on a pre-trained test question classification model, so as to obtain an objective question answer area and a subjective question answer area; identifying at least one objective question in the objective question answer area and an objective question answers for each of the at least one objective question; identifying at least one subjective question in the subjective question answer area and a subjective question answers for each of the at least one subjective question; and determining a total score value of the test paper based on the objective question, the objective question answer, the subjective question and the subjective question answer.