Research‎ > ‎


MELOW: Machine learning framework for Embedded LOW-power system

posted Mar 31, 2017, 1:24 AM by 김병주

2016.12.01~2020.11.30 (4 years). This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIP) (No.2016R1E1A1A01942642).

This project aims for development of machine learning technology optimized for low-power embedded system. Currently researchers tend to use complex models to obtain highly accurate models via machine learning from big data. However, the training process consumes more power as the size of model increases, which becomes the main obstacle for utilizing machine learning in embedded system with limited power. Therefore, this project suggests developing machine learning technology minimizing power consumption during training while keeping accuracy of the model by exploiting various techniques such as model compression, utilizing various computing resources(flash memory, GPU), and ultimately the framework (MELOW: Machine laerning for Embedded LOW-power system)combining those methodologies. Furthermore, this project will show the practicability of MELOW by developing machine learning applications running on the MELOW framework.

Development of Enabling Software Technology for Big Data Mining

posted Jan 2, 2013, 9:35 AM by Hwanjo Yu   [ updated Jan 2, 2013, 9:38 AM ]

2012.07.01 - 2017.06.30 (5 years). This research was supported by Next-Generation Information Computing Development Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education, Science and Technology (No. 2012M3C4A7033344) 

The goal of this project is the development of enabling software technologies for big data mining. Through this project, we will research data mining techniques for big data in natural sciences and social networks. We will also develop personalized service technologies based on unstructured big data analysis and customer behavior models. Furthermore, we will produce well-trained software engineers who are experts in big data mining.

Developing Search and Mining Technologies for Mobile Devices (*우수 국과 과제 선정)

posted Jan 2, 2013, 9:28 AM by Hwanjo Yu   [ updated Mar 10, 2017, 1:05 PM ]

2011.05.01 - 2014.04.30 (3 years). This work was supported by the Brain Korea 21 Project in 2010 and Mid-career Researcher Program through NRF grant funded by the MEST (No. KRF-2011-0016029).

Combining the highly profitable information search industry and the mobile computing paradigm, mobile information search industry has been growing rapidly despite the global economy recession. Thus, development of mobile search technology will impact on the economy positively. This project aims at advancing the technologies in the areas of mobile search and mining, low-power consumption utility mining, and mining for mobile online advertizing.

User-Friendly Search Engine for PubMed

posted Jan 2, 2013, 9:27 AM by Hwanjo Yu   [ updated Jan 2, 2013, 9:27 AM ]

2009.05.01 - 2012.02.28 (3 years). This work was supported by the Brain Korea 21 Project and Mid-career Researcher Program through NRF grant funded by the MEST (No. KRF-2009-0080667).

PubMed MEDLINE, a database of biomedical and life science journal articles, is one of the most important information source for medical doctors and bio-researchers. Finding the right information from the MEDLINE is nontrivial because it is not easy to express the intended relevance using the current PubMed query interface, and its query processor focuses on fast matching rather than accurate relevance ranking. This project develop techniques for building a user-friendly MEDLINE search engine.

Novel Recommendation for Digital TV

posted Jan 2, 2013, 9:26 AM by Hwanjo Yu   [ updated Jan 2, 2013, 9:26 AM ]

2011.02.01 - 2012.01.31 (1 years). This work was supported by Samsung Electronics.

Existing recommendation systems (e.g., the Netflix competition) focus on an accurate prediction of purchase, as the systems are evaluated based on the prediction accuracy. However, such systems tend to recommend popular items. Recommending popular items, however, might not be effective or affective on users' purchase decisions, as users likely already know the items and likely have pre-made decisions on the purchase of items, e.g., recommend to watch Star Wars or Titanic. Effective recommendation must recommend unexpected or novel items that could surprise users and affect users' purchase decision. This project is to develop an effective recommendation for digital TV customers.

Feature Weighting for Ranking

posted Oct 11, 2011, 9:58 PM by Hwanjo Yu   [ updated Jan 2, 2013, 9:31 AM ]

This work is supported by MSRA (Microsoft Research Asia).
Feature weighting for ranking has not been researched as extensively as for classification.  This project develops various feature weighting methods for ranking by leveraging existing methods for classification. The developed methods are used on the Live Search query log data to identify key features that determine the users’ click-through behaviors. The developed methods will also be used to build the feature selection component of RefMed -- relevance feedback PubMed search engine.

