Dr Ash Rahimi

Lecturer in Computer Science

School of Information Technology and Electrical Engineering
Faculty of Engineering, Architecture and Information Technology

Overview

Dr. Ash (Afshin) Rahimi is a lectuer at the School of Information Technology and Electrical Engineering, University of Queensland. He received his B.S. degree in computer science and M.S. degree in computational linguistics both from Sharif University of technology, Tehran, Iran, in 2006 and 2013, respectively. He achieved his PhD from The University of Melbourne in 2018 supervised by Timothy Baldwin and Trevor Cohn, focusing on social media user location inference from text and graph data. As a research fellow in 2019, he worked on NLP for low-resource languages, and also NLP in health domain. He regularly publishes his research results in top NLP conferences such as ACL, EMNLP, and NAACL (gScholar).

Research Interests

  • NLP, Social Media, Health
    My research interests fall within the fields of Natural Language Processing, Social Network Analysis and Machine Learning. I am specifically interested in exploiting both structured and unstructured data to help machines understand conversational language in Emergency Situations and Health Informatics.
  • Social Media Analysis and User Privacy
    Example project: Protecting Private Information in Public Domain: Defending Against Attribute Inference Helping social media users keep their attributes such as location private by building explainable attribute inference models. The models will be able to explain to the user why they're able to identify their private attributes e.g. by a post on Twitter or their follower list.
  • NLP for Low-resource Languages
    Building NLP models for low-resource languages where we don't have enough annotated training data through transfer learning techniques. Example project: In few years, most devices will be voice-operated, but support will be missing for low-resource languages such as indigenous languages of Australia. It is vital to create minimal support for these languages to have an inclusive society. One example of such methods is to build spoken language identification methods that work with limited annotated speech examples so that a device can at least have a backup strategy for speakers of these languages e.g. reply back in the identified language "Unfortunately, we don't support your language".
  • NLP for Health Domain
    Example project: Mapping public knowledge-bases such as WikiData to specialised knowledge-bases such as UMLS (a medical knowledge-base) to help patients understand medical terms. Assume a patient getting a discharge summary full of complex medical terms and difficult to understand through Google search. It'd be great if these terms could be linked to a publicly accessible knowledge source such as Wikipedia.

Qualifications

  • Docotor of Philosophy, The University of Melbourne

Publications

  • Rahimi, Afshin, Li, Yuan and Cohn, Trevor (2019). Massively multilingual transfer for NER. 57th Annual Meeting of the Association for Computational Linguistics (ACL), Florence, Italy, Jul 28-Aug 02, 2019. Stroudsburg, United States :Association for Computational Linguistics -ACL. doi: 10.18653/v1/p19-1015

  • Hovy, Dirk, Rahimi, Afshin, Baldwin, Timothy and Brooke, Julian (2019). Visualizing regional language variation across Europe on Twitter. Handbook of the Changing World Language Map. edited by Stanley D Brunn and Roland Kehrein.Cham, Switzerland: Springer International Publishing. doi:10.1007/978-3-319-73400-2_175-1

  • Darwish, Kareem, Magdy, Walid, Rahimi, Afshin, Baldwin, Timothy and Abokhodair, Norah (2018). Predicting online islamophobic behavior after #ParisAttacks. Journal of Web Science, 4 (2) 34-52. doi:10.1561/106.00000013

View all Publications

Supervision

  • Doctor Philosophy

View all Supervision

Available Projects

  • Project Description

    People don’t live in a flat society, they live in overlapping and interconnected communities sharing attributes such as geography, language, identity, and political opinion. Online communities mirror offline ones and are highly influential in determining how information is propagated and how opinion is shaped. The goal of this project is to utilise both linguistic clues and online social interactions to identify the social and linguistic structure of communities, their interactions and how these communities influence and shape opinions. Hidden factors that determine why users connect to each other on social media will be inferred and used to detect and measure communities, improving on traditional graph-based community detection methods.

    The communication within and between these communities will be studied to understand the spread of cultural and linguistic innovation and how information and political ideas travel along these networks. For example, we are interested in studying the relationship between the core and the periphery, like London and the North of England or Sydney and Western Australia, answering questions like: How do the concerns of people in the periphery reach decision and policy makers in the core? How does culture like news, memes and hashtags spread around these networks? What biases do people in the core have about the periphery and vice-versa?

    This will advance our knowledge on how information spreads and provide tools to magnify marginalised voices. The ideal student for this project will have both a keen interest in social and political issues as well as a strong technical background which they will build on in this project, becoming experts in NLP, network analysis and deep learning techniques.

    Details:

    The project will be supervised by Rudy Arthur at University of Exeter and me.

    The selected student will spend at least one year at University of Exeter in the UK.

    Application deadline: 15 June 2020.

View all Available Projects

Publications

Book Chapter

  • Hovy, Dirk, Rahimi, Afshin, Baldwin, Timothy and Brooke, Julian (2019). Visualizing regional language variation across Europe on Twitter. Handbook of the Changing World Language Map. edited by Stanley D Brunn and Roland Kehrein.Cham, Switzerland: Springer International Publishing. doi:10.1007/978-3-319-73400-2_175-1

Journal Article

Conference Publication

  • Rahimi, Afshin, Li, Yuan and Cohn, Trevor (2019). Massively multilingual transfer for NER. 57th Annual Meeting of the Association for Computational Linguistics (ACL), Florence, Italy, Jul 28-Aug 02, 2019. Stroudsburg, United States :Association for Computational Linguistics -ACL. doi: 10.18653/v1/p19-1015

  • Rahimi, Afshin, Cohn, Trevor and Baldwin, Timothy (2018). Semi-supervised User Geolocation via Graph Convolutional Networks. 56th Annual Meeting of the Association for Computational Linguistics (ACL), Melbourne, Australia, 15-20 July, 2018. Stroudsburg, United States :Association for Computational Linguistics -ACL.

