A Fast Method for Mining Frequent Patterns from Big Data in Distributed Computing Environments

• Main Authors: CHEN, YUNG-LIN, 陳永霖
• Other Authors: CHANG, WENG-LONG
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/uasdw3

碩士 === 國立高雄應用科技大學 === 資訊工程系 === 104 === Frequent pattern mining is important field in data mining researches, through this technique can mine the hidden useful information from transaction databases. For example, a company can apply data mining to the discovery of association rule from transaction databases to greatly increase the profits of enterprises by understanding the customer behavior. Unfortunately, as the volume of database gets larger day by day, most of the frequent pattern mining algorithms in literature become ineffective due to too huge low computing performance, out of memory or too much communication. In this thesis, we propose a distributed method that is able to mine the frequent patterns for big data under limited memory. Through various simulation conditions, our proposed method is shown to deliver excellent performance in terms of execution efficiency and scalability.