Mengxin Wang

Data stream classification is a critical challenge in data mining, where models must rapidly adapt to evolving data distributions and concept drift in real time, while extreme learning machines offer fast training and strong generalization, most existing methods struggle to jointly address multi-class imbalance, concept drift, and the high cost of label acquisition in streaming settings. In this paper, we present the Adaptive Broad Learning System for Online Imbalanced Classification (ABLS-OIC), which introduces three core innovations: (1) a Class-Adaptive Weight Matrix (CAWM) that dynamically adjusts sample weights according to class distribution, sample density, and difficulty; (2) a Hybrid Memory Retention Mechanism (HMRM) that selectively retains representative samples based on importance and diversity; and (3) a Multi-Objective Adaptive Optimization Framework (MAOF) that balances classification accuracy, class balance, and computational efficiency. Extensive experiments on ten benchmark datasets with varying imbalance ratios and drift patterns show that ABLS-OIC consistently outperforms state-of-the-art methods, with improvements of 5.9% in G-mean, 6.3% in F1-score, and 3.4% in AUC. Furthermore, a real-world credit fraud detection case study demonstrates the practical effectiveness of ABLS-OIC, highlighting its value for early detection of rare but critical events in dynamic, high-stakes applications.

In recent years, China has made significant strides in developing the new energy vehicle industry, providing substantial financialsupport to this sector. The production and sales of new energy vehicles have experienced explosive growth, accompanied by notable improvements in technology and product quality. However, as the industry evolved, the effectiveness of financial subsidies in enhancing the production and technological capabilities of enterprises has gradually diminished. To foster the rapid development of new energy vehicle enterprises and facilitate the rational planning of relevant government financial subsidies, this study utilizes data from listed Chinese new energy vehicle companies spanning from 2017 to 2022 to investigate the intrinsic relationship between financial subsidies and the quality development of enterprises. The study reveals several key findings. First, when fiscal subsidies are maintained within a reasonable range, they significantly enhance the quality development of state-owned enterprises (SOEs). However, once this threshold is surpassed, further increases in fiscal subsidies do not promote quality development of enterprises, a phenomenon that does not apply to non-state enterprises. Second, the threshold effect of financial subsidies on quality development is particularly pronounced between newly listed companies and those listed for more than three years. Third, for SOEs, financial subsidies play a critical role in fostering high-quality development by boosting R&D investment; this relationship is not observed in non-SOEs. Consequently, this study advocates the establishment of differentiated financial support policies for the new energy vehicle industry to maximize their effectiveness. Received: 24 July 2024 | Revised: 21 November 2024 | Accepted: 24 December 2024 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The data that support the findings of this study are openly available in [Wind] at https://www.wind.com.cn/portal/zh/WFT/ index.html Author Contribution Statement Leyan Zuo: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization, Project administration. Songlan Zhou: Funding acquisition. Mengxin Wang: Conceptualization, Methodology, Software, Resources, Data curation, Writing – review & editing, Supervision, Project administration. Siyu Zhang: Methodology, Software, Validation.

Multi-view subspace clustering (MVSC) can effectively group multi-view data distributed around several low-dimensional subspaces. Although encouraging results, most existing methods suffer from two typical limitations, resulting in clustering performance degradation. They ignore high-order correlations underlying the multi-view data, leading to degeneration of complementary power; in addition, they rely on much prior knowledge (e.g., pairwise constraints) for clustering enhancement. In this paper, a novel algorithm called Enhanced Multi-view Subspace Clustering (EMVSC) is proposed to address both limitations. EMVSC can effectively exploit high-order correlations and optimally use limited prior knowledge for better clustering performance. Specifically, EMVSC imposes twist tensor nuclear norm on multi-view tensor representation constructed by stacking view-specific self-representations; in addition, EMVSC exploits prior knowledge of pairwise constraints from whole dataset by employing constraint propagation, which propagates limited constraint knowledge from constrained samples to unconstrained samples. To efficiently optimize EMVSC, an extended intact augmented Lagrangian method is derived with good convergence. Experimental results on seven standard multi-view databases demonstrate its efficacy.

