Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks150 from public data
- Visual-Textual Joint Relevance Learning for Tag-Based Social Image Search465
An approach that simultaneously utilizes both visual and textual information to estimate the relevance of user tagged images is proposed, and the relevance estimation is determined with a hypergraph learning approach.
- 369
- 365
- N Doping to ZnO Nanorods for Photoelectrochemical Water Splitting under Visible Light: Engineered Impurity Distribution and Terraced Band Structure259
N gradient-doped ZnO NRAs can not only utilize the visible light irradiation but also efficiently drive photo-induced electron and hole transfer via the terraced band structure and provide novel insights into doped photoelectrode materials for solar water splitting.
- 241
- 233
- SMR: Medical Knowledge Graph Embedding for Safe Medicine Recommendation224
Safe Medicine Recommendation (SMR) is the first to learn embeddings of a patient-disease-medicine graph for medicine recommendation in the world and jointly embeds diseases, medicines, patients, and their corresponding relations into a shared lower dimensional space.
- Maximum Correntropy Unscented Kalman Filter for Spacecraft Relative State Estimation111
A new algorithm called maximum correntropy unscented Kalman filter (MCUKF) is proposed and applied to relative state estimation in space communication networks and can enhance the robustness of UKF against impulsive noises.
- 102
- A query expansion method for retrieving online BIM resources based on Industry Foundation Classes100
This work develops a prototype semantic search engine, named BIMSeek, for retrieving online BIM resources, and presents an automatic query expansion method, which outperforms traditional keyword-based methods and WordNet-based query expansion methods.
- Efficiently embedding dynamic knowledge graphs93
This paper proposes a new context-aware Dynamic Knowledge Graph Embedding method, DKGE, which supports the embedding learning in an online fashion and can also learn KG embedding from scratch.
- 85
- Safe Medicine Recommendation via Medical Knowledge Graph Embedding73
Safe Medicine Recommendation (SMR) is the first to learn embeddings of a patient-disease-medicine graph for medicine recommendation in the world and jointly embeds diseases, medicines, patients, and their corresponding relations into a shared lower dimensional space.
- 70
- A Survey of Techniques for Constructing Chinese Knowledge Graphs and Their Applications67
This paper aims to introduce the techniques of constructing Chinese knowledge graphs and their applications, as well as analyse the impact of knowledge graph on OBOR.
- CdS/CdSe Core–Shell Nanorod Arrays: Energy Level Alignment and Enhanced Photoelectrochemical Performance63
CdS/CdSe core-shell arrays gave a maximum photocurrent of 5.3 mA/cm(2), which was 4 and 11 times as large as bare CdS and CdSe, respectively.
- EgoTextVQA: Towards Egocentric Scene-Text Aware Video Question Answering61
This work introduces EgoTextVQA, a novel and rigorously constructed benchmark for egocentric QA assistance involving scene text, and comprehensively evaluates 10 prominent multimodal large language models.
- 58
- 57
- 55
- 51
- Towards Empty Answers in SPARQL: Approximating Querying with RDF Embedding48
An RDF graph embedding based framework to solve the SPARQL empty-answer problem in terms of a continuous vector space and can significantly improve the quality of approximate answers and speed up the generation of alternative queries.
- 44
- Fair Representation Learning for Recommendation: A Mutual Information Perspective43
This paper re-define recommendation fairness with a novel two-fold mutual information objective, and proposes novel mutual information (upper/lower) bounds with contrastive information estimation for model optimization.
- 42
- 41
- 37
- PDD Graph: Bridging Electronic Medical Records and Biomedical Knowledge Graphs via Entity Linking34
This paper proposes a novel framework to extract important medical entities from MIMIC-III (Medical Information Mart for Intensive Care III) and automatically link them with the existing biomedical knowledge graphs, including ICD-9 ontology and DrugBank.
- Leveraging Knowledge Graph Embeddings for Natural Language Question Answering31
A novel framework, called KemQA, is proposed, which stands on recent advances in relation phrase dictionaries and knowledge graph embedding techniques to address the mapping problem and construct graph-structured queries of natural language questions.
- Self-Supervised Dynamic Hypergraph Recommendation based on Hyper-Relational Knowledge Graph30
This framework establishes a cross-view hypergraph self-supervised learning mechanism for KG enhancement and model hyper-relational facts in KGs to capture interdependencies between entities under complete semantic conditions to alleviate the effects of over-smoothing and supervision signal sparsity.
