Abstractive Thai Opinion Summarization

Abstract:

With the advancement in the Internet technology, customers can easily share opinions on services and products in forms of reviews. There can be large amounts of reviews for popular products. Manually summarizing those reviews for important issues is a daunting task. Automatic opinion summarization is a solution to the problem. The task is more complicated for reviews written in Thai language. Thai words are written continuously without space, there is no symbol to signify the end of a sentence, and many reviews are written informally, thus accurate word identification and linguistic annotation cannot be relied upon. This research proposes a novel technique to generate abstractive summaries of customer reviews written in Thai language. The proposed technique, which consists of the local and the global models, is evaluated using actual reviews of fifty randomly selected products from a popular cosmetic website. The results show that the local model outperforms the other model and the two baseline methods both quantitatively and qualitatively.

• Periodical: Advanced Materials Research (Volumes 971-973)
• Pages: 2273-2280
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.971-973.2273
• Online since: June 2014
• Authors: Orawan Chaowalit*, Ohm Sornil
• Keywords: Abstractive Opinion Summarization, Hopfield Network Algorithm, Thai Language

References

[1] O. Sornil and K. Gree-ut: An Automatic Text Summarization Approach using Content-Based and Graph-Based Characteristics in 2006 IEEE Conference on Cybernetics and Intelligent Systems (2006), p.1–6.

DOI: 10.1109/iccis.2006.252361

[2] K. Ganesan, C. Zhai, and J. Han: Opinogsis A Graph-based Approach to Abstractive Summarization of Highly Redundant Opinions, in Proceedings of the 23rd International Conference on Computational Linguistics, Stroudsburg, PA, USA (2010), p.340–348.

[3] K. Filippova: Multi-sentence Compression Finding Shortest Paths in Word Graphs, in Proceedings of the 23rd International Conference on Computational Linguistics, Stroudsburg, PA, USA (2010), p.322–330.

[4] K. Ganesan, C. Zhai, and E. Viegas: Micropinion Generation An Unsupervised Approach to Generating Ultra-concise Summaries of Opinions, in Proceedings of the 21st International Conference on World Wide Web, New York, NY, USA (2012), p.869–878.

DOI: 10.1145/2187836.2187954

[5] P. Bheganan, R. Nayak, and Y. Xu: Thai Word Segmentation with Hidden Markov Model and Decision Tree, in Advances in Knowledge Discovery and Data Mining, Springer Berlin Heidelberg (2009), p.74–85.

DOI: 10.1007/978-3-642-01307-2_10

[6] Information on http: /links. nectec. or. th/download. php.

[7] Kirk Baker: Singular Value Decomposition, in Singular Value Decomposition Tutorial, 2nd ed., (2013), p.21–22.

[8] H. Chen and T. Ng: An algorithmic approach to concept exploration in a large knowledge network (automatic thesaurus consultation): symbolic branch-and-bound search vs. connectionist Hopfield net activation, J. Am. Soc. Inf. Sci., vol. 46 (1995).

DOI: 10.1002/(sici)1097-4571(199506)46:5<348::aid-asi6>3.0.co;2-1

[9] O. P. Damani: Improving Pointwise Mutual Information (PMI) by Incorporating Significant Co-occurrence, Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, Seattle, Washington, USA(2013), pp.163-169.

[10] Michael T. Goodrich, Roberto Tamassia: Algorithm Design Foundations, Analysis, and Internet Examples. (2001).

[11] Anna Huang: Similarity Measures for Text Document Clustering, in Proceedings of the Sixth New Zealand Computer Science Research Student Conference (NZCSRSC2008), vol. May (2008), p.49–56.

[12] C. Lin: Rouge: a package for automatic evaluation of summaries, in In Proceedings of the Workshop on Text Summarization Branches Out (WAS 2004), (2004), p.25–26.