A5101671137- Remote Sensing and Land Use
- Software System Performance and Reliability
- Remote Sensing in Agriculture
- Network Security and Intrusion Detection
- Software Engineering Research
- Advanced biosensing and bioanalysis techniques
- Dengue and Mosquito Control Research
- Machine Learning and Algorithms
- Software Testing and Debugging Techniques
- Climate change impacts on agriculture
- Text and Document Classification Technologies
- IoT and Edge/Fog Computing
- Data Stream Mining Techniques
- Land Use and Ecosystem Services
- Cloud Computing and Resource Management
- Cloud Data Security Solutions
- Remote-Sensing Image Classification
- Anomaly Detection Techniques and Applications
- Genetics, Aging, and Longevity in Model Organisms
- Environmental Quality and Pollution
- CRISPR and Genetic Engineering
Shanghai Jiao Tong University
2025
Microsoft (Norway)
2023
Nanjing University
2022
Institute of Remote Sensing and Digital Earth
2014-2015
Chinese Academy of Sciences
2014
Taking advantage of multiple new remote sensing data sources, especially from Chinese satellites, the CropWatch system has expanded scope its international analyses through development indicators and an upgraded operational methodology. The approach adopts a hierarchical covering four spatial levels detail: global, regional, national (thirty-one key countries including China) “sub-countries” (for nine largest countries). thirty-one encompass more that 80% both production exports maize, rice,...
Crop condition assessment in the early growing stage is essential for crop monitoring and yield prediction. A normalized difference vegetation index (NDVI)-based method employed to evaluate by inter-annual comparisons of both spatial variability (using NDVI images) seasonal dynamics (based on profiles). Since this type will generate false information if there are changes rotation, cropping area or phenology, cropped/uncropped arable land integrated improve accuracy monitoring. The study...
Performance anomaly alerting based on trace data plays an important role in assuring the quality of online service systems. However, engineers find that many anomalies reported by existing techniques are not interest for them to take further actions. For a large scale with hundreds different microservices, current methods either fire lots false alarms applying simple thresholds temporal metrics (i.e., latency), or run complex end-to-end deep learning model limited interpretability. Engineers...
In microservice systems, the identification of root causes anomalies is imperative for service reliability and business impact. This process typically divided into two phases: (i)constructing a dependency graph that outlines sequence structure system components are invoked, (ii) localizing cause using graph, traces, logs, Key Performance Indicators (KPIs) such as latency. However, both phases not straightforward due to highly dynamic complex nature system, particularly in large-scale...
Regulation of the target DNA cleavage activity CRISPR/Cas has naturally evolved in a few bacteria or bacteriophages but is lacking higher species. Thus, identification bioactive agents and mechanisms that can suppress Cas9 urgently needed to rebalance this new genetic pressure. Here, we identify four specific inhibitors by screening 4607 compounds could inhibit endonuclease via three distinct mechanisms: substrate-competitive protospacer adjacent motif (PAM)-binding site-occupation;...
The aim of this study was to investigate the utility datasets composed texture measures and other features for classification vegetation cover, specifically wetlands. QUEST decision tree classifier applied a SPOT-5 image sub-scene covering typical wetlands area in Three River Sources region Qinghai province, China. dataset used comprised of: (1) spectral data components principal component analysis; (2) derived from pixel basis; (3) DEM ancillary research area. Image textures is an important...
Global education industry informatization wave with the maturing of cloud computing force, introduction is to promote school transformation. Through analysis computer resources update difficult and high cost, inconvenience unified management maintenance, low resource utilization, traditional information system itself cannot be Shared, security, defects such as energy consumption. To solve above problems, FusionSphere, Huawei's commercial version platform based on open source OpenStack...
æåºåä½ç©é¿å¿ç»¼åçæµæ¹æ³,å©ç¨å«æé¥æå¾å°çNDVIæ¶é´åºåæ°æ®,综åéç¨å®æ¶çæµãè¿ç¨çæµåæ¶é´åºåèç±»çæµæ¹æ³,æç¡®ä¸åæ¹æ³éç¨ççæµå°ºåº¦åçæµç®ç,对ä¸åèå´åä½ç©é¿å¿è¿è¡çæµãæ¹è¿äºCrop Watchå ¨çåæ é¥æéæ¥ç³»ç»è¿è¡åä½ç©é¿å¿çæµæ¹æ³,å æäºåæä½ç©é¿å¿çæµä¸å®æ¶çæµæ¹æ³æ æ³åæ ç¸ååºåèæ å¨æ´ä¸ªçé¿è¿ç¨ä¸çè¿ç»ååæ...
Pre-trained models (PTMs) have achieved great success in various Software Engineering (SE) downstream tasks following the ``pre-train then fine-tune'' paradigm. As fully fine-tuning all parameters of PTMs can be computationally expensive, a widely used solution is parameter-efficient (PEFT), which freezes while introducing extra parameters. Though work has been done to test PEFT methods SE field, comprehensive evaluation still lacking. This paper aims fill this gap by evaluating...
Classifying user requests is the first step of transforming them into structured requirements specifications. Manual classification time-consuming, labor-intensive and error prone. Many supervised learning methods therefore have been proposed for automatic classification. These need to manually annotate large amount data instances train a model automatically classify others. To reduce workload manual annotation, Active Learning (AL) proposed. Few works done employ AL in However, these could...