Chieh-Jen Wang (王界人)

Ph.D. in Computer Science and Information Engineering,
National Taiwan University

• Ph.D. Dissertation: Enhancing Effectiveness of Internet Search and Advertising with Web Log Mining
• Advisor: Distinguished Prof. Hsin-Hsi Chen (陳信希)

Industrial Technology Research Institute (ITRI)2013/07-present

• Technical Manager/Senior Engineer
• • Current Position: Technical Manager at the Industrial Technology Research Institute, with over 15 years of experience in AI core technology development
  • Technical Expertise: Specialized in Artificial Intelligence (AI), focusing on Large Language Models (LLM) and Computer Vision (CV) technologies
  • R&D Achievements: Secured over NT$35 million in research funding and obtained 4 patents with 10 additional patent applications
  • Industry Applications: Led digital transformation efforts in smart manufacturing and retail, assisting in securing over NT$50 million in government grants
  • Project Experience: Led 11 cross-functional projects with extensive experience in research and project management

National Taiwan University2010/02-2013/06

• Teaching Assistant and Web Administrator

Academia Sinica2006/10-2007/09

• Research Assistant

Kingston Technology Far East Corp2002/07-2002/08

• Summer Intern

Large Language Models (LLM)

• Retrieval-Augmented Generation (RAG)
• Natural Language Generation (NLG)
• Natural Language Processing (NLP)
• Summarization

Deep Computer Vision (DCV)

• Anomaly Detection
• Object Detection and Segmentation
• Image Synthesis
• 3D Reconstruction

Intelligent Manufacturing (IM)

• Prognostics and Health Management (PHM)
• Root Cause Analysis and Tracking
• Operation Instruction Optimization
• Demand Forecasting in Supply Chain

Smart Retail (SR)

• Customer Behavior Analysis
• Personalized Recommendations
• Retail Media Network
• Sales Forecasting

Deep Learning Frameworks

• PyTorch, TensorFlow, Keras

Software Development

• DevOps: GitLab, Jenkins
• MLOps: Hugging Face, MLflow
• Project Management: Jira
• Software Quality: SonarQube

Programming Languages

• Python, Java, C++, SQL

Intelligent Manufacturing (IM)

• Industrial Knowledge-Based Large Language Model (LLM)
• • Extracted key knowledge from structured and unstructured data
  • Identified causal relationships to build a knowledge graph in the semiconductor domain
  • Developed and implemented a knowledge discovery model based on Retrieval-Augmented Generation (RAG)
  • Achieved a knowledge discovery precision of over 93.6% and a response accuracy exceeding 98.7%, representing a 13% improvement compared to the original KMS
  • Use case: Winbond KMS system


• Knowledge Graph-Based Maintenance Manual Optimization
• • Optimized original operating instructions to be semantically complete and reduce cognitive load, improving comprehension for maintenance engineers.
  • Integrated with mobile or wearable devices, providing visual guidance to ensure accurate assembly and repair operations
  • Expert evaluations showed an 87.84% improvement in the comprehension of the optimized instructions
  • Use case: Successfully implemented in Marketech semiconductor equipment


• Prognostics and Health Management (PHM)
• • Collected and analyzed IoT sensor data from machinery and equipment
  • Developed multi-level ensemble learning techniques, combining advanced machine learning algorithms with domain knowledge
  • Predicted semiconductor equipment/component failures and remaining useful life, with up to 48 hours advance notice
  • Achieved accuracy higher than 95% with a false alarm rate below 1%
  • Use case: Winbond PHM system (https://tinyurl.com/ye994bgj)

Smart Retail and Smart Logistics

• User Behavior-Based Personalized Ad Generation Using Gen-AI
• • Achieved 95% accuracy in facial recognition for Asian demographics
  • Utilized Gen-AI to analyze customer interactions and produce personalized ads
  • Optimized and trained localized Traditional Chinese marketing copy with Breeze-7B
  • Developed marketing content generation models using LLM and Stable Diffusion
  • Use case: Partnered with Hi-Life, attracting 210k visitors in a month and increasing revenue by NT$500k


