Graph Regularized Transductive Classification on Heterogeneous Information Networks

By Ming Ji, Yizhou Sun, Marina Danilevsky, Jiawei Han, Jing Gao. 2010

261 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics, Graph Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Philip S. Yu
2. Mohit Bansal
3. Jiawei Han
4. Jianfeng Gao
5. Percy Liang
6. Iryna Gurevych
7. Quoc V. Le
8. Luke S. Zettlemoyer

In Statistical and Nonlinear Physics

1. Chuan Shi
2. Yizhou Sun
3. Jiawei Han
4. Philip S. Yu
5. Irwin King
6. M. Murty
7. Sharad Nandanwar
8. Zhilin Yang

In Graph Learning

1. Chenyi Zhuang
2. Qiang Ma
3. Zhilin Yang
4. Jiawei Han
5. Yizhou Sun
6. Ruslan Salakhutdinov
7. Chuan Shi
8. Philip S. Yu

Citations in the corpus

• Revisiting Semi-Supervised Learning with Graph Embeddings (2016, background): “The graph A can either be derived from distances between instances (Zhu et al., 2003), or be explicitly derived from external data, such as a knowledge graph (Wijaya et al., 2013) or a citation network between documents (Ji et al., 2010).”
• Dual Graph Convolutional Networks for Graph-Based Semi-Supervised Classification (2018, background): “The similarity can be derived by calculating the distances among data points [33], or may be explicitly given by structured data, such as knowledge graphs [29], citation graphs [14], hyperlinks between documents [20], and so on.”
• Heterogeneous Metric Learning with Joint Graph Regularization for Cross-Media Retrieval (2013, background): “2011), heterogeneous multi task learning(Zhang and Yeung 2011), and transductive classification on heterogeneous information networks(Ji et al. 2010).”
• HeteSim: A General Framework for Relevance Measure in Heterogeneous Networks (2013, methodology): “DBLP dataset [27]: The DBLP dataset is a sub-network collected from DBLP website4 involving major conferences in four research areas: database, data mining, information retrieval and artificial intelligence, which naturally form four classes.”
• Augmented SBERT: Data Augmentation Method for Improving Bi-Encoders for Pairwise Sentence Scoring Tasks (2020, background): “We remove all overlaps and ensure that a question in one split of the dataset does not appear in any other split to mitigate the transductive classification problem (Ji et al., 2010).”
• MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding (2020, methodology): “We adopt a subset of DBLP extracted by [12, 15], containing 4057 authors, 14328 papers, 7723 terms, and 20 publication venues after data preprocessing.”
• Mining heterogeneous information networks: a structural analysis approach (2013, methodology): “These ideas lead to effective algorithms, such as GNetMine [10] and RankClass [9].”
• Structural Neighborhood Based Classification of Nodes in a Network (2016, background): “Ming Ji et al. [7] proposed a framework for transductive learning in heterogeneous networks based on two assumptions: lo-cal consistency (class assignment of neighbors is similar) and ground truth (pre-assigned class labels are correct).”