Nature Of Data Mining Research Articles

BISON: bio-interface for the semi-global analysis of network patterns

BackgroundThe large amount of genomics data that have accumulated over the past decade require extensive data mining. However, the global nature of data mining, which includes pattern mining, poses difficulties for users who want to study specific questions in a more local environment. This creates a need for techniques that allow a localized analysis of globally determined patterns.ResultsWe developed a tool that determines and evaluates global patterns based on protein property and network information, while providing all the benefits of a perspective that is targeted at biologist users with specific goals and interests. Our tool uses our own data mining techniques, integrated into current visualization and navigation techniques. The functionality of the tool is discussed in the context of the transcriptional network of regulation in the enteric bacterium Escherichia coli. Two biological questions were asked: (i) Which functional categories of proteins (identified by hidden Markov models) are regulated by a regulator with a specific domain? (ii) Which regulators are involved in the regulation of proteins that contain a common hidden Markov model? Using these examples, we explain the gene-centered and pattern-centered analysis that the tool permits.ConclusionIn summary, we have a tool that can be used for a wide variety of applications in biology, medicine, or agriculture. The pattern mining engine is global in the way that patterns are determined across the entire network. The tool still permits a localized analysis for users who want to analyze a subportion of the total network. We have named the tool BISON (Bio-Interface for the Semi-global analysis Of Network patterns).

Clustering Techniques in Data Mining—A Survey

In last few years there has been tremendous research interest in devising efficient data mining algorithms. Clustering is a very essential component of data mining techniques. Interestingly, the special nature of data mining makes the classical clustering algorithms unsuitable. These characteristics are usually very large datasets; the dataset need not be necessarily numeric and hence importance should be given to efficient I/O operations instead of algorithmic complexity. As a result in last few years a number of clustering algorithms are proposed for data mining. The present paper gives a brief overview of these algorithms. The first part of the paper discusses numerical clustering which are classified into partitioned clustering and hierarchical clustering. In the second part the paper discusses the clustering algorithms for categorical data.