Sugarcane is vital to the national sugar industry and food security; however, its productivity is significantly affected by environmental factors, including temperature, light intensity, soil moisture, and pH. Fluctuations in these variables frequently lead to erratic yields and diminished sugar quality. Data obtained from IoT-based monitoring systems is often affected by noise, absent values, and outliers, complicating analysis. This research employs exploratory data analysis (EDA) on IoT-based sensor data to obtain comprehensive insights into environmental factors influencing sugarcane growth. The dataset contains 1,811 non-null entries from sensors that measure temperature, light, soil moisture, and pH. Data preparation encompassed cleansing, addressing missing values, and eliminating outliers. Univariate and multivariate analyses were conducted to evaluate variable distributions and correlations. The findings indicated that eliminating outliers improved data consistency and showed that temperature and pH had near-normal distributions, whereas light and soil moisture were skewed. A correlation study revealed moderate associations between light and pH, while regression analysis confirmed a favorable relationship between light intensity and pH. This research emphasizes enhancing the dependability and interpretability of IoT-based monitoring data through EDA, providing significant insights for precision agriculture. Future research may concentrate on predictive modeling and real-time decision-support systems to enhance farming operations.
Sari et al. (Wed,) studied this question.