Sanghyun Seo

Ph.D Candidate

Name: Sanghyun Seo(서상현)

Status: Ph.D. Student, Department of Computer Engineering at Dongguk University

Tel: +82-2-2290-1421

Address: New Engineering building 5116, Dongguk University, 30, Pildong-ro 1-gil, Jung-gu, Seoul, Republic of Korea.

E-mail: shseo@dongguk.edu

Research Areas: Machine Learning, Deep Learning, Meta Learning

Master Thesis: A Study on Weights Quantization for Improving Efficiency of Deep Neural Networks


Academic Career

2016.02: Dongguk University, Bachelor of Arts in Korea Language and literature
2018.02: Dongguk University, Master of Engineering in Computer Engineering and Science
2018.02 ~ Present: Dongguk University, Ph.D. Student in Computer Engineering and Science

 


Publications (Google Scholar Page)

2020

  • [SCI] Sanghyun Seo, Sanghyuck Na, Juntae Kim, “HMTL: Heterogeneous Modality Transfer Learning for Audio-visual Sentiment Analysis”, IEEE Access. July 02, 2020.[PDF]

2019

  • [SCI(E)] Sanghyun Seo, Juntae Kim, “Hierarchical Semantic Loss and Confidence Estimator for Visual-Semantic Embedding-Based Zero-Shot Learning, Applied Sciences 9.15 (2019): 3133-2,August 2019. [PDF]
  • [SCI(E)] Sanghyun Seo, Juntae Kim, “Efficient Weights Quantization of Convolutional Neural Networks Using Kernel Density Estimation based Non-uniform Quantizer, Applied Sciences 9.12 (2019): 2559—23,June 2019. [PDF]
  • [SCI] Seongchul Park, Sanghyun Seo, Changhoon Jeong, Juntae Kim, The weights initialization methodology of unsupervised neural networks to improve clustering stability”, The Journal of Supercomputing (2019): 1-17. [PDF]
  • [SCI(E)] Seongchul Park, Sanghyun Seo, Changhoon Jeong, Juntae Kim, “Online Eigenvector Transformation Reflecting Concept Drift for Improving Network Intrusion Detection”, Expert Systems. Nov 14, 2019. [PDF]
  • Sanghyun Seo,  Juntae Kim, “Heterogeneous Data Integration using Confidence Estimation of Unseen Visual Data for Zero-shot Learning, ICSIM 2019 Proceedings of the 2nd International Conference on Software Engineering and Information Management, pp. 171-174, Bali, Indonesia — Jan 10 – 13, 2019. [PDF]
  • Sanghyuck Na, Daeung Kim, Sanghyun Seo, Juntae Kim, Improvement of Cross-Modal Retrieval Performance for Visual-Semantic Data Through Adversarial Learning”, 7th International Conference on Big Data Applications and Services (BIGDAS2019), Jeju Island, South Korea — August 21-24, 2019.
  • Seongchul Park, Sanghyun Seo, Changhoon Jeong, Juntae Kim, “Method for the Initial Value of Weights to Obtain Stable Results of Unsupervised Neural Network, FUTECH 2019 Proceedings of The International Workshop on Future Technology, pp. 55-57, Taichung, Taiwan— Jan 19- 21, 2019. [PDF]

