A5101960445- Natural Language Processing Techniques
- Topic Modeling
- Marine Bivalve and Aquaculture Studies
- Data Management and Algorithms
- Advanced Computational Techniques and Applications
- Advanced Text Analysis Techniques
- Speech Recognition and Synthesis
- Genetic and phenotypic traits in livestock
- Physical Unclonable Functions (PUFs) and Hardware Security
- Meat and Animal Product Quality
- Advanced Malware Detection Techniques
- Cryptographic Implementations and Security
- Aquatic Invertebrate Ecology and Behavior
- Metaheuristic Optimization Algorithms Research
- Speech and dialogue systems
- Hydrology and Watershed Management Studies
- Text and Document Classification Technologies
- Web Data Mining and Analysis
- Face and Expression Recognition
- Data Mining Algorithms and Applications
- Genetic diversity and population structure
- Advanced Multi-Objective Optimization Algorithms
- Biomedical Text Mining and Ontologies
- Artificial Immune Systems Applications
- Digital Media Forensic Detection
Xiamen University
2015-2021
Beijing University of Posts and Telecommunications
2019-2021
Institute of Botany
2015
Chinese Academy of Sciences
2015
Shandong Women’s University
2015
Guilin University of Electronic Technology
2015
Hainan Agricultural School
2013
Fujitsu (China)
2008
In attempting to overcome the limitation of current methods solve complicated constrained optimization problems, this paper proposes an adaptive hybrid particle swarm multi-objective (AHPSOMO) algorithm. early stage, algorithm initializes individuals in a population even manner using good point set (GPS) theory so that diversity can be guaranteed. process local search, differential evolution (DE) is introduced for updating optimal individuals. Particle method further adopted conduct global...
Compared to the injection of a transient fault, time synchronization and accuracy are not required for process persistent fault. However, known <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">persistent fault analyses</i> (PFAs) do work on SM4 implementations because linear transformation layer hides position where an error occurs during encryption process. We present first analysis</i> against implemented with S-box by combining inverse...
K-medoids clustering algorithm is an efficient in classifying cluster categories. Based on analysis, this paper first improves the selection of K center point and then sets up a web model ontology data set object with aim demonstrating through experiment evaluation that improved can greatly enhance accuracy results under semantic web.
The emerging success of large language models (LLMs) heavily relies on collecting abundant training data from external (untrusted) sources. Despite substantial efforts devoted to cleaning and curation, well-constructed LLMs have been reported suffer copyright infringement, poisoning, and/or privacy violations, which would impede practical deployment LLMs. In this study, we propose a simple easily implementable method for purifying the negative effects caused by uncurated data, namely,...
In recent years, the unit selection based concatenative speech synthesis system that uses large database has become popular because it can produce high quality synthesized speech.However, using such a is not practical for many applications as those ported on embedded devices with storage requirement and computational complexity involved in searching it.In this paper, proposed context pruning algorithm waveform adjustment effect to compact database.At last, presents experimental results discussion.
Cluster ensemble combines multiple partitions of a set objects into stable and robust one. To obtain good performance, base clusterings are required to take account quality diversity. Recently, in spite some researches focus on consensus function prove quality, how produce high-diversity high-quality at general step without global screening remains an open problem. For high-dimensional data, the clustering algorithm suitable for common data sets is extremely inefficient there basically no...
This paper discusses the ways English is translated into Chinese by applying some sentences with special verbal predicates peculiar to language.In this methods involves constructions in series, and pivotal sentences.This introduces knowledge of grammar briefly then mentions what sentence patterns or structures can be using knowledge.
Lots of research findings have been made from home and abroad on clustering algorithm in recent years. In view the traditional partition method K-means algorithm, this paper, after analyzing its advantages disadvantages, combines it with ontology-based data set to establish a semantic web model. It improves existing various constraint conditions aim demonstrating that improved has better efficiency accuracy under web.
In this paper, a global prosodic word grouping probability estimation method is proposed.By using some statistical probabilities of four kinds lexical position types, the optimum path sentence can be obtained with dynamic programming approach.In addition, and rule punish or encourage strategies have been used to improve accuracy grouping.Lastly, experiment results discussion are presented.