A5045807579- Human Mobility and Location-Based Analysis
- Data Management and Algorithms
- Energy Efficient Wireless Sensor Networks
- Underwater Vehicles and Communication Systems
- Peer-to-Peer Network Technologies
- Sparse and Compressive Sensing Techniques
- Caching and Content Delivery
- Geographic Information Systems Studies
- Indoor and Outdoor Localization Technologies
- Astronomical Observations and Instrumentation
- Image Retrieval and Classification Techniques
- Metal-Organic Frameworks: Synthesis and Applications
- Recommender Systems and Techniques
- Advanced Vision and Imaging
- Video Analysis and Summarization
- Advanced Image and Video Retrieval Techniques
- Stellar, planetary, and galactic studies
- Advanced Graph Neural Networks
- Enzyme Catalysis and Immobilization
- Astronomy and Astrophysical Research
- Water Quality Monitoring Technologies
- Topic Modeling
- Time Series Analysis and Forecasting
- Public Relations and Crisis Communication
- Anomaly Detection Techniques and Applications
Xiamen University of Technology
2025
Tsinghua University
2008-2024
Nanjing University of Posts and Telecommunications
2015-2024
Hunan University of Science and Technology
2024
Data Assurance and Communication Security
2019-2023
University of Chinese Academy of Sciences
2023
Nanjing Normal University
2012-2017
Shanghai Research Center for Wireless Communications
2014-2017
University of Wisconsin–Madison
2014-2016
Nanjing Institute of Geography and Limnology
2006
Social media datasets have been widely used in disaster assessment and management. When a occurs, many users post messages variety of formats, e.g., image text, on social platforms. Useful information could be mined from these multimodal data to enable situational awareness support decision making during disasters. However, the collected contain lot irrelevant misleading content that needs filtered out. Existing work has mostly unimodal methods classify messages. In other words, treated...
Enzyme immobilization in hierarchical macro–microporous metal–organic frameworks (MOFs) is greatly desirable but challenging due to the poor stability of MOFs a practical biocatalysis process. Herein, we prepared series single-crystalline ordered zeolitic imidazolate framework-8 (SOM-ZIF-8) with different macropore sizes 180, 270, and 360 nm for lipase investigated their performance biodiesel production Under an ethanol-assisted infiltration strategy, potential SOM-ZIF-8 enzyme was unleashed...
The taxonomy of galaxy morphology is critical in astrophysics as the morphological properties are powerful tracers evolution. With upcoming Large-scale Imaging Surveys, billions images challenge astronomers to accomplish classification task by applying traditional methods or human inspection. Consequently, machine learning, particular supervised deep has been widely employed classify morphologies recently due its exceptional automation, efficiency, and accuracy. However, learning requires...
In astronomy, it is important to categorize celestial bodies by classifying collected spectral data. The currently available methods present unsatisfactory classification accuracy and incur high computing costs. We propose a network based on residual attention convolutional (RAC-Net). this network, convolution operations can extract shallow deep features of data classify them without relying redshifts. mechanism augment the depth make training more efficient. allows focus specific bands...
With the development of urban science, researches on mining big data have attracted more and attention. One typical microcosm is taxi trajectory data. Predicting travel time between two specified points accurately great significance for applications, such as plan. However, current approach just uses limited modality or single model without considering their one-sidedness. This paper puts forward to one optimized method estimate time, which based ensemble with multi-modality data, namely...
Vitamin E is easily oxidized by light, air, oxidizing agents and heat, limiting its application in many ways. Compared to vitamin E, ester derivatives exhibit improved stability a stronger antioxidant capacity, even gain new biological functions. In recent years, enzymatic synthesis of has received increasing attention due environmental friendliness, high catalytic efficiency, inherent selectivity. This paper reviews the related progress lipase-mediated preparation derivatives. The function...
We address the problem of data acquisition in large distributed wireless sensor networks (WSNs). propose a method for using hierarchical routing and compressive sensing WSNs. Only few samples are needed to recover original signal with high probability since sparse representation technology is exploited capture similarities differences signal. To collect effectively WSNs, framework use proposed, randomized rotation cluster-heads evenly distribute energy load among sensors network....
