Bu kaydın yasal hükümlere uygun olmadığını düşünüyorsanız lütfen sayfa sonundaki Hata Bildir bağlantısını takip ederek bildirimde bulununuz. Kayıtlar ilgili üniversite yöneticileri tarafından eklenmektedir. Nadiren de olsa kayıtlarla ilgili hatalar oluşabilmektedir. MİTOS internet üzerindeki herhangi bir ödev sitesi değildir!

SENTIMENT ANALYSIS FOR TURKISH SOCIAL MEDIA: A CASE STUDY ON TWITTER

Oluşturulma Tarihi: 09-11-2016

Niteleme Bilgileri

Tür: Tez

Alt Tür: Yüksek Lisans Tezi

Yayınlanma Durumu: Yayınlanmamış

Dosya Biçimi: PDF

Dil: İngilizce

Konu(lar): TEKNOLOJİ, Teknik bilgiler iletişimi,

Yazar(lar): Yurtalan, Gökhan (Yazar),

Emeği Geçen(ler): Koyuncu, Murat (Danışman),

Anahtar Kelimeler

Sentiment Analysis, Text Classification, Machine Learning, DataMining, Turkish Text.


Özet

Sentiment Analysis tries to resolve the sense which a writer or a speaker wants to giveto the people as positive, negative or neutral. The attempts to resolve emotions occur afterautomatically classifying the written text of the author or speaker. By means of using internetand other social media actively, people easily imply their attitude and emotion towards anitem, a person, a political party, a country or a brand name. In this way, writers, artists, brandowners, political party directors have an opportunity to know what is going around aboutthemselves or their brands easily. Such a platform that has highly active usage and hugeamount of data daily makes nearly impossible to process by hand so emotional classificationbecomes important.In recent years, there have been many successful studies for the English language. Inthese studies, there were many words and word groups which set emotion polarities that stemfrom the English grammar structure. Correspondingly, there are datasets which are used totest performance of those studies. But, our study notes that studies done for Turkish havelower performance compared to the studies for English. There are various reasons for thissuch as the translation of datasets from English to Turkish and ignoring special grammarstructures in Turkish. In this thesis, by using newly constructed Turkish emotional polaritywords, we try to develop a best fitting methodology for Turkish grammar with the help of aTurkish linguist. We collect instant data by Twitter API and perform the analysis with thisnew methodology.


İçindekiler



Açıklamalar



Haklar



Notlar



Kaynakça


Atıf Yapanlar

Gözat Sayfasına Dön

 

Sosyal Medya ve Araçlar

İstatistikler

  • Kayıt
    • Bu ay: 3
    • Toplam: 2417
  • Online
    • Ziyaretçi: 19
    • Üye: 0
    • Toplam: 19

Detaylı İstatistikler