检索结果-南通市图书馆

Fairness in machine learning: definition, testing,debugging, and application

在线全文

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第09期67卷 41-61页

作者： Xuanqi GAO Chao SHEN Weipeng JIANG Chenhao LIN Qian LI Qian WANG Qi LI Xiaohong GUAN faculty of Electronic and Information engineering Xi'an Jiaotong University School of Cyber science and engineering Wuhan University Institute for Network sciences and Cyberspace Tsinghua University

In recent years, artificial intelligence technology has been widely used in many fields, such as computer vision, natural language processing and autonomous driving. Machine learning algorithms, as the core technique of AI, have significantly facilitated people's lives. However, underlying fairness issues in machine learning systems can pose risks to individual fairness and social security. Studying fairness definitions, sources of problems, and testing and debugging methods of fairness can help ensure the fairness of machine learning systems and promote the wide application of artificial intelligence technology in various fields. This paper introduces relevant definitions of machine learning fairness and analyzes the sources of fairness problems. Besides, it provides guidance on fairness testing and debugging methods and summarizes popular datasets. This paper also discusses the technical advancements in machine learning fairness and highlights future challenges in this area.

关键词： artificial intelligence security machine learning security machine learning fairness model testing model debugging

同方期刊数据库

Drug-Target Interactions Prediction Based on Signed Heterogeneous Graph Neural Networks

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2024年第1期33卷 231-244页

作者： Ming CHEN Yajian JIANG Xiujuan LEI Yi PAN Chunyan JI Wei JIANG College of Information science and engineering Hunan Normal University School of Computer science Shaanxi Normal University faculty of Computer science and Control engineering Shenzhen Institute of Advanced TechnologyChinese Academy of Sciences Computer science Department BNU-HKBU United International College

Drug-target interactions(DTIs) prediction plays an important role in the process of drug *** computational methods treat it as a binary prediction problem, determining whether there are connections between drugs and targets while ignoring relational types information. Considering the positive or negative effects of DTIs will facilitate the study on comprehensive mechanisms of multiple drugs on a common target, in this work, we model DTIs on signed heterogeneous networks, through categorizing interaction patterns of DTIs and additionally extracting interactions within drug pairs and target protein pairs. We propose signed heterogeneous graph neural networks(SHGNNs), further put forward an end-to-end framework for signed DTIs prediction, called SHGNN-DTI,which not only adapts to signed bipartite networks, but also could naturally incorporate auxiliary information from drug-drug interactions(DDIs) and protein-protein interactions(PPIs). For the framework, we solve the message passing and aggregation problem on signed DTI networks, and consider different training modes on the whole networks consisting of DTIs, DDIs and PPIs. Experiments are conducted on two datasets extracted from Drug Bank and related databases, under different settings of initial inputs, embedding dimensions and training modes. The prediction results show excellent performance in terms of metric indicators, and the feasibility is further verified by the case study with two drugs on breast cancer.

关键词： Drugs Training Proteins Measurement Message passing Prediction methods Graph neural networks

Attaining synergetic equilibrium of electrical conductivity and tensile strength in GQDs@GN/Cu composites through multi-scale intragranular and intergranular reinforcements

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Rare Metals 2024年第1期43卷 366-379页

作者： Shuang-Yin Zhang Liang Liu Rui Bao Jian-Hong Yi Sheng-Da Guo faculty of Materials science and engineering Kunming University of Science and TechnologyKunming 650093China Key Laboratory of Advanced Materials of Yunnan Province Kunming University of Science and TechnologyKunming 650093China faculty of Materials Metallurgy and Chemistry Jiangxi University of Science and TechnologyGanzhou 341000China

