Data Sheet 1_Research on an intelligent tutoring system based on automatic construction of

Chao Deng's research proposes a framework for an Intelligent Tutoring System using automatic construction of multimodal knowledge graphs and Retrieval-Augmented Generation. The 30.9 KB document, uploaded to figshare on 2026-03-17, describes a pipeline integrating FFmpeg, Whisper, OCR, and layout analysis to extract knowledge from course videos and textbook PDFs. The study reports experimental results on knowledge graph construction, Q&A accuracy, and user satisfaction.

Use Cases

• Develop automated knowledge graph construction pipelines based on the described integration of FFmpeg, Whisper, OCR, and layout analysis.
• Benchmark intelligent tutoring system performance based on metrics like Q&A accuracy and user satisfaction mentioned in the experimental results.
• Implement cross-modal fusion mechanisms for educational content based on the described alignment of auditory explanations with visual-textual textbook knowledge.
• Design RAG-driven educational Q&A systems based on the framework combining graph and vector retrieval strategies.

Strengths

• Proposes a fully automatic pipeline for knowledge graph construction from videos and PDFs, addressing a noted bottleneck of expert-dependent manual construction.
• Integrates multiple technologies (FFmpeg, Whisper, OCR, layout analysis) for cross-modal knowledge fusion from auditory, visual, and textual sources.
• Released under a permissive CC-BY-4.0 license, allowing for broad reuse and adaptation.

Limitations

• The dataset is a 30.9 KB DOCX file describing a research framework; the actual constructed knowledge graphs, Q&A data, or system outputs are not included.
• Row count and column-level documentation for any underlying data are unknown, limiting suitability assessment.
• Description metadata is limited; actual data quality and implementation details require manual inspection of the document.

Provenance

Source

Chao Deng via figshare.

Collection Method

Describes a research framework and experimental results; the method for gathering any underlying data is not specified in the provided metadata.

Freshness

Last updated 2026-03-17 13:05:07; freshness should be verified.