Enabling Relevance Ranking in Databases for User-Friendly Data Retrieval

posted Aug 8, 2011, 4:26 AM by Hwanjo Yu   [ updated Aug 8, 2011, 4:33 AM ]

2008.07.01 - 2011.06.30 (3 years). This work is supported by the Brain Korea 21 Project and the Korea Research Foundation Grant funded by the Korean Government (KRF-2008-331-D00528).

    Most online data retrieval systems, built based on relational database management systems(RDBMS), support fast processing of Boolean queries but offer little support for relevance or preference ranking. A unified support of Boolean and ranking constraints in a query is essential for user-friendly data retrieval. This project develops foundational techniques that enable such data retrieval systems in which users intuitively express ranking constraints and the system efficiently process the queries.

    Development of Kernel Based Real-time Recommender System through Structured Web Data Analysis

    posted Sep 4, 2010, 5:51 PM by Hwanjo Yu   [ updated Aug 8, 2011, 4:34 AM ]

    With Prof. Jaewook Lee (PI), 2008.07.01 - 2010.06.31 (2 years). This work was supported by the Brain Korea 21 Project and the Korea Research Foundation Grant funded by the Korean Government (KRF-2008-314-D00483).

    Support Vector Machine with Limited Memory

    posted Sep 11, 2009, 9:43 PM by Taehoon Kim   [ updated Sep 4, 2010, 6:29 PM by Hwanjo Yu ]

    Support vector machines (SVMs) have been promising methods for classification, regression, ranking analysis due to their solid mathematical
    foundations, which include two desirable properties: margin maximization and nonlinear classification using kernels. However, despite these prominent properties, SVMs are usually not chosen for large-scale data mining problems because their training complexity is highly dependent on the data set size. Unlike traditional pattern recognition and machine learning, real-world data mining applications often involve a huge number of data records that does not fit in main memory and a multiple scans of the data set is often too expensive. Through this project, we developed techniques for approximately training SVMs in one scan of the database.

    Representative Publications
    • H Yu, J Yang, J Han & X Li, "Making SVMs Scalable to Large Data Sets using Hierarchical Cluster Indexing", Data Mining and Knowledge Discovery, Springer, 11(3): 295-321, 2005. (DAMI'05) (SCI)
    • H Yu, J Yang & J Han, "Classifying Large Data Sets Using SVM with Hierarchical Clusters", Proc. of ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, 2003. (KDD'03 full paper, 13% accepted, received student scholarship award)

    SCC: Single-Class Classification or Classification Without Negative Examples

    posted Sep 11, 2009, 9:37 PM by Taehoon Kim   [ updated Sep 4, 2010, 6:18 PM by Hwanjo Yu ]

    Single-Class Classification (SCC) seeks to distinguish one class of data from universal set of multiple classes. We call the target class positive and the complement set of samples negative. In SCC problems, it is assumed that a reasonable sample of the negative data is not available. Since it is not natural to collect the "non-interesting'' objects (i.e., negative data) to train the concept of the "interesting'' objects (i.e., positive data), SCC problems are prevalent in the real world where positive and unlabeled data are widely available but negative data are hard or expensive to acquire. We developed SCC algorithms which compute the boundary functions of the target class from positive and unlabeled data (without labeled negative data). The basic idea is to exploit the natural "gap'' between positive and negative data by incrementally labeling negative data from the unlabeled data using the margin maximization property of SVM. Our SCC algorithms build classification functions very close to the SVM with fully labeled data when the positive data is not much under-sampled.

    Representative Publications
    • H. Yu, "Single-Class Classification with Mapping Convergence", Machine Learning, Springer, 61:49-69, 2005. (ML'05)
    • H. Yu, J. Han & K. C.-C. Chang, "PEBL: Web Page Classification without Negative Examples", IEEE Transaction on Knowledge and Data Engineering, Special Issue on Mining and Searching the Web, IEEE Computer Society, 16(1): 70-81, 2004. (TKDE'04 Special Issue, 11% accepted) 
    • H. Yu, "SVMC: Single-Class Classification With Support Vector Machines", Proc. of Int. Joint Conf. on Artificial Intelligence, 2003. (IJCAI'03 full paper, 20% accepted, received student scholarship award)
    • H. Yu, J. Han & K. C.-C. Chang, "PEBL: Positive Example Based Learning for Web Page Classification Using SVM", Proc. of ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, 2002. (KDD'02 full paper, 14% accepted, received student scholarship award)

    1-10 of 12