Exporters From Japan
Wholesale exporters from Japan   Company Established 1983
CARVIEW
Select Language

Alumni

  • Mincheol Kang, Undergrad Intern (DGIST), Spring 2024 - Summer 2024
  • Jiheon Kang, Undergrad Intern (DGIST), Spring 2024 - Summer 2024
  • Yibin Moon, Undergrad Intern (DGIST), Spring 2024
  • Nayoung Kim, Undergrad Intern (DGIST), Winter 2023 - Spring 2024
  • Jaeyoung Choi, Undergrad Intern (DGIST), Winter 2023 - Spring 2024
  • Suyeon Shin, Undergrad Intern (DGIST), Summer 2023 - Spring 2024
  • Sungmin Ju, Undergrad Intern (DGIST), Summer 2023 - Spring 2024
  • Jeongeon Park, Researcher, Fall 2023 - Spring 2024
  • Sunoo Kim, Undergrad Intern (DGIST), Winter 2024 - Spring 2024
  • Youngjin Park, Undergrad Intern (DGIST), Winter 2024
  • Dain Kim, Undergrad Intern (DGIST), Winter 2023 - Fall 2023
  • Sihwan Seok, Undergrad Intern (DGIST), Winter 2023 - Fall 2023
  • Subeen Jo, Undergad Intern (DGIST), Winter 2023 - Fall 2023
  • Minseong Kim, Undergrad Intern (DGIST), Spring 2023 - Summer 2023
  • Huidam Woo, Undergrad Intern (DGIST), Spring 2023
  • Chan Lee, Undergrad Intern (DGIST), Spring 2023
  • Ahyeon Shin, Undergrad Intern (DGIST), Winter 2023
  • Sangeun Seo, Undergrad Intern (UNIST), Summer 2022, Winter 2023
  • Chanyu Moon, Undergrad Intern (DGIST), Winter 2023
  • Gunuk Nam, Undergrad Intern (Handong Global University), Winter 2023
  • Nayoung Kim, Undergrad Intern (DGIST), Winter 2022 - Winter 2023
  • Hojin Jin, Undergrad Intern (DGIST), Winter 2022 - Fall 2022
  • Hyeonho Kwon, Undergrad Intern (DGIST), Winter 2022 - Fall 2022
  • Giwon Lee, Undergrad Intern (DGIST), Winter 2022 - Fall 2022
  • Giwa Osaruiyobo Henrietta, Graduate Intern (DGIST), Summer 2022

Publications

Conference and Journal Papers



Posters, Demos, and Workshop Papers

Projects

Our research projects include but are not limited to the following list. We are open to discuss new project ideas in the topic of Human-Computer Interaction, Human-AI Interaction, Crowdsourcing, Human Computation, or any other related domains. Please feel free to contact us if you have exciting ideas.


Facilitating Online Community Moderation using AI Techniques

Communities on social platforms have a group of users who volunteer to moderate their communities, called online moderators. They respond to the behavior of community members that violate rules and work to improve overall interaction experiences between community members. This project aims to build human-AI interaction tools to support moderators to more easily and transparently moderate their online communities.



carview.php?tsp=

Improving Facial Emotion Recognition for both Human and AI

This project focuses on enhancing Facial Emotion Recognition(FER) for individuals who struggle with recognizing facial expressions, as well as for current models with low FER performance. To address these challenges, we developed a gamified application called "Find the Bot!" that re-labels FER datasets and trains individuals with difficulties in recognizing facial emotions simultaneously. Our vision is to improve FER abilities for individuals with weak FER, leading to enhanced relationship quality, social acceptance, and increased social productivity, while also contributing to the development of reliable multi-labeling for facial expression datasts.





carview.php?tsp=

Reducing Negative Emotions for Crowd Workers Exposed to Disturbing Content

Often times crowd workers who do content moderation or data annotations are exposed to harmful content in their daily routines. This project explores UI intervention techniques that can prevent crowd workers from getting too much negative emotional impacts when conducting these tasks. Through this study, we aim to find ways to maintain the quality and cost of the crowdsourced work while protecting the emotions and mental health of crowd workers.



carview.php?tsp=



Continual Learning from a Data Science Perspective

Implementing a model created in a controlled environment into the real world requires solving the problem of Catastrophic Forgetting. To solve that problem, we seek to draw inspiration from the Human Complementary Learning Systems (CLS) from a data science perspective. Through this research, we hope that the model created in a controlled environment can continuously learn in the real world, and through this, the controlled model can be applied to daily life.





carview.php?tsp=



Augmenting QnA Dataset for Technical Documents with Crowdsourcing

This project proposes a new sentence structuring method and crowdsourcing interface to augment rich Korean dataset in text documents specialized in technology and science. Our novel authoring tool enables to efficiently generating various question and answer pairs for technical documentation by breaking down sentences into elements that can be recombined to form new sentences.


carview.php?tsp=

Contact

​ We welcome collaborations!
If you have any questions or would like to hear more about our research, please feel free to shoot us an email: jeansong@yonsei.ac.kr


Copyright © diag.kr 2025