The Correlation Between Individualism Level and the Trend of Using Rare Characters in Personal Names: an Exploration Based on Big Data

The intricate relationship between individual behavior and cultural norms provides a rich canvas for sociological research. A subtle yet persuasive reflection of this interaction is the choice of personal names, which inherently captures the zeitgeist of cultural values and identity expression. Therefore, exploring the relationship between individualism levels and the trends in the use of rare characters in names has become an important task. In this context, individualism is a cultural dimension that emphasizes personal autonomy and self-expression, thought to influence the rarity of characters chosen in names. This study initiates an empirical analysis, harnessing the computational prowess of neural networks and vast databases, to explore the link between individualism and the preference for the use of rare characters in personal names. Against this backdrop, examining the correlation between levels of individualism and the trend of using rare characters in names becomes an important task. This work combines it with neural networks, proposing a deep learning network model to predict the level of individualism based on the usage of rare characters in names. In order to extract more detailed feature information, the technique uses a lightweight Xception network as its foundational model and then employs multi-scale depthwise separable convolution modules of varying sizes for feature extraction and fusion. As an added bonus, the multi-scale convolutional neural network model can zero in on discriminative features thanks to a convolutional attention module that is integrated on top of it. Our method achieved an accuracy of 95.2% and a recall rate of 96.9% when validating the relationship between individualism levels and the trends in the use of rare characters in names, demonstrating the superiority of this approach.