Feedback driven improvement of data preparation pipelines

Towards Data Driven Model Improvement

In the area of student knowledge assessment, knowledge tracing is a model that has been used for over a decade to predict student knowledge and performance. Many modifications to this model have been proposed and evaluated, however, the modifications are often based on a combination of intuition and experience in the domain. This method of model improvement can be difficult for researchers with...

The RDF Pipeline Framework: Automating Distributed, Dependency-Driven Data Pipelines

Semantic web technology is well suited for large-scale information integration problems such as those in healthcare involving multiple diverse data sources and sinks, each with its own data format, vocabulary and information requirements. The resulting data production processes often require a number of steps that must be repeated when source data changes -often wastefully if only certain porti...

Feedback-Driven Concurrency Improvement and Refinement of Performance Models

Within the design stage of software engineering, the performance of a system should be evaluated with regards to its performance requirements. Models have to be built to be able to predict performance, because the performance of the system cannot yet be measured. To achieve a proper prediction accuracy of the model, it is gradually refined by adding details (introducing new model elements or sp...

Ontology-Driven Data Preparation for Association Mining

Ontologies can convey domain semantics to various phases of a KDD application through a mapping established between ontology entities and columns of the data matrix. The approach implemented in the Ferda tool focuses on providing support for the data preparation phase. Information about important data values and column groupings, once injected into a domain ontology, can be repeatedly used for ...

Using a data grid to automate data preparation pipelines required for regional-scale hydrologic modeling