The configuration and quality of reinforcements, as well as the robustness of interfacial bonding,holding a critical significance in determining the concurrence between electrical conductivity and mechanical strength in metal matrix composites. In this study, citric acid was employed as the precursor for synthesizing multiscale carbon nanomaterials(graphene quantum dots and graphene, abbreviated as GQDs and GN). The GQDs@GN/Cu composites were fabricated through a segmented ball milling process in conjunction with subsequent spark plasma sintering(SPS). The intragranular GQDs and intergranular GQDs@GN had synergistically reinforced Cu composites through Orowan strengthening, load transfer strengthening and refinement strengthening. Furthermore,the robust interface bonding between GQDs@GN and Cu effectively mitigated interfacial impedance stemming from electron-boundary scattering. The yield strength and ultimate tensile strength of the GQDs@GN/Cu composites were recorded as 270 and 314 MPa, respectively, representing an improvement of 92 and 28% over pure Cu, while maintaining electrical conductivity at a level comparable to that of pure Cu. This study advances the understanding of the possibility of realizing a synergistic compatibility between electrical conductivity and mechanical strength in Cu composites.

关键词： Cu matrix composites Interface bonding Mechanical property Graphene quantum dots(GQDs) Electrical conductivity

In-situ growth of MIL-53(Fe) on charcoal sponge as a highly efficient and recyclable photocatalyst for removal of Cr(Ⅵ)

在线全文

学校读者我要写书评

暂无评论

Rare Metals 2024年第9期43卷 4344-4355页

作者： Zhuo-Fan Yu Ye Yang Hai-Feng Zhuang Sheng-Dao Shan Mihail-Simion Beldean-Galea Qing-Quan Xue Xiao-Feng Shen Shi-Jie Li Key Laboratory of Recycling and Eco-Treatment of Waste Biomass of Zhejiang Province School of Environmental and Natural Resources Zhejiang University of Science and Technology faculty of Environmental science and engineering Babes-Bolyai University Key Laboratory of Pollution Exposure and Health Intervention of Zhejiang Province Interdisciplinary Research Academy (IRA) Zhejiang Shuren University National engineering Research Center for Marine Aquaculture Institute of Innovation & Application Zhejiang Ocean University

The application of metalDorganic frameworks(MOFs) is hindered by several crucial factors, including low photocatalytic efficiency, poor stability and challenges in recycling. In this study, MIL-53(Fe) was grown in situ on a charcoal sponge(CS) with a large surface area,excellent mechanical properties, and high photocatalytic efficiency using a solvothermal method, thus addressing the above-mentioned shortcomings of MOF ***, CS/MIL-53(Fe) has an amazing light-harvesting capacity in the visible range, leading to a significant improvement in the utilization of light. The results of the experiments indicate that CS/MIL-53(Fe) displays a higher photocatalytic activity(87% Cr(VI)) when exposed to visible light than CS(58%) or MIL-53(Fe)(27%),attributing to the increased separation of photogenerated carriers. e-is the main reactive radical and dominates the photocatalytic reduction process, rooting from that e-can reduce Cr(VI) to low-toxicity Cr(III). This study presents a novel method for constructing photocatalysts with large surface areas, superior mechanical characteristics, and enhanced photocatalytic performance. Additionally, it achieves the resource utilization of agricultural and forestry waste.

关键词： Charcoal sponge Cr(VI) Light absorption MIL-53(Fe) Photocatalysis

High efficiency removal of NO using waste calcium carbide slag by facile KOH modification

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Environmental sciences 2024年第5期139卷 182-192页

作者： Fang Wang Hui Li Jiyun Gao Na Geng Enzhu Jiang Futing Xia Mingwu Xiang Lijuan Jia Ping Ning School of Chemistry and Environment Yunnan Minzu UniversityKunming 650504China faculty of Environmental science and engineering Kunming University of Science and TechnologyKunming 650500China

Waste calcium carbide slags(CS),which are widely applied to desulfurisation,are not typically used in ***,to well achieve waste control by waste,a facile and highefficiency denitration strategy is developed using KOH to modify the calcium carbide slags(KCS).Various KCS samples were investigated using a series of physical and chemical *** performance test results showed that the KOH concentration and reaction temperature are the main factors affecting the denitration efficiency of KCS,and CS modified with 1.5 mol/L KOH(KCS-1.5)can achieve 100% denitration efficiency at 300℃.Such excellent removal efficiency is due to the catalytic oxidation of the oxygen-containing functional groups derived from the *** studies showed that KOH treatment significantly increased the concentration of oxygen vacancies,nitro compounds,and basic sites of *** study provides a novel strategy for the resource utilisation of waste CS in the future.

