University of Twente Student Theses

Login

A method for enhancing reproducibility of geoprocessing workflows using semantic annotation

Beyene, Dawit Haile (2021) A method for enhancing reproducibility of geoprocessing workflows using semantic annotation.

[img] PDF
2MB
Abstract:In their everyday lives, people deal with spatial problems, whether or not they are conscious. These spatial problems could be specific, simple, straightforward, or complicated, which necessitate integrating various processes and data. In recent years, web technologies and cloud computing advancements have highlighted the importance of chaining geospatial operations and data to solve complex problems. Well-orchestrated sequential methods encapsulated as workflows can be used to allow the integration of datasets to solve complex scientific problems. The reproducibility of scientific applications has become increasingly important to advance computational science because it enables the original