Resume

Projects

Research projects

• Social dimensions impact individual sleep quantity and quality (June 2023)

We study how Sleep Dimensions like quality and quantity are explained by cultural norms and values. Here we analyze 52 million sleeping readings aggregated from wearable logs from 19 cities between 2014 and 2017 and examine how sleep is affected by culture. Accepted to Scientific Reports 2023.

[Paper]

• User-Chatbot Conversations During the COVID-19 Pandemic: A Study Based on Topic Modeling and Sentiment Analysis. (2023)

This work examined user-chatbot interactions on a live platform and provides insights into people's informational and emotional needs during a global health crisis. Users sought health-related information and shared emotional messages with the chatbot, indicating the potential of future chatbots to provide accurate health information and emotional support.  Accepted to Journal of Medical Internet Research (JMIR), To Appear (IF=7.08).

[Paper]

• Unique Characteristics of Human-Chatbot Conversations and Their Potential For Mental Health Support (May 2022)

In this work, we investigate how differently users express emotions online when interacting with a commercially available chatbot (Simsimi) and posting on a public social media platform (Twitter).  This is an ongoing project in collaboration with Simsimi corporation. The short paper is published at '"KCC 2021 Web Conference".

[PDF] [Paper]

• Political personalization on Twitter (Summer 2022)

In this work, I tried to understand how leaders and countries are associated with each other during international disasters among English-speaking Twitter users. Under review.

[Presentation] 

• Emotion Bubbles (April 2022) 

In this project, we use sentiment classifiers to analyze the public discourse on Twitter and compare the average sentiments in posts of individuals before and after the outbreak of the pandemic.  Published at '"KCC 2021 Web Conference". Presented at "BIEN 2021: The International Conference of Women Scientists and Engineers Conference on BT, IT, ET, and NT". Accepted to "TheWebConf22" (WWW'22), Lion, France.

[Paper] [Presentation]

• On the Social-Relational Moral Standing of AI: An Empirical Study Using AI-Generated Art (Summer 2021)  

We shed light on how empirical studies can contribute to the AI and robot rights debate by revealing the public perception of this issue and conducting several experiments with GAN-generated images. Published in the "Frontiers in Robotics and AI" journal, section Ethics in Robotics and Artificial Intelligence.

[Paper] [Data]

• COVID-19 Simulation Using Probabilistic Programming (Spring 2021)  

We present a generative simulation model that accounts for the economic and human capital cost of adopting some strategies against COVID-19, and provide an end-to-end pipeline to simulate the virus spread and the incurred loss of various policy combinations. Front. Public Health, 21 November 2022 Sec. Infectious Diseases: Epidemiology and Prevention

[Code] [Paper] [Poster]

Course Projects

• ClefClub (Fall 2021). 

App for crowdsourced music generation.

[Code] [Prototype]

• River-LSTM (Spring 2021).

Predicting river flows from satellite images. This project got 1st place at JunctionX Hackathon Seoul 2021. 

[Presentation]

• Diffusion in GNN (Spring 2021). 

In these experiments, I compare different diffusion kernels for Graph diffusion Convolution. Original paper: Diffusion Improves Graph Learning by Klicpera et al. 

[Code] [Presentation]

• Iterative Null-space Projection for bias mitigation (Spring 2021). 

In these experiments, we apply the INLP method for mitigating gender, race, and religious biases in the toxicity classification tasks.

[Code] [Presentation]

• QuessIt (Spring 2021). 

In this project, we developed an ML model using Qiskit for solving quantum problems. Qiskit Hackathon 2021. 

[Code]

• Lecture Summarizer (Fall 2020). 

Summarization of YouTube lectures using BERT for better remote education. Coded the backend of a website for human-AI cooperation. Used tools: Flask, React, Python. 

[Code] [Prototype]

• Soccer Analysis (Fall 2020). 

Analysis of Kaggle Soccer data. Conducted preprocessing and statistical analysis of the data. Predicted key features for successful soccer teams using R.

[Code] [Report]

• PANicDEMIC (Spring 2020). 

NLP project on sentiment analysis of COVID-19 tweets. Implemented transfer learning for increasing performance of emotion classifier from 82% to 83%. Used tools: Python, NLTK.

[Code] [Report] [Video]

Recognitions & Awards

• 1st place in the SIA track, JunctionX Seoul Hackathon (May 2021). 🥇 Predicting river flows from satellite images.

• KCS 2020 Conference Excellence Award (Feb 2021). 🏆  Response to COVID-19 with probabilistic programming.

• Bronze Medal (Summer 2017). 🥉 49th International Chemistry Olympiad in Thailand.

• Silver Medal (Spring 2017). 🥈 51st International Mendeleev Olympiad in Astana.

• Creativity Award and Gold Medal (Spring 2017). 🥇 International Bekturov Olympiad in Pavlodar.

• Gold Medal (Spring 2017). 🥇 National Chemistry Olympiad in Pavlodar.

• Gold Medal (Winter 2012). 🥇 National Ballroom Championship among Juniors in Karaganda.

Hobbies

Reading 📖  : writing book reviews in personal Telegram channel (t.me/SemkaReads)

Volunteering: Member of the volunteering club in Daejeon (Silver Lining)

Yoga 🧘‍♀️ : still can't stand still

Ukulele 🎶  : learned to play some Beatles songs during quarantine

Writing ✍️ : hoping to publish some crazy and boring novels