Studies have shown that land and climate interact in complex ways through multiple biophysical and biogeochemical feedbacks. In this interaction mechanism, the carbon fixation effect among different land-use types and objective conditions among different regions have significant gaps, leading to the heterogeneous impact of land-use on climate change. This study takes temperature change as the observation index to reflect climate change, and analyzes the process of land use type adjustment affecting vegetation cover and climate change. Based on the data of 214 countries from 1990 to 2018, this paper uses the spatial Durbin model with temperature lag to verify the heterogeneous impact of land-use on climate change in two dimensions of land-use type (Agriculture, forestry and their subdivision structure) and region (latitude and land-sea difference). The following conclusions are drawn: 1) The impact of different land-use types on climate change is heterogeneous. The impact of agricultural land on climate change is not significant, but the increase of the forest land proportion will help to restrain the rise of national temperature. 2) The impact of land-use on climate change has regional heterogeneity. There is heterogeneity in the impact on climate change among sample countries of different latitudes. The geographical differences make the mechanism of land-use affecting climate change between island countries and mainland countries also have heterogeneity, mainly in that island countries are not affected by the land-use structure adjustment of neighboring countries. 3) A country’s climate change is affected by both its own land-use structure and the land-use structure of neighboring countries, and the latter is more critical. The conclusions in this study provide helpful supplementary evidence for the importance of international climate cooperation and provide a reference for proposing international initiatives to address climate change or establishing an international convention to address climate change.

The effect of digital financial inclusion on farmers’ income growth is worth theoretical discussion and empirical testing. Based on the sample data of 30 provinces in the Chinese mainland from 2011 to 2019, this paper adopted the spatial Durbin model and the panel threshold model to empirically study the impact of digital financial inclusion on the growth of farmers’ income; furthermore, the heterogeneity of the impact was studied in terms of the difference of provincial economic development level. The results showed that: first, digital financial inclusion can significantly promote farmers’ income growth; second, digital financial inclusion has a positive spatial spillover effect on farmers’ income growth; third, the impact of digital financial inclusion on farmers’ income growth has a double threshold effect: farmers’ income growth increases with the development of digital financial inclusion; and fourth, the impact of digital financial inclusion on farmers’ income growth is heterogeneous in terms of provincial economic development level.

With the full integration of digital information technology and financial services, digital finance has developed rapidly. As there are significant differences in the development level of FinTech and traditional financial sectors in different cities, it is important to evaluate the development level of urban digital finance. This study aimed to compile an index of urban digital finance to present an accurate and in-depth depiction of how urban digital finance has developed in China. Our sample covers 278 cities in China, over the period 2010–2020. This paper firstly constructs the urban digital financial index system from the three dimensions of digital financial services, digital financial technology, and digital financial operating environment, and then adopts a combination of subjective and objective methods to measure the urban digital financial index. This paper study revealed that China's urban digital finance has been on an upward trend from 2010 to 2020, and the digital finance operating environment is an important driving force for the growth of the urban digital finance index. The convergence of China's urban digital finance is decreasing, indicating that the gap in digital financial development between cities is increasing. Urban digital finance has positive spatial agglomeration, but this spatial agglomeration is decreasing.

The impact of high-standard basic farmland construction policies on agricultural eco-efficiency has been extensively considered. Using the Chinese provincial panel data from 2007–2017, we first measure the level of agricultural eco-efficiency in China by employing data envelopment analysis. Then, using difference-in-difference models, we analyze the impact of high-standard basic farmland construction policies on agricultural eco-efficiency and test whether there is heterogeneity of this impact. Finally, we further explore the specific channels through which the polices of high-standard basic farmland construction affect agricultural eco-efficiency. The empirical results indicate that 1) the implementation of high-standard farmland construction policies can significantly improve agricultural eco-efficiency, 2) the heterogeneity of the impact of high-standard farmland construction policies on agricultural eco-efficiency is manifested in both regional and efficiency aspects and 3) high-standard farmland construction policies promote agricultural eco-efficiency through the interaction between the new land scale and the replanting index.

We first derive the stellar population properties: age and metallicity for ∼43,000 low redshift galaxies in the DR7 of the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST) survey, which have no spectroscopic observations in the Sloan Digital Sky Survey (SDSS). We employ a fitting procedure based on the small-scale features of galaxy spectra so as to avoid possible biases from the uncertain flux calibration of the LAMOST spectroscopy. We show that our algorithm can successfully recover the average age and metallicity of the stellar populations of galaxies down to signal-to-noise ratio ≥5 through testing on both mock galaxies and real galaxies comprising LAMOST and their SDSS counterparts. We provide a catalog of the age and metallicity for ∼43,000 LAMOST galaxies online. As a demonstration of the scientific application of this catalog, we present the Holmberg effect on both age and metallicity of a sample of galaxies in galaxy pairs.