Research focus: Ethics of natural language processingMentors: Syed Ishtiaque Ahmed (Dynamic Graphics Project) and Shion Guha (Human-Centered Data Science)

Al-Adala (justice, fairness, and equality) Lab, Advisor: Dr. Bryan SemaanProject: Through colonialism, external forces can alter and shift social structures and practices, thus, profoundly marginalizing the collective identities of local and indigenous populations. Decolonization is the undoing of the impacts of colonial domination. While sociotechnical systems (e.g., online platforms) can support the identity work of such marginalized communities, scholars have also discussed how these systems impose values and exhibit colonial impulses. By employing a mixed method approach (interviews, trace ethnography, speculative and participatory design, network analysis, NLP), we are interested in understanding how colonially marginalized people decolonize their identities and practices on different online platforms (e.g., Quora, YouTube, Facebook); and how we can make these platforms more equitable, accessible, just, and fair to these communities.Publications: CHI 2022, COMPASS 2022, GROUP 2022, CSCW 2021

With GPTI (Graduate Part-Time Instructor) role, I am the instructor of record for the INFO 1301 (Statistics for Information Science) course for Spring 2023. The course Introduces concepts and techniques for characterizing and quantifying data, teaches to summarize, visualize, and interpret data with descriptive statistics, and the foundations of statistical inference and modeling.

INFO 1201: Computational Reasoning 1 (Fall 2021: 160 students, Spring 2022: 140 students)Responsibilities include attending two recitations, helping students debug codes during theory classes, holding office hours, setting quiz questions, and grading.

Research practicum:Advised by Dr. Carsten ØsterlundProject: Sociomateriality of Governance in Online PlatformsResearch experience:Cognition, Computation, Communication & Complexity (C4) Lab, Supervisor: Dr. Josh IntroneProject: We have developed a framework for exploring how social forces interact to guide the evolution of beliefs amongst groups of people which we call the belief landscape framework (BLF). As with a fitness landscape, the belief landscape can be imagined as a surface where some points are more ‘optimal’ than others - more coherent systems of belief. We want to understand the structure of BLF, predict individuals' likely future belief states, whether people in different regions of a belief space exhibit different information behaviors and how social influence contributes to these variations.Teaching experience/practicum:* IST 649: Human Interaction with Computers (Fall 2019: 8 students), mentored by Dr. Bryan Semaan. Responsibilities included holding office hours and leading discussions in graduate-level classes.* IST 421: Information Visualization (Spring 2021: 33 students), mentored by Dr. Josh Introne. Responsibilities included designing the syllabus and assignments. updating codes, and teaching class.* IST 343: Data in Society (Spring 2020: 86 students), mentored by Dr. Jennifer Stromer-Galley. Responsibilities included leading and grading online and in-class discussions and preparing quiz questions.* IST 300: Digital Platforms (Fall 2020), mentored by Dr. John Jordan. Responsibilities included designing the syllabus, reading list, and assignments.

Research experience:Autonomous Robotics and Complex Systems (ARCS) Lab, Advisor: Dr. Anthony J. ClarkProject: While natural language processing is heavily used to analyze sentiment online, sarcasm has often been overlooked as an exceptional case of user sentiment which can lead to misinterpretation of user interaction. Moreover, most of sentiment analysis approaches highlight on the use of English text data, and thus, cannot utilize the large amount of multimedia data (e.g. images, video) which can have cues applicable across a wider range of languages besides English. With a mixed method study, we propose a multimodal approach to sarcasm detection on social media.Publications: ICMI 2018, ICCBD 2018, HCC 2019, TransAI 2019Teaching experience:* CSC 131: Computational Thinking (Spring 2018: 42 students, Spring 2019: 45 students)* CSC 130: The World of Computer Science (Fall 2018: 36 + 32 students, Spring 2019: 37 students)Responsibilities included assisting in classes, tutoring, grading.

Taught courses:* CSE 401+402: Compiler Design+Lab (two sections: 33 + 49 students)* CSE 301+302: Algorithm Design and Analysis+Lab (54 students)* CSE 209: Computer Organization and Architecture (59 students)Responsibilities included lecturing in theory classes (odd-numbered courses), conducting labs (even-numbered courses), grading, and attending office hours.

Undergraduate thesis project, advised by Drs. Anindya Iqbal and Rifat ShahriyarProject: On StackOverflow, the users contributing answers to different questions are awarded similarly irrespective of the difficulty of the questions. To facilitate a question difficulty aware rewarding system, we proposed SOQDE (Stack Overflow Question Difficulty Estimation), a supervised learning based Question difficulty estimation model for the StackOverflow, based on different (pre-hoc and post-hoc) textual and contextual features features.Publication: APSEC 2018