In general, the number of frequent itemsets in a data set is very large. In order to represent them in more compact notation, closed or maximal frequent itemsets (MFIs) are used. However, the characteristics of a data stream make such a task be more difficult. For this purpose, this paper proposes a method called estMax that can trace the set of MFIs over a data stream. The proposed method maintains the set of frequent itemsets by a prefix tree and extracts all of MFIs without any additional superset/subset checking mechanism. Upon processing a newly generated transaction, its longest matched frequent itemsets are marked in a prefix tree as candidates for MFIs. At the same time, if any subset of these newly marked itemsets has been already marked as a candidate MFI, it is cleared as well. By employing this additional step, it is possible to extract the set of MFIs at any moment. The performance of the proposed method is comparatively analyzed by a series of experiments to identify its various characteristics.

The online analytical processing (OLAP) is a process where an end user directly accesses multi-dimensional information to analyze information in a dialogue type for his or her decision making. It is limited to apply the method to the recent big data environment including smart phones and SNS. In this study, OLAP analysis-based clustering method is proposed by combining the data cube modeling technique with multi-dimensional clustering structure. The proposed method clusters the attribute values of big data streams so it is possible its precision may be slightly lower. But it reduces memory and time spent to analyze and process multi-dimensional data streams.

In a ubiquitous computing environment of sensor networks, monitoring the frequent contexts of a user occurred by those sensors is very important to provide a proactive service to the user. Such frequent contexts of the sensors can be recognized by associations between the set of sensor values and a set of actuator operations. This paper proposes a method of generating a new type of an association rule, called a context association rule, over an online sensor/actuator transactional data stream in order to invoke proper operations of actuators relevant to values of the sensors. To enumerate context association rules, a new prefix tree structure is introduced. It maintains all frequent context itemsets over the current data steam of sensor networks, so that a set of frequently co-occurred sensors and actuators items can be captured efficiently.

No abstract

Most of emerging applications deal with an infinite data stream in an incessant, immense and volatile manner. Consequently, it is very important to analyze not only the varying characteristics of a source data stream in a short-term period but also those in a longterm period. For this purpose, this paper demonstrates an OLAP system, DS-Cuber (Data Stream Cuber) for the analysis of data streams. The proposed system consists of two analytic components: short-term and long-term, so that it can provide an integrated analysis environment for infinite data streams. Furthermore, each of these two components supports diversified exception detection methods which can be used for the automatic identification of abnormality in the data elements of a data stream in order to guide the data cube navigation of a user effectively. Network traffic flow streams are used to demonstrate the features of the DS-Cube system.