Polarity Inducing Latent Semantic Analysis

By Wen-tau Yih, Geoffry Zweig, John C. Platt. 2012

82 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Multi-Relational Latent Semantic Analysis (2013, methodology): “Similar to LSA or PILSA when applied to lexical semantics, each word is still mapped to a vector in the latent space.”
Deep Learning for Answer Sentence Selection (2014, background): “They combine a group of word matching features with semantic features obtained from a wide range of linguistic resources including WordNet, polarity-inducinglatent semantic analysis (PILSA) model [24] and different vector space models.”
Improving Lexical Semantics for Sentential Semantics: Modeling Selectional Preference and Similar Words in a Latent Variable Model (2013, background): “There are latent variable models designed for lexical semantics, such as word senses (Boyd-Graber et al., 2007; Guo and Diab, 2011), function words (Grifﬁths et al., 2005), selectional preference (Ritter et al., 2010), synonyms and antonyms (Yih et al., 2012), etc.”
Distributional Models of Word Meaning (2018, methodology): “2014b) to using hybrid models in which DSMs are enriched with information extracted from lexical resources (Yih et al. 2012).”
Revisiting Word Embedding for Contrasting Meaning (2015, methodology): “For comparison, we list the results reported in (Yih et al., 2012) and (Zhang et al.”
Combining Heterogeneous Models for Measuring Relational Similarity (2013, methodology): “Instead, we leverage the recently proposed polarity-inducing latent semantic analysis (PILSA) model (Yih et al., 2012), which speciﬁcally estimates the degree of synonymy and antonymy.”
Semantic Specialisation of Distributional Word Vector Spaces using Monolingual and Cross-Lingual Constraints (2017, background): “Other ones modify the skip-gram (Mikolov et al., 2013b) objective function by introducing semantic constraints (Yih et al., 2012; Liu et al., 2015) to train word vectors which emphasise word similarity over relatedness.”
Word Semantic Representations using Bayesian Probabilistic Tensor Factorization (2014, methodology): “Mohammad et al. (2008) reproted a 23% F-score look-up performance of WordNet which support this claim as well.”
Question Answering Using Enhanced Lexical Semantic Models (2013, methodology): “Following the setting described in (Yih et al., 2012), we construct a PILSA model based on the Encarta thesaurus and enhance it with a discriminative projection matrix training method.”