Identifying Text Polarity Using Random Walks
Hassan, Ahmed and Radev, Dragomir R.

Article Structure

Abstract

Automatically identifying the polarity of words is a very important task in Natural Language Processing.

Introduction

Identifying emotions and attitudes from unstructured text is a very important task in Natural Language Processing.

Related Work

Hatzivassiloglou and McKeown (1997) proposed a method for identifying word polarity of adjectives.

Word Polarity

We use a Markov random walk model to identify polarity of words.

Experiments

We performed experiments on the General Inquirer lexicon (Stone et al., 1966).

Conclusions

Predicting the semantic orientation of words is a very interesting task in Natural Language Processing and it has a wide variety of applications.

Topics

WordNet

Appears in 19 sentences as: WordNet (19)
In Identifying Text Polarity Using Random Walks
  1. (2004) construct a network based on WordNet synonyms and then use the shortest paths between any given word and the words ’good’ and ’bad’ to determine word polarity.
    Page 2, “Related Work”
  2. ’ good’ and ’bad’ themselves are closely related in WordNet with a 5-long sequence “good, sound, heavy, big, bad”.
    Page 2, “Related Work”
  3. Hu and Liu (2004) use WordNet synonyms and antonyms to predict the polarity of words.
    Page 2, “Related Work”
  4. For any word, whose polarity is unknown, they search WordNet and a list of seed labeled words to predict its polarity.
    Page 2, “Related Work”
  5. WordNet is used to expand these lists.
    Page 3, “Related Work”
  6. A similar method is presented in (Andreevskaia and Bergler, 2006) where WordNet synonyms, antonyms, and glosses are used to iteratively expand a list of seeds.
    Page 3, “Related Work”
  7. One such important source is WordNet (Miller, 1995).
    Page 3, “Word Polarity”
  8. WordNet is a large lexical database of English.
    Page 3, “Word Polarity”
  9. The simplest approach is to connect words that occur in the same WordNet synset.
    Page 3, “Word Polarity”
  10. We can collect all words in WordNet , and add links between any two words that occurr in the same synset.
    Page 3, “Word Polarity”
  11. The resulting graph is a graph G(W, E) where W is a set of word / part-of-speech pairs for all the words in WordNet .
    Page 3, “Word Polarity”

See all papers in Proc. ACL 2010 that mention WordNet.

See all papers in Proc. ACL that mention WordNet.

Back to top.

semi-supervised

Appears in 9 sentences as: semi-supervised (9)
In Identifying Text Polarity Using Random Walks
  1. The method could be used both in a semi-supervised setting where a training set of labeled words is used, and in an unsupervised setting where a handful of seeds is used to define the two polarity classes.
    Page 1, “Abstract”
  2. It outperforms the state of the art methods in the semi-supervised setting.
    Page 1, “Abstract”
  3. Previous work on identifying the semantic orientation of words has addressed the problem as both a semi-supervised (Takamura et al., 2005) and an unsupervised (Turney and Littman, 2003) learning problem.
    Page 1, “Introduction”
  4. In the semi-supervised setting, a training set of labeled words
    Page 1, “Introduction”
  5. The proposed method could be used both in a semi-supervised and in an unsupervised setting.
    Page 2, “Introduction”
  6. Empirical experiments on a labeled set of words show that the proposed method outperforms the state of the art methods in the semi-supervised setting.
    Page 2, “Introduction”
  7. This view is closely related to the partially labeled classification with random walks approach in (Szummer and J aakkola, 2002) and the semi-supervised learning using harmonic functions approach in (Zhu et al., 2003).
    Page 4, “Word Polarity”
  8. This method could be used in a semi-supervised setting where a set of labeled words are used and the system learns from these labeled nodes and from other unlabeled nodes.
    Page 5, “Experiments”
  9. The proposed method can be used in a semi-supervised setting where a training set of labeled words is used, and in an unsupervised setting where only a handful of seeds is used to define the two polarity classes.
    Page 8, “Conclusions”

See all papers in Proc. ACL 2010 that mention semi-supervised.

See all papers in Proc. ACL that mention semi-supervised.

Back to top.

co-occurrence

Appears in 8 sentences as: co-occurrence (8)
In Identifying Text Polarity Using Random Walks
  1. To get co-occurrence statistics, they submit several queries to a search engine.
    Page 2, “Related Work”
  2. They construct a network of words using gloss definitions, thesaurus, and co-occurrence statistics.
    Page 2, “Related Work”
  3. Another source of links between words is co-occurrence statistics from corpus.
    Page 4, “Word Polarity”
  4. We study the effect of using co-occurrence statistics to connect words later at the end of our experiments.
    Page 4, “Word Polarity”
  5. The spin model approach uses word glosses, WordNet synonym, hypernym, and antonym relations, in addition to co-occurrence statistics extracted from corpus.
    Page 5, “Experiments”
  6. Adding co-occurrence statistics slightly improved performance, while using glosses did not help at all.
    Page 5, “Experiments”
  7. No glosses or co-occurrence statistics are used.
    Page 5, “Experiments”
  8. A possible solution to this might be identifying those words and adding more links to them from glosses of co-occurrence statistics in corpus.
    Page 7, “Experiments”

See all papers in Proc. ACL 2010 that mention co-occurrence.