- 28
- 27
- Time-Sensitve Retrieval-Augmented Generation for Question Answering25
This work proposes a novel approach that integrates supervised contrastive learning with tailored negative sample pairs for temporal constraints to train the retriever of an RAG system, along with query-side fine-tuning and routing techniques.
- Probing the Impacts of Visual Context in Multimodal Entity Alignment25
This work proposes to learn entity representations from graph structures and visual context, and combine feature similarities to find alignments at the output level, and explores a mechanism which utilizes classification techniques and entity types to remove potentially un-helpful images during alignment learning and inference.
- 25
- Answering why-not questions on SPARQL queries25
A novel explanation model is proposed, called answering why-not questions on SPARQL (ANNA) to answer why- not questions using a divide-and-conquer strategy and results show that the proposed model and algorithms can provide high-quality explanations in terms of both effectiveness and efficiency.
- Joint Edge-Model Sparse Learning is Provably Efficient for Graph Neural Networks24
This paper provides the first theoretical characterization of joint edge-model sparse learning from the perspective of sample complexity and convergence rate in achieving zero generalization error and proves analytically that both sampling important nodes and pruning neurons with the lowest-magnitude can reduce the sample complexityand improve convergence without compromising the test accuracy.
- 24
- AHNG: Representation learning on attributed heterogeneous network22
A unified embedding model which represents each node in an AHN with a G aussian distribution (AHNG), which fuses multi-type nodes/links and diverse attributes through a two-layer neural network and captures the uncertainty by embedding nodes as Gaussian distributions.
- 21
- Modulation of the MAPKs pathways affects Aβ-induced cognitive deficits in Alzheimer’s disease via activation of α7nAChR20
The results indicated that inhibition of p38 or JNK MAPKs could alleviate Aβ-induced spatial memory deficits through regulating activation of α7nAChR via recovering memory-related proteins.
- CdSe-sensitized branched CdS hierarchical nanostructures for efficient photoelectrochemical solar hydrogen generation20
The branched hierarchical nanostructure improved the optical absorption by increasing the optical path via additional light trapping, as well as increasing the contact area between the electrode and electrolyte for more reactive sites, contributing to the higher photoelectrochemical performance than that obtained for the rod-like nanostructures.
- 19
- Reliability Analysis and Design of MMC Based on Mission Profile for the Components Degradation17
A system level MMC reliability analysis and design method based on MMC mission profile and insulated-gate bipolar transistor (IGBT) lifetime degradation and multi-objective function optimization is presented.
- 17
- Hybrid long-range dependency-aware graph convolutional network for node classification15
This work proposes a novel method, called hybrid long-range dependency-aware graph convolutional network (HLDGCN), which employs a flexible node selection process that considers both topological and attribute features across diverse graph structures and incorporates both high- and low-similarity nodes, ensuring the preservation of diverse frequency information within the graph.
- Effect of thermal expansion additives on alleviating the ash deposition of high-sodium coal15
The results show that the molten deposit obtained by drop-tube furnace at 1373 K was transformed into weakly-condensed deposit and strongly-sticky deposit respectively when vermiculite and perlite were added separately.
- 15
- 14
- 14
- Triple Context-Based Knowledge Graph Embedding14
This paper defines a novel context-dependent knowledge graph representation model named triple-context-based knowledge embedding, which is based on the notion of triple context used for embedding entities and relations, and achieves significant improvements over the state-of-the-art methods.
- 13
- SOAP: Enhancing Spatio-Temporal Relation and Motion Information Capturing for Few-Shot Action Recognition12
A novel plug-and-play architecture for FSAR called Spatio-tempOral frAme tuPle enhancer (SOAP) is proposed and SOAP-Net achieves new state-of-the-art performance across well-known benchmarks such as SthSthV2, Kinetics, UCF101, and HMDB51.
- 12
- 12
- 11
- Research trend of microbiota-gut-brain axis in Alzheimer’s disease based on CiteSpace (2012–2021): A bibliometrics analysis of 608 articles11
The study revealed that the research and development of MGBA in AD rapidly progressed, but no breakthrough has been made in the past decade, and it still needs to be closely combined with multidisciplinary technology to grasp the frontier hotspots.
- 10
- 10
- 10
- 9
- Fast Human Action Recognition via Millimeter Wave Radar Point Cloud Sequences Learning9
This paper proposes a fast human action recognition framework based on 3D point cloud sequences generated by commercial 4D millimeter wave imaging radar systems that significantly enhances inference speed while retaining competitive inference accuracy.