Abstract Astronomical outliers, such as unusual, rare or unknown types of astronomical objects phenomena, constantly lead to the discovery genuinely unforeseen knowledge in astronomy. More unpredictable outliers will be uncovered principle with increment coverage and quality upcoming survey data. However, it is a severe challenge mine unexpected targets from enormous data human inspection due significant workload. Supervised learning also unsuitable for this purpose because designing proper...
MapReduce is a programming model and an associated implementation for processing parallel data, which widely used in Cloud computing environments. However, the traditional system based on centralized master-slave structure. While, along with increase of number jobs submitted scale, master node will become bottleneck system. To improve this problem, we have proposed new named ChordMR, designed to use peer-to-peer Chord network manage churn failure decentralized way. More importantly, propose...
Abstract Geographic services based on GPS trajectory data, such as location prediction and recommender services, have received increasing attention because of their potential social commercial benefits. In this study, a Service Recommender Model (GSRM) is proposed, which loosely comprises three essential steps. Firstly, sequences are obtained through clustering operation locations. To improve efficiency, programming model with distributed algorithm employed to accelerate the clustering....
Urban big data include various types of datasets, such as air quality data, meteorological and weather forecast data. Air index is broadly used in many countries an indicator to measure the pollution status. This has a great impact on outdoor activities urban residents, long-distance cycling, running, jogging, walking. However, for routes planning activities, there still lack comprehensive consideration quality. In this paper, prediction model (namely airQP-DNN) its application are proposed...
In this paper, we study the challenge of image-to-video retrieval, which uses query image to search relevant frames from a large collection videos. A novel framework based on convolutional neural networks (CNNs) is proposed perform large-scale video retrieval with low storage cost and high efficiency. Our consists key-frame extraction algorithm feature aggregation strategy. Specifically, takes advantage clustering idea so that redundant information removed in data greatly reduced. The...
Abstract The exponential growth of astronomical datasets provides an unprecedented opportunity for humans to gain insight into the Universe. However, effectively analyzing this vast amount data poses a significant challenge. In response, astronomers are turning deep learning techniques, but these methods limited by their specific training sets, leading considerable duplicate workloads. To overcome issue, we built framework general analysis galaxy images based on large vision model (LVM) plus...
Underwater acoustic sensor networks (UASNs) have attracted much attention in both the academic and industrial fields. Routing design of a UASN is an important part research, yet routing algorithms with low energy consumption high data delivery ratios are not established completely. This paper proposed Cluster-Based Adaptive Algorithm (CBAR) to fulfill demands large-scale underwater networks. It optimized network architecture by introducing concept cluster cellular wireless communication....
Location-based social networks (LBSN) allow users to socialize with friends by sharing their daily life experiences online. In particular, a large amount of check-ins data generated LBSNs capture the visit locations and open new line research spatio-temporal big data, i.e., next point-of-interest (POI) recommendation. At present, while some advanced methods have been proposed for POI recommendation, existing work only leverages temporal information two consecutive LBSN check-ins....
Location-Based Social Network is a kind of online social network developed on the basis traditional network. Location cornerstone its functions and services. The large number user data that networks collected provides more reliable guarantee for exploring studying development human society. Behavior Pattern some inherent way can be abstracted generalized from actual behaviors. Mining behavior find activities in law provide theoretical many aspects, such as urban planning, commercial...
Wireless Sensor Networks (WSNs) are widely applied in many fields, which air quality monitoring is a very important application. In this paper we build an system using WSN, consists of sensor nodes, gateway mobile nodes and center. Furthermore, solar energy power supply module designed to increase the lifetime. Moreover, order ensure communication reliability position accuracy, Global System for Mobile communications (GSM) Position (GPS) integrated into system. addition, method improved by...
Accelerating energy consumption and increasing data traffic have become prominent in large-scale wireless sensor networks (WSNs). Compressive sensing (CS) can recover through the collection of a small number samples with efficiency. General CS theory has several limitations when applied to WSNs because high complexity its [Formula: see text]-based conventional convex optimization algorithm large storage space required by Gaussian random observation matrix. Thus, we propose novel solution...