This study explores the intricate relationship between emotional cues present in food delivery app reviews, normative ratings, and reader engagement. Utilizing lexicon-based unsupervised machine learning, our aim is to identify eight distinct emotional states within user reviews sourced from the Google Play Store. Our primary goal is to understand how reviewer star ratings impact reader engagement, particularly through thumbs-up reactions. By analyzing the influence of emotional expressions in user-generated content on review scores and subsequent reader engagement, we seek to provide insights into their complex interplay. Our methodology employs advanced machine learning techniques to uncover subtle emotional nuances within user-generated content, offering novel insights into their relationship. The findings reveal an inverse correlation between review length and positive sentiment, emphasizing the importance of concise feedback. Additionally, the study highlights the differential impact of emotional tones on review scores and reader engagement metrics. Surprisingly, user-assigned ratings negatively affect reader engagement, suggesting potential disparities between perceived quality and reader preferences. In summary, this study pioneers the use of advanced machine learning techniques to unravel the complex relationship between emotional cues in customer evaluations, normative ratings, and subsequent reader engagement within the food delivery app context.

Mangrove forests are vital to coastal protection, biodiversity support, and climate regulation. In the Niger Delta, these ecosystems are increasingly threatened by oil spill incidents linked to intensive petroleum activities. This study investigates the extent of mangrove degradation between 1986 and 2022 in the lower Niger Delta, specifically the region between the San Bartolomeo and Imo Rivers, using remote sensing and machine learning. Landsat 5 TM (1986) and Landsat 8 OLI (2022) imagery were classified using the Support Vector Machine (SVM) algorithm. Classification accuracy was high, with overall accuracies of 98% (1986) and 99% (2022) and Kappa coefficients of 0.97 and 0.98. Healthy mangrove cover declined from 2804.37 km2 (58%) to 2509.18 km2 (52%), while degraded mangroves increased from 72.03 km2 (1%) to 327.35 km2 (7%), reflecting a 354.46% rise. Water bodies expanded by 101.17 km2 (5.61%), potentially due to dredging, erosion, and sea-level rise. Built-up areas declined from 131.85 km2 to 61.14 km2, possibly reflecting socio-environmental displacement. Statistical analyses, including Chi-square (χ2 = 1091.33, p < 0.001) and Kendall's Tau (τ = 1, p < 0.001), showed strong correlations between oil spills and mangrove degradation. From 2012 to 2022, over 21,914 barrels of oil were spilled, with only 38% recovered. Although paired t-tests and ANOVA results indicated no statistically significant changes at broad scales, localized ecological shifts remain severe. These findings highlight the urgent need for integrated environmental policies and restoration efforts to mitigate mangrove loss and enhance sustainability in the Niger Delta.

Photovoltaic systems have shown significant attention in energy systems due to the recent machine learning approach to addressing photovoltaic technical failures and energy crises. A precise power production analysis is utilized for failure identification and detection. Therefore, detecting faults in photovoltaic systems produces a considerable challenge, as it needs to determine the fault type and location rapidly and economically while ensuring continuous system operation. Thus, applying an effective fault detection system becomes necessary to moderate damages caused by faulty photovoltaic devices and protect the system against possible losses. The contribution of this study is in two folds: firstly, the paper presents several categories of photovoltaic systems faults in literature, including line-to-line, degradation, partial shading effect, open/close circuits and bypass diode faults and explores fault discovery approaches with specific importance on detecting intricate faults earlier unexplored to address this issue; secondly, VOSviewer software is presented to assess and review the utilization of machine learning within the solar photovoltaic system sector. To achieve the aims, 2258 articles retrieved from Scopus, Google Scholar, and ScienceDirect were examined across different machine learning and energy-related keywords from 1990 to the most recent research papers on 14 January 2025. The results emphasise the efficiency of the established methods in attaining fault detection with a high accuracy of over 98%. It is also observed that considering their effortlessness and performance accuracy, artificial neural networks are the most promising technique in finding a central photovoltaic system fault detection. In this regard, an extensive application of machine learning to solar photovoltaic systems could thus clinch a quicker route through sustainable energy production.

