Transfer Learning Using Twitter Data for Improving Sentiment Classification of Turkish Political News

Date

2013-10-29

Author

Kaya, Mesut
Fidan, Guven
Toroslu, İsmail Hakkı

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

297
views

0
downloads

In this paper, we aim to determine the overall sentiment classification of Turkish political columns. That is, our goal is to determine whether the whole document has positive or negative opinion regardless of its subject. In order to enhance the performance of the classification, transfer learning is applied from unlabeled Twitter data to labeled political columns. A variation of self-taught learning has been proposed, and implemented for the classification. Different machine learning techniques, including support vector machine, maximum entropy classification, and Naive-Bayes has been used for the supervised learning phase. In our experiments we have obtained up to 26 % increase in the accuracy of the classification with the inclusion of the Twitter data into the sentiment classification of Turkish political columns using transfer learning.

Subject Keywords

Support vector machine, Sentiment analysis, Unlabeled data, Transfer learning, Feature ranking

URI

https://hdl.handle.net/11511/36586

DOI

https://doi.org/10.1007/978-3-319-01604-7_14

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Sentiment Analysis of Turkish Political News Kaya, Mesut; Fidan, Guven; Toroslu, İsmail Hakkı (2012-12-07) In this paper, sentiment classification techniques are incorporated into the domain of political news from columns in different Turkish news sites. We compared four supervised machine learning algorithms of Naive Bayes, Maximum Entropy, SVM and the character based N-Gram Language Model for sentiment classification of Turkish political columns. We also discussed in detail the problem of sentiment classification in the political news domain. We observe from empirical findings that the Maximum Entropy and N-Gr...
Clustering based personality prediction on turkish tweets Tutaysalgir, Esen; Karagöz, Pınar; Toroslu, İsmail Hakkı (2019-08-30) In this paper, we present a framework for predicting the personality traits by analyzing tweets written in Turkish. The prediction model is constructed with a clustering based approach. Since the model is based on linguistic features, it is language specific. The prediction model uses features applicable to Turkish language and related to writing style of Turkish Twitter users. Our approach uses anonymous BIGS questionnaire scores of volunteer participants as the ground truth in order to generate personalit...
Intellectual hegemony of Justice and Development Party: a Gramscian pespective Yaman, Murat; Saraçoğlu, Cenk; Political Science and International Relations (2012-9) This thesis attempts to examine the relationship between liberal intellectuals and the Justice and Development Party (Adalet ve Kalkınma Partisi, AKP) in Turkish politics. The Justice and Development Party was elected to power in the 2002 general elections, and has been continually in power since. The rise of the AKP was interpreted as a real historical break in Turkish politics since the foundation of the Republic of Turkey, and sparked the emergence of new public discussions and academic debates on...
Spatial analysis of contemporary Turkish elections: a comprehensive approach Özen, İlhan Can (2017-01-01) This study offers a comprehensive approach to spatial analysis of parliamentary elections in Turkey since 2002. Using advanced spatial models, we find that electoral competitiveness and concentration mostly stabilized in the Western subprovinces whereas they are still in flux in the Eastern and Southern regions. There is an increasing level of geographically dependent concentration and competitiveness in recent elections, particularly in the 2015 elections (June and November). Our analyses also show that wh...
Discourses of Kemalism and Islamism on the political dimension of EU - Turkey Relations Çilingir, Sevgi; Dağı, İhsan Duran; Department of International Relations (2005) This thesis aims at defining and analyzing the positions of Kemalism and Islamism about the relationship between the European Union and Turkey, with respect to identity and political reform. The study is conducted by the usage of academic literature and examples of the writings of intellectuals from both positions; in order to analyze their discourses on the issue.With respect to the political dimension of EU - Turkey relations, the problems and EU demands on democracy, human rights, minority rights - with ...

Citation Formats

M. Kaya, G. Fidan, and İ. H. Toroslu, “Transfer Learning Using Twitter Data for Improving Sentiment Classification of Turkish Political News,” 2013, vol. 264, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/36586.