Daeseung Park

AI Engineer & Software Systems Engineer

Profile

Experience

2023.05 - present Research Engineer Korea Advanced Institute of Science and Technology (KAIST)
KAIST Information & Electronics Research Institute
KAIST Convergence Research Center for College of Engineering
Daejeon Metropolitan City-KAIST Strategic Project Research Center
2020.08 - 2023.05 Participating Researcher Industry-Academic Cooperation Foundation of Kwangwoon University
2020.03 - 2023.05 Associate Researcher Industry-Academic Cooperation Foundation of Namseoul University
2014.02 - 2015.03 Participating Researcher Ministry of Science, ICT and Future Planning - 3D Smart Media/Augmented Reality Project

Skills

Back-end Front-end Database Android C# C/C++ Java Python Ruby UI/UX Bootstrap Photoshop Vegas

Awards

2010.06The Bronze Award of Information & CommunicationCommercial Practical Competency Competition
2009.10The Excellence Award of Information & CommunicationLocal Skills Competition
2008.06The Bronze Award of Information & CommunicationCommercial Practical Competency Competition

Certifications

정보처리기사 (Engineer for Information Processing) 2014.08
정보처리산업기사 (Industrial Engineer for Information Processing) 2013.08
윈도우포렌식전문가 (Windows Forensic Specialist) 2013.06
리눅스마스터 (Linux Master) 2009.12
네트워크관리사 (Network Management Specialist) 2009.12
Cisco Certified Network Associate (CCNA) 2009.02
Linux Professional Institute Certification (LPIC) 2009.02
정보처리기능사 (Craftsman for Information Processing) 2002.05

Special Credentials (Google)

Google Workspace(G Suite) Deployment Service Specialist
Google Education Deployment Specialist
Google Cloud Sales Specialist
Google Education Sales Specialist

