Learning Representation From Concurrence-Words Graph For Aspect Sentiment Classification

Article Geochemistry & Geophysics

Registration-Is-Evaluation: Robust Point Set Matching With Multigranular Prior Assessment

Yucheng Shu et al.

Summary: The article introduces a new framework for point set registration, which dynamically evaluates the matching status by utilizing a probabilistic model to improve the accuracy and robustness of registration.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Unsupervised Spectral Feature Selection With Dynamic Hyper-Graph Learning

Xiaofeng Zhu et al.

Summary: The study proposes a novel feature selection method that dynamically constructs a hyper-graph based Laplacian matrix in the framework of sparse feature selection, resulting in interpretable and discriminative results. Compared to previous methods, it outperforms in clustering and segmentation tasks.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Efficient Utilization of Missing Data in Cost-Sensitive Learning

Xiaofeng Zhu et al.

Summary: The study introduces a novel Data-driven Incremental Imputation Model (DIM) to effectively and economically impute missing values using all available information in the dataset. By considering both economical criteria and effective imputation information, DIM outperforms comparison methods in terms of prediction accuracy and classification accuracy on UCI datasets.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Global-Feature Encoding U-Net (GEU-Net) for Multi-Focus Image Fusion

Bin Xiao et al.

Summary: A global-feature encoding U-Net (GEU-Net) is proposed for multi-focus image fusion, which treats the generation of focus map as a global two-class segmentation task using the U-Net network. The GEU-Net introduces global feature pyramid extraction module (GFPE) and global attention connection upsample module (GACU) to effectively utilize global semantic and edge information, enhancing performance significantly.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Pyramidal Multiple Instance Detection Network With Mask Guided Self-Correction for Weakly Supervised Object Detection

Yunqiu Xu et al.

Summary: The proposed pyramidal multiple instance detection network (P-MIDN) addresses the issue of local optima often encountered in multiple instance detection networks. By incorporating multiple MIDNs in a sequence and using proposal removal during training to reduce exposure to local discriminative regions, the P-MIDN enables better coverage of target objects. The combination of P-MIDN with an online instance classifier refinement framework and a mask guided self-correction method results in state-of-the-art performance on various benchmark datasets.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2021)

添加到收藏夹

Article Computer Science, Information Systems