A5110372015- Matrix Theory and Algorithms
- Advanced Text Analysis Techniques
- Topic Modeling
- Sentiment Analysis and Opinion Mining
- Stability and Control of Uncertain Systems
- Numerical methods for differential equations
- Financial Markets and Investment Strategies
- Credit Risk and Financial Regulations
- Mathematics and Applications
- Machine Learning and Data Classification
- Nanoparticle-Based Drug Delivery
- Adaptive Control of Nonlinear Systems
- Machine Learning and Algorithms
- Iterative Methods for Nonlinear Equations
- Spam and Phishing Detection
- Peptidase Inhibition and Analysis
- Stock Market Forecasting Methods
- Vehicle Routing Optimization Methods
- Software Engineering Research
- Radiopharmaceutical Chemistry and Applications
- Monoclonal and Polyclonal Antibodies Research
- Financial Risk and Volatility Modeling
- Web Data Mining and Analysis
- Algebraic and Geometric Analysis
- Control and Stability of Dynamical Systems
Sichuan University
2024
State Key Laboratory of Biotherapy
2024
Chinese Academy of Medical Sciences & Peking Union Medical College
2024
Xiamen University
2005-2023
Viewpoints Research Institute
2018
Hunan University
2008
In this paper we consider a credit scoring problem. We compare three powerful models: genetic programming (GP), backpropagation neural networks (BP) and support vector machines (SVM) when applied to problem, then give combined model. The results show that the model produces good classification results.
We present Variational Aspect-based Latent Topic Allocation (VALTA), a family of autoencoding topic models that learn aspect-based representations reviews. VALTA defines user-item encoder maps bag-of-words vectors for combined reviews associated with each paired user and item onto structured embeddings, which in turn define per-aspect weights. model individual manner by inferring an aspect assignment sentence given review, where the weights obtained serve to mixture over topics, conditioned...
This paper uses interpolation polynomial method to approximate Nonlinear System. The Interpolation adopted in this methodology is a combination of Lagrange's polynomials, which are obtained by Aitken formula step step. During the constructing polynomial, we infer sufficient condition such that System has limited L <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> -gain. group linear matrix inequations, can be solved LMI toolbox Matlab....
As the Internet is progressively becoming larger and more intricate, users of various social media choose to post their comments express opinions thinking on those platforms. Analyzing emotions contained in user holds great business value, helping accurately perceive consumption habits improve service levels. However, use emoticons stickers has increased dramatically recent years, which brings new challenges text sentiment analysis based natural language processing. In this paper, order...
<h3>Background</h3> Hepatocellular carcinoma (HCC) is the sixth most common cancer and fourth leading cause of cancer-related death worldwide. Effective early diagnosis hampered by lack noninvasive methods, with approximately 80% cases diagnosed late. Mesenchymal-epithelial transition factor (c-Met) plays a critical role in HCC progression metastasis serves as biomarker for tumor aggressiveness. Current PET tracers show poor vivo performance due to low liver background uptake. In this study,...
Purpose This study aims to tackle the critical issue of detecting stock market manipulation, which undermines integrity and stability financial markets globally. Even enhanced with machine learning, traditional statistical methods often struggle analyze high-frequency trading data effectively due inherent noise limited availability publicly known manipulation cases. leads poor model generalization a tendency toward over-fitting. Focusing on China's securities market, our introduces an...
In order to develop a cigarette which is less hazardous,the permeability of paper and tipping paper,filter length,cut tobacco composition,deliveries tar seven harmful smoke components,including CO,HCN,NNK,NH3,benzo[a]pyrene,phenol crotonaldehyde,of thirty domestic imported samples were analyzed,the hazard indices(HIs) these calculated compared.The results showed that: 1) Of the same type,the HIs lower than those ones.2) For all samples,the blended type higher Virginia type.3) The slightly...
Image classification for long-tailed scenarios has attracted more attention because its distribution is similar to real-world image data. From the perspective of solving imbalance at sample level, we propose a simple but effective method, named Sample-aware Knowledge Distillation (SAKD), which includes Selective module and Stable Feature Center Learning module. The former conducts knowledge distillation sample-level by selecting samples, in whether needs be distilled what extent determined...
Online shopping and fierce competition among logistics companies have led to unprecedented demand for services. However, recent studies on the Vehicle Routing Problem (VRP) considered only simple constraints. In this paper, we combine Three-Dimensional Loading Capacitated (3L-CVRP) with Time Windows (VRPTW) describe a variant of VRP, named 3L-CVRPTW. Apparently, 3L-CVRPTW accounts actual needs in industry. To address problem, first propose an innovative method loading problems, applying...
By employing the inverse function theory, a class of Non-Affine Nonlinear System was converted into Affine System. Under assumption that input satisfy inequality, sufficient condition such Systems is asymptotically stable and has limited <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</i> <sub xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> -gain provided according to Hamiltor Jacobi-inequality subject Moreover, state feedback control law presented.