- Blockchain Technology Applications and Security
- Face and Expression Recognition
- Flow Measurement and Analysis
- Remote-Sensing Image Classification
- Machine Learning and ELM
- Text and Document Classification Technologies
- Domain Adaptation and Few-Shot Learning
- Ultrasonics and Acoustic Wave Propagation
- Air Quality Monitoring and Forecasting
- Statistical Methods in Clinical Trials
- Caching and Content Delivery
- Neural Networks and Applications
- Sentiment Analysis and Opinion Mining
- IoT and Edge/Fog Computing
- Bayesian Modeling and Causal Inference
- Advanced Text Analysis Techniques
- Hydrological Forecasting Using AI
- Cancer-related molecular mechanisms research
- Imbalanced Data Classification Techniques
- Supply Chain Resilience and Risk Management
- Digital Marketing and Social Media
- Time Series Analysis and Forecasting
- Energy and Environment Impacts
- Statistical Distribution Estimation and Applications
- Anomaly Detection Techniques and Applications
Guiyang Medical University
2019-2025
Jiangxi Agricultural University
2021-2023
Affiliated Hospital of Guizhou Medical University
2023
Chongqing University
2019-2021
Ministry of Education of the People's Republic of China
2019-2021
Given the issues of low efficiency agricultural Internet Things (IoT) data collection and storage security, this study proposes a fast reliable method for IoT based on blockchain. Firstly, it performs RC5 encryption in sensor module. Secondly, aggregates same batch collected gateway into transaction reconstructs Merkle ordered tree to verify integrity. Finally, modifies configuration rules blockchain improve storage. Compared with experimental results hash values stored itself blockchain,...
This paper presents an ultrasonic thermometry method for thermal image distribution reconstruction in a heating environment. It is based on determining the time-of-flight (ToF) of waves along large number effective paths measured area, these ToFs depend temperature paths. On this basis, can be reconstructed by some appropriate methods. During process, flight time measurement system analog front end TDC1000 and time-to-digital converter TDC7200 built to obtain ToF data, reliable...
Abstract Bladder cancer (BLCA) typically has a poor prognosis due to high rates of relapse and metastasis. Although the emergence immunotherapy brings hope for patients with BLCA, not all will benefit from it. Identifying some markers predict treatment response is particularly important. Here, we aimed determine clinical value ribonuclease/angiogenin inhibitor 1 (RNH1) in BLCA therapy based on functional status analysis. First, found that RNH1 aberrantly expressed multiple cancers but...
The data credibility tracing for agricultural product transactions is affected by all organizations in the entire supply chain. For traditional centralized transaction management, are only stored one organization chain, making management non-transparent and not credible. However, blockchain connects information blocks into a tamper-proof distributed chain ledger Hash Function, which can help solve problem. But access performance still needs improvement. With Hyperledger Fabric as underlying...
The categorical attributes in mixed data are represented as high-quality numerical attributes, which have been widely concerned by the processing community. existing representation methods do not subdivide into ordinal and nominal (i.e., these consider order of attributes), resulting loss ordering relationship within may fail to achieve desired performance. To this end, we take ordinal, object study, fuse various intra- inter-attribute feature relationships from three propose a Categorical...
As companies continue to generate merchandise sales, the data is important for their marketing planning, market analysis, and logistics so it accurately predict sales. This paper combines a recently proposed attention mechanism multiple time series regression model, CNN-BiLSTM-Attention product sales next three months. The model can fix overfitting problem, volume of products with high accuracy.
In order to save on test costs, the optimal truncated sequential (OTST) for parameter of exponential distribution is studied. According Bayes decision theory, dynamic programming methods and procedures are established solve OTST. Through comparison with plans provided by international standard IEC 61124 Russian national GOST R 27.402, results show that OTSTs solved our new method can control error probabilities strictly more synthetical expected time (SETT). The also compared near-optimal...
The traditional Raft algorithm has such an issue as "vote snatching" among candidates that results in abnormal elections the absence of a majority. To address this issue, random forest-based method is introduced for identification authorized nodes. First, attributes authorization are to original be label nodes, i.e., only nodes eligible candidacy. Second, forest classify according their term, broadcasting duration, timeout and other characteristics divide all into preferred ordinary In end,...
Traditional ring signature algorithms suffer from large data capacity and low speed of verification during collective signing. In this work, we propose a representative algorithm based on smart contracts. By collecting the opinions signatory multiparty secure computation, proposed technique protects privacy interaction process in consortium chain. Moreover, method uses contracts to organize formulate strategy "one encryption per signature" prevent forgery. It Hyperledger Fabric framework as...
With the increasing application of blockchain technology in decentralized data management systems (DMS), prevailing problems poor query performance relative to its conventionally centralized counterparts are urgent need reasonable solutions. Bloom filters efficient frameworks that adopt multiple hash functions map target database into one-dimensional arrays (one bit per array cell), leading high-efficiency information extraction and fulfilling optimization needs blockchain's performance. In...
In the mobile network, user can share their trip stories at any time and place, massive data provides an opportunity to mining users' travel preferences. Tourism recommendation has been studied by lots of researchers, although they took into account social relation users need travelers, but if multi-dimensional information seasonality tourism service are considered, it may lead better personality efficiency for system. this paper, we propose a novel personalized model termed GA-LSTM_CSInf,...
Many researchers have applied neural networks to mangrove ecosystem health research. However, it is challenging find efficient training algorithms for networks, so the prediction results are often difficult meet needs. The dynamic group evolution (DGE) algorithm a recently proposed metaheuristic algorithm, which exhibits rapid convergence rate and good performance in searching avoiding local optima. In present study, we use DGE back propagation (BP) feed-forward network build DGBPNN model,...