A5100657320- Algorithms and Data Compression
- Wireless Communication Networks Research
- Advanced Wireless Communication Techniques
- Blind Source Separation Techniques
- Machine Fault Diagnosis Techniques
- Natural Language Processing Techniques
- Data Management and Algorithms
- Data Mining Algorithms and Applications
- Topic Modeling
- Fault Detection and Control Systems
- Engineering and Test Systems
- Advanced Data Storage Technologies
- Industrial Technology and Control Systems
- Network Packet Processing and Optimization
- Genomics and Phylogenetic Studies
- Simulation and Modeling Applications
- Speech and dialogue systems
- Web Data Mining and Analysis
- Stock Market Forecasting Methods
- Time Series Analysis and Forecasting
- Remote Sensing and Land Use
- Software System Performance and Reliability
- Wireless Communication Security Techniques
- Chaos control and synchronization
- Water Resources and Sustainability
Kobe University
2022-2024
Chinese Academy for Environmental Planning
2024
Kunming Medical University
2024
Shijiazhuang Tiedao University
2010-2022
Anhui University
2021
Inner Mongolia Agricultural University
2012
Michigan State University
2003-2009
Naval University of Engineering
2009
Beijing University of Posts and Telecommunications
2002-2005
Effective feature extraction is crucial for accurate fault diagnosis of rolling bearings. A novel method called hierarchical dispersion entropy (HDE) based on analysis proposed in this study. The includes the following three steps: 1) bearing vibration signal decomposed into a series subband components; 2) entropies components different frequency bands are calculated as original vector; and 3) joint approximate diagonalization eigenmatrices (JADE) used to extract fusion features from...
Abstract The Beautiful China Initiative (BCI) is a vivid embodiment of the harmonious coexistence between humans and nature during modernization. Implementing BCI an effective method for achieving goals building beautiful China, while offering “Chinese solution” to global sustainable development. This article summarizes progress main experiences BCI, as well analyzing primary challenges facing its future Finally, five policy recommendations are proposed, which emphasize importance top-level...
Similarity searches in multidimensional Non-ordered Discrete Data Spaces (NDDS) are becoming increasingly important for application areas such as bioinformatics, biometrics, data mining and E-commerce. Efficient similarity require robust indexing techniques. Unfortunately, existing methods developed (ordered) Continuous (CDS) the R-tree cannot be directly applied to an NDDS. This is because some essential geometric concepts/properties minimum bounding region area of a CDS no longer valid...
There is an increasing demand for similarity searches in a multidimensional non-ordered discrete data space (NDDS) from application areas such as bioinformatics and mining. The nature of NDDS raises new challenges developing efficient indexing methods searches. In this article, we propose technique, called the NSP-tree , to support NDDS. As know, overlap causes performance degradation (e.g., R-tree) continuous space. NDDS, problem even worse due limited number elements available on each...
A mathematical model of the test selection with unreliable is created and a heuristic function point are established based on tolerances detection, coverage reliance in this paper. In order to overcome computational explosion, genetic algorithm proposed settle problem test. The illustrated tested using range real-word system. examples show that significantly better than simple algorithm. It also feasible effective, especially large-scale
Abstract The periodic‐phase‐diagram similarity method is proposed to identify the frequency of weak harmonic signals. key technology find a set optimal coefficients for Duffing system, which leads periodic motion under influence signal and strong noise. Introducing phase diagram similarity, influences noise on are discussed. principle highest with same detected by discussing persistence constructed. signals early fault input into system obtain identified system. stochastic subharmonic...
In recent years, generation-based dialogue systems using state-of-the-art (SoTA) transformer-based models have demonstrated impressive performance in simulating human-like conversations. To improve the coherence and knowledge utilization capabilities of systems, knowledge-based integrate retrieved graph into models. However, dialog sometimes generate responses without knowledge. this work, we propose a method which system can constantly utilize text infilling . Text is task predicting...
Test sequencing is a binary identification problem wherein one needs to develop minimal expected cost testing procedure determine which of finite number possible failure sources, if any, present. directed towards the test in sequential fault diagnosis for systems, converted searching complete based on ant algorithm. In order overcome computational explosion optimal sequence searched by setting up state transfer rule and feedback pheromones. The better performance feasibility proposed method...
This paper introduces a novel multimodal framework for economic time series forecasting, integrating textual information with historical price data to enhance predictive accuracy. The proposed method employs multi-head attention mechanism dynamically align embeddings temporal data, capturing previously unrecognized cross-modal dependencies and enhancing the model’s ability interpret event-driven market dynamics. enables model complex behaviors in unified effective manner. Experimental...
Abstract In financial markets, the sentiment expressed in news articles plays a pivotal role interpreting and forecasting market trends, which also holds true for task of summarization (FNS). Leveraging AI models to analyze social science data, this paper employs improve FNS effectiveness by introducing novel method that combines polarity extracted from with prompt augmentation techniques ensure generated summaries are emotionally consistent source articles. Specifically, detected sentiments...
In this paper, a novel blind adaptive multiuser detection is proposed. The computational complexity of algorithm O(NL), where N the spread gain, and L window length. has good steady state performance very fast convergence. Simulations indicate that better than CMOE can be used in dynamic multiple-access channel interference may enter or exit channel.
The profile hidden Markov model (PHMM) has received increasing attention in the field of protein homology detection, since profile-based methods are much more sensitive detecting distant homologous relationships than pairwise methods. Pure dynamic-programming-based systems often used for PHMM searches. However, these dynamic-programming- based very time consuming a large database. For instance, it may take approximately 15 minutes to search short length 12 GenBank sequence Instead searching...
The problem for blind adaptive multi-user to detect the direct-sequence code-division multiple-access (DS/CDMA) signals in multipath fading channels is investigated this paper. A novel detection algorithm based on constant modulus (CMA) proposed. Analysis shows that satisfies decorrelating conditions. This reduces complexity remarkably as compared with traditional SVD subspace algorithm. Simulation results demonstrate offers higher detecting performance and better resistance both channel...
Based on the rectification of detectable domain method and elimination blind method, variable scale-convex-peak for identification frequency weak signal is improved. For initial phase to be detected, analytic expression chaotic threshold obtained by using stochastic Melnikov method. The influence clarified, which leads that can identified. Specifically, in a period divided into domain. In domain, when deviates, introduced. Newton interpolation used give calculation formula rectify deviation,...