A5026995541- Data Quality and Management
- Web Data Mining and Analysis
- Topic Modeling
- Advanced Graph Neural Networks
- Advanced Database Systems and Queries
- Data Management and Algorithms
- Caching and Content Delivery
- Complex Network Analysis Techniques
- Semantic Web and Ontologies
- Recommender Systems and Techniques
- Privacy-Preserving Technologies in Data
- Service-Oriented Architecture and Web Services
- Data Mining Algorithms and Applications
- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- Peer-to-Peer Network Technologies
- Graph Theory and Algorithms
- Network Security and Intrusion Detection
- Text and Document Classification Technologies
- Blockchain Technology Applications and Security
- Web visibility and informetrics
- Cryptography and Data Security
- Image Retrieval and Classification Techniques
- Spam and Phishing Detection
- Advanced Computational Techniques and Applications
Northeastern University
2016-2025
Nankai University
2024
Civil Aviation Flight University of China
2024
Universidad del Noreste
2023
Tianjin University
2016-2018
Tianjin University of Technology
2018
Beijing University of Posts and Telecommunications
2015-2016
Southwest Jiaotong University
2012
China Railway Group (China)
2009
Northeastern University
2008
Pt-Decorated highly porous flower-like Ni particles with nanopores and well-dispersed small Pt grains on petals show high activity for ammonia electro-oxidation.
As one of the advanced cobalt-based materials, cobalt sulfides with novel architecture have attracted huge interest due to low cost, easy availability, and promising bifunctional activity for both oxygen evolution reaction (OER) reduction (ORR), which is essential next-generation energy storage devices. Herein, we demonstrated a facile clean electrochemical technique directly synthesize CoS nanosheets high purity onto surface carbon cloth, quick thermal treatment was performed further...
Data imbalance is a common phenomenon in machine learning. In the imbalanced data classification, minority samples are far less than majority samples, which makes it difficult for to be effectively learned by classifiers. A synthetic oversampling technique (SMOTE) improves sensitivity of classifiers synthesizing without repetition. However, process new SMOTE algorithm may lead problems such as "noisy samples" and "boundary samples." Based on above description, we propose based Gaussian...
The primary challenge of developing clean energy conversion/storage systems is to exploit an efficient bifunctional electrocatalyst both for oxygen reduction reaction (ORR) and evolution (OER) with low cost good durability. Here, we synthesized chlorine-doped Co(OH)2 in situ grown on carbon cloth (Cl-doped Co(OH)2) as integrated electrode by a facial electrodeposition method. anodic potential was then applied the Cl-doped alkaline solution remove chlorine atoms (electro-oxidation...
Column semantic-type detection is a crucial task for data integration and schema matching, particularly when dealing with large volumes of unlabeled tabular data. Existing methods often rely on supervised learning models, which require extensive labeled In this paper, we propose SNMatch, an unsupervised approach based Siamese network detecting column semantic types without training The novelty SNMatch lies in its ability to generate the embeddings columns by considering both format features...
We study the problem of self-supervised and interpretable data cleaning, which automatically extracts repair rules from dirty data. In this paper, we propose a novel framework, namely Garf, based on sequence generative adversarial networks (SeqGAN). One key information Garf tries to capture is (for example, if city "Dothan", then county should be "Houston"). employs SeqGAN consisting generator G discriminator D that trains learn dependency relationships ( e.g. , given value "Dothan" as...
Collaborative Filtering has achieved great success in capturing users' preferences over items. However, existing techniques only consider limited collaborative signals, leading to unsatisfactory results when the user-item interactions are sparse. In this paper, we propose a Cross-grained Neural model (CNCF), which enables recommendation more accurate and explainable. Specifically, first construct four kinds of interaction graphs both fine-grained signals coarse-grained can better compensate...
Hadoop is widely deployed distributed computing framework and makes creating applications much easier. However, unlike text data, there no existing video r/w interface for Hadoop, many analytic implemented in C/C++ are not compatible with framework. In this paper, we propose an open source processing HVPI to extend support applications. It provides easy-to-use developers quickly build large-scale based on native help users easily port written into platform. We also present two typical use...
Finding a team that is both competent in performing the task and compatible working together has been extensively studied. However, most methods for formation tend to rely on set of skills only. In order solve this problem, we present an efficient method based Constrained Pattern Graph (called CPG). Unlike traditional methods, our takes into account structure constraints communication members, which can better meet requirements users. First, CPG preprocessing proposed normalize represent it...
MapReduce has been proven to be a highly desirable platform for scalable parallel data analysis. The task scheduling in is very crucial the job execution and marked impact on system performance. To best of our knowledge, previous algorithms rarely consider job-intensive environments are not able provide high throughput. Hence this paper proposes novel technique improve Firstly, by making an in-depth analysis environments, we sum up 4 major factors which affect Secondly, based factors,...
Nowadays, people usually participate in multiple social networks simultaneously, e.g., Facebook and Twitter. Formally, the correspondences of accounts that belong to same user are defined as anchor links, aligned by links can be denoted networks. In this paper, we study problem link prediction (ALP) across a pair based on network structure. First, three similarity metrics (CPS, CCS, CPS+) proposed. Different from previous works, focus theoretical guarantees our metrics. We prove...