Complexity

With eco-friendly green agriculture becoming the development trend of modern agriculture, how to make green investments and how to coordinate the supply chain become the key issues of agricultural green development. Using game theory and optimization theory, this paper studies the green investment decision in a two-echelon agricultural supply chain composed of a risk-averse farmer and a risk-neutral retailer under different power structures including three kinds of decentralized decision making and three kinds of cooperative decision making and conducts the supply chain coordination based on generalized Nash bargaining model. The results show that under decentralized decision making, Nash vertical, farmer-led, and retailer-led maximizes green investment level, the expected utility of farmer and retailer, respectively. In addition, the cooperative decision increases the marginal revenue, sales price, and the expected utility of the retailer and decreases the expectations of farmers. Except for retailer-led cooperative decisions, all cooperative decisions have increased the level of green investment and wholesale prices; among the six decision models, the green investment level is negatively correlated with risk aversion, while it is positively correlated with the cost-sharing contract. The optimal cost-sharing ratio is positively correlated with risk aversion and bargaining power. The cost-sharing contracts are invalid when farmers have full bargaining power. Numerical analysis shows that a cost-sharing contract with equal bargaining power can achieve perfect coordination in the supply chain.

Imbalanced datasets pose significant challenges in the field of machine learning, as they consist of samples where one class (majority) dominates over the other class (minority). Although AdaBoost is a popular ensemble method known for its good performance in addressing various problems, it fails when dealing with imbalanced data sets due to its bias towards the majority class samples. In this study, we propose a novel weighting factor to enhance the performance of AdaBoost (called IMBoost). Our approach involves computing weights for both minority and majority class samples based on the performance of classifier on each class individually. Subsequently, we resample the data sets according to these new weights. To evaluate the effectiveness of our method, we compare it with six well-known ensemble methods on 30 imbalanced data sets and 4 synthetic data sets using ROC, precision-eecall AUC, and G-mean metrics. The results demonstrate the superiority of IMBoost. To further analyze the performance, we employ statistical tests, which confirm the excellence of our method.

In this paper, a class of integrodifferential equations with the Caputo fractal-fractional derivative is considered. We study the exact and numerical solutions of the said problem with a fractal-fractional differential operator. The abovementioned operator is arising widely in the mathematical modeling of various physical and biological problems. In our scheme, first, the integrodifferential equation with the fractal-fractional differential operator is converted to an integrodifferential equation with the Riemann–Liouville differential operator. Additionally, the obtained integrodifferential equation is then converted to the equivalent integrodifferential equation involving the Caputo differential operator. Then, we convert the integrodifferential equation under the Caputo differential operator using the Laplace transform to an algebraic equation in the Laplace space. Finally, we convert the obtained solution from the Laplace space into the real domain. Moreover, we utilize two techniques which include analytic inversion and numerical inversion. For numerical inversion of the Laplace transforms, we have to evaluate five methods. Extensive results are presented. Furthermore, for numerical illustration of the abovementioned methods, we consider three problems to demonstrate our results.

The existence of modern humanity directly depends on environmental security. Human society, biodiversity, ecosystems, and climate safety are interdependent. The anthropogenic influence that causes irreversible climate change threatens both the ecosystem’s existence and humans’ survival. To maintain a balance between human well-being and a safe environment, it is important to have a diverse knowledge of the interrelations between the technological impact on the natural environment and climate change and an understanding of action strategies to mitigate climate change and ensure sustainable development. Applying a scientific approach, data analytics, and data science tools can effectively support climate change mitigation and prevent a climate disaster. Based on the components of the climate change performance index (CCPI) 2023 for 59 countries and the EU, a canonical discriminant model was built to identify the significant factors that influence the assessment of the effectiveness of climate protection in a particular country or region and the assessment of climate risks. It can be used to assess the level of climate protection effectiveness of countries that have not defined the CCPI 2023. Based on empirical data, we have determined the real weights of the relevant CCPI components in relation to the effectiveness of actions aimed at reducing global warming. We have established that there are additional factors important for assessing climate protection, but the CCPI rating does not consider them. We conducted a comparative analysis of the CCPI index and the sustainable development goals (SDG) index. The study establishes that the differences in environmental protection among the world’s countries do not determine the assessment of the level of sustainable development of the world’s countries. The obtained results can provide information to support decision-making in developing effective strategies and urgent actions to ensure climate protection.

The establishment of a civil-military integration supply chain system is the cornerstone of China’s strategic development in military-civilian integration. It is essential to explore cooperative innovation and development between upstream civilian enterprises and downstream military enterprises within the supply chain to optimise resource allocation and promote the sustainable use of civil-military resources. This exploration is a prerequisite for accelerating the formation of the civil-military integration supply chain system and holds significant importance for realising the internal synergy between the civilian industry and the military industry. However, utilizing the evolutionary game model as a foundation, this study delves into the impact of absorption capacity, transformation and integration capability, network synergy, and change and innovation capacity on the vertical cooperation and innovation behaviour within the supply chain of civil-military integration enterprises. Firstly, civilian enterprises are more cost-sensitive concerning collaborative innovation investments compared to military enterprises. Excessive costs can discourage collaboration between civilian and military entities. Secondly, strong exploratory and absorptive capabilities, along with network synergies, can enhance the benefits of cooperation and innovation among these enterprises, but they also introduce the risk of opportunistic “free-rider” behaviour. Thirdly, the dynamics of the technology and product chains are influenced by an excess supply for civilian enterprises, while the opposite is true for military enterprises. Finally, a strong capacity for transformation and integration fosters cooperative and innovative behaviours among enterprises, with civilian enterprises exhibiting greater responsiveness. This study brings new research perspectives to the forefront, exploring vertical cooperation and innovative development within supply chain enterprises, particularly through the lens of supply and demand dynamics. Additionally, it offers practical recommendations aimed at helping the government expedite the establishment of integrated military-civilian supply chains and foster the synergistic development of the two key sectors: the military and civilian economies.

Decision tree models are widely used for classification tasks in data mining. However, privacy becomes a significant concern when training data contain sensitive information from different parties. This paper proposes a novel framework for secure two-party decision tree classification that enables collaborative training and evaluation without leaking sensitive data. The critical techniques employed include homomorphic encryption, function secret sharing (FSS), and a custom secure comparison protocol. Homomorphic encryption allows computations on ciphertexts, enabling parties to evaluate an encrypted decision tree model jointly. FSS splits functions into secret shares to hide sensitive intermediate values. The comparison protocol leverages FSS to securely compare attribute values to node thresholds for tree traversal, reducing overhead through efficient cryptographic techniques. Our framework divides computation between two servers holding private data. A privacy-preserving protocol lets them jointly construct a decision tree classifier without revealing their respective inputs. The servers encrypt their data and exchange function secret shares to traverse the tree and obtain the classification result. Rigorous security proofs demonstrate that the protocol protects data confidentiality in a semihonest model. Experiments on benchmark datasets confirm that the approach achieves high accuracy with reasonable computation and communication costs. The techniques minimize accuracy loss and latency compared to prior protocols. Overall, the paper delivers an efficient, modular framework for practical two-party secure decision tree evaluation that advances the capability of privacy-preserving machine learning.