The marriage of federated learning and recommender system (FedRec) has been widely used to address the growing data privacy concerns in personalized recommendation services. In FedRecs, users’ attribute information and behavior data (i.e., user-item interaction data) are kept locally on their personal devices, therefore, it is considered a fairly secure approach to protect user privacy. As a result, the privacy issue of FedRecs is rarely explored. Unfortunately, several recent studies reveal that FedRecs are vulnerable to user attribute inference attacks, highlighting the privacy concerns of FedRecs. In this paper, we further investigate the privacy problem of user behavior data (i.e., user-item interactions) in FedRecs. Specifically, we perform the first systematic study on interaction-level membership inference attacks on FedRecs. An interaction-level membership inference attacker is first designed, and then the classical privacy protection mechanism, Local Differential Privacy (LDP), is adopted to defend against the membership inference attack. Unfortunately, the empirical analysis shows that LDP is not effective against such new attacks unless the recommendation performance is largely compromised. To mitigate the interaction-level membership attack threats, we design a simple yet effective defense method to significantly reduce the attacker’s inference accuracy without losing recommendation performance. Extensive experiments are conducted with two widely used FedRecs (Fed-NCF and Fed-LightGCN) on three real-world recommendation datasets (MovieLens-100K, Steam-200K, and Amazon Cell Phone), and the experimental results show the effectiveness of our solutions.

Purpose: Wood decay fungi are the only forms of life capable of degrading wood to its initial constituents, greatly contributing to the soil ecosystem. This study summarizes the current research status and development characteristics of global wood decay fungi research, in order to better understand their role in soils. Methods: A bibliometric analysis was applied to the literature from 1913 to 2020, based on data from the Web of Science (WOS) Core Collection. For this, various bibliometric analysis methods, R (Biblioshiny package), and VOSviewer were applied. Results: A total of 8089 documents in this field were identified in the WOS Core Collection. The annual number of publications tended to increase, with exponential growth after 2008. Researchers in this field were mainly concentrated in North Europe, the USA, and China. Biotechnology, applied microbiology, environmental sciences, and microbiology were the most popular WOS categories. Bioresource Technology and Applied Environmental Microbiology were the top two journals with the most citations. The top three authors with the most published papers were Dai YC, Martinez AT, and Cui BK. Co-occurrence analysis of author keywords identified six clusters, mainly divided into three categories: the classification and diversity, the degradation mechanisms, and the ecological functions of wood decay fungi. Clustering results further showed that the lignin degradation process and the application of wood decay fungi in industrial production and soil contamination remediation are current research hotspots. Conclusions: We present a comprehensive and systematic overview of research related to wood decay fungi and provide a deep perspective to understand the associated research progress. This is important for facilitating the development of a profound understanding of the contribution of wood decay fungi to soil systems and the degradation of soil contaminants.

Purpose: Accurately assessing soil organic carbon (SOC) content is vital for ecosystem services management and addressing global climate challenges. This study undertakes a comprehensive bibliometric analysis of global estimates for SOC using remote sensing (RS) and machine learning (ML) techniques. It showcases the historical growth and thematic evolution in SOC research, aiming to amplify the understanding of SOC estimation themes and provide scientific support for climate change adaptation and mitigation. Materials and Methods: Employing extensive literature database analysis, bibliometric network analysis, and clustering techniques, the study reviews 1,761 articles on SOC estimation using RS technologies and 490 articles on SOC employing both RS and ML technologies. Results and Discussion: The results indicate that satellite-based RS, particularly the Landsat series, is predominant for estimation of SOC and other associated studies, with North America, China, and Europe leading in evaluations with Africa is having low evaluations adopting RS technology. Trends in the research demonstrate an evolution from basic mapping to advanced topics such as carbon (C) sequestration, complex modeling, and big data utilization. Thematic clusters from co-occurrence analysis suggest the interplay between technology development, environmental surveys, soil properties, and climate dynamics. Conclusion: The study highlights the synergy between RS and ML, with advanced ML techniques proving to be critical for accurate SOC estimation. These findings are crucial for comprehensive ecosystem SOC estimation, informed environmental management and strategic decision-making.

Grassland remote sensing (GRS) is an important research topic that applies remote sensing technology to grassland ecosystems, reflects the number of grassland resources and grassland health promptly, and provides inversion information used in sustainable development management. A scientometrics analysis based on Science Citation Index-Expanded (SCI-E) was performed to understand the research trends and areas of focus in GRS research studies. A total of 2692 papers related to GRS research studies and 82,208 references published from 1980 to 2020 were selected as the research objects. A comprehensive overview of the field based on the annual documents, research areas, institutions, influential journals, core authors, and temporal trends in keywords were presented in this study. The results showed that the annual number of documents increased exponentially, and more than 100 papers were published each year since 2010. Remote sensing, environmental sciences, and ecology were the most popular Web of Science research areas. The journal Remote Sensing was one of the most popular for researchers to publish documents and shows high development and publishing potential in GRS research studies. The institution with the greatest research documents and most citations was the Chinese Academy of Sciences. Guo X.L., Hill M.J., and Zhang L. were the most productive authors across the 40-year study period in terms of the number of articles published. Seven clusters of research areas were identified that generated contributions to this topic by keyword co-occurrence analysis. We also detected 17 main future directions of GRS research studies by document co-citation analysis. Emerging or underutilized methodologies and technologies, such as unmanned aerial systems (UASs), cloud computing, and deep learning, will continue to further enhance GRS research in the process of achieving sustainable development goals. These results can help related researchers better understand the past and future of GRS research studies.

Purpose: Wildfire is one of the most important natural disturbances in forest and multi-vegetation ecosystems, directly or indirectly affecting the structural processes and functions of forest ecosystems with varying degrees. Wildfire releases vast amounts of carbon dioxide and other substances by destroying vegetation, making itself an important topic for the study of global change and environmental impacts. Therefore, a deeper understanding of this topic is particularly crucial for managing forest ecosystems. Methods: This paper was based on a literature search of the Web of Science database for international forest wildfire research, utilizing bibliometric and quantity statistical analysis methods. Results: The results show that forest wildfire research has been rapidly growing over the last 20 years, with the number of relevant articles generally increasing yearly at an average annual growth rate of about 22.45%. The US tops the list in terms of total and independent publications, with a total of 3111 articles (49.88%). The key journals publishing on this topic include 12 journals, Stephens S.L., Bergeron Y., and Lindenmayer D.B. are the key contributing authors to the field, and research institutions are primarily concentrated in the US Forest Service. Keyword co-occurrence analysis shows that current forest wildfire research is focused on seven main areas. This paper systematically reviewed the progress and hotspots of international forest wildfire research in recent decades, mainly focusing on occurrences, severity, management, and warning techniques for wildfires, as well as the impact of climate change and human activities on wildfires. Conclusions: The study concludes that research trends in this field have undergone a significant evolution in recent decades. The future forest wildfire research moves towards a combination of typical mechanisms and large-scale effects across spatial and temporal scales, deep integration of aerospace and earth observations and precise simulations, discipline fusion, and couplings research. We believe that this study provides a comprehensive and systematic overview for future forest wildfire observation, prediction, management, and investigation of ecological effects.