Adjectival Phrases as the Sentiment Carriers in the Urdu Text

In this paper we present a comprehensive overview of the structures of the adjectival phrases in the Urdu language with respect to the task of sentiment analysis. Urdu is a widely spoken but one of the least explored languages by the computational linguistics community. After a detailed analysis of adjectival phrases in Urdu text we conclude that this language is orthographically, morphologically and grammatically different from other well established languages, like English and hence, it requires updated or different approaches and algorithms for the task of sentiment analysis. We present our approach in which the adjectival phrases are combined with polarity shifters, and conjunctions to make sentiment expressions in the opinionated sentences. We label these sentiment expressions as the SentiUnits. We apply shallow parsing based chunking to extract the SentiUnits. The overall polarity of a sentence in a given review can be determined by computing the polarity of these expressions. Adjectives are the head words, which appear with modifiers and postpositions. The experimentation based evaluation of the model with a sentiment-annotated lexicon of Urdu words and two corpuses of reviews as test-beds, shows encouraging achievement in terms of sentimental analysis and accuracy. [Afraz Z. Syed, Aslam Muhammad, Martinez-Enriquez A. M. Adjectival Phrases as the Sentiment Carriers in the Urdu Text. Journal of American Science 2011;7(3):644-652]. (ISSN: 1545-1003). http://www.americanscience.org.

[1]  Vasileios Hatzivassiloglou,et al.  Towards the Automatic Identification of Adjectival Scales: Clustering Adjectives According to Meaning , 1993, ACL.

[2]  Marilyn A. Walker,et al.  Learning to Generate Naturalistic Utterances Using Reviews in Spoken Dialogue Systems , 2006, ACL.

[3]  Sabine Bergler,et al.  Semantic Tag Extraction from WordNet Glosses , 2006, LREC.

[4]  Peter D. Turney.  Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews , 2002, ACL.

[5]  Grzegorz Kondrak,et al.  A Comparison of Sentiment Analysis Techniques: Polarizing Movie Blogs , 2008, Canadian Conference on AI.

[6]  Vasileios Hatzivassiloglou,et al.  Predicting the Semantic Orientation of Adjectives , 1997, ACL.

[7]  Shlomo Argamon,et al.  Unsupervised Extraction of Appraisal Expressions , 2010, Canadian Conference on AI.

[8]  Janyce Wiebe,et al.  Recognizing subjectivity: a case study in manual tagging , 1999, Natural Language Engineering.

[9]  Smruthi Mukund,et al.  Using Cross-Lingual Projections to Generate Semantic Role Labeled Corpus for Urdu-A Resource Poor Language , 2010 .

[10]  Nadir Durrani,et al.  Urdu Word Segmentation , 2010, NAACL.

[11]  Ellen Riloff,et al.  Learning Extraction Patterns for Subjective Expressions , 2003, EMNLP.

[12]  Ana María Martínez Enríquez,et al.  Lexicon Based Sentiment Analysis of Urdu Text Using SentiUnits , 2010, MICAI.

[13]  Kashif Riaz,et al.  Challenges in Urdu stemming: a progress report , 2007 .

[14]  Ellen Riloff,et al.  Learning subjective nouns using extraction pattern bootstrapping , 2003, CoNLL.

[15]  Bing Liu,et al.  Mining and summarizing customer reviews , 2004, KDD.

[16]  Janyce Wiebe,et al.  Effects of Adjective Orientation and Gradability on Sentence Subjectivity , 2000, COLING.

[17]  Casey Whitelaw.  Using Appraisal Taxonomies for Sentiment Analysis , 2005 .