Year: 2024
Type: conference-paper
Abstract: With the deepening implementation of the “Double First Class” construction strategy, China's graduate education has entered a new stage of high-quality development. As an important indicator for measuring graduate students' academic literacy and innovation ability, the importance of the quality monitoring mechanism for master's theses is becoming increasingly prominent. Faced with the limitations of subjectivity, low efficiency, and inconsistent standards in traditional manual review models, exploring a more scientific, objective, and efficient thesis evaluation system has become an urgent task. This article innovatively proposes and designs a graduate thesis quality inspection system driven by artificial intelligence (AI), deeply integrating deep learning (DL) technology and big data analysis capabilities. By constructing a multi-level natural language processing (NLP) model, the system can automatically extract key information from these, evaluate academic contributions, identify citation norms and potential plagiarism behavior, and conduct comparative analysis based on massive academic literature data, achieving comprehensive and accurate evaluation of qualities. The experimental results show that the system can not only significantly improve the efficiency of paper evaluation and reduce bias caused by human factors but also provide personalized feedback for graduate students.
Authors: Jingwei Lv, Nana Zhao, Kainan Li, Shikui Cao
Institution: Changchun University of Chinese Medicine
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: gold
OpenAlex ID: W4412524842