Cheul Young Park

I am a Data Manager at Upstage. I work on:

automating Data Management Operations (DMOps),
and developing LLM-based applications.

Previously, I was:

an AI Researcher at Silvia Health. I trained models to detect cognitive impairments from speech.
a Master’s student in Interactive Computing Lab at KAIST. I worked on:
- AdaptiveESM – an active learning approach towards a balanced and efficient collection of emotions in the wild.
- K-EmoCon Dataset – a multimodal, biometric sensor dataset for recognizing emotions in the wild.

CV: [pdf] (last updated: Jan. ‘22)

Updates

Publications

* Works in this section are subject to changes until their official publication

[CHI]

Understanding Emotion Changes in Mobile Experience Sampling Soowon Kang, Cheulyoung Park, Narae Cha, Auk Kim, and Uichin Lee In CHI 2022, to appear

2021

[EMBC]

Arousal-Valence Classification from Peripheral Physiological Signals Using Long Short-Term Memory Networks M Sami Zitouni, Cheulyoung Park, Uichin Lee, Leontios Hadjileontiadis, and Ahsan Khandoker In 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) [abstract] [pdf] [web]

The automated recognition of human emotions plays an important role in developing machines with emotional intelligence. However, most of the affective computing models are based on images, audio, videos and brain signals. There is a lack of prior studies that focus on utilizing only peripheral physiological signals for emotion recognition, which can ideally be implemented in daily life settings using wearables, e.g., smartwatches. Here, an emotion classification method using peripheral physiological signals, obtained by wearable devices that enable continuous monitoring of emotional states, is presented. A Long Short-Term Memory neural network-based classification model is proposed to accurately predict emotions in real-time into binary levels and quadrants of the arousal-valence space. The peripheral sensored data used here were collected from 20 participants, who engaged in a naturalistic debate. Different annotation schemes were adopted and their impact on the classification performance was explored. Evaluation results demonstrate the capability of our method with a measured accuracy of >93% and >89% for binary levels and quad classes, respectively. This paves the way for enhancing the role of wearable devices in emotional state recognition in everyday life.

2020

[Sci. Data]

K-EmoCon, a multimodal sensor dataset for continuous emotion recognition in naturalistic conversations Cheulyoung Park, Narae Cha, Soowon Kang, Auk Kim, Ahsan Khandoker, Leontios Hadjileontiadis, Alice Oh, Yong Jeong, and Uichin Lee Nature Scientific Data [abstract] [pdf] [web] [data] [code]

Recognizing emotions during social interactions has many potential applications with the popularization of low-cost mobile sensors, but a challenge remains with the lack of naturalistic affective interaction data. Most existing emotion datasets do not support studying idiosyncratic emotions arising in the wild as they were collected in constrained environments. Therefore, studying emotions in the context of social interactions requires a novel dataset, and K-EmoCon is such a multimodal dataset with comprehensive annotations of continuous emotions during naturalistic conversations. The dataset contains multimodal measurements, including audiovisual recordings, EEG, and peripheral physiological signals, acquired with off-the-shelf devices from 16 sessions of approximately 10-minute long paired debates on a social issue. Distinct from previous datasets, it includes emotion annotations from all three available perspectives: self, debate partner, and external observers. Raters annotated emotional displays at intervals of every 5 seconds while viewing the debate footage, in terms of arousal-valence and 18 additional categorical emotions. The resulting K-EmoCon is the first publicly available emotion dataset accommodating the multiperspective assessment of emotions during social interactions.

