초대의 말씀
  Program at a Glance
  초청강연
  기조연설
  협력워크샵
  공통워크샵
  분과워크샵
  특별세션I
  특별세션II
  튜토리얼
  논문발표
  조직 및 후원
  행사장소/교통/숙박
 
 
HOME > 행사안내 > Text Mining Enhanced by Heterogeneous ...
 
  Text Mining Enhanced by Heterogeneous ...

 

 

 

 

 

 

 

 

 

 

 김영훈 교수 (한양대학교 ERICA캠퍼스)
 
학력: 박사: 서울대학교 전기컴퓨터공학부(2013)

         학사: 서울대학교 컴퓨터공학부(2007)

 

 관심연구: 텍스트마이닝, MapReduce기반 병렬데이터마이닝, 문자열매칭

               쿼리프로세스 최적화

 

 대표논문: * Latent Ranking Analysis using Pairwise Comparisons, ICDM 2014

               * Efficient top-k algorithms for approximate substring matching.,

                 SIGMOD 2013

               * DIGTOBI: a recommendation system for Digg articles using

                 probabilistic modeling, WWW 2013

               * Parallel Top-K Similarity Join Algorithms Using MapReduce.

                 ICDE 2012

 

제목

Text Mining Enhanced by Heterogeneous Information Network

요약

Text mining analyzes unstructured text data to extract useful information. It includes various topics such as named entity recognition, sentiment analysis and fact extraction. Among them, this talk focuses on topic modeling. Due to the nature of unstructured text consisting of natural language words, deriving high quality information from text could be a very tricky problem. In this talk, I introduced my recent work to utilize associated heterogeneous types of data available in social network services for fine granularity topic modeling. The techniques to be presented tackle text mining problems from various applications using probability modeling techniques.

 

 
 
 
Untitled Document