- 9
- Newtonian-Type Adaptive Filtering Based on the Maximum Correntropy Criterion9
The experimental results show that, compared to the conventional MCC adaptive filter, the MCC-based Newtonian-type method converges faster and still maintains a good steady-state performance under impulsive noise.
- 9
- Fine-Grained Evaluation of Knowledge Graph Embedding Model in Knowledge Enhancement Downstream Tasks8
This paper conducts comprehensive experiments on different KG embedding models in KG based question answering, recommendation and natural language processing tasks, and tries to mine the characteristics of each KGE model in actual application scenarios, and provides guidance for the research of KGE models and knowledge-enhanced downstream tasks.
- 8
- 8
- Quadratic Sparse Gaussian Graphical Model Estimation Method for Massive Variables8
This work proposes a novel method, called Fast and Scalable Inverse Covariance Estimator by Thresholding (FST), which obtains a graph structure by applying a generalized threshold to the sample covariance matrix and solves multiple block-wise subproblems via element-wise thresholding.
- Building Chinese Legal Hybrid Knowledge Network8
This paper proposes a framework for constructing a legal hybrid knowledge network from Chinese encyclopedia and legal judgments, and builds a legal knowledge graph through Chinese legal judgments which captures the strict logical connections in the legal judgments.
- 7
- Fully-Convolutional Intensive Feature Flow Neural Network for Text Recognition7
The recognition results on large-scale Chinese string and MNIST datasets show that the IntensiveNet can deliver enhanced recog-nition results, compared with other related deep models.
- Structured query construction via knowledge graph embedding7
This paper proposes a novel framework that first encodes the underlying knowledge graph into a low-dimensional embedding space by leveraging generalized local knowledge graphs and uses the learned embedding representations of the knowledge graph to compute the query structure and assemble vertices/edges into the target query.
- 7
- 6
- 6
- Knowledge and data-driven prediction of organ failure in critical care patients6
The use of an improved TCN model and a medical knowledge graph led to substantial improvement in prediction accuracy, providing generalizability and an independent explanation for organ failure prediction in critical care patients and show the potential of incorporating prior domain knowledge into machine learning models to inform care and service planning.
- Efficient search over incomplete knowledge graphs in binarized embedding space6
A novel end- to-end learning-to-hash framework for encoding incomplete KGs and graph queries in a Hamming space and is compared to state-of-the-art methods commonly used in real-world KGs.
- Cosine-Based Embedding for Completing Schematic Knowledge6
A novel model named CosE is proposed, which outperforms state-of-the-art methods and successfully preserve the transitivity and symmetry of axioms in schematic knowledge.
- Improving Chinese Sentiment Analysis via Segmentation-Based Representation Using Parallel CNN6
A novel sentiment-analysis model, which is capable of sentiment analysis task in Chinese, and which concatenate the representation of each segment to obtain the final representation of the text which does not only contain the semantic and syntactic features but also retains the essential sequential information.
- Answering why-not questions on semantic multimedia queries6
A novel explanation model is proposed to generate logical explanations at the basic graph pattern level, the filter constraint level, or the multimedia function level, respectively, which helps users refine their initial queries.
- Quality Prediction of Newly Proposed Questions in CQA by Leveraging Weakly Supervised Learning6
This work considers that a good question is popular and answerable in CQA websites and trains a weakly supervised model based on Convolutional Neural Network to recognize good newly proposed questions.
- 5
- ANNA: Answering Why-Not Questions for SPARQL5
A novel explanation system, ANNA (Answering why-Not questioNs for spArql), is presented to explain why-not questions using a divide-and-conquer strategy and Experimental results on DBpedia prove that ANNA can generate high-quality explanations within a reasonable amount of time.
- A Hash Learning Framework for Search-Oriented Knowledge Graph Embedding4
This paper encodes knowledge graph into Hamming space and introduces a novel HAsh Learning Framework (HALF) for search-oriented knowledge graph embedding and gets a bonus of speed-up in the searching embeddings while the accuracy and scalability of the original model are basically maintained.
- 4
- 4
- 4
- 4
- 4
- Preparation and properties of MC-Cys-MEK Self -healing Hydrogel with improved mechanical Performence3
- 3
- 3
- Dispute Generation in Law Documents via Joint Context and Topic Attention3
This paper proposes a novel Seq2Seq model with two dispute detection modules, which captures relationships among the PA and the DA in two ways and shows that the context-level and the topic-level detection modules can improve the accuracy and coverage of generated disputes.
