The week's pick
Reseach Article
An Algorithm for the Reconstruction of Enthymemes for Effective Machine Translation
| International Journal of Applied Information Systems |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 12 - Number 31 |
| Year of Publication: 2020 |
| Authors: Enikuomehin Oluwatoyin, Odunowo Adebisi T., Mustapha Oluwatoyin S. |
10.5120/ijais2020451867
|
Enikuomehin Oluwatoyin, Odunowo Adebisi T., Mustapha Oluwatoyin S. . An Algorithm for the Reconstruction of Enthymemes for Effective Machine Translation. International Journal of Applied Information Systems. 12, 31 ( July 2020), 1-7. DOI=10.5120/ijais2020451867
Abstract
Enthymeme, which is arguments with missing premises or conclusions, is common in natural language text. Enthymeme reconstruction, the art of reformulating arguments with missing propositions, has not been effective in argument classification and consequently, rhetorical Algorithms have yielded poor result. They cannot discover features, text orientation, intent and sentiments in enthymematic arguments. This has led to poor performance of enthymematic Natural language toolkits. Hence, generating new context of enthymematic data reconstruction will provide better and useable insight. The aim of this research is to build a manual annotation framework for enthymemes to enable appropriate tagging and effective classification in argumentation. Manual Annotation technique is used in this experiment to manually separate statements that contain an aspect (enthymemes) from ArguAna corpus of hotel reviews from TripAdvisor.com to know the opinion from the statements. A total of 1201 reviews gave 5575 opinions which were then annotated with defined conclusions. The linear Support Vector Machine (SVM) and fastText classifier were used to train and test data while Valence Aware Dictionary for sEntiment Reasoning (VADER) was used to assign scores for each word based on sentiments. MATLAB and Python programming language were used for model implementation. The supervised learning approach showed the best performance results on the test set with a macro averaged F1-scores of 0.72 and 0.94 for explicit and implicit stances respectively. The identified implicit stances are explicit premises of either complete arguments or enthymemes. (If they are premises of complete arguments, there are other, additional premises.) The identified explicit stances can represent common knowledge information for the implicit premises, thus becoming explicit premises to fill in the gap present in the respective enthymemes. The experimental framework shows that manual annotation of enthymeme data can provide better and useable insight in machine based annotation.
References
- Lippi, M., and Torroni, P.: ‘Context-independent claim detection for argument mining’, Twenty-Fourth International Joint Conference on Artificial Intelligence, 2015
- Sergeant, A.: ‘Automatic argumentation extraction’, in Editor (Ed.)^(Eds.): ‘Book Automatic argumentation extraction’ (Springer, 2013, edn.), pp. 656-660
- Mebane, W.: ‘Detection of claims and supporting evidence in wikipedia articles on controversial topics’, 2017
- Mailly, J.-G.: ‘Using enthymemes to fill the gap between logical argumentation and revision of abstract argumentation frameworks’, arXiv preprint arXiv:1603.08789, 2016
- Walton, D.: ‘The three bases for the enthymeme: A dialogical theory’, Journal of Applied Logic, 2007, 6, (2008), pp. 361-379
- Brock, K.: ‘Enthymeme as Rhetorical Algorithm’, Present Tense Journal, 2014, 4, (1), pp. 2 - 8
- Pedersen, T., and Dyrkolbotn, S.K.: ‘The legally mandated approximate language about AI’, Open Journal Systems, 2018
- da Rocha, G.F.: ‘ArgMine: Argumentation Mining from Text’, 2016
- Hu, M., and Liu, B.: ‘Mining opinion features in customer reviews’, in Editor (Ed.)^(Eds.): ‘Book Mining opinion features in customer reviews’ (2004, edn.), pp. 755-760
- Wachsmuth, H., Trenkmann, M., Stein, B., and Engels, G.: ‘Modeling review argumentation for robust sentiment analysis’, in Editor (Ed.)^(Eds.): ‘Book Modeling review argumentation for robust sentiment analysis’ (2014, edn.), pp. 553-564
- Hutto, C.J., and Gilbert, E.: ‘Vader: A parsimonious rule-based model for sentiment analysis of social media text’, in Editor (Ed.)^(Eds.): ‘Book Vader: A parsimonious rule-based model for sentiment analysis of social media text’ (2014, edn.), pp.
- Amgoud, L., and Besnard, P.: ‘A formal analysis of logic-based argumentation systems’, in Editor (Ed.)^(Eds.): ‘Book A formal analysis of logic-based argumentation systems’ (Springer, 2010, edn.), pp. 42-55
- Feng, V.W., and Hirst, G.: ‘Classifying arguments by scheme’, in Editor (Ed.)^(Eds.): ‘Book Classifying arguments by scheme’ (Association for Computational Linguistics, 2011, edn.), pp. 987-996
- Levy, R., Bilu, Y., Hershcovich, D., Aharoni, E., and Slonim, N.: ‘Context dependent claim detection’, in Editor (Ed.)^(Eds.): ‘Book Context dependent claim detection’ (2014, edn.), pp. 1489-1500
- Boltužic, F., and Šnajder, J.: ‘Back up your stance: Recognizing arguments in online discussions’, in Editor (Ed.)^(Eds.): ‘Book Back up your stance: Recognizing arguments in online discussions’ (2014, edn.), pp. 49-58
- Ghosh, A., Li, G., Veale, T., Rosso, P., Shutova, E., Barnden, J., and Reyes, A.: ‘Semeval-2015 task 11: Sentiment analysis of figurative language in twitter’, in Editor (Ed.)^(Eds.): ‘Book Semeval-2015 task 11: Sentiment analysis of figurative language in twitter’ (2015, edn.), pp. 470-478
- Park, J., and Cardie, C.: ‘Identifying appropriate support for propositions in online user comments’, in Editor (Ed.)^(Eds.): ‘Book Identifying appropriate support for propositions in online user comments’ (2014, edn.), pp. 29-38
- Habernal, I., Eckle-Kohler, J., and Gurevych, I.: ‘Argumentation Mining on the Web from Information Seeking Perspective’, in Editor (Ed.)^(Eds.): ‘Book Argumentation Mining on the Web from Information Seeking Perspective’ (2014, edn.), pp.
- Rajendran, P., Bollegala, D., and Parsons, S.: ‘Contextual stance classification of opinions: A step towards enthymeme reconstruction in online reviews’, in Editor (Ed.)^(Eds.): ‘Book Contextual stance classification of opinions: A step towards enthymeme reconstruction in online reviews’ (2016, edn.), pp. 31-39
- Wachsmuth, H., Trenkmann, M., Stein, B., Engels, G., and Palakarska, T.: ‘A review corpus for argumentation analysis’, in Editor (Ed.)^(Eds.): ‘Book A review corpus for argumentation analysis’ (Springer, 2014, edn.), pp. 115-127
- Fort, K., Nazarenko, A., and Rosset, S.: ‘Modeling the complexity of manual annotation tasks: a grid of analysis’, in Editor (Ed.)^(Eds.): ‘Book Modeling the complexity of manual annotation tasks: a grid of analysis’ (2012, edn.), pp.
- Petrillo, M., and Baycroft, J.: ‘Introduction to manual annotation’, Fairview Research, 2010.
Index Terms
Keywords