Research Projects

전문기관 (Professional Organization): 행정안전부 (Ministry of the Interior and Safety) 신종 감염병 해외유입 예측 및 지능적 차단 기술개발 Development of Technology for Predicting and Intelligently Blocking Overseas Inflow of New Infectious Diseases
신종 감염병 대응을 위한 역학조사관의 업무수행 보조 에이전트 연구, 급격한 수치 변화에 적응하는 위상 지연 감소 감염자 시계열 예측 모델 연구 및 높은 추론 성능 달성 Research on an agent to assist epidemiologists in responding to new infectious diseases, research on a time-series prediction model for infected cases with reduced phase delay that adapts to rapid numerical changes, achieving high inference performance.
수행기관 (Executing Organization): 한국수력원자력-일주지앤에스/위탁 (KHNP-Ilju GNS/Consignment) 한수원 원전 빅데이터 플랫폼 및 기반 기술 개발 수행을 위한 AI 분석 모델 개발 및 딥러닝 처리 선행 모델 조사 및 자문 Development of AI Analysis Models and Advisory on Deep Learning Preceding Models for KHNP Nuclear Power Plant Big Data Platform and Infrastructure Development
원자로 발전량 예측에서 원자로 특성에 따른 다변량 학습 및 환경변수 기반 추론을 위해 자연어처리 및 기계번역에서 주로 사용되는 Seq2Seq 모델을 시계열 예측에 도입해 높은 추론 성능 달성 (Encoder-Decoder 구조를 환경변수 입력으로 사용) Achieved high inference performance in reactor power generation prediction by introducing the Seq2Seq model, commonly used in NLP and machine translation, for time-series forecasting to enable multivariate learning based on reactor characteristics and inference based on environmental variables (using an Encoder-Decoder structure for environmental variable input).
수행기관 (Executing Organization): 대전광역시청-LX/KAIST컨소시움 (Daejeon Metropolitan City-LX/KAIST Consortium) 융복합 데이터 활용 실감형 소방안전도시 구축 Establishment of a Realistic Fire-Safe City Utilizing Converged Data
지하시설물 파손을 예측하고 보전방안을 제안하는 언어생성 모델 연구 수행, 운용 환경 특성상 학습 이후 GPU 없이 CPU 기반 추론이 가능하도록 경량언어 모델 구성 연구, 토지 정보를 활용해 정보를 추론 및 생성하는 GeoLLM 연구, 일반적인 신경망의 학습 성능 향상을 위한 가변 레이어 학습 기술 연구 등 Conducted research on a language generation model to predict damage to underground facilities and propose maintenance plans, research on configuring a lightweight language model for CPU-based inference without a GPU post-training due to operational environment constraints, research on GeoLLM for inferring and generating information using land data, and research on variable layer learning technology to improve the training performance of general neural networks.
전문기관 (Professional Organization): 과학기술정보통신부 (Ministry of Science and ICT) 광역권 도시를 위한 차세대 AI 융합 모빌리티 시뮬레이션 및 예측/활용 기술 개발 Development of Next-Gen AI-Converged Mobility Simulation, Prediction, and Utilization Technology for Metropolitan Areas
AI 기반 교통 예측 시뮬레이션 개발이 목표이며 연구수행 기간이 얼마 되지 않아 아직 성과 없음 The goal is to develop an AI-based traffic prediction simulation. No significant results yet as the research period has been short.
2021.06.01 ~ 2022.07.31 (주참여 연구 / Primary Research) 학제간융합연구 / 독거 중고령자의 디지털 헬스케어를 위한 인공지능기반 치료적 의사소통 챗봇 연구 Interdisciplinary Research / A Study on an AI-based Therapeutic Communication Chatbot for Digital Healthcare of Single, Middle-aged and Elderly Households 전문기관 (Professional Organization): 한국연구재단 (National Research Foundation of Korea)
상기 연구에서는 인공지능 챗봇을 바탕으로 디지털 헬스케어, 멘탈 케어등을 연구하기 위해 자연어처리, 인공신경망, 언어생성 모델, 각종 센서기반 패턴분석, 우울진단, 긍부정판단, 감정인식 등 인공지능 및 자연어처리 핵심기술을 연구하였고, SCI 1건, KCI 1건 및 기타 학술논문과 인공지능 챗봇 관련 특허 2건 출원, 데이터 수집 관련 프로그램성과물 4건을 등록함. 본 연구를 통해 인공지능 알고리즘, Transformer, 활성함수, Python, Tensorflow, Pytorch 등 각종 라이브러리 사용, 인공지능 챗봇등 인공지능 기반 응용서비스 기술에 대한 역량을 강화 This research focused on core AI and NLP technologies such as neural networks, language generation models, sensor-based pattern analysis, depression diagnosis, sentiment analysis, and emotion recognition for digital and mental healthcare via AI chatbots. It resulted in 1 SCI paper, 1 KCI paper, other academic articles, 2 patent applications related to AI chatbots, and 4 registered programs for data collection. This enhanced capabilities in AI algorithms, Transformer, activation functions, Python, Tensorflow, Pytorch, and AI-based application services.
2020.04.01 ~ 2022.07.31 (보조참여 연구 / Auxiliary Research) 시스템반도체핵심IP개발 / 모바일 기기에 내장 가능한 동적 실시간 3D 오디오를 위한 사운드 렌더링IP 개발 Core System-on-Chip IP Development / Development of Sound Rendering IP for Dynamic Real-time 3D Audio Embeddable in Mobile Devices 전문기관 (Professional Organization): 한국산업기술평가관리원 (Korea Evaluation Institute of Industrial Technology)
상기 연구에서는 오디오를 3차원 랜더링 하는 칩셋을 개발하고 또한 개발한 칩셋을 사용할 수 있는 소프트웨어 개발을 목표로 하였습니다. 리눅스 기반 소프트웨어 개발, 칩셋에 사용되는 API 프로토콜 설계, 드라이버 관련 소프트웨어 개발등의 업무 수행 및 연구논문 작성을 수행 This research aimed to develop a chipset for 3D audio rendering and the accompanying software. Responsibilities included Linux-based software development, designing API protocols for the chipset, developing driver-related software, and writing research papers.
2020.09.01 ~ 2022.08.31 (보조참여 연구 / Auxiliary Research) 대학ICT연구센터육성지원사업 / 실감콘텐츠 단말 기술 연구 개발 University ICT Research Center Fostering Support Program / Research and Development of Immersive Content Terminal Technology 전문기관 (Professional Organization): 정보통신기획평가원 (Institute for Information & communications Technology Planning & Evaluation)
상기 연구에서는 대형 디스플레이 사이니지의 단말기술, 디스플레이 연동 및 확장기술, 서버리스 기반 단말제어 기술, 상황인지 기반 콘텐츠 관리 및 배포 기술, 실감콘텐츠 기술등을 연구하였고, KCI 1건, 기타 학술논문과 특허 1건, 프로그램성과물 1건을 등록 This research focused on terminal technology for large display signage, display linkage and expansion, serverless-based terminal control, context-aware content management and distribution, and immersive content technologies. It resulted in 1 KCI paper, other academic articles, 1 patent, and 1 registered program.