关键词： Waste Carbide slag KOH Effective denitration

Hydrothermal carbonization pretreatment makes a remarkable difference in activation of rice and lettuce in food waste

在线全文

学校读者我要写书评

暂无评论

Green Chemical engineering 2024年第3期5卷 348-363页

作者： Lijun Zhang Yifan Sun Chao Li Shu Zhang Mortaza Gholizadeh Xun Hu School of Material science and engineering University of Jinan Joint International Research Laboratory of Biomass Energy and Materials College of Materials Science and Engineering Nanjing Forestry University faculty of Chemical and Petroleum engineering University of Tabriz

Cooked rice and the vegetables like lettuce are common kitchen waste, which are carbonaceous materials and have the potential as feedstock for the production of activated carbon. Cooking is similar to hydrothermal treatment（HTC）, which might impact the subsequent activation of kitchen waste. In this study, the HTC of lettuce, rice, or their mixture and the activation of the resulting hydrochars were conducted. The results indicated that cross-polymerization between the N-containing organics from lettuce and the sugar derivatives from rice took place in their co-HTC, which significantly increased the hydrochar yield. Activation of the hydrochar from the coHTC generated the AC with a yield of 2 times that from direct activation of mixed lettuce/rice. However, the coHTC facilitated aromatization, reducing reactivity with K2C2O4in activation and producing the AC with main micropores and low specific surface area. Activation of the hydrochar from HTC of rice followed the above trend,while that from lettuce was the opposite. The organics in lettuce were thermally unstable and could not undergo sufficient aromatization. The activation of hydrochar from HTC of lettuce thus generated the AC with the lowest yield, but the highest specific surface area（1684.9 m2/g）, abundant mesopores, and superior capability for adsorption of tetracycline. However, the environmental impacts and energy consumption for the production of AC from the hydrochar of lettuce were higher than that from hydrochar of co-HTC.

关键词： Food waste Hydrothermal pretreatment Hydrochar Chemical activation Products properties

Advances in noble metal-modified g-C3N4 heterostructures toward enhanced photocatalytic redox ability

在线全文

学校读者我要写书评

暂无评论

International Journal of Minerals, Metallurgy and Materials 2024年第11期31卷 2368-2389页

作者： Zhang, Xiao Yang, Ping faculty of Chemical engineering and Technology Cracow University of Technology Krakow Poland School of Material science & engineering University of Jinan Jinan China

The photocatalytic activity of catalysts depends on the energy-harvesting ability and the separation or transport of photogenerated carriers. The light absorption capacity of graphitic carbon nitride (g-C3N4)-based composites can be enhanced by adjusting the surface plasmon resonance (SPR) of noble metal nanoparticles (e.g., Cu, Au, and Pd) in the entire visible region. Adjustments can be carried out by varying the nanocomponents of the materials. The SPR of noble metals can enhance the local electromagnetic field and improve interband transition, and resonant energy transfer occurs from plasmonic dipoles to electron–hole pairs via near-field electromagnetic interactions. Thus, noble metals have emerged as relevant nanocomponents for g-C3N4 used in CO2 photoreduction and water splitting. Herein, recent key advances in noble metals (either in single atom, cluster, or nanoparticle forms) and composite photocatalysts based on inorganic or organic nanocomponent-incorporated g-C3N4 nanosheets are systematically discussed, including the applications of these photocatalysts, which exhibit improved photoinduced charge mobility in CO2 photoconversion and H2 production. Issues related to the different types of multi-nanocomponent heterostructures (involving Schottky junctions, Z-/S-scheme heterostructures, noble metals, and additional semiconductor nanocomponents) and the adjustment of dimensionality of heterostructures (by incorporating noble metal nanoplates on g-C3N4 forming 2D/2D heterostructures) are explored. The current prospects and possible challenges of g-C3N4 composite photocatalysts incorporated with noble metals (e.g., Au, Pt, Pd, and Cu), particularly in water splitting, CO2 reduction, pollution degradation, and chemical conversion applications, are summarized.