2018

  • Sanghyun Seo, Yongjin Jeon, Juntae Kim, “Meta Learning for Imbalanced Big Data Analysis by using Generative Adversarial Networks”, ICBDC ’18 Proceedings of the 2018 International Conference on Big Data and Computing, pp. 5-9, Shenzhen, China — April 28 – 30, 2018. [PDF]
  • Sanghyun Seo, Seongchul Park, Changhoon Jeong, Juntae Kim, “Knowledge Distillation based Online Learning Methodology using Unlabeled Data Stream”, MLMI18 Proceedings on 2018 International Conference on Machine Learning and Machine Intelligence, Hanoi, Vietnam — September 28 – 30, 2018. [PDF]
  • Sanghyun Seo, Juntae Kim, “Image-Text Embedding with Hierarchical Knowledge for Cross-Modal Retrieval”, CSAI 2018 Proceedings on 2018 International Conference on Computer Science and Artificial Intelligence, Shenzhen, China — December 19 – 21, 2018. [PDF]
  • Sanghyun Seo, Juntae Kim, “Hybrid Approach for Efficient Quantization of Weights in Convolutional Neural Networks”, Proceedings of 2018 IEEE International Conference on Big Data and Smart Computing, pp. 638-641, Shanghai, China, 2018. [PDF]
  • Seongchul Park, Sanghyun Seo, Changhoon Jeong, Juntae Kim, “Network Intrusion Detection through Online Transformation of Eigenvector Reflecting Concept Drift”, Data’18 Proceedings on International Conference on Data Science, E-learning and Information Systems 2018, Madrid, Spain— October 1 – 3, 2018. [PDF]
  • 서상현, 기효진, 전용진, 박현민, 마고트라, 김준태, “멀티모달 이종 빅데이터 통합 임베딩을 위한 Quintet Networks”, 2018 한국컴퓨터종합학술대회 논문집 (KCC 2018), pp.844~846, Jeju Korea, 2018. [PDF]
  • 박성철, 서상현, 정창훈, 김준태, “생성적 적대 신경망을 이용한 정형 데이터 증강에 대한 연구”, 2018 한국컴퓨터종합학술대회 논문집 (KCC 2018), pp. 844~846, Jeju Korea, 2018. [PDF]

2017

  • Sanghyun Seo, Seongchul Park, Injea Hwang, Juntae Kim, “ADSTREAM: Anomaly Detection in Large-Scale Data Streams Using Local Outlier Factor Based on Micro-Cluster”, Advanced Science Letters Volume 23 Number 10, pp. 9907-9911, 2017. [PDF]
  • Seongchul Park, Sanghyun Seo, Juntae Kim, “Network Intrusion Detection Using Stacked Denoising Autoencoder”, Advanced Science Letters Volume 23 Number 10, pp. 10204-10209, 2017. [PDF]
  • Seongchul Park, Sanghyun Seo, Juntae Kim, “A Study on Network Intrusion Detection using Autoencoder”, Proceedings of The 2’nd International Conference on Advanced Science and Information Technology (ASCIT 2017), pp. 29-32, Cheonan Korea,  2017.
  • 나훈엽, 서상현, 윤지상, 정창훈, 전용진, 김준태, 텍스트 기반 상담시스템의 효율성 제고를 위한 합성곱신경망을 이용한 자동답변추천 시스템”, 제29회 한글 및 한국어 정보처리학회 논문집, pp. 272-275, Daegu Korea, 2017.
  • 서상현, 전용진, 이종수, 정호재, 김준태, 불균형 빅데이터의 효율적인 분류를 위한 생성적 적대 신경망 기반 오버샘플링 기법”, 한국정보과학회 2017 한국소프트웨어종합학술대회 논문집, pp. 1030-1032, Busan Korea, 2017. [PDF]
  • 알렉스 샤이코니, 서상현, 권영식, “딥러닝 기반의 다범주 감성분석 모델 개발”, 한국IT서비스학회지 제16권 4호, pp. 149-160. 2017. [PDF]

2016

  • 서상현, 김준태,  “딥러닝 기반 감성분석 연구동향”, 한국멀티미디어학회, 제20권 제3호, pp.8~22, 2016 9. [PDF]
  • Sanghyun Seo, Seongchul Park, Juntae Kim, “Improvement of Network Intrusion Detection Accuracy by using Restricted Boltzmann Machine”, 2016 8th International Conference on Computational Intelligence and Communication Networks, Tehri India, 2016. [PDF]
  • Seongchul Park, Sanghyun Seo, Gilsik Park, Juntae Kim, “A Study on Intelligent Financial Counseling Services by using Case-Based Reasoning”, International Symposium on Advanced and Applied Convergence 2016(ISAAC2016), pp. 11~13, Jeju Island, 2016. 11.