About me

Dongwon Kim is a software engineer conducting data analysis and applying machine‑learning techniques, cloud computing techniques. He focused on context-aware systems providing indoor/outdoor localization with managing and optimizing the energy consumption for mobile devices into Linux kernel. Now, he is a part of a music recommendation team, while contributing to the recommender system by applying log-based/content-based machine learning techniques. Consequently, his activities include machine learning, cloud computing, and mobile sensing.

Interests	Cloud computing, Data engineering, Recommender system
Programming languages	Python, C, Unix shell scripts, Scala, Java
Technical keywords	Spark, MapReduce, Hive, Presto, Hbase, Airflow, Consul & Nomad, Kafka, RabbitMQ, Numpy, Faiss, Annoy

Industrial experience

Music recommender system

About data

Log data pipeline to extract item's features from log data
Content data pipeline to extract audio's features from audio with various models automatically

About items

Song and artist recommendation which you may also like based on word2vec and topic model (item to item)
Song recommendation from natural language based on word embeddings
Similarity search between millions of items while maintaining up-to-date information
Musical instrument classification with audio content data and convolutional neural networks

About users

Personalized recommendation with collaborative filtering
Artist suggestion which would be preferred to cold users

Playlist generation

Discover Songs: providing songs for each user which is not listened yet applying a mixture of collaborative filtering and content-based filtering method

Realtime music recommender system from songs, artists, or natural language

Mobile sensing system

Indoor and outdoor localization systems with WiFi fingerprinting
Action recognition with only accelerometers on smartwatch
Long-lived fitness tracking systems on smartwatch

Academic experience

Energy management and mobile sensing for mobile devices

Content-centric display energy management by modifying Linux kernel
Differentiation between active and passive WiFi scanning for efficient WiFi fingerprinting
Wake Lock management for Android devices

Power analysis and kernel activity monitoring for mobile devices

AppScope: Application Energy Metering Framework for Android Smartphone using Kernel Activity Monitoring
DevScope: A Nonintrusive and Online Power Analysis Tool for Smartphone

Education

M.Sc. in Computer Science, Yonsei University, Korea, August 2015 (Advisor: Prof. Hojung Cha)
B.S. in Computer Science, Yonsei University, Korea, February 2011

Publications

B. Jeon, A. Kim, C. Kim, D. Kim, J. Park, and J. Ha, “Music Emotion Recognition via End-to-End Multimodal Neural Networks”, in Proceeding of the 9th Conference on Recommender Systems (RecSys), ACM, Aug, 2017.
J. Ha, A. Kim, D. Kim, C. Kim, and J. Park, “Music Highlight Extraction via Convolutional Recurrent Attention Networks”, Machine Learning for Music Discovery Workshop in International Conference on Machine Learning (ICML), Aug, 2017.
J. Ha, A. Kim, D. Kim, J. Kim, J. Kim, J. Park, B. Ryu, “Predicting high-risk prognosis from diagnostic histories of adult disease patients via deep recurrent neural networks”, 2017 IEEE International Conference on Big Data and Smart Computing (BigComp), IEEE, Jeju, South Korea, Feb, 2017.
D. Kim, Y. Chon, W. Jung, Y. Kim, H. Cha, “Accurate Prediction of Available Battery Time for Mobile Applications”, ACM Transactions on Embedded Computing Systems (TECS), DOI: 10.1145/2875423
T. Choi, Y. Chon, Y. Kim, D. Kim, H. Cha, “Enhancing WiFi-Fingerprinting Accuracy Using RSS Calibration in Dual-Band Environments”, Pervasive and Mobile Computing (PMC), DOI: 10.1016/j.pmcj.2015.10.017
D. Kim, N. Jung, Y. Chon, H. Cha, “Content-Centric Energy Management of Mobile Displays”, IEEE Transactions on Mobile Computing (TMC), DOI: 10.1109/TMC.2015.2467393
S. Park, D. Kim, H. Cha, “Reducing Energy Consumption of Alarm-induced Wake-ups on Android Smartphones,” ACM The 16th International Workshop on Mobile Computing Systems and Applications (HotMobile 2015), ACM, Santa Fe, USA, Feb, 2015.
W. Jung, Y. Chon, D. Kim, H. Cha, “Powerlet: An Active Battery Interface for Smartphones,” in Proceeding of 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2014), ACM, Seattle, USA, Sep, 2014.
Y. Chon, S. Kim, S. Lee, D. Kim, Y. Kim, H. Cha, “Sensing WiFi Packets in the Air: Practicality and Implications in Urban Mobility Monitoring,” in Proceeding of 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2014), ACM, Seattle, USA, Sep, 2014.
D. Kim, N. Jung, H. Cha, “Contents-Centric Display Energy Management for Mobile Devices”, in Proceeding of 2014 ACM/IEEE Design Automation Conference (DAC 2014), San Francisco, USA, June, 2014.
N.D. Lane, Y. Chon, L. Zhou, Y. Zhang, F. Li, D. Kim, G. Ding, F. Zhao, H. Cha, “Piggyback CrowdSensing (PCS): Energy Efficient Crowdsourcing of Mobile Sensor Data by Exploiting Smartphones App Opportunities.” in Proceeding of 11th ACM Conference on Embedded Networked Sensor Systems (SenSys’13), ACM, Rome, Italy, Nov 2013.
D. Kim, W. Jung, H. Cha, “Runtime Power Estimation of Mobile AMOLED Displays”, in Proceeding of ACM Conference on 2013 Design, Automation and Test in Europe (DATE), ACM, Grenoble, France, 2013.
W. Jung, C. Kang, C. Yoon, D. Kim. H. Cha, “DevScope: A Nonintrusive and Online Power Analysis Tool for Smartphone Hardware Components”, in Proceeding of ACM International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS), ACM, Tampere, Finland, 2012.
C. Yoon, D. Kim, W. Jung, C. Kang, H. Cha, “AppScope: Application Energy Metering Framework for Android Smartphone using Kernel Activity Monitoring”, in Proceeding of 2012 USENIX Annual Technical Conference (USENIX ATC), Usenix, Boston, MA, USA, 2012.
W. Jung, D. Kim, H. Cha, “Observing Thermal Characteristics of Energy-Aware Mobile Devices”, in Proceeding of IEEE 18th International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA), IEEE, Seoul, Korea, 2012.