Fire hazard is often mapped as a static conditional probability of fire characteristics’ occurrence. We developed a dynamic product for operational risk management to forecast the probability of occurrence of fire radiative power in the locally possible near-maximum fire intensity range. We applied standard machine learning techniques to remotely sensed data. We used a block maxima approach to sample the most extreme fire radiative power (FRP) MODIS retrievals in free-burning fuels for each fire season between 2001 and 2020 and associated weather, fuel, and topography features in northwestern south America. We used the random forest algorithm for both classification and regression, implementing the backward stepwise repression procedure. We solved the classification problem predicting the probability of occurrence of near-maximum wildfire intensity with 75% recall out-of-sample in ten annual test sets running time series cross validation, and 77% recall and 85% ROC-AUC out-of-sample in a twenty-fold cross-validation to gauge a realistic expectation of model performance in production. We solved the regression problem predicting FRP with 86% r² in-sample, but out-of-sample performance was unsatisfactory. Our model predicts well fatal and near-fatal incidents reported in Peru and Colombia out-of-sample in mountainous areas and unimodal fire regimes, the signal decays in bimodal fire regimes.

Soil salinization is a difficult challenge for agricultural productivity and environmental sustainability, particularly in arid and semi-arid coastal regions. This study investigates the spatial variability of soil electrical conductivity (EC) and its relationship with key cations and anions (Na⁺, K⁺, Ca²⁺, Mg²⁺, Cl⁻, CO₃²⁻, HCO₃⁻, SO₄²⁻) along the southeastern coast of the Caspian Sea in Iran. Using a combination of field-based soil sampling, laboratory analyses, and Landsat 8 spectral data, linear Multiple Linear Regression and Partial Least Squares Regression (MLR, PLSR) and nonlinear Artifician Neural Network and Support Vector Machine (ANN, SVM) modeling approaches were employed to estimate and map soil EC. Results identified Na⁺ and Cl⁻ as the primary contributors to salinity (r = 0.78 and r = 0.88, respectively), with NaCl salts dominating the region’s soil salinity dynamics. Secondary contributions from Potassium Chloride KCl and Magnesium Chloride MgCl₂ were also observed. Coastal landforms such as lagoon relicts and coastal plains exhibited the highest salinity levels, attributed to geomorphic processes and anthropogenic activities. Among the predictive models, the SVM algorithm outperformed others, achieving higher R² values and lower RMSE (RMSE_Test = 27.35 and RMSE_Train = 24.62, respectively), underscoring its effectiveness in capturing complex soil-environment interactions. This study highlights the utility of digital soil mapping (DSM) for assessing soil salinity and provides actionable insights for sustainable land management, particularly in mitigating salinity and enhancing agricultural practices in vulnerable coastal systems.

Creating a crop type map is a dominant yet complicated model to produce. This study aims to determine the best model to identify the wheat crop in the Haridwar district, Uttarakhand, India, by presenting a novel approach using machine learning techniques for time series data derived from the Sentinel-2 satellite spanned from mid-November to April. The proposed methodology combines the Normalized Difference Vegetation Index (NDVI), satellite bands like red, green, blue, and NIR, feature extraction, and classification algorithms to capture crop growth's temporal dynamics effectively. Three models, Random Forest, Convolutional Neural Networks, and Support Vector Machine, were compared to obtain the start of season (SOS). It is validated and evaluated using the performance metrics. Further, Random Forest stood out as the best model statistically and spatially for phenology parameter extraction with the least RMSE value at 19 days. CNN and Random Forest models were used to classify wheat crops by combining SOS, blue, green, red, NIR bands, and NDVI. Random Forest produces a more accurate wheat map with an accuracy of 69% and 0.5 MeanIoU. It was observed that CNN is not able to distinguish between wheat and other crops. The result revealed that incorporating the Sentinel-2 satellite data bearing a high spatial and temporal resolution with supervised machine-learning models and crop phenology metrics can empower the crop type classification process.