关键词： g-C3N4 heterostructure noble metal photocatalysis H_(2)generation CO_(2)

An Encoding-Decoding Framework Based on CNN for circ RNA-RBP Binding Sites Prediction

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2024年第1期33卷 256-263页

作者： Yajing GUO Xiujuan LEI Yi PAN School of Computer science Shaanxi Normal University faculty of Computer science and Control engineering Shenzhen Institute of Advanced TechnologyChinese Academy of Sciences Department of Computer science Georgia State University

Predicting RNA binding protein(RBP) binding sites on circular RNAs(circ RNAs) is a fundamental step to understand their interaction mechanism. Numerous computational methods are developed to solve this problem, but they cannot fully learn the features. Therefore, we propose circ-CNNED, a convolutional neural network(CNN)-based encoding and decoding framework. We first adopt two encoding methods to obtain two original matrices. We preprocess them using CNN before fusion. To capture the feature dependencies, we utilize temporal convolutional network(TCN) and CNN to construct encoding and decoding blocks, respectively. Then we introduce global expectation pooling to learn latent information and enhance the robustness of circ-CNNED. We perform circ-CNNED across 37 datasets to evaluate its effect. The comparison and ablation experiments demonstrate that our method is superior. In addition, motif enrichment analysis on four datasets helps us to explore the reason for performance improvement of circ-CNNED.

关键词： Proteins RNA Neural networks Encoding Robustness Decoding Convolutional neural networks

同方期刊数据库

Highly efficient ternary organic solar cells with excellent open-circuit voltage and fill factor via precisely tuning molecular stacking and morphology

在线全文

学校读者我要写书评

暂无评论

science China Chemistry 2024年第3期67卷 963-972页

作者： Dinghong Sun Zhenyu Chen Jianfeng Zhang Wei Song Jingyu Shi Jintao Zhu Yuanyuan Meng Fei Jin Shuncheng Yang Ziyi Ge faculty of Electrical engineering and Computer science Ningbo UniversityNingbo 315211China faculty of Material science and Chemical engineering Ningbo UniversityNingbo 315211China Zhejiang Provincial engineering Research Center of Energy Optoelectronic Materials and Devices Ningbo Institute of Materials Technology&EngineeringChinese Academy of SciencesNingbo 315201China Center of Materials science and Optoelectronics engineering University of Chinese Academy of sciences Beijing 100049China Department of Chemical and Environmental engineering University of Nottingham NingboNingbo 315100China

The micro-morphology and molecular stacking play a key role in determining the charge transport process and nonradiative energy loss, thus impacting the performances of organic solar cells(OSCs). To address this issue, a non-fullerene acceptor PhC6-IC-F with alkylbenzene side-chain, possessing optimized molecular stacking, complementary absorption spectra and forming a cascade energy level alignment in the PM6:BTP-eC9 blend, is introduced as guest acceptor to improve efficiency of ternary OSCs. The bulky phenyl in the side-chain can regulate crystallinity and optimizing phase separation between receptors in ternary blend films, resulting in the optimal phase separations in the ternary films. As a result, high efficiencies of 18.33% as photovoltaic layer are obtained for PhC6-IC-F-based ternary devices with excellent fill factor(FF) of 78.92%. Impressively, the ternary system produces a significantly improved open circuit voltage(V_(oc)) of 0.857 V compared with the binary device,contributing to the reduced density of trap states and suppressed non-radiative recombination result in lower energy loss. This work demonstrates an effective approach for adjusting the aggregation, molecular packing and fine phase separation morphology to increase V_(oc) and FF, paving the way toward high-efficiency OSCs.

关键词： alkylbenzene side-chain morphology regulation ternary blends organic solar cells