Portfolio

Untact Medical Technology R&D

Development of realistic content terminal technology including Smart Mirrors and Digital Signage, focusing on non-contact vital-sign measurement using Photoplethysmography (PPG).

Government Project Smart Mirror Non-contact Vital-sign PPG

Mobile 3D Audio Sound Rendering IP

Development of sound rendering IP for dynamic, real-time 3D audio embeddable in mobile devices. A government-funded project focusing on SoC and API technology.

Government Project Sound Rendering SoC IP Development

Google Android App: Mini's Lottery

Developed and launched a commercial Android application, achieving over 50,000 downloads. Responsible for UI/UX design, development, and market strategy.

Android 50k+ Downloads UI/UX Bootstrap

Publications

2024.02.22 [SCIE] 주저자 (First Author)

A Study on Performance Improvement of Prompt Engineering for Generative AI with a Large Language Model
Journal of Web Engineering
2023.10 [International Conference] 주저자 (First Author)

Short-term Korea East-sea Temperature Forecasting Approach based on Seq2Seq Model using Multi Parameters
2023 14th International Conference on Information and Communication Technology Convergence (ICTC)
2023.10 [International Conference] 제2저자 (Second Author)

A Comparative Analysis of Time Series Forecasting Methods for Short-Term Electricity Demand Prediction
2023 14th International Conference on Information and Communication Technology Convergence (ICTC)
2022.09.22 [SCIE] 제2저자 (Second Author)

A Study of BERT-Based Classification Performance of Text-Based Health Counseling Data
CMES-COMPUTER MODELING IN ENGINEERING & SCIENCES
2022.04.01 [KCI] 주저자 (First Author)

단어그룹 확장 기법을 활용한 순환신경망 알고리즘 성능개선 연구
A Study on Performance Improvement of Recurrent Neural Network Algorithm using Word Group Expansion Technique
산업융합연구 / 대한산업경영학회 (Journal of Industrial Convergence / The Korean Society of Industrial Management)
2021.09.01 [KCI] 주저자 (First Author)

이기종 디바이스를 이용한 인터렉티브 디지털 사이니지 시스템 연구
A Study on Interactive Digital Signage System using Heterogeneous Devices
반도체디스플레이학회 (The Korean Society of Semiconductor & Display Technology)
2022.10.01 [Domestic Conference] 주저자 (First Author)

지수 선형 단위와 로지스틱 회귀 모델의 비대칭 조합 함수를 활용한 순환신경망 성능 개선 연구
A Study on Improving RNN Performance using an Asymmetric Combination of ELU and Logistic Regression
한국산업정보학회논문지/춘계학술대회/한국산업정보학회 (Journal of the Korea Industrial Information Systems Research)
2022.12.20 [International Conference] 주저자 (First Author)

Implementation of Multi-Turn Chatbot with Weather Information Based on Dialogflow
The International Association of Scientific & Engineering Researcher - E-CS
2022.12.20 [International Conference] 주저자 (First Author)

A Study on DVI (Dialogue Voice Interface) based on AI Chatbot for Immersive Digital Signage Content
The International Association of Scientific & Engineering Researcher - E-CS
2022.12.20 [International Conference] 주저자 (First Author)

A Study on Web-Database-based Immersive Digital Signage Content Control
The International Association of Scientific & Engineering Researcher - E-CS
2022.07.20 [International Conference] 주저자 (First Author)

