Integration of Data Mining into Scientific Data Analysis Processes

Abstract

In recent years, using advanced semi-interactive data analysis algorithms such as those from the field of data mining gained more and more importance in life science in general and in particular in bioinformatics, genetics, medicine and biodiversity. Today, there is a trend away from collecting and evaluating data in the context of a specific problem or study only towards extensively collecting data from different sources in repositories which is potentially useful for subsequent analysis, e.g. in the Gene Expression Omnibus (GEO) repository of high throughput gene expression data. At the time the data are collected, it is analysed in a specific context which influences the experimental design. However, the type of analyses that the data will be used for after they have been deposited is not known. Content and data format are focused only to the first experiment, but not to the future re-use. Thus, complex process chains are needed for the analysis of the data. Such process chains need to be supported by the environments that are used to setup analysis solutions. Building specialized software for each individual problem is not a solution, as this effort can only be carried out for huge projects running for several years. Hence, data mining functionality was developed to toolkits, which provide data mining functionality in form of a collection of different components. Depending on the different research questions of the users, the solutions consist of distinct compositions of these components.<br /> Today, existing solutions for data mining processes comprise different components that represent different steps in the analysis process. There exist graphical or script-based toolkits for combining such components. The data mining tools, which can serve as components in analysis processes, are based on single computer environments, local data sources and single users. However, analysis scenarios in medical- and bioinformatics have to deal with multi computer environments, distributed data sources and multiple users that have to cooperate. Users need support for integrating data mining into analysis processes in the context of such scenarios, which lacks today. Typically, analysts working with single computer environments face the problem of large data volumes since tools do not address scalability and access to distributed data sources. Distributed environments such as grid environments provide scalability and access to distributed data sources, but the integration of existing components into such environments is complex. In addition, new components often cannot be directly developed in distributed environments. Moreover, in scenarios involving multiple computers, multiple distributed data sources and multiple users, the reuse of components, scripts and analysis processes becomes more important as more steps and configuration are necessary and thus much bigger efforts are needed to develop and set-up a solution. <br /> In this thesis we will introduce an approach for supporting interactive and distributed data mining for multiple users based on infrastructure principles that allow building on data mining components and processes that are already available instead of designing of a completely new infrastructure, so that users can keep working with their well-known tools. <br /> In order to achieve the integration of data mining into scientific data analysis processes, this thesis proposes an stepwise approach of supporting the user in the development of analysis solutions that include data mining. <br /> We see our major contributions as the following: first, we propose an approach to integrate data mining components being developed for a single processor environment into grid environments. By this, we support users in reusing standard data mining components with small effort. The approach is based on a metadata schema definition which is used to grid-enable existing data mining components. Second, we describe an approach for interactively developing data mining scripts in grid environments. The approach efficiently supports users when it is necessary to enhance available components, to develop new data mining components, and to compose these components. Third, building on that, an approach for facilitating the reuse of existing data mining processes based on process patterns is presented. It supports users in scenarios that cover different steps of the data mining process including several components or scripts. The data mining process patterns support the description of data mining processes at different levels of abstraction between the CRISP model as most general and executable workflows as most concrete representation.