A5102889910- Topic Modeling
- Advanced Graph Neural Networks
- Text and Document Classification Technologies
- Natural Language Processing Techniques
- Face and Expression Recognition
- Machine Learning and Data Classification
- Data Mining Algorithms and Applications
- Advanced Text Analysis Techniques
- Imbalanced Data Classification Techniques
- Sentiment Analysis and Opinion Mining
- Complex Network Analysis Techniques
- Face recognition and analysis
- Catalysts for Methane Reforming
- Anaerobic Digestion and Biogas Production
- Biofuel production and bioconversion
- Multimodal Machine Learning Applications
- Graph Theory and Algorithms
- Gene expression and cancer classification
- Advanced Clustering Algorithms Research
- Data Stream Mining Techniques
- Hybrid Renewable Energy Systems
- Spam and Phishing Detection
- Advanced Image Processing Techniques
- Video Surveillance and Tracking Methods
- Explainable Artificial Intelligence (XAI)
State Key Laboratory of Pollution Control and Resource Reuse
2025
Nanjing University
2025
Henan Agricultural University
2024
Shanghai Institute of Technology
2023
Nanjing Normal University
2023
JDSU (United States)
2019-2022
University of Michigan–Ann Arbor
2017-2021
Second Affiliated Hospital of Zhejiang University
2019
Xi'an Jiaotong University
2012-2017
Daqing City People's Hospital
2011
Multi-hop reading comprehension (RC) across documents poses new challenge over single-document RC because it requires reasoning multiple to reach the final answer. In this paper, we propose a model tackle multi-hop problem. We introduce heterogeneous graph with different types of nodes and edges, which is named as Heterogeneous Document-Entity (HDE) graph. The advantage HDE that contains granularity levels information including candidates, entities in specific document contexts. Our proposed...
Interpretable multi-hop reading comprehension (RC) over multiple documents is a challenging problem because it demands reasoning information sources and explaining the answer prediction by providing supporting evidences. In this paper, we propose an effective interpretable Select, Answer Explain (SAE) system to solve multi-document RC problem. Our first filters out answer-unrelated thus reduce amount of distraction information. This achieved document classifier trained with novel pairwise...
Self-attention mechanism in graph neural networks (GNNs) led to state-of-the-art performance on many representation learning tasks. Currently, at every layer, attention is computed between connected pairs of nodes and depends solely the two nodes. However, such does not account for that are directly but provide important network context. Here we propose Multi-hop Attention Graph Neural Network (MAGNA), a principled way incorporate multi-hop context information into layer computation. MAGNA...
Distance-based knowledge graph embeddings have shown substantial improvement on the link prediction task, from TransE to latest state-of-the-art RotatE. However, complex relations such as N-to-1, 1-to-N and N-to-N still remain challenging predict. In this work, we propose a novel distance-based approach for prediction. First, extend RotatE 2D domain high dimensional space with orthogonal transforms model relations. The transform embedding keeps capability modeling symmetric/anti-symmetric,...
Clustering is one of the research hotspots in field data mining and has extensive applications practice. Recently, Rodriguez Laio [1] published a clustering algorithm on Science that identifies centers an intuitive way clusters objects efficiently effectively. However, sensitive to preassigned parameter suffers from identification "ideal" number clusters. To overcome these shortages, this paper proposes new can detect automatically via statistical testing. Specifically, proposed first...
Long non-coding RNAs have recently become a key regulatory factor for cancers, whereas FER1L4, newly discovered long RNA, has been mostly studied in gastric carcinoma and colon cancer cases. The functions molecular mechanism of FER1L4 rarely reported glioma malignant phenotypes. In this study, it was found that the expression LncRNA is upregulated high-grade gliomas than low-grade cases high predicts poor prognosis gliomas. Meanwhile, vitro study suggests with SiRNA knockdown obviously...
Xiaochen Hou, Peng Qi, Guangtao Wang, Rex Ying, Jing Huang, Xiaodong He, Bowen Zhou. Proceedings of the 2021 Conference North American Chapter Association for Computational Linguistics: Human Language Technologies. 2021.
Abstract Unsupervised feature selection is an important problem, especially for high‐dimensional data. However, until now, it has been scarcely studied and the existing algorithms cannot provide satisfying performance. Thus, in this paper, we propose a new unsupervised algorithm using similarity‐based clustering, Feature Selection‐based Clustering (FSFC). FSFC removes redundant features according to results of clustering based on similarity. First, clusters their A proposed, which overcomes...
Question answering over temporal knowledge graphs (KGs) efficiently uses facts contained in a KG, which records entity relations and when they occur time, to answer natural language questions (e.g., “Who was the president of US before Obama?”). These often involve three time-related challenges that previous work fail adequately address: 1) do not specify exact timestamps interest “Obama” instead 2000); 2) subtle lexical differences time “before” vs “after”); 3) off-the-shelf KG embeddings...
Abstract This article proposes a novel approach to address the issues of low accuracy in fault diagnosis and difficulty installing sensors on rolling bearings mechanical electrical equipment systems. To accomplish bearings, network structure algorithm based convolutional neural (CNN) support vector machine (SVM) is presented, which incorporates electric motor current signal. Firstly, collected signal subjected wavelet filter with soft-hard threshold eliminate noise. Secondly, processed data...
<p>Anthropogenic climate and environmental changes increasingly threaten the sustainability of life on Earth, hindering sustainable development human societies. These detrimental ecological are driven by activities that have elevated atmospheric levels greenhouse gases toxic substances, increased inorganic organic pollutants in water bodies, led to accumulation solid waste soils. Over next two three decades, impacts change, pollution, soil contamination expected intensify, posing...
Abstract In response to the challenges posed by complex operating conditions of electromechanical systems, non-stationary noise interference, difficulties in sensor installation on bearings, and scarcity real fault sample data, this paper proposes an ad-vanced diagnosis method for rolling bearings. This employs enhanced stacking ensemble learning model, im-proved with wavelet soft hard threshold algorithms, utilizes both motor current vibration signals. Initially, prepro-cessing raw signals...
Graph Attention Networks (GATs) are the state-of-the-art neural architecture for representation learning with graphs. GATs learn attention functions that assign weights to nodes so different have influences in feature aggregation steps. In practice, however, induced prone over-fitting due increasing number of parameters and lack direct supervision on weights. also suffer from over-smoothing at decision boundary nodes. Here we propose a framework address their weaknesses via margin-based...
Document-level relation extraction is a challenging task, requiring reasoning over multiple sentences to predict set of relations in document. In this paper, we propose novel framework E2GRE (Entity and Evidence Guided Relation Extraction) that jointly extracts the underlying evidence by using large pretrained language model (LM) as input encoder. First, guide LM’s attention mechanism focus on relevant context probabilities additional features for prediction. Furthermore, instead feeding...
Commonsense knowledge graph (CKG) is a special type of (KG), where entities are composed free-form text. Existing CKG completion methods focus on transductive learning setting, all the present during training. Here, we propose first inductive setting for completion, unseen may appear at test time. We emphasize that crucial CKGs, because frequently introduced due to fact CKGs dynamic and highly sparse. InductivE as framework targeted task. ensures capability by directly computing entity...
Self-attention mechanism in graph neural networks (GNNs) led to state-of-the-art performance on many representation learning tasks. Currently, at every layer, attention is computed between connected pairs of nodes and depends solely the two nodes. However, such does not account for that are directly but provide important network context. Here we propose Multi-hop Attention Graph Neural Network (MAGNA), a principled way incorporate multi-hop context information into layer computation. MAGNA...