Heriot watt

Eshrag Refaee

 

 

Links

Home

About me

Contact me

 

Resaerch Interests

Subjectivity and sentiment analysis of Arabic content of social media.

 

Resources I have developed

1)Refaee & Rieser (R&R): An Arabic Twitter Corpus annotated for subjectivity and sentiment analysis download via LREC repository

To download the corpus, look it up under "Shared LRs" tab by using the search text box (e.g. type Twitter or Arabic). Alternatively, you can just scroll down the LRs.

How to find the corpus

If use the corpus, please cite: Bib

2)A dialectal Arabic subjectivity lexicon - manually annotated

3)A translated version of the MPQA subjectivity lexicon - manually filtered download the lexica . To find out about original MPQA click here

If use the lexicon, please cite: Bib

4) A corpus of 1.2M Arabic tweets automatically annotated for sentiment using emoticons. download the corpus or download the corpus + 51 features (1.84Gbytes).

If use the corpus, please cite: Bib

5) A corpus of 130.2k Arabic tweets automatically annotated for sentiment using sentiment-bearing hashtags. download the corpus or download the corpus + 51 features (230Kbytes).

If use the corpus, please cite: Bib

6) A corpus of 487.5k Arabic tweets automatically annotated for sentiment using a lexicon-based method. download the corpus or download the corpus + features (0.5Gbytes).

If use the corpus, please cite: Bib

7) A corpus of 34k Arabic tweets automatically annotated for sentiment using ML models. download the corpus

If use the corpus, please cite: Bib

8) A list of social signals (Arabic): laughter, consents, dazzle, prayers, etc. download here .

If use the lexicon, please cite: Bib

9) A java-based system for retrieving tweets from the Twitter stream about a given query and automatically assigning them sentiment labels. download trained models and codes.

If use the system, please cite: Bib. To obtain the jar file of the system, please contact me.

 

Conferences, Workshops and Events

- The 7th Saudi Scientific Conference, 1-2 Feb 2014, Edinburgh, Scotland. link to my abstract

- The 9th edition of the Language Resources and Evaluation Conference, 26-31 May 2014, Reykjavik, Iceland

- The 5th International Workshop on EMOTION, SOCIAL SIGNALS, SENTIMENT & LINKED OPEN DATA, 26-27 May 2014, Reykjavik, Iceland

- The Workshop on Free/Open-Source Arabic Corpora and Corpora Processing Tools, 27 May 2014, Reykjavik, Iceland

- SemDial2014-DialWatt THE 18th WORKSHOP ON THE SEMANTICS AND PRAGMATICS OF DIALOGUE (as an organiser), 1-3 September 2014, Edinburgh, Scotland

- The Arabic Natural Language Processing Workshop co-located with EMNLP 2014, Doha, Qatar 25-29 October 2014.

- The 8th Saudi Scientific Conference, 31 Jan - 1 Feb 2015, London, UK. link to my abstract (*** best presentation award)

- The North American Chapter of the Association for Computational Linguistics (NAACL HLT 2015). Denver, Colorado, USA. 31 May - 5 June 2015.

- MACS PhD poster event. Heriot-Watt University, Edinburgh, 8th April 2016. (*** best poster award).

 

Publications

[1] Refaee, E. and Rieser, V. (2014). An Arabic Twitter Corpus for Subjectivity and Sentiment Analysis. The 9th edition of the Language Resources and Evaluation Conference LREC'2014. The European Language Resources Association. Reykjavik, Iceland 26-31 May 2014.

link to my [poster] [paper] Bib

[2] Refaee, E. and Rieser, V. (2014). Can we Read Emotions from a smiley face? Emoticon-based distant supervision for subjectivity and sentiment analysis of Arabic Twitter feeds. In the 5th International Workshop on Emotion, Social Signals, Sentiment and Linked Open Data. LREC'2014. Reykjavik, Iceland 26-31 May 2014.

link to my [presentation] [paper] Bib

[3] Refaee, E. and Rieser, V. (2014). Subjectivity and Sentiment Analysis of Arabic Twitter feeds with limited resources. In Workshop on Free/Open-Source Arabic Corpora and Corpora Processing Tools. LREC'2014. Reykjavik, Iceland 26-31 May 2014.

link to my [presentation] [paper] Bib

[4] Refaee, E. and Rieser, V. (2014). Evaluating Distant Supervision for Subjectivity and Sentiment Analysis on Arabic Twitter Feeds. In The Arabic Natural Language Processing Workshop co-located with EMNLP 2014, Doha, Qatar 25-29 October 2014.

link to my [presentation] [paper] Bib

[5] Refaee, E. and Rieser, V. (2015). Benchmarking Machine Translated Sentiment Analysis for Arabic Tweets. In the North American Chapter of the Association for Computational Linguistics - NAACL 2015 Student Research Workshop (SRW) (NAACL HLT 2015). Denver, Colorado, USA. 31 May - 5 June 2015.

link to my [poster] [paper] Bib

[6] Malmasi, S. Refaee, E. Dras, M (2015). Arabic Dialect Identification using a Parallel Multidialectal Corpus. In Proceedings of the 14th Conference of the Pacific Association for Computational Linguistics (PACLING 2015). 19-21 May 2015. Bali, Indonesia.

[7] Refaee, E. and Rieser, V. (2016). iLab-Edinburgh at SemEval-2016 Task 7: A Hybrid Approach for Determining Sentiment Intensity of Arabic Twitter Phrases. In Proceedings of the International Workshop on Semantic Evaluation, SemEval '16, co-located with NAACL'16, San Diego, California, June 2016. (*** Top system in the Arabic Twitter subtask )

link to my [paper] Bib

Non peer-reviewed publications

[1] Rob Pooley, Eshrag Refaee and Tessa Berg (2013). Ontologies and Rich Pictures - An unlikely collaboration. 22nd International Conference on Information Systems Development (ISD2013), Sevilla, Spain, September 2-4, 2013. Available online: http://iwt2.org/isd2013/program.php

[2] Refaee, E. and Rieser, V. (2015). No Bad Feelings: Distant Supervision Helps Subjectivity but not Sentiment Analysis of Arabic Twitter Feeds. In the 8th Saudi Scientific Conference, 31 Jan - 1 Feb 2015, London, UK.