- 3
- 3
- A robust blind adaptive multiuser detection based on maximum correntropy criterion in satellite CDMA systems3
This paper improves the robustness of MUD methods under impulsive noise by bringing forward a new cost function based on the maximum correntropy criterion and deduces the expression of the blind adaptive detector.
- Adversarial Discriminative Denoising for Distant Supervision Relation Extraction3
An adversarial discriminative denoising framework is proposed, which provides an effective way of introducing human supervision and exploiting it along with the potentially useful information underlying the noisy data in a unified framework and employs a continuous approximation of sampling action to guarantee the holistic denOising framework to be differentiable.
- Faceted fusion of RDF data3
FF aggregates distributed RDF data on a specific topic according to different facets, based on two topological properties of a topic-specific RDF Graph (TRG), and outperforms three other approaches based on data distance, structure distance and structure/attribute respectively.
- WPIS: From In-the-Wild Web Images to Physics-Aware 3D Scene Graphs for Physical Reasoning2
This work introduces WPIS (Web- and Physics-Informed Scene-understanding), a pipeline that compiles Physics-aware 3D Scene Graphs from web imagery by fusing open-vocabulary instance/mask cues with relative geometry, augmenting nodes with real-valued liquid states and fine-grained hand–object interaction (HOI) subgraphs, and attaching concise natural-language functional relations.
- VQFT: A Visual Query Approach Based on Full-Text Search for Knowledge Graphs2
A Visual Query approach based on Full-Text search for knowledge graphs, called VQFT, which simplifies the process of querying knowledge graphs for users, and is more user-friendly and easier to learn than existing methods.
- 2
- Explaining similarity for SPARQL queries2
Four explainable models based on Linear Regression, Support Vectorregression, Ridge Regressive, and Random Forest Regression are implemented to provide quantitative weights to different dimensional SPARQL features, which helps users and machines understand the result of similarity measures in different query scenarios and can be used in many downstream tasks.
- Fine-Grained Evaluation of Knowledge Graph Embedding Models in Downstream Tasks2
Modifying embeddings by considering more complex KG structural information may not achieve improvements in practical applications, such as updating TransE to TransR, and modeling KG embeddins in non-euclidean space can effectively improve the performance of downstream tasks.
- Characterizing Robotic and Organic Query in SPARQL Search Sessions2
This paper defines and partition SPARQL queries into different sessions, design an algorithm to detect loop patterns, which is an important characteristic of robotic queries, in a given query session, and employs a pipeline method that leverages loop pattern features and query request frequency to distinguish the robotic and organic SParQL queries.
- Combining Knowledge Graph Embedding and Network Embedding for Detecting Similar Mobile Applications2
A novel framework by knowledge graph (KG) techniques and a hybrid embedding strategy to fill above gaps and preliminary results indicate the effectiveness of the method comparing to existing models in terms of reciprocal ranking and minimum ranking.
- 2
- Faceted Exploring for Domain Knowledge over Linked Open Data2
KFM is demonstrated, which can aggregate the distributed RDF data of a topic according to the facets of this topic and provides a new way for users to obtain and explore domain knowledge in the LOD cloud.
- 2
- Every Subtlety Counts: Fine-grained Person Independence Micro-Action Recognition via Distributionally Robust Optimization1
The Person Independence Universal Micro-action recognition Framework is proposed, which integrates Distributionally Robust Optimization principles to learn person-agnostic representations and outperforms existing methods in both accuracy and robustness, achieving stable generalization under fine-grained conditions.
- 1
- 1
- NoTeNet: Normalized Mutual Information-Driven Tuning-free Dynamic Dependence Network Inference Method for Multimodal Data1
NoTeNet provides a promising paradigm that can integrate two different data modalities to enhance prediction accuracy and demonstrates superior prediction accuracy and efficiency without the need for hyperparameter tuning, making it potential for a wide range of web data applications.
- Factor Model-Based Large Covariance Estimation from Streaming Data Using a Knowledge-Based Sketch Matrix1
This work proposes a novel method - KEEF: Knowledge-based Time and Memory Efficient Covariance Estimator in Factor Model and its extended variation, which leverages historical data to train a knowledge-based sketch matrix, which is used to accelerate the factor analysis of streaming data and directly estimates the covariance matrix from the sketched data.