• Gen-AI Based Process Automation Robot (RPA)
• • Utilized generative AI to automatically generate various forms, improving table detection and structure recognition accuracy to 90%.
  • Developed AI handwriting OCR technology based on deep transfer learning, achieving 92.38% recognition accuracy
  • Integrated the handwriting recognition model with LINE BOT to reduce the operation threshold
  • Automated inventory and procurement processes by integrating with ERP systems
  • Use Case: Successfully implemented in Carrefour’s inventory system, reducing backend labor by 60% and significantly shortening processing time


• Large-Scale Cloud E-Commerce Search Engine
• • Analyzed query semantics and user search behavior to understand intent, delivering precise product search results with 98.7% precision
  • Built a scalable, cloud-based parallel system for a real-time, high-efficiency, stable, and expandable search engine environment
  • Enhanced search engine performance to handle high traffic during events like Double 11, supporting over 5,000 QPS (11 times the previous system)
  • Helped businesses increase revenue by 5-10%, with single-day sales reaching NT$21 billion during promotions, an 80% growth over previous years
  • Use case:Successfully implemented for Taiwan's leading e-commerce company MOMO

Interactive Recommendation System and Method

• Taiwan (2019/01/11：TWI647638B)
• United States (2021/01/05：US10885568)

Drug-Screening System and Drug-Screening Method

• Taiwan (2021/09/21：TWI740415B)
• United States (2023/05/30：US11664094)
• China (2021/06/29：CN113053470A)

Graph-Based Natural Language Optimization System and Method

• Taiwan (2023/05/01：TW202318251A)
• United States (2024/07/16：US12039268)
• China (2023/05/05：CN116069940A)

Self Propelled Vehicle for Following Target and Method Thereof

• Taiwan (2024/06/16：TW202424675A)
• China (2024/06/14：CN118192533A)

Applications of AI-driven Innovation in Smart Retail

• National Yang Ming Chiao Tung University, 2024

Applications of AI-driven Innovation in Communication Technology

• National Yang Ming Chiao Tung University, 2024

Applications of Digital Image Processing Driving Industry Innovation

• National Taiwan University of Science and Technology, 2024

Generative AI-Driven Industry Innovation

• Examination Yuan, 2024

Natural Language Processing and Its Applications

• Ministry of Science and Technology, 2017

Big Data Trends and Analytics

• Chung Cheng Institute of Technology, 2017
• Kaohsiung First University of Science and Technology, 2015
• Hua Fan University, 2013
• Yuan Ze University, 2013

Big Data for Internet Advertising

• Shih Chien University, 2013

• Chieh-Jen Wang and Hsin-Hsi Chen. “Intent Mining in Search Query Logs for Automatic Search Script Generation.” Knowledge and Information Systems, Volume 39, Issue 3, pp. 513-542, 2014.
• Chieh-Jen Wang, Yung-Wei Lin, Ming-Feng Tsai and Hsin-Hsi Chen. “Mining Subtopics from Different Aspects for Diversifying Search Results.” Information Retrieval, Volume 16, Issue 4, pp. 452-483, 2013.
• Chieh-Jen Wang, and Hsin-Hsi Chen. “Intent Shift Detection Using Search Query Logs.” International Journal of Computational Linguistics and Chinese Language Processing, Volume 3, Number 3-4, pp. 61-76, 2011.
• Cheng-Lung Huang, Mu-Chen Chen and Chieh-Jen Wang. “Credit Card Scoring with a Data Mining Approach Based on Support Vector Machine.” Expert Systems with Applications, Volume 34, Issue 3, pp. 847-856, 2007.
• Cheng-Lung Huang and Chieh-Jen Wang. “A GA-Based Attribute Selection and Parameter Optimization for Support Vector Machine.” Expert Systems with Applications, Volume 31, Issue 2, pp.231-240, 2006.
• 黃承龍，陳穆臻，王界人，「支援向量機於信用評等之應用」，計量管理期刊, Volume 1, Issue 2, pp.155-172, 2005.

