A5078773589- Catalytic Processes in Materials Science
- Control Systems in Engineering
- Iterative Learning Control Systems
- Software Reliability and Analysis Research
- Software Engineering Research
- Topic Modeling
- Carbon dioxide utilization in catalysis
- Mercury impact and mitigation studies
- Advanced Photocatalysis Techniques
- Metallurgy and Material Forming
- Nanomaterials for catalytic reactions
- Catalysis and Oxidation Reactions
- Software Testing and Debugging Techniques
- Software System Performance and Reliability
- Industrial Gas Emission Control
- Advanced Text Analysis Techniques
- Natural Language Processing Techniques
- Ergonomics and Human Factors
- Thermochemical Biomass Conversion Processes
- Microwave Dielectric Ceramics Synthesis
- Catalysis for Biomass Conversion
- Personality Disorders and Psychopathology
- Dielectric materials and actuators
- Environmental and Industrial Safety
- Ammonia Synthesis and Nitrogen Reduction
Chengdu University
2024-2025
Yanshan University
2022-2024
Institute of Electrical Engineering
2022-2023
Guangdong University of Technology
2022-2023
Sichuan University
2023
Institute of Process Engineering
2013-2022
Yunnan University of Traditional Chinese Medicine
2022
Norsk Hydro (Slovakia)
2020
Chinese Academy of Sciences
2020
Hunan University
2017-2018
Software fault prediction is a valuable exercise in software quality assurance to best allocate limited testing resources. Classification one of the effective methods for prediction. The classification models are trained based on datasets obtained by mining historical repositories. However, performance depends datasets. In this paper, we propose novel two-stage data preprocessing approach which incorporates both feature selection and instance reduction. Specifically, stage, first perform...
Software defect prediction can classify new software entities into either buggy or clean. However the effectiveness of existing methods is influenced by irrelevant and redundant features. In this paper, we propose a feature selection framework FECAR using Feature Clustering And Ranking. This firstly partitions original features k clusters based on FF-Correlation measure. Then it selects relevant from each cluster FC-Relevance empirical study, choose Symmetric Uncertainty as measure,...
Rhodiola crenulata (Hook.f. & Thomson) H.Ohba, a member of the Crassulaceae family, is traditional Chinese medicine recognized as original source Rhodiolae Crenulatae Radix et Rhizoma in 2020 edition China Pharmacopoeia. It has been widely used both Asia and Europe to enhance stress resistance reduce fatigue. However, classification species can lead confusion, raising safety concerns herbal market. The cleaved amplified polymorphic sequence (CAPS) RT-PCR was identify single nucleotide...
Software fault prediction is valuable in predicting proneness of software modules and then limited test resources can be effectively allocated for quality assurance. Researchers have proved that either feature selection or instance reduction improve the performance classification models used prediction. However, to best our knowledge, few researchers combined them study effects on models. Therefore we propose a novel two-stage data preprocessing approach, which incorporates both reduction....
Polygonatum kingianum Collett & Hemsl is one of the famous traditional Chinese herbs with satisfactory therapeutic effects on invigorating Qi, nourishing Yin and moistening lungs, in which steroidal saponins are class important active substances. The main purpose to determine optimal extraction technology evaluate quality P. planted five different areas. ultrasonic-assisted (UAE) was established by using single-factor experiments response surface methodology (RSM), determination method...
Abstract The hydrogenation of ethylene carbonate to co‐produce commodity methanol and glycol has attracted growing interests due the potential chemical utilization CO 2 in large scale. In this work, we report a novel facile protocol for preparation mesoporous Cu@SiO catalysts successfully applied as‐synthesized carbonate. were characterized detail by means N physisorption, O titration, XRD, FT‐IR, H ‐TPR, TEM XPS (XAES). This strategy is an effective method fabricating unique flower‐like...
A series of Ru/g-C<sub>3</sub>N<sub>4</sub> materials with highly dispersed Ru were firstly prepared by an ultrasonic impregnation method using carbon nitride as a support.
