Scaling up all pairs similarity search pdf

2015

Citations in the corpus

• Asymmetric LSH (ALSH) for Sublinear Time Maximum Inner Product Search (MIPS) (2014, background): “In a typical matrix factorization model, a user i is associated with a latent user characteristic vector u i , and similarly, an item j is associated with a latent item characteristic vector v j .”
• Matrix Profile I: All Pairs Similarity Joins for Time Series: A Unifying View That Includes Motifs, Discords and Shapelets (2016, methodology): “In the text domain the algorithm has applications in a host of problems, including community discovery, duplicate detection, collaborative filtering, clustering, and query refinement [1].”
• Improved Densification of One Permutation Hashing (2014, background): “…hashing is a popular indexing technique for a variety of large-scale data processing applications, which include duplicate detection [4, 13], all-pair similarity [3], fast linear learning [19], temporal correlation [10], 3-way similarity & retrieval [17, 23], graph algorithms [6, 11, 21], and more.”
• Asymmetric Minwise Hashing (2014, background): “…for creating hash buckets leading to efficient algorithms for numerous applications including spam detection [6], collaborative filtering [4], news personalization [15], compressing social networks [13], graph sampling [14], record linkage [24], duplicate detection [20], all pair similarity [5], etc.”
• Densifying One Permutation Hashing via Rotation for Fast Near Neighbor Search (2014, unknown): “, 2007; Henzinger, 2006), all-pair similarity (Bayardo et al., 2007), similarity join/record-linkage (Koudas & Srivastava, 2005), temporal correlation (Chien & Immorlica, 2005), etc.”
• Min-Max Kernels (2015, methodology): “arities, it is not the focus of our paper (and may hurt the repeatability of the experiments if we try to alter the data). (ii) Several datasets downloaded from the LIBSVM site were already scaled to [-1, 1]. To make use of these datasets, we simply transform them by (z + 1)/2, where z is the origin”
• Inner Product Similarity Search using Compositional Codes (2014, methodology): “The latent factor models widely used in recommendation systems [10] and document matching [11], such as matrix factorization [10], latent semantic index [12], and so on, also rely on inner product similarity search to find the best matches.”
• Cascade Ranking for Operational E-commerce Search (2017, background): “Somework of search efficiency have also been proposed [3, 12, 14, 17, 21] as the data size of real-world applications is getting bigger and bigger.”