STRUCTURED APPROACH TO VISUAL DATA ANALYSIS IN INFORMATION PROCESSING TASKS USING PYTHON

Abstract

This study introduces a structured approach to visual data analysis in information processing tasks using Python libraries, addressing the growing need for effective and reproducible visualization methods in data science. The research systematically evaluates key Python libraries – Matplotlib, seaborn, Pandas Visualization, and Plotly—analyzing their strengths and limitations in relation to data types and analytical objectives. A comprehensive methodology comprising five iterative stages is proposed: rapid preliminary overview, in-depth univariate analysis, bivariate and multivariate analysis, targeted visualization, and final result presentation. These stages ensure systematic data exploration, from identifying initial patterns to generating polished, audience-specific visualizations. The methodology was validated using the “Titanic” dataset, which includes numerical (age, ticket price) and categorical (gender, cabin class) variables. Through the application of the proposed stages, key factors influencing passenger survival were identified, such as gender, cabin class, and ticket price. Visual tools like histograms, scatterplots, heatmaps, and interactive dashboards were employed to uncover trends, correlations, and anomalies, demonstrating the methodology’s ability to support data-driven decision-making. Recommendations for selecting appropriate libraries based on data type, analysis goals, and audience were formulated, enhancing the approach’s practical utility.

The structured approach integrates theoretical principles of data visualization with practical Python-based tools, offering flexibility, reproducibility, and cost-effectiveness compared to commercial platforms like Tableau. By leveraging open-source libraries, the methodology overcomes limitations in automation and customization, making it suitable for diverse applications in scientific research, business analytics, and technological processes.

Future research directions include integrating the methodology with machine learning techniques to automate graph selection, optimizing library performance for large datasets, and developing standardized templates for automated dashboards. This approach has the potential to advance data visualization practices, supporting digital transformation and efficient information processing across industries.