Search

Clustering high-dimensional data stream is a difficult and important problem. Grid-based algorithms are easily influenced by the size and borders of the grid. To overcome the weakness, we propose a new Irregualr Grid-based Clustering algorithm for high-dimensional data streams, called IGDCL. This method incorporates an irregular grid structure and subspace clustering algorithm. In this paper, an irregular grid structure is generated by means of splitting each dimension into different grid cells. With new data arriving, the irregular grid structure is dynamically adjusted. We assign a fading density function for each data point to embody the evolution of data streams. The final clusters are obtained in subspaces which are formed by dimensions associated with corresponding clusters. Experimental results demonstrate that IGDCL has higher clustering quality than CluStream.

Frequent itemsets mining is a crucial problem in the field of data mining. Although many related studies have been suggested, these algorithms may suffer from high computation cost and spatial complexity in dense database, especially when mining long frequent itemsets or support threshold is very lower. To address this problem, a new data structure called P Array is proposed. P Array makes use of data horizontally and vertically like Bit Table FI, and those itemsets that co_occurence with single frequent items are found by computing intersection in P Array. Then, a new algorithm, call MFIPA, is proposed based on P Array. Some frequent itemsets which have the same supports as single frequent item can be found firstly by connecting the single frequent item with every nonempty subsets of its projection, then all other frequent itemsets can be found by using depth-first search strategy. The experimental results show that the proposed algorithm is superior to Bit Table FI in execution efficiency and memory requirement, especially for dense database.

Clustering high-dimensional data stream is a difficult and important issue. In this paper, we propose MStream, a new clustering algorithm based on matrix over high dimensional data stream. MStream algorithm incorporates a synopsis structure, called GC (Grid Cell Structure), and grid matrix technique. The algorithm adopts the two-phased framework. In the online component, the GC is employed to monitor one-dimensional statistics data distribution of each dimension independently. Sparse GCs which need to be deleted are checked by predefined threshold. In the offline component, it is possible to tracing multi-dimensional clusters by dense GCs which are maintained in the online component. Grid matrix technique is introduced to generate the final multi-dimensional clusters in the whole data space. Experimental results show that our algorithm has the flexible scalability and higher clustering quality.