AnnexML
Author:
Open Access
- 4 August 2017
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 455-464
Abstract
Extreme multi-label classification methods have been widely used in Web-scale classification tasks such as Web page tagging and product recommendation. In this paper, we present a novel graph embedding method called \"AnnexML\". At the training step, AnnexML constructs a k-nearest neighbor graph of label vectors and attempts to reproduce the graph structure in the embedding space. The prediction is efficiently performed by using an approximate nearest neighbor search method that efficiently explores the learned k-nearest neighbor graph in the embedding space. We conducted evaluations on several large-scale real-world data sets and compared our method with recent state-of-the-art methods. Experimental results show that our AnnexML can significantly improve prediction accuracy, especially on data sets that have larger a label space. In addition, AnnexML improves the trade-off between prediction time and accuracy. At the same level of accuracy, the prediction time of AnnexML was up to 58 times faster than that of SLEEC, which is a state-of-the-art embedding-based method.Keywords
This publication has 16 references indexed in Scilit:
- Erratum to: Pruned Bi-directed K-nearest Neighbor Graph for Proximity Search2016
- Extreme Multi-label Loss Functions for Recommendation, Tagging, Ranking & Other Missing Label Applications2016
- Robust Extreme Multi-label Learning2016
- Goal-Directed Inductive Matrix Completion2016
- Distributed Balanced Partitioning via Linear Embedding2016
- On Approximately Searching for Similar Word Embeddings2016
- Fast Label Embeddings via Randomized Linear Algebra2015
- FastXML2014
- Learning deep structured semantic models for web search using clickthrough data2013
- Maximum inner-product search using cone trees2012