期刊名称：International Journal of Advanced Computer Research
印刷版ISSN：2249-7277
电子版ISSN：2277-7970
出版年度：2012
卷号：2
期号：7
页码：157-161
出版社：Association of Computer Communication Education for National Triumph (ACCENT)
摘要：From the advent of association rule mining, it has become one of the most researched areas of data exploration schemes. In recent years, implementing association rule mining methods in extracting rules from a continuous flow of voluminous data, known as Data Stream has generated immense interest due to its emerging applications such as network-traffic analysis, sensor-network data analysis. For such typical kinds of application domains, the facility to process such enormous amount of stream data in a single pass is critical. Nowadays, many organizations generate and utilize vast data streams (Huang, 2002). Employing data mining schemes on such massive data streams can unearth real-time trends and patterns which can be utilized for dynamic and timely decisions. Mining in such a high speed, enormous data streams significantly differs from traditional data mining in several ways. Firstly, the response time of the mining algorithm should be as small as possible due to the online nature of the data and limited resources dedicated to mining activities (Charikar, 2004). Second, the underlying data is highly volatile and subject to change over period of time (Chang, 2003). Moreover, since there is no time for preprocessing the data in order to remove noise, the streamed data can have noise inherent in it. Due to all aforementioned problems, data stream mining is receiving increasing attention and current research is now focused on the efficient resolution to the problem cited above. Although, the field of data stream mining is being heavily investigated, there is still a lack of a holistic and generic approach for mining association rules from data streams. Thus, this research attempts to fill this gap by integrating ideas from previous work in data stream mining. This investigation focuses on the degree of effectiveness of using a probabilistic approach of sampling in the data stream together with an incremental approach to maintenance of frequent itemsets in a data stream environment. The following paper describes the design and experimentation conducted with a novel association rule mining algorithm that can be deployed on a high speed data stream.
关键词：Data stream; Association Rule mining; Frequent Patten data stream.