See all papers in Proc. ACL that mention co-occurrence.

Back to top.

cross validation

Appears in 7 sentences as: cross validation (7)
In Identifying Text Polarity Using Random Walks
  1. We used 10-fold cross validation for all tests.
    Page 5, “Experiments”
  2. Table 2 compares the performance using 10-fold cross validation .
    Page 5, “Experiments”
  3. Table 2: Accuracy for SO-PMI with different dataset sizes, the spin model, and the random walks model for 10-fold cross validation and 14 seeds.
    Page 6, “Experiments”
  4. We perform 10-fold cross validation using the General Inquirer lexicon.
    Page 6, “Experiments”
  5. Figure 4 shows the accuracy for 10-fold cross validation and for using only 14 seeds at different thresholds.
    Page 7, “Experiments”
  6. that the top 60% words are classified with an accuracy greater than 99% for 10-fold cross validation and 92% with 14 seed words.
    Page 7, “Experiments”
  7. We also looked at the classification accuracy for different parts of speech in Figure 5. we notice that, in the case of 10-fold cross validation , the performance is consistent across parts of speech.
    Page 7, “Experiments”

See all papers in Proc. ACL 2010 that mention cross validation.

See all papers in Proc. ACL that mention cross validation.

Back to top.

hypernyms

Appears in 5 sentences as: hypernym (2) hypernyms (3)
In Identifying Text Polarity Using Random Walks
  1. For example, we can use other WordNet relations: hypernyms , similar to,...etc.
    Page 4, “Word Polarity”
  2. The spin model approach uses word glosses, WordNet synonym, hypernym , and antonym relations, in addition to co-occurrence statistics extracted from corpus.
    Page 5, “Experiments”
  3. The proposed method achieves better performance by only using WordNet synonym, hypernym and similar to relations.
    Page 5, “Experiments”
  4. We build a network using only WordNet synonyms and hypernyms .
    Page 6, “Experiments”
  5. We use a network built from WordNet synonyms and hypernyms only.
    Page 6, “Experiments”

See all papers in Proc. ACL 2010 that mention hypernyms.

See all papers in Proc. ACL that mention hypernyms.

Back to top.

synset

Appears in 4 sentences as: synset (2) Synsets (1) synsets (1)
In Identifying Text Polarity Using Random Walks
  1. Nouns, verbs, adjectives and adverbs are grouped into sets of cognitive synonyms ( synsets ), each expressing a distinct concept (Miller, 1995).
    Page 3, “Word Polarity”
  2. Synsets are inter-linked by means of conceptual-semantic and lexical relations.
    Page 3, “Word Polarity”
  3. The simplest approach is to connect words that occur in the same WordNet synset .
    Page 3, “Word Polarity”
  4. We can collect all words in WordNet, and add links between any two words that occurr in the same synset .
    Page 3, “Word Polarity”

See all papers in Proc. ACL 2010 that mention synset.

See all papers in Proc. ACL that mention synset.

Back to top.

Natural Language

Appears in 3 sentences as: Natural Language (3)
In Identifying Text Polarity Using Random Walks
  1. Automatically identifying the polarity of words is a very important task in Natural Language Processing.
    Page 1, “Abstract”
  2. Identifying emotions and attitudes from unstructured text is a very important task in Natural Language Processing.
    Page 1, “Introduction”
  3. Predicting the semantic orientation of words is a very interesting task in Natural Language Processing and it has a wide variety of applications.
    Page 8, “Conclusions”

See all papers in Proc. ACL 2010 that mention Natural Language.

See all papers in Proc. ACL that mention Natural Language.

Back to top.

semantically related

Appears in 3 sentences as: semantically related (3)
In Identifying Text Polarity Using Random Walks
  1. For example, the synonyms of any word are semantically related to it.
    Page 3, “Word Polarity”
  2. The intuition behind that connecting semantically related words is that those words tend to have similar polarity.
    Page 3, “Word Polarity”
  3. We construct a network where two nodes are linked if they are semantically related .
    Page 3, “Word Polarity”

See all papers in Proc. ACL 2010 that mention semantically related.

See all papers in Proc. ACL that mention semantically related.

Back to top.

PDF available from www.aclweb.org.
Send us an email