• Publications
  • Influence
SIDE: Representation Learning in Signed Directed Networks
TLDR
This paper proposes SIDE, a general network embedding method that represents both sign and direction of edges in the embedding space and carefully formulates and optimizes likelihood over both direct and indirect signed connections.
BEAR: Block Elimination Approach for Random Walk with Restart on Large Graphs
TLDR
BEAR is proposed, a fast, scalable, and accurate method for computing RWR on large graphs that significantly outperforms other state-of-the-art methods in terms of preprocessing and query speed, space efficiency, and accuracy.
MapReduce Triangle Enumeration With Guarantees
TLDR
This work is the first to give guarantees on the maximum load of each reducer for an arbitrary input graph, and is competitive with existing methods improving the performance by a factor up to 2X, and can significantly increase the size of datasets that can be processed.
PTE: Enumerating Trillion Triangles On Distributed Systems
TLDR
Experimental results show that PTE provides up to 47 times faster performance than recent distributed algorithms on real world graphs, and succeeds in enumerating more than 3 trillion triangles on the ClueWeb12 graph, which any previous triangle computation algorithm fail to process.
Fully Scalable Methods for Distributed Tensor Factorization
TLDR
This paper proposes two distributed tensor factorization methods, CDTF and SALS, which are scalable with all aspects of data and show a trade-off between convergence speed and memory requirements.
Personalized Ranking in Signed Networks Using Signed Random Walk with Restart
TLDR
Signed Random Walk with Restart is proposed, a novel model for personalized ranking in signed networks that provides proper rankings reflecting signed edges based on the signed surfer and achieves the best accuracy for sign prediction and predicts trolls 4× more accurately than other ranking models.
Scalable Tucker Factorization for Sparse Tensors - Algorithms and Discoveries
TLDR
P-Tucker is proposed, a scalable Tucker factorization method for sparse tensors that successfully discover hidden concepts and relations in a large-scale real-world tensor, while existing methods cannot reveal latent features due to their limited scalability or low accuracy.
Fast and Scalable Distributed Boolean Tensor Factorization
TLDR
DBTF is proposed, a distributed algorithm for Boolean tensor factorization running on the Spark framework that decomposes up to 163–323 larger tensors than existing methods in 68–382 less time, and exhibits near-linear scalability in terms of tensor dimensionality, density, rank, and machines.
BePI: Fast and Memory-Efficient Method for Billion-Scale Random Walk with Restart
TLDR
BePI is proposed, a fast, memory-efficient, and scalable method for computing RWR on billion-scale graphs that exploits the best properties from both preprocessing methods and iterative methods.
Knowledge Extraction with No Observable Data
TLDR
KegNet (Knowledge Extraction with Generative Networks), a novel approach to extract the knowledge of a trained deep neural network and to generate artificial data points that replace the missing training data in knowledge distillation is proposed.
...
...