  • Miyazaki, Taro, Rahimi, Afshin, Cohn, Trevor and Baldwin, Timothy (2018). Twitter geolocation using knowledge-based methods. 2018 EMNLP Workshop W-NUT: The 4th Workshop on Noisy User-generated Text, Brussels, Belgium, November 2018. Stroudsburg, PA, United States :Association for Computational Linguistics. doi: 10.18653/v1/w18-6102

  • Rahimi, Afshin, Cohn, Trevor and Baldwin, Timothy (2017). A neural model for user geolocation and lexical dialectology. 55th Annual Meeting of the Association for Computational Linguistics (ACL), Vancouver, Canada, 30 July - 4 August, 2017. Stroudsburg, PA, United States :Association for Computational Linguistics -ACL. doi: 10.18653/v1/p17-2033

  • Rahimi, Afshin, Baldwin, Timothy and Cohn, Trevor (2017). Continuous representation of location for geolocation and lexical dialectology using mixture density networks. 2017 Conference on Empirical Methods in Natural Language Processing, Copenhagen, Denmark, September 2017. Stroudsburg, PA, USA :Association for Computational Linguistics. doi: 10.18653/v1/d17-1016

  • Magdy, Walid, Darwish, Kareem, Abokhodair, Norah, Rahimi, Afshin and Baldwin, Timothy (2016). #ISISisNotIslam or #DeportAllMuslims? Predicting unspoken views. WebSci '16: 8th ACM Conference on Web Science, Hannover, Germany, 22-25 May 2016. New York, NY, United States :Association for Computing Machinery. doi: 10.1145/2908131.2908150

  • Rahimi, Afshin, Cohn, Trevor and Baldwin, Timothy (2016). Pigeo: A Python geotagging tool. 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016, Berlin, Germany, 7 - 12 August 2016. Stroudsburg, United States :Association for Computational Linguistics (ACL).

  • Rahimi, Afshin, Vu, Duy, Cohn, Trevor and Baldwin, Timothy (2015). Exploiting text and network context for geolocation of social media users. NAACL HLT 2015 - Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Denver, Colorado, United States, 31 May - 5 June 2015. Stroudsburg, United States :Association for Computational Linguistics (ACL). doi: 10.3115/v1/n15-1153

  • Rahimi, Afshin, Cohn, Trevor and Baldwin, Timothy (2015). Twitter user geolocation using a unified text and network prediction model. 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing of the Asian Federation of Natural Language Processing, ACL-IJCNLP 2015, Beijing, China, 26 - 31 July 2015. Stroudsburg, United States :Association for Computational Linguistics (ACL). doi: 10.3115/v1/p15-2104

  • Rahimi, Afshin, Sahlgren, Magnus, Kerren, Andreas and Paradis, Carita (2014). The STAVICTA group report for RepLab 2014 Reputation Dimensions task. CLEF 2014 Conference and Labs of the Evaluation Forum, Sheffield, United Kingdom, 15-18 September 2014. Aachen, Germany :Rheinisch-Westfaelische Technische Hochschule Aachen.

  • Neshati, Mahmood, Alijamaat, Ali, Abolhassani, Hassan, Rahimi, Afshin and Hoseini, Mehdi (2007). Taxonomy learning using compound similarity measure. IEEE/WIC/ACM International Conference on Web Intelligence (WI'07), Fremont, CA, United States, 2-5 November 2007. Piscataway, NJ, United States :IEEE. doi: 10.1109/WI.2007.135

PhD and MPhil Supervision

Current Supervision

  • Doctor Philosophy — Associate Advisor

Possible Research Projects

Note for students: The possible research projects listed on this page may not be comprehensive or up to date. Always feel free to contact the staff for more information, and also with your own research ideas.

  • Project Description

    People don’t live in a flat society, they live in overlapping and interconnected communities sharing attributes such as geography, language, identity, and political opinion. Online communities mirror offline ones and are highly influential in determining how information is propagated and how opinion is shaped. The goal of this project is to utilise both linguistic clues and online social interactions to identify the social and linguistic structure of communities, their interactions and how these communities influence and shape opinions. Hidden factors that determine why users connect to each other on social media will be inferred and used to detect and measure communities, improving on traditional graph-based community detection methods.

    The communication within and between these communities will be studied to understand the spread of cultural and linguistic innovation and how information and political ideas travel along these networks. For example, we are interested in studying the relationship between the core and the periphery, like London and the North of England or Sydney and Western Australia, answering questions like: How do the concerns of people in the periphery reach decision and policy makers in the core? How does culture like news, memes and hashtags spread around these networks? What biases do people in the core have about the periphery and vice-versa?

    This will advance our knowledge on how information spreads and provide tools to magnify marginalised voices. The ideal student for this project will have both a keen interest in social and political issues as well as a strong technical background which they will build on in this project, becoming experts in NLP, network analysis and deep learning techniques.

    Details:

    The project will be supervised by Rudy Arthur at University of Exeter and me.

    The selected student will spend at least one year at University of Exeter in the UK.

    Application deadline: 15 June 2020.