A STUDY OF MULTI-MODEL CHATBOT STRUCTURE
The International Association of Scientific & Engineering Researcher - E-CS
2021.07.21 [International Conference] 주저자 (First Author)

A Study of Designing Conversational AI Chatbot with Emotion Using NLP and NLU
The International Association of Scientific & Engineering Researcher - E-CS
2021.07.21 [International Conference] 주저자 (First Author)

A Study of Indoor Interactive Digital Signage with Context-Awareness Recommendation
The International Association of Scientific & Engineering Researcher - E-CS
2020.12.20 [International Conference] 제3저자 (Third Author)

A Contactless Personal Health Monitoring System Using Raspberry Pi Based Smart Mirror
The International Association of Scientific & Engineering Researcher - E-CS
2020.12.20 [International Conference] 주저자 (First Author)

A Study of Smart Contents Management System Using Machine Learning for Digital Signage
The International Association of Scientific & Engineering Researcher - E-CS
2020.12.20 [International Conference] 제2저자 (Second Author)

A Study of Proxy Based Information Crawling on Multi-Screen Digital Signage
The International Association of Scientific & Engineering Researcher - E-CS
2021.08.31 [Domestic Conference] 제4저자 (Fourth Author)

AI기반 대화형 챗봇 설계를 위한 오픈소스 플랫폼 사례 연구
A Case Study of Open Source Platform for AI-Based Conversational Chatbot Design
2021 한국정보통신설비학회 하계학술대회 (KIIFCE 2021 Summer Conference)
2021.08.31 [Domestic Conference] 제4저자 (Fourth Author)

Dialogflow 대화형 플랫폼을 활용한 헬스케어 챗봇 구조 연구
A Study of Healthcare Chatbot Architecture With Conversational Platform Dialogflow
2021 한국정보통신설비학회 하계학술대회 (KIIFCE 2021 Summer Conference)

Patents & Software

2022.10.12특허출원 (Patent Application)

선택적 모델 적용 기능을 구비한 멀티 모델 헬스케어 챗봇 시스템
Multi-model Healthcare Chatbot System with Selective Model Application Function
출원번호 (Application No.): 10-2022-0130207
2022.06.02특허등록 (Patent Registered)

이기종 디바이스 간 디지털 사이니지 듀얼 스크린 시스템
Digital Signage Dual Screen System between Heterogeneous Devices
출원번호 (Application No.): 10-2022-0067648 (2025년 등록 완료 / Registration completed in 2025)
2022.03.22특허출원 (Patent Application)

선택적 적시 챗봇 프로비저닝 기능을 구비한 계층적 다중 챗봇 디지털 헬스케어 시스템
Hierarchical Multi-Chatbot Digital Healthcare System with Selective Just-in-Time Chatbot Provisioning
출원번호 (Application No.): 10-2022-0035339
2022.12.06프로그램 등록 (Program Registration)

대형 디스플레이 디지털 사이니지용 웹기반 템플릿
Web-based Template for Large Display Digital Signage
등록번호 (Registration No.): C-2022-050983
2022.11.22프로그램 등록 (Program Registration)

건강상담사이트 질의응답 빅데이터 구성 소프트웨어
Software for Constructing Q&A Big Data from Health Counseling Websites
등록번호 (Registration No.): C-2022-047223
2022.11.22프로그램 등록 (Program Registration)

포탈사이트 건강 질의응답 빅데이터 구성 소프트웨어
Software for Constructing Health Q&A Big Data from Portal Sites
등록번호 (Registration No.): C-2022-047222
2022.01.03프로그램 등록 (Program Registration)

챗봇 Dialogflow 학습데이터 분석 및 작성 지원 도구
Support Tool for Analyzing and Creating Training Data for Chatbot Dialogflow
등록번호 (Registration No.): C-2022-000182
2022.01.03프로그램 등록 (Program Registration)

챗봇 Rasa 학습데이터 분석 및 작성 지원 도구
Support Tool for Analyzing and Creating Training Data for Chatbot Rasa
등록번호 (Registration No.): C-2022-000181