A5100393253- Advanced Image and Video Retrieval Techniques
- Behavioral Health and Interventions
- Image Retrieval and Classification Techniques
- Physical Activity and Health
- Face and Expression Recognition
- Matrix Theory and Algorithms
- Recommender Systems and Techniques
- Mobile Health and mHealth Applications
- Web Data Mining and Analysis
- Safety Systems Engineering in Autonomy
- Flexible and Reconfigurable Manufacturing Systems
- Gene expression and cancer classification
- Stability and Control of Uncertain Systems
- Machine Learning in Bioinformatics
- Model Reduction and Neural Networks
- Nuclear Engineering Thermal-Hydraulics
- Advanced Vision and Imaging
- COVID-19 and Mental Health
- Software Reliability and Analysis Research
- Traffic and Road Safety
- Formal Methods in Verification
- Control Systems and Identification
- Robotics and Sensor-Based Localization
- Advanced Decision-Making Techniques
- Numerical methods for differential equations
Shandong University of Science and Technology
2024
University of Glasgow
2019-2022
Shanghai University of Sport
2022
Henan University
2021-2022
Beijing Institute of Technology
2015
Northwestern University
2013
Tsinghua University
2008
Feng Chia University
1995
Osaka University
1994
Factorization Machine (FM) is an effective solution for context-aware recommender systems (CARS) which models second-order feature interactions by inner product. However, it insufficient to capture high-order and nonlinear interaction signals. While several recent efforts have enhanced FM with neural networks, they assume the embedding dimensions are independent from each other model in a rather implicit manner. In this paper, we propose Convolutional (CFM) address above limitations....
The inner-outer factorization of a non-square proper function which has infinite and finite jomega -axis zeros no poles in the open right-half plane is studied this paper via concept an zero compensator. For stabilizable detectable realization above with full normal column rank, eigenstructures system matrix pencils its spectral density are discussed based on Kronecker's theory, simple state-space solution to given solutions generalized eigenvalue problems terms original function.
Web service recommendation is one of the most important fields research in area computing. The two core problems are prediction unknown QoS property values and evaluation overall according to user preferences. Aiming address these their current challenges, we propose efficient approaches solve problems. First, were predicted by modeling high-dimensional data as tensors, utilizing an tensor operation, i.e., composition, predict values. Our method, which considers all dimensions integrally...
Factorization machine (FM) is an effective model for feature-based recommendation which utilizes inner product to capture second-order feature interactions. However, one of the major drawbacks FM that it couldn't complex high-order interaction signals. A common solution change function, such as stacking deep neural networks on top FM. In this work, we propose alternative approach signals in embedding level, namely Generalized Embedding Machine (GEM). The used GEM encodes not only information...
Abstract Singular (J, J')-lossless factorization for strictly proper functions is studied. The method of infinite zero compensation derived by Copeland and Safonov used to extend the results Kimura singular systems. Since contains both inner–outer Wiener–Hopf spectral as special cases, obtained here unify previous on problems. Additional informationNotes contributorsXIN XIN Current address: Research Institute Automation, Southeast University, Nanjing 210018, People's Republic China.
We address the problem of academic conference homepage understanding for Semantic Web. This consists three labeling tasks - function pages, blocks, and attributes. Different from traditional information extraction tasks, data in homepages has complex structural dependencies across multiple Web pages. In addition, there are logical constraints data. this paper, we propose a unified approach, Constrained Hierarchical Conditional Random Fields, to accomplish simultaneously. can be well...
Industry 4.0 adopts Internet of Things (IoT) and service-oriented architectures to integrate Cyber-Physical Systems Enterprise Planning into manufacturing operations. This kind integration consists a combination connected sensors run-time control algorithms. Consequential decisions are driven by sensor-generated data. Hence, the trustworthiness sensor network readings is increasingly crucial guarantee performance quality task. However, existing methodologies test such systems often do not...
This paper proposes a Content Based Image Retrieval (CBIR) application for searching landmarks and buildings in city using smartphone. A user can snap picture of the building his The is able to quickly accurately find name along with many other interesting information, such as history its Wi-Fi availability. We present novel client-server CBIR that combines Laplacian-SIFT feature descriptor, multiple kd-trees indexing two levels geometric verification. back-end front-end Application...
Cyber-Physical Systems (CPSs) are the core of Industry 4.0 applications, integrating advanced technologies such as sensing, data analytics, and artificial intelligence. This kind combination typically consists networked sensors decision-making processes in which sensor-generated drive control decisions. Hence, trustworthiness is essential to guarantee performance, safety quality during operation. Formal model verification techniques a valuable tool allowing strong reasoning about high-level...
<sec> <title>BACKGROUND</title> Evidence suggests low physical activity (PA) levels among college students could be addressable by both intrapersonal fitness tracking (e.g., step-counting) and interpersonal support tailored to the campus environment. However, no studies have examined appeal influence of intra- components in campus-based PA interventions. </sec> <title>OBJECTIVE</title> Retrospectively observe a promotion program compare usage intervention between faculty/staff, then test...
Abstract Objectives: This study aimed to analyze the relationship between physical activity level, executive function, and trait anxiety focused on investigating mediating role of subcomponents functions different levels intensity anxiety. Methods: The International Physical Activity Questionnaire, State-Trait Anxiety Inventory, Stroop task, 1-back More-odd shifting task were used 248 college students by one-way ANOVA, linear regression analysis, structural equation modeling analysis....
A comparison theorem for the solutions of two generalized algebraic Riccati equations (GAREs) coming from different systems is presented in this paper. It then shown that so-called strong solutions, whose related pencils have all their finite eigenvalues closed left half plane, are maximal. The results obtained paper generalize existing monotonicity equations. An application derivation parameterization GARE to singular spectral factorization a proper transfer function with and infinite...