A Slovenian public research organization has developed a crowd-sourced workflows platform on the cloud. It is an open-source data processing workflow platform with a user-friendly graphical interface that can run in any browser and does not require installation on client computers. The platform offers software components supporting data analytics and pattern discovery and workflow components allowing graphical user interaction during runtime and visualization of results by implementing views in any format and rendered in a web browser. Data processing in the platform is managed by connecting processing components into a workflow executable on the cloud, and the graphical user interface for constructing workflows follows a visual programming paradigm that simplifies the representation of complex procedures into spatial arrangements of building blocks.
Similar and related platforms exist, but no workflow construction tool would match the ease of use and workflow sharing abilities of the platform. Currently, probably the most advanced workflow management system (used primarily for bioinformatics research) is conceived as a suite of tools used to design and execute scientific workflows. A multilingual Internet service platform for supporting Intercultural collaboration is based on a service-oriented architecture and supports a web-oriented version of the pipeline architecture typically employed by natural language processing tools. Yet another platform, a more recent development, enables workflows to have interactive components, where the execution of the workflow pauses to receive input from the user.
The basic unit of the platform proposed is a processing component, which is graphically represented as a widget. Considering its inputs and parameters every such component performs a task and stores the results as outputs. Different processing components are linked via connections through which data is transferred from a widget's output to another's input. Alternative widget inputs are parameters, which the user enters into a widget's text fields.
The platform is easy-to-use, requires no expertise to construct new workflows while allowing complex data mining analyses to be performed on the input data set.
The researchers come from a Slovenian public research organization. Their technologies have been successfully applied to many practical problems, including earthquake prediction, selection of applicants for loans of the National Housing Fund, analysis of UK traffic accidents, medical diagnosis, analysis of the Slovenian public healthcare system, scientific digital editions of Slovene literature.
The partners are sought among data mining practitioners and developers for:
• Technical cooperation agreements in the implementation of existing workflows.
• Research cooperation: development of new workflows.
