DataMiner

Data Analytics

DataMiner addresses the computational and analytical needs of researchers, providing a platform for efficient data processing. It generates provenance information and enables collaboration through easy processes sharing.
 

Key Features

DataMiner offers a user-friendly cloud-based platform for executing a wide array of pre-configured and custom statistical algorithms in parallel, focusing on geospatial data analysis and trend detection, all accessible through a standard WPS REST API without requiring in-depth technical knowledge.
 

Advanced Pre-configured Algorithms

DataMiner offers a collection of state-of-the-art data mining and versatile algorithms, such as Clustering, Principal Component Analysis, and Artificial Neural Networks, ready for use as a service. It also enables the generation and analysis of data trends, including the examination of time series and the application of elementary signal processing to detect patterns in trends. 
 

Simplified Algorithm Integration

DataMiner allows for the straightforward integration and sharing of new algorithms, regardless of the programming language, without users needing to understand the complexities of cloud computing, Web Processing Service (WPS), or distributed computing systems.
 

Reproducibility

The automatically generated provenance information by DataMiner is pivotal, as it ensures the reproducibility and reuse of scientific processes, fostering transparency and reliability in research. All executions are archived and can be repeated, reused and shared at any time. 
 

Multi-tenant and distributed computing

DataMiner, a multi-tenant service, facilitates the concurrent execution of diverse statistical algorithms, utilizing a map-reduce framework to enable robust distributed computing, all seamlessly accessible through an intuitive WPS REST API.
 

How it works

Integrating your unique process is straightforward with the custom Software Importer interface, or you may opt for one of the pre-integrated processes. Simply set your parameters, choose your input datasets, and then leave the rest to DataMiner. It will run the process, compile all necessary information, create PROV-O provenance data, save the outcomes and logs in your personal workspace, and promptly inform you upon completion.
 

Success Stories

These success stories demonstrate the wide range of applications and the positive impact of D4Science in various scientific domains. The platform's core strengths, including data integration, cloud computing, customisable VREs, and collaborative tools, are highlighted as key factors contributing to the successful outcomes of these projects.