Abstract
Data mining is an important method for understanding various phenomena and extracting useful knowledge from data.The use of data mining has become common in companies or research institutes.Recently,huge amount of data called Big Data has been gaining much attention for generating new business opportunities.Big Data Mining,that is to say,applying data mining techniques to big data also have been popular.However,because big data mining poses many difficulties,only simple analysis like aggregation has been performed.One reason is that the amount of data is so large that traditional data mining method is inapplicable.Another reason is that there exists not only persistent data like customer information,but also dynamically generated data likesensor data or stock price data.For dynamic data,data mining need to be performed in real-time,but traditional data mining methods achieve it only partially.Recently,methods such as on-line learning and data compression using Locality Sensitive Hashing(LSH) have been proposed for dealing with big data and used in practical applications.In this paper,I first introduce data mining.Next I present the definition of Big Data and difficulties in analyzing Big Data.Then we introduce on-line learning and data compression using LSH.
