Web-Scale Distributional Similarity and Entity Set Expansion

By Patrick Pantel, Eric Crestan, Arkady Borkovsky, Ana-Maria Popescu, Vishnu Vyas. Conference on Empirical Methods in Natural Language Processing, 2009

311 citations recorded by the source corpus.

Computing the pairwise semantic similarity between all words on the Web is a computationally challenging task. Parallelization and optimizations are necessary. We propose a highly scalable implementation based on distributional similarity, implemented in the MapReduce framework and deployed over a 200 billion word crawl of the Web. The pairwise similarity between 500 million terms is computed in 50 hours using 200 quad-core nodes. We apply the learned similarity matrix to the task of automatic set expansion and present a large empirical study to quantify the effect on expansion performance of corpus size, corpus quality, seed composition and seed size. We make public an experimental testbed for set expansion analysis that includes a large collection of diverse entity sets extracted from Wikipedia.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/00fce98c3fda59bcb84b6d0626fb3137d2fbb984

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. Lei Zhang
2. B. Liu
3. Hoifung Poon
4. Heiko Paulheim
5. Bing Liu
6. Lei Shu
7. Hu Xu
8. Arjun Mukherjee

Citations in the corpus

• Sentiment Analysis and Opinion Mining (2012, methodology): “The classic methods for solving this problem in NLP are based on distributional similarity (Lee, 1999; Pantel et al., 2009).”
• Knowledge graph refinement: A survey of approaches and evaluation methods (2016, methodology): “Here, entity set expansion methods, which have been deeply investigated in the NLP field [75,91,103], would be an interesting fit to further increase the coverage of knowledge graphs, especially for less well-known long tail entities.”
• A Survey of Opinion Mining and Sentiment Analysis (2012, methodology): “It was shown that learning from positive and unlabeled examples provides a more effective method than the traditional distributional similarity methods [53, 79] and the machine learning technique Bayesian Sets [25] which was designed specifically for set expansion.”
• Grounded Unsupervised Semantic Parsing (2013, methodology): “Since case information is important for parsers and taggers, we first truecased the sentences us-ing DASH (Pantel et al., 2009), which stores the case for each phrase in Wikipedia.”