A5100421950- Machine Learning and ELM
- Brain Tumor Detection and Classification
- Power System Optimization and Stability
- Advanced Technologies in Various Fields
- Robotic Path Planning Algorithms
- Educational Technology and Assessment
- Power System Reliability and Maintenance
- Ferroelectric and Negative Capacitance Devices
- Stochastic Gradient Optimization Techniques
- Topic Modeling
- Advanced Manufacturing and Logistics Optimization
- Neural Networks and Applications
- Energy Load and Power Forecasting
- Hydrological Forecasting Using AI
- Water resources management and optimization
- Electric Power System Optimization
- Homelessness and Social Issues
- Hand Gesture Recognition Systems
- Face and Expression Recognition
- Disaster Management and Resilience
- Smart Grid Security and Resilience
- Fuel Cells and Related Materials
- Community Health and Development
- Electrocatalysts for Energy Conversion
- Robotics and Automated Systems
Tianjin University of Technology and Education
2022-2025
China University of Mining and Technology
2021-2023
Shenyang Jianzhu University
2023
Shandong Management University
2022
Kunming University of Science and Technology
2021
Chinese Academy of Sciences
2019
Institute of Mountain Hazards and Environment
2019
Changsha University of Science and Technology
2019
Nanyang Technological University
2012-2013
University of Newcastle Australia
2013
Artificial Neural Network (ANN) has been recognized as a powerful method for short‐term load forecasting (STLF) of power systems. However, traditional ANNs are mostly trained by gradient‐based learning algorithms which usually suffer from excessive training and tuning burden well unsatisfactory generalization performance. Based on the ensemble strategy, this paper develops an model promising novel technology called extreme machine (ELM) high‐quality STLF Australian National Electricity...
Extreme learning machines (ELMs) have been proposed for generalized single-hidden-layer feedforward networks which need not be neuron alike and perform well in both regression classification applications. The problem of determining the suitable network architectures is recognized to crucial successful application ELMs. This paper first proposes a dynamic ELM (D-ELM) where hidden nodes can recruited or deleted dynamically according their significance performance, so that only parameters...
Introduction The COVID-19 pandemic has been a global public health emergency, and countries worldwide have responded to it through vast array of pre-planned, adaptively devised ad-hoc measures. In China, emergency plans - the expected drive response epidemics or pandemics demonstrated concerning tendency towards “ritualization.” “Ritualization” denotes practice be reliably developed so that formal requirement is met, while being implemented selectively not at all in response. Methods This...
Credit risk evaluation has become an increasingly important field in financial management for institutions, especially banks and credit card companies. Many data mining statistical methods have been applied to this field. Extreme learning machine (ELM) classifier as a type of generalized single hidden layer feed-forward networks used many applications achieve good classification accuracy. Thus, we use ELM (kernel based) tool perform the paper. The simulations are done on two datasets with...
Intelligent systems (IS) have gained popularity in facilitating very fast dynamic security assessment (DSA). However, conventional IS methods are limited their ability to be updated with current system operation conditions online due the excessive training time and complex parameters tuning required for updates. In this paper, an sequential extreme learning machine (ELM) based method is proposed enable efficient real-time DSA model updating. To enhance performance of ELMs, feature selection...
The lack of real-time tools capable detecting risk blackouts is one the contribution factors to recent large occurred around world. In terms dynamic security assessment (DSA), artificial intelligence and data mining techniques have been widely applied facilitate very fast DSA for enhanced situational awareness insecurity. However, many current state-of-the-art models usually suffer from excessive training time complex parameters tuning problems, leading their inefficiency implementation....
With the development of human-computer interaction technology, gesture recognition is becoming more and important. At same time, due to rapid automotive intelligence, introduction technology into intelligent vehicles has increasingly become an important work. Aiming at problems low accuracy, efficiency weak anti-interference ability previous applications in driving scenes. This paper presents improved yolov5 algorithm. By adding optimized k-means++ clustering optimization algorithm, unstable...
<sec> <title>BACKGROUND</title> Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language processing tasks, including medical question-answering (QA). However, individual LLMs often exhibit varying performance across different QA datasets, highlighting the need for strategies that can harness their collective strengths. Ensemble learning methods, which combine multiple models to improve overall accuracy and reliability, offer a promising approach address this...
Recently, a novel learning algorithm called the broad system (BLS) has been established by using theory of pseudoinverse and compressed sensing technology. As an alternative to deep learning, BLS gained extensive attention in data analytics. Unlike existing structures based tools (such as neural networks), it is able rapidly achieve incremental does not require tedious retraining remodel system. However, given image modeling tasks, 1-D requires conventional vectorization operation, dimension...
In image classification and recognition, in order to obtain higher accuracy, it is necessary extract more accurate expressive features of semantic information. This paper proposes an improved model based on convolutional neural network. By improving optimizing AlexNet from both the network architecture internal structure, we can further improve expression ability features. introducing extreme learning machine full connection layer, not only time are improved, but also structure has better...
Abstract Before COVID-19, although the online assessment platform has developed, it is relatively slow, and people prefer to organize on-site examinations. After outbreak of epidemic, realize urgency necessity information construction in all walks life. More more researchers begin pay attention explore use advanced machine learning methods improve practicability examinations platform. The test paper generation function core link, a good can ensure quality paper. In this paper, an...