Skip to content

Latest commit

 

History

History
78 lines (66 loc) · 6.41 KB

index.adoc

File metadata and controls

78 lines (66 loc) · 6.41 KB
Raw

Chiwan Park

Profile image of Chiwan Park

Chiwan Park

Research Engineer @ Kakao Corporation

Biography

I’m a research engineer at the ART (Advanced Recommendation Technology) team of Kakao Corporation, where I’m building recommender systems and machine learning applications for various mobile and web services of Kakao. My research interests include interpretable learning for recommender systems, scalable representation learning for user modeling, and large-scale graph processing systems. I received my M.Sc. in Computer Science and Engineering at Seoul National University, where I worked on large-scale graph processing using distributed systems under the guidance of Prof. U Kang. Here is a full Curriculum Vitae.

Just for fun, I have developed data-related products. SolveSQL, a web-based SQL learning platform for data analysts, is publicly available. I have contributed to open-source data processing tools such as Apache Flink, Apache Sqoop, and Apache MRQL. You can see my hobby development activities on Github.

News

Education

Seoul National University (Mar. 2016 - Feb. 2018)

M.Sc. in Computer Science and Engineering Thesis: Pre-partitioned Matrix-Vector Multiplication for Scalable Graph Mining Advisor: Prof. U Kang

Yonsei University (Mar. 2010 - Feb. 2016)

B.Sc. in Earth System Sciences B.Eng. in Computer Science and Engineering (double major)

Publications

Simple and Efficient Recommendation Strategy for Warm/Cold Sessions for RecSys Challenge 2022

Hyunsung Lee, Sungwook Yoo, Andrew Yang, Wonjun Jang, and Chiwan Park RecSys Challenge Workshop at ACM RecSys 2022 [paper | github]

FlexGraph: Flexible partitioning and storage for scalable graph mining

Chiwan Park, Ha-Myung Park, and U Kang PLoS ONE 15(1): e0227032 [paper | github]

PegasusN: A Scalable and Versatile Graph Mining System

Ha-Myung Park, Chiwan Park, and U Kang AAAI 2018 (demo paper) [paper | homepage]

A Distributed Vertex Rearrangement Algorithm for Compressing and Mining Big Graphs

Namyong Park, Chiwan Park, and U Kang Journal of KIISE (Vol. 43, 2016, domestic) [paper | homepage]

Experience

Kakao R&D Center (Apr. 2018 - May. 2021)
Kakao Advanced Recommendation Technology Team (May. 2021 - Now)

  • Recommender Systems for Audio-only Social Network Service

    • mm is an audio-only social network service. Users can talk with each other by joining a chat room. They also can follow others and invite them to talk privately. I develop personalized recommender systems on whom to follow and which audio chat room to join based on graph-based recommendation models. Several methods, such as graph pruning and regularization, are applied to the models to improve recommendation quality.

  • Product Recommender Systems for Gift

    • Kakaotalk Gift is an e-Commerce platform to send gifts through Kakaotalk, a well-known messaging platform in Korea. Choosing gifts depends on multiple contexts like a relationship between sender and receiver, demographical information, and product popularity. To consider the contexts, I am developing a recommender system based on neural networks, graph embedding, and collaborative filtering.

  • Recommender Systems for Digital Comic Service

    • Kakao Webtoon and Piccoma are digital comic platforms served in Korea and Japan, respectively. I mainly work on combining content-based representation learning and collaborative filtering for cold-start items. Our recommender systems are treated as a user-targeted marketing tool for new comics and optimize the first conversion rate.

  • Large-scale Machine Learning Applications for e-Commerce Platform

    • shoppinghow is an e-Commerce platform like eBay and Amazon. Users can search for the products and compare their prices. I develop and maintain several machine learning applications for product categorization and product matching based on neural networks and graph algorithms. Various parallel computing techniques are applied to the models to handle billion-scale data.

Contacts