- 1
- ST-ECP: A Novel Spatial-Temporal Framework for Energy Consumption Prediction of Vehicle Trajectory1
A Spatial-Temporal Framework for Energy Consumption Prediction of Vehicle Trajectories (ST-ECP) is introduced, which constructs a heterogeneous interaction graph that captures the complex relationships between vehicles, environments, and drivers, effectively characterizing the dynamic attributes of trajectories across various conditions.
- 1
- 1
- 1
- Research on Hotspots of Educational Application of Natural Language Processing Based on LDA Topic Model1
This study takes the papers from the workshop “Innovative Use of NLP for Building Educational Applications” as data and uses LDA topic model for semantic analysis to identify the research topics automatically and analyzes the topic intensity, topic novelty and topic evolution.
- 1
- Improving Core Path Reasoning for the Weakly Supervised Knowledge Base Question Answering1
A CPR model is introduced that aligns questions and paths in a step-by-step reasoning manner from explicit text semantic matching and implicit knowledge bases structure matching and a two-stage learning strategy to alleviate the spurious path problem efficiently is proposed.
- Link Trustworthiness Evaluation over Multiple Heterogeneous Information Networks1
This work proposes a novel multi-HIN link trustworthiness evaluation model that leverages information across multiple related HINs to accomplish linkTrustworthiness evaluation tasks inherently and efficiently.
- 1
- 1
- Machine learning-driven optimization of specific, compact, and efficient base editors via single-round diversification–
A single round of diversification of newly identified TadA orthologs by DNA shuffling created a small set of novel and specific cytosine and adenosine deaminases that were markedly distinct in sequence from published base editor deaminases.
- –
- Identification of Influential Node Group in Attributed Graph through Explaining Graph Neural Network–
This work introduces Global Graph UNderstanding (GGUN), a perturbation-based framework leveraging the explanatory power of Graph Neural Networks that fills the gap between Deep Neural Network gradient-based feature importance analysis and discrete structure in the graph, which is formulated as a combinatorial optimization problem.
- STCKGE:Continual Knowledge Graph Embedding Based on Spatial Transformation–
A novel CKGE framework SoTCKGE grounded in Spatial Offset Transformation is proposed, which enhances the model’s ability to represent complex relational structures but also allows for the embedding update of both new and old knowledge through simple spatial offset transformations, without the need for continuous learning methods.
- –
- –
- AVB-aware fault-tolerance routing of time-triggered traffic in time-sensitive networks–
An AVB-aware fault-tolerant routing (AAFTR) is proposed that flexibly combines temporal and spatial redundancy mechanisms based on the redundancy needs of TT traffic, while also taking AVB traffic into account and improves AVB scheduling performance.
- A cell-interacting and multi-correcting method for automatic circulating tumor cells detection–
A new cell-interacting and multi-correcting detector called CMD is developed, and H&E-stained slide images are applied to detect CTCs automatically for the first time, and the superiority of the method over state-of-the-art cell detection methods is confirmed.
- –
- –
- –
- Graph Inference via the Energy-efficient Dynamic Precision Matrix Estimation with One-bit Data–
This work proposes a novel estimator to estimate the covariance matrix from one-bit data, and then utilize the covariances matrices to capture the dynamic structure of the graph structure, and theoretically demonstrates the effectiveness of the estimators.
- –
- –
- –
- –
- –
- –
- A META-ANALYSIS OF SERIOUS MOBILE GAMES TO SUPPORT MUSEUM’S INFORMAL EDUCATIONAL–
5 main categories of IoT technologies used for serious mobile games in museums are presented and how each type of well-known types of games implements their functions and features through IoT is illustrated.
- –
- –
- Web Mapping - Ramsey County Construction Tool–
Presentation completed by students enrolled in GIS 5574: Web GIS and Services, taught by Justin Hansen in Spring 2019.
- –
- DFTL: a Description Language for Transformation of Data Formats–
Experimental results show that it is possible to transform data with various formats to the data in XML by DFTL,thus a unified XML-view becomes true for the accession of data with heterogeneous formats.
- Generic and Indexed Programming (Project Paper)–
The EPSRC-funded Generic and Indexed Programming project will explore the interaction between datatype-generic programming (DGP) — programs parametrized by the shape of their data — and indexed programming (IP) — lightweight dependentlytyped programming, with programs indexed by type-level representations of properties.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.