[IMWUT]

Hello There! Is Now a Good Time to Talk? Opportune Moments for Proactive Interactions with Smart Speakers Narae Cha, Auk Kim, Cheulyoung Park, Soowon Kang, Mingyu Park, Jae-Gil Lee, Sangsu Lee, and Uichin Lee Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. [abstract] [pdf] [web]

Increasing number of researchers and designers are envisioning a wide range of novel proactive conversational services for smart speakers such as context-aware reminders and restocking household items. When initiating conversational interactions proactively, smart speakers need to consider users’ contexts to minimize disruption. In this work, we aim to broaden our understanding of opportune moments for proactive conversational interactions in domestic contexts. Toward this goal, we built a voice-based experience sampling device and conducted a one-week field study with 40 participants living in university dormitories. From 3,572 in-situ user experience reports, we proposed 19 activity categories to investigate contextual factors related to interruptibility. Our data analysis results show that the key determinants for opportune moments are closely related to both personal contextual factors such as busyness, mood, and resource conflicts for dual-tasking, and the other contextual factors associated with the everyday routines at home, including user mobility and social presence. Based on these findings, we discuss the need for designing context-aware proactive conversation management features that dynamically control conversational interactions based on users’ contexts and routines.

2018

[SOUPS]

Share and Share Alike? An Exploration of Secure Behaviors in Romantic Relationships Cheulyoung Park, Cori Faklaris, Siyan Zhao, Alex Sciuto, Laura Dabbish, and Jason Hong In Fourteenth Symposium on Usable Privacy and Security (SOUPS) [abstract] [pdf] [web] [press]

Security design choices often fail to take into account users’ social context. Our work is among the first to examine security behavior in romantic relationships. We surveyed 195 people on Amazon Mechanical Turk about their relationship status and account sharing behavior for a cross-section of popular websites and apps (e.g., Netflix, Amazon Prime). We examine differences in account sharing behavior at different stages in a relationship and for people in different age groups and income levels. We also present a taxonomy of sharing motivations and behaviors based on the iterative coding of open-ended responses. Based on this taxonomy, we present design recommendations to support end users in three relationship stages: when they start sharing access with romantic partners; when they are maintaining that sharing; and when they decide to stop. Our findings contribute to the field of usable privacy and security by enhancing our understanding of security and privacy behaviors and needs in intimate social relationships.

Research

Interactive Computing Lab, KAIST KSE (Aug ‘18 ~ Feb ‘21)

Mobile Computing, Affective Computing (Advised by Prof. Uichin Lee)

CHIMPS Lab, CMU HCII (Jul ‘17 ~ Jul ‘18)

Usable Privacy and Security (Advised by Prof. Jason Hong)

Infant Language and Learning Lab, CMU Psychology (May ‘16 ~ Dec ‘16)

Developmental Psychology, Cognitive Psychology (Advised by Prof. Erik Thiessen)

Teaching

KAIST IT Academy - Image Deep Learning with PyTorch (Feb ‘20, co-instructed with Jisu Han)

A 30-hour course on the basics of deep learning using PyTorch with a focus on image processing, taught in Korean & English.

KAIST IT Academy - Python Programming Basics (Oct ‘19 & Jan ‘20)

A 30-hour course on the fundamentals of programming with Python, taught in Korean & English.

📌 Dec 9, 2021	A new paper “Arousal-Valence Classification from Peripheral Physiological Signals Using Long Short-Term Memory Networks” has been published on IEEE EMBC.
📌 Nov 16, 2021	A new paper has been conditionally accepted with minor revisions to CHI’22!
📌 Feb 1, 2021	Joined the amazing team at Silvia Health as an AI Researcher 🎉
📌 Dec 23, 2020	Successfully defended my master’s thesis “Towards Adaptive Sampling of Emotions in the Wild with Active Learning”
📌 Nov 22, 2020	My past work on password sharing was featured in the WIRED UK article “You need to stop sharing your passwords with your partner” by Victoria Turk.
📌 Sep 8, 2020	A new paper “K-EmoCon, a multimodal sensor dataset for continuous emotion recognition in naturalistic conversations” has been published on Nature Scientific Data.
📌 Sep 7, 2020	A new paper “Hello There! Is Now a Good Time to Talk?: Opportune Moments for Proactive Interactions with Smart Speakers” has been published on IMWUT.