FENS Seminars – Dr. Rahim Dehkharghani

21 October 2022
KHAS Cibali Campus Fener Hall

Dr. Rahim Dehkharghani from Iu015fu0131k University will be the guest of KHAS Faculty of Engineering and Natural Sciencesu00a0seminar series with his speech u201cSentiment Analysis on Turkish: Challenges, Resources and Techniquesu201d on Friday, October 21 at 2 pm.

The event will take place inu00a0Fener Hallu00a0and is open to anyone interested.

Abstract: Sentiment analysis which attempts to extract the sentiment polarity from data, has attracted growing research interest in recent years, especially in the context of social media. While most of this research has focused on English, there is ample data and research work for resource-lean languages such as Turkish. In this presentation, a comprehensive approach to sentiment analysis on Turkish will be explained. Also we will cover different granularity levels of sentiment analysis such as aspect, sentence, and document levels as well as some linguistic issues such as conjunction, negation and intensification. Furthermore, the largest polarity lexicon in Turkish, as one of the outputs of my PhD thesis, called SentiTurknet, will be introduced.

About the Speaker: Rahim Dehkharghani is an assistant professor at Isik University, Istanbul since September 2021. He was also a faculty member at (national) University of Bonab in Iran from 2016 to 2021. He has been graduated with a PhD degree in artificial intelligence from Sabanci University and with a MSc degree in software engineering from (national) Shahid Beheshti University of Tehran, Iran. His outstanding achievements during his educational life includes winning the Sumo prize 2007 for building the best domain ontology for SUMO, and also ranking 24 among all Computer engineering students in the national computer engineering Olympiad in Iran at 2006. His research areas include but are not limited to Natural language processing, data mining, and machine and deep learning.

FENS Seminars – Dr. Zahra Barzegar

Leave A Reply

Your email address will not be published.