The separation of ethylene glycol (EG) and 1,2-butanediol (1,2-BDO) azeotrope in the synthesis process EG via coal biomass is becoming increasing commercial environmental importance. Selective adsorption deemed as most promising methods because energy saving environment favorable. In this paper, NaY zeolite was used to separate 1,2-BDO from EG, its properties then investigated. isotherms vapor liquid phases 298 328 K indicated that they fitted Langmuir model quite well, absorbent favored...
In this study, nanosheet g-C3N4-H2 was prepared by thermal exfoliation of bulk g-C3N4 under hydrogen. A series Ru/g-C3N4-H2 catalysts with Ru species supported on the were synthesized via ultrasonic assisted impregnation-deposition method. Ultrafine nanoparticles (<2 nm) highly dispersed g-C3N4-H2. Strong interaction due to Ru-Nx coordination facilitated uniform distribution species. Meanwhile, involvement surface basicity derived from abundant nitrogen sites favourable for enhancing...
There has recently been a surge of research on neural network-inspired algorithms to produce numerical vector representations words, based contextual information. In this paper, we present an approach improve such word embeddings by first mining cognitively salient relationships from text and then using stochastic gradient descent jointly optimize the reflect information, in addition regular information captured word2vec CBOW objective. Our findings show that new training regime leads...
This study presents a clean, novel, and highly efficient system for the decomposition of methyl N-phenyl carbamate (MPC) to phenyl isocyanate (PI). MPC conversion high as 89% PI selectivity almost 100% were obtained using chlorobenzene solvent under pressure without any catalyst. The effects temperature time on investigated, apparent kinetics model reaction was established. Results indicate that is pseudo-first-order with Arrhenius parameters Ea = 71.28 kJ/mol A 2.74 × 105 min–1, correlated...
Word embeddings have recently proven useful in a number of different applications that deal with natural language.Such succinctly reflect semantic similarities between words based on their sentence-internal contexts large corpora.In this paper, we show information extraction techniques provide valuable additional evidence relationships can be exploited when producing word embeddings.We propose joint model to train both regular context and more explicit extractions.The vectors obtained from...
Classification based software fault prediction methods aim to classify the modules into either fault-prone or non-fault-prone. Feature selection is a preprocess step used improve data quality. However most of previous research mainly focus on feature relevance analysis, there little work focusing redundancy analysis. Therefore we propose two-stage framework for solve this issue. In particular, during phase, adopt three different measures obtain relevant subset. Then analysis use...
Purpose The main supporting frame of steel structure buildings is steel, and the beam-column joints directly affect stability strength frame. Design/methodology/approach This paper briefly introduced which are used for ensuring in building, selected fabricated were different from traditional welding methods, tested with reaction jack detected influence thickness length splice plate on mechanical properties joints. Findings results showed that joint stress displacement vertical direction...
In this work, the isomer mixture of 4,4'-diphenylmethane diisocyanate (MDI) and 2,4'-MDI was separated purified by dynamic falling film melt crystallization, 99.3% purity 50.8% yield 4,4'-MDI could be obtained under optimized conditions. The separation mechanism simulated density functional theory (DFT) molecular dynamics (MD) simulation. Results showed that compared with 2,4'-MDI, molecules form a more stable symmetrical crystal structure due to their stronger charge symmetry electrostatic...
Persistent key-value (KV) stores are widely used by cloud services at ByteDance as local storage engines, and RocksDB to be the de facto implementation since it can tailored a variety of workloads requirements. In this paper, we provide key insights into engine usage ByteDance, explain why combination highly write-intensive stringent requirements on cost efficiency point lookup tail latency may pose challenges general-purpose such RocksDB, present design LavaStore , high-performance...
In this paper, a novel robust low-complexity vibration suppression control scheme is proposed for the vertical-torsional-horizontal coupling of rolling mill. To facilitate design, overall system decoupled into vertical-horizontal subsystem and torsional subsystem. For subsystem, controller designed via constructed error transformations to provide strong robustness against uncertainties disturbances. new state feedback backstepping technique ensure angle motor rotor within given constraints...