Internal structural attributes and external locational factors are commonly used as predictive variables in housing price models, enabling the identification of key determinants for assessing whether a property is fairly priced or represents a good investment. However, prior research has largely overlooked alternative applications of such models that could further enhance the identification of advantageous purchase opportunities. This study addresses this gap by applying lateral thinking to identify a “green zone”, thereby supporting more informed housing purchase decisions. This study investigates the impact of geographic features on housing prices, based on 229 transactions recorded between September 2023 and September 2024 in the North District of Taichung City, Taiwan, using data from the Ministry of the Interior’s real estate platform. Building upon an existing regression model that includes structural variables, such as the lot size, number of rooms, age, number of floors, availability of an elevator, and the presence of a garage, we introduced six additional geographic variables that reflect the proximity to a metro station, hospital, museum, CBD, funeral home, and clothing outlets. The revised model demonstrates improved explanatory power, with the R² increasing from 0.756 to 0.791. Among the newly included variables, proximity to the CBD and the museum exhibit the strongest positive effects on housing prices, contributing increases of TWD 9.8834 million and TWD 6.1644 million, respectively. These effects are both statistically significant. In contrast, proximity to a metro station has a significant negative impact of – TWD 7.2272 million , a finding attributed to a boundary constraint, since only sales south of the station fall within the district, creating a data bias. Using the escape technique of lateral thinking, these results also provide practical guidance for buyers by identifying cost-effective areas located just outside the 500 m proximity zones of high-value amenities. One such example is located near the intersection of Zhongming Road and Section 1, Zhongqing Road, with price savings of at least TWD 7.36 million. A residual analysis reveals clustering in the Yizhong Business District, suggesting that other unobserved value factors may warrant future investigation.

Agricultural productivity is frequently threatened by crop diseases, leading to substantial economic losses and hindering the adoption of sustainable farming practices. Early detection and timely intervention are therefore critical to mitigate these risks. This paper presents a novel hybrid deep learning method (HDLM) for accurate classification of cotton leaf diseases by integrating the strengths of two fine-tuned deep learning models-Visual Geometry Group 16 and Inception v3-through a stacking ensemble strategy that combines their predictions at the output level. The model was trained and validated on a carefully curated dataset of 3,000 images, manually labeled into six classes: Aphids, Armyworm, Bacterial Blight, Powdery Mildew, Target Spot, and Healthy Leaves, with 2,400 images used for training and 600 reserved for validation. To enhance generalization and robustness against real-world variations in leaf orientation, illumination, and background, extensive data augmentation techniques were applied, including rotations, flips, zooming, translations, and brightness adjustments. The proposed HDLM achieved a classification accuracy of 98.56%, significantly outperforming benchmark models such as AlexNet, DenseNet-121, ResNet-50, LeNet-5, and a 7-layer convolutional neural network, which achieved accuracies of 90–95%. In addition, the model incorporates a disease management recommendation system that provides actionable guidance to farmers to mitigate diseases and improve crop yields. This research demonstrates the efficacy of ensemble deep learning techniques in plant disease detection, providing a scalable, robust, and practical solution for precision agriculture. Future work should focus on expanding the dataset with heterogeneous sources, integrating advanced augmentation strategies, and exploring real-time feedback mechanisms to further enhance model adaptability, predictive performance, and applicability across diverse agricultural environments.

Cloud computing environments require high availability and scalability, making proactive failure management essential for ensuring system reliability, security, and consistent performance. Effective failure prediction significantly reduces downtime, improves disaster recovery processes, and maintains uninterrupted service delivery. This paper presents an optimized machine learning framework for predicting failures in cloud infrastructures by integrating principal component analysis (PCA) with advanced ensemble learning models. The study employs three prominent models-random forest (RF), categorical boosting (CatBoost), and light gradient boosting machine (LightGBM)-enhanced through PCA to improve feature representation and overall predictive accuracy. Key operational metrics, including class scheduling, memory usage, central processing unit utilization, event instances, and task priority, are used as features. The Google 2019 cluster dataset is utilized, and preprocessing steps involve handling missing data, scaling numerical attributes, and encoding categorical variables to ensure data quality. Experimental results reveal that PCA-enhanced RF, CatBoost, and LightGBM achieve superior accuracies of 94.31%, 97.17%, and 98.36%, respectively, outperforming their standard counterparts. These outcomes highlight the effectiveness of PCA-integrated ensemble learning and underscore its potential for real-time cloud failure prediction and automated fault monitoring in large-scale distributed environments.

This study examines how ambidextrous innovation and market orientation jointly shape digital servitisation and, through relational capabilities, customer lifetime value (CLV) in Vietnamese manufacturing small and medium-sized enterprises (SMEs). Survey data from 205 firms are analysed with partial least squares structural equation modelling using a two-stage hierarchical component model for digital servitisation, market orientation, inter-functional alignment, and customer participation in value co-creation. The results show that both exploitative and exploratory innovation are positively associated with digital servitisation, with exploratory innovation showing the stronger association. Market orientation strengthens these associations. Digital servitisation is also positively associated with inter-functional alignment and customer participation in value co-creation, which in turn are positively associated with CLV. The study conceptualises digital servitisation as an integrative dynamic capability that links upstream ambidextrous innovation and market orientation to downstream financial value in an emerging-economy SME context, and highlights relational mechanisms that open the “black box” between digital strategies and long-term customer value.