• Chieh-Jen Wang,Yung-Ping Tien, Yun-Wei Hung (2022). "Language Model Based Chinese Handwriting Address Recognition." In Proceedings of the 34th annual Conference on Computational Linguistics and Speech Processing (ROCLING).
• 王界人、沈民新 (2015). "運用關聯分析探勘民眾關注議題與發展方向: 以環保議題為例." In Proceedings of the 27th annual Conference on Computational Linguistics and Speech Processing (ROCLING).
• Yang-Yin Lee, Chih-Chieh Shao, Yen-Pin Chiu, Yong-Siang Shih, Hsin-Hsi Chen, Chieh-Jen Wang, and Sen-Chia Chang (2014). "Question Type Analysis for Question-Answering Applications in Education." In Proceedings of the 22nd International Conference on Computers in Education (ICCE).
• Chieh-Jen Wang and Hsin-Hsi Chen (2012). “Learning to Predict the Cost-Per-Click for Your Ad Words.” In Proceedings of the 21st ACM International Conference on Information and Knowledge Management (CIKM), pp. 2291-2294, Maui, Hawaii. (Top-tier Conference)
• Chieh-Jen Wang, Hung-Sheng Huang and Hsin-Hsi Chen (2012). “Automatic Construction of an Evaluation Dataset from Wisdom of the Crowds for Information Retrieval Applications.” In Proceedings of IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp. 490-495, Seoul, Korea.
• Chieh-Jen Wang, Shuk-Man Cheng, Lung-Hao Lee, Hsin-Hsi Chen,Wen-shen Liu, Pei-Wen Huang and Shih-Peng Lin (2012). “NTUSocialRec: An Evaluation Dataset Constructed from Microblogs for Recommendation Applications in Social Networks.” In Proceedings of the 8th International Conference on Language Resources and Evaluation (LREC), pp.2328-2333, Istanbul, Turkey.
• Chieh-Jen Wang, Yung-Wei Lin, Ming-Feng Tsai and Hsin-Hsi Chen (2011). “NTU Approaches to Subtopic Mining and Document Ranking at NTCIR-9 Intent Task.” In the 9th NTCIR Workshop Meeting, pp.129-134, Tokyo, Japan.
• Chieh-Jen Wang and Hsin-Hsi Chen (2011). “Learning User Behaviors for Advertisements Click Prediction.” In the 34rd international ACM SIGIR conference on Research and development in information retrieval Workshop on Internet Advertising (IA2011), Beijing, China.
• Chieh-Jen Wang and Hsin-Hsi Chen (2011). “Search Scripts Mining from Wisdom of the Crowds.” In Proceedings of IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp.878-883, Anchorage, Alaska.
• Kevin Hsin-Yih Lin, Chieh-Jen Wang and Hsin-Hsi Chen (2011). “Predicting Next Search Actions with Search Engine Query Logs.” In Proceedings of 2011 IEEE/WIC/ACM International Conference on Web Intelligence (WI), pp.227-234, Lyon, France.
• Chieh-Jen Wang, Kevin Hsin-Yih Lin and Hsin-Hsi Chen (2010). “Intent Boundary Detection in Search Query Logs.” In Proceedings of the 33rd international ACM SIGIR conference on Research and development in information retrieval (SIGIR), pp. 749-750, Geneva, Switzerland.
• Shih-Wei Lin, Shih-Chieh Chen, Chieh-Jen Wang and Kung-Ying Hwang (2005). “Data classification techniques for Credit Cards Promotion.” In Proceedings of the 12th ISPE International Conference on Concurrent Engineering: Research and Applications, pp. 195-200, Dallas, USA.
• 黃承龍，楊俊傑，應哲磊，王界人，「使用支援向量機預測台灣期貨指數」，2005年第十一屆資訊管理暨實務研討會，實踐大學，2005.
• 林詩偉，陳士杰，王界人，「應用資料探勘技術提昇信用卡促銷效益之研究」，第一屆創新與管理學術研討會，實踐大學，2004.
• 黃承龍，王界人，「應用支援向量機於信用卡資料之分類」，2004年第一屆中華國際機率統計與計量管理學術研討會，國立台北科技大學，2004.
• 黃承龍，王界人，「遺傳演算法應用於支援向量機之參數調整與屬性篩選」，第九屆人工智慧與應用研討會，國立政治大學，2004.
• 黃承龍，曾綜源，楊雯珺，王界人，「支援向量機於糖尿病診斷之應用」，中國工業工程學會九十三年度年會暨學術研討會，國立成功大學，2004.
• 曾綜源，黃承龍，王界人，「優勢關係評估法應用於企業跨國投資之研究」，第一屆台灣作業研究學會學術研討會暨2004年科技與管理學術研討會，國立台北科技大學，2004.
• 黃承龍，王界人，「遺傳演算法應用於支援向量迴歸之參數調整與屬性篩選」，第十屆資訊管理暨實務研討會，國立台中技術學院，2004.