Contextual phrase-level polarity analysis using lexical affect scoring and syntactic N-grams

Abstract

We present a classifier to predict contextual polarity of subjective phrases in a sentence. Our approach features lexical scoring derived from the Dictionary of Affect in Language (DAL) and extended through WordNet, allowing us to automatically score the vast majority of words in our input avoiding the need for manual labeling. We augment lexical scoring with n-gram analysis to capture the effect of context. We combine DAL scores with syntactic constituents and then extract ngrams of constituents from all sentences. We also use the polarity of all syntactic constituents within the sentence as features. Our results show significant improvement over a majority class baseline as well as a more difficult baseline consisting of lexical n-grams.

Columbia NLP: Sentiment Detection of Sentences and Subjective Phrases in Social Media

“…We include POS tags and the top n-gram features as described in prior work (Agarwal et al, 2009;Rosenthal and McKeown, 2013). The DAL and other dictionaries are used along with a negation state machine (Agarwal et al, 2009) to determine the polarity for each word in the sentence.…”

“…The DAL and other dictionaries are used along with a negation state machine (Agarwal et al, 2009) to determine the polarity for each word in the sentence. We include all the features described in the original system (Agarwal et al, 2009) DAL scores, polar chunk n-grams, and count of syntactic chunks with their prior polarity based on the chunks position. Finally, we include several lexical-stylistic features that can occur in all datasets.…”

“…As in the phrase-based system we too build off of prior work (Agarwal et al, 2009) by using the DAL and augmenting it with Wordnet. However, we do not follow the earlier method but use it as motivation.…”

“…We developed a phrase based sentiment detection system geared towards Social Media by augmenting the state of the art system developed by Agarwal et al (2009) to include additional dictionaries such as Wiktionary and new features such as word lengthening (e.g. helllllloooo) and emoticons (e.g.…”

Iterative Constrained Clustering for Subjectivity Word Sense Disambiguation

“…Subjectivity lexicons (e.g., (Turney, 2002;Whitelaw et al, 2005;Riloff and Wiebe, 2003;Yu and Hatzivassiloglou, 2003;Kim and Hovy, 2004;Bloom et al, 2007;Andreevskaia and Bergler, 2008;Agarwal et al, 2009)) play an important role in opinion, sentiment, and subjectivity analysis. These systems typically look for the presence of clues in text.…”