Word and Phrase Features in Graph Convolutional Network for Automatic Question Classification

2024-09-04

arXiv.org

AI Summary

To address low accuracy in automatic question classification across skill, difficulty, and competency dimensions in educational settings, this paper proposes Phrase Question-GCN (PQ-GCN): a framework that models questions as linguistic graphs integrating dependency syntax and semantic relations, with words and phrases as nodes and syntactic/semantic associations as edges. PQ-GCN introduces phrase-level embeddings and a multi-granularity graph construction mechanism to jointly enhance the Graph Convolutional Network’s (GCN) capacity for structured semantic modeling. This work is the first to achieve deep integration of phrase-level features with graph-structured representation learning, significantly improving fine-grained classification performance under low-resource conditions. On multiple educational QA benchmarks, PQ-GCN achieves an average 5.2% absolute accuracy gain over conventional word-embedding–based classifiers, with particularly strong improvements in skill identification and difficulty-level discrimination tasks.

Technology Category

• Natural Language Processing: Question Answering
• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Community question answering

Abstract

Effective question classification is crucial for AI-driven educational tools, enabling adaptive learning systems to categorize questions by skill area, difficulty level, and competence. This classification not only supports educational diagnostics and analytics but also enhances complex tasks like information retrieval and question answering by associating questions with relevant categories. Traditional methods, often based on word embeddings and conventional classifiers, struggle to capture the nuanced relationships in natural language, leading to suboptimal performance. To address this, we propose a novel approach leveraging graph convolutional networks (GCNs), named Phrase Question-Graph Convolutional Network (PQ-GCN) to better model the inherent structure of questions. By representing questions as graphs -- where nodes signify words or phrases and edges denote syntactic or semantic relationships -- our method allows GCNs to learn from the interconnected nature of language more effectively. Additionally, we explore the incorporation of phrase-based features to enhance classification accuracy, especially in low-resource settings. Our findings demonstrate that GCNs, augmented with these features, offer a promising solution for more accurate and context-aware question classification, bridging the gap between graph neural network research and practical educational applications.

Problem

• Complex Relationships
• Accuracy Improvement
• Educational Analysis

Innovation

• PQ-GCN
• phrase features
• problem classification