Research on Statistical Analysis of Best-selling Books Sales Data Based on Python

Abstract

To accurately identify core operational characteristics and development patterns in the bestseller market and provide data-driven support for industry decision-making, this study analyzes 2,000 sales records of best-selling books from 1982 to 2023. Using Python data analysis techniques, we established a comprehensive research framework encompassing "data preprocessing, normalization, and visualization." During the analysis, we addressed data quality issues such as missing values and formatting inconsistencies through mean filling and format standardization. Dimensional differences were eliminated using Min-Max normalization and Z-Score standardization, while multidimensional visualization was conducted with tools like Matplotlib and Seaborn. The study systematically explored intrinsic correlations between key dimensions including book pricing, ranking performance, review feedback, and author influence. Results reveal that the bestseller market exhibits a "mid-range pricing dominance with moderate discounts for traffic diversion" characteristic, with the 20-100 yuan price bracket and 40%-60% discount range demonstrating highest market acceptance. Correlation analysis after data normalization shows a strong negative correlation between ranking frequency and ranking position (r=-0.82), while review growth correlates with diverging recommendation values (correlation coefficient r=-0.51). Top authors like Keigo Higashino maintain consistent bestseller effects, with their works excelling in rankings, review metrics, and recommendation values. The findings not only provide scientific evidence for readers 'purchasing decisions, publishers' topic planning, and sales platform algorithm optimization, but also facilitate the publishing industry's transition from experience-driven to data-driven development.