AIGC Check Service System of CNKI

Overview

AIGC Check Service System of CNKI relies on the high-quality literature resources of CNKI which are structured, fragmented, and knowledge-based. It combines advanced knowledge enhancement large model technology and natural language processing algorithms to rapidly and accurately identify varying degrees of suspected AI-generated contents in academic texts in both Chinese and English.

Services

AIGC Check Service System of CNKI can quickly and accurately identify the AI-generated contents in academic texts including academic papers, dissertations, academic monographs and scientific research achievements by comparing language models and semantic logic differences, and display the suspected AI-generated rate with clear and intuitive reports.

Features

Abundant literature comparison sources: The system has built a massive, professional, complete, frequently updated, and clearly copyrighted full-text comparison resource library. It covers all the contents of China Integrated Knowledge Resources Database, foreign language literature resources, minority language literature resources, academic literature resources of Hong Kong, Macao and Taiwan, Internet resources, literature resources under priority publication, book resources, CNKI-edited original works resources, joint comparison database for university students’ papers etc., and supports open resource construction.

Advanced semantic analysis and comparison technology: The system adopts the AMLFP feature detection technology independently developed by CNKI, and it is supported by mature SmartTextMiner knowledge mining technology and international advanced NLPE technology. It can effectively combine content-based information fingerprint technology with semantic analysis technology to provide users with professional and efficient detection services.

Features

Multilingual and cross-language check: The system supports the check of literature in simplified Chinese, traditional Chinese, English, Japanese, French and other languages, and uses large-scale Chinese-English corpuses for analysis to automatically generate translation rules and effectively detect translation plagiarism. It supports the check of Chinese texts, English texts and Chinese-English texts and can effectively identify the AIGC-generated content.

Multi-type and Multi-format Text Check: The system supports the detection of various types of literature texts, including journal articles, dissertations and theses, conference proceedings, and books. Users can directly input texts for detection, or upload files in various formats, including doc, docx, pdf, txt, caj, kdh, nh, and rtf, to improve the convenience and efficiency of detection.

Features

Multi-level and Visualized Report Presentation: The system offers two types of reports, a comprehensive report that provides detailed information on the detection process to enable users to conduct thorough analysis and comprehension, and a succinct report that emphasizes key points and facilitates a rapid understanding on the detection results.

Text and User Data Protection: The system adopts advanced encryption techniques to ensure that the papers uploaded by users will not be disclosed or abused in the check process. At the same time, the system strictly adheres to relevant laws, regulations and privacy policies to ensure data security and privacy.

Slide Down