Mining arabic Jordanian dialect sentiment

العناوين الأخرى

تنقيب الآّراء باللهجة العربية الأردنية

مقدم أطروحة جامعية

Numan, Mays Imad

مشرف أطروحة جامعية

Atum, Jalal Yusuf

الجامعة

جامعة الأميرة سمية للتكنولوجيا

الكلية

كلية الملك الحسين لعلوم الحوسبة

القسم الأكاديمي

قسم علم الحاسوب

دولة الجامعة

الأردن

الدرجة العلمية

ماجستير

الملخص الإنجليزي

Social media has played a tremendous role in the research areas, and become a very popular communication tool, where people from different places communicate and connect through.

Sentiment Analysis (SA) also has the researches interest, it become one of the growing areas of research related to data mining and natural language processing.

This research thesis focuses on the opinion mining of Arabic Jordanian Dialect.

We use Twitter as the source of collecting our data, we annotated these data on three different classes; positive, negative, and neutral.

We used these data to classify the sentiment analysis of the Arabic Jordanian Dialect.

Six phases were taken for the preprocessing task; cleaning noisy tweets stage, normalization, tokenization, Named Entity Recognition, removing stop words, and stemming.

The experimental results show encouraging outcomes when Arabic light stemmer/segment was applied on the tweets. However, applying the same mechanism on rooted tweets and tweets without stemming, the results showed the importance of applying light stemmer on the Arabic Jordanian Dialect tweets.

We used both machine learning classification tools Support Vector Machines (SVM) and Naïve Bayes (NB) on supervised learning data.

The results presented that SVM Tweet Classification is playing an efficient role on dialects.

Additionally, experimentation outcomes reveal new guidelines and insights that could be addressed in future work.

التخصصات الرئيسية

تكنولوجيا المعلومات وعلم الحاسوب
اللغة العربية وآدابها

الموضوعات

• الأردن
• معالجة البيانات
• اللهجات
• اللغة العربية

عدد الصفحات

75

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-413726