AI tools facilitate research on text data augmentation technology
ZHAO Bingrui
LIU Dongmei
ZHANG Hongnuo
WANG Yang
ZHANG Hejing
LIANG Zheng
QIAO Yixun
Abstract：To address issues such as the scarcity of text data and data distortion,this paper takes Deep Seek as an example,uses netizens'comments as experimental samples,and employs sentiment analysis as verifica-tion methods.It focuses on the application of AI tools in text data enhancement and explores its technical im-plementation,advantages,and challenges.Through experimental verification,the emotion value distribution of the enhanced sample data after data augmentation is similar to that of the original sample,still remaining between-10 and 10 points.The proportion of positive emotion data to negative emotion data still roughly re-mains at 4:6.In the environment where the main body of the text data center remains unchanged,the senti-ment words as subordinate relationships have been expanded,the network relationship of the original text has been enhanced,and the diversity of the text data has been increased.Through experimental verification and other methods,the effects that data augmentation approaches based on AI tools can achieve in AGI develop-ment are demonstrated,and the AI tools and interaction interface methods are analyzed to facilitate subsequent AGI developers in conducting relevant code conversions and provide valuable references for researchers in re-lated fields.
Keywords：data augmentationnatural language processingartificial intelligencelow resources
Publication Date：2025-12-30
Online Publishing Date：2025-12-19(First online date of this platform, not the publication date of the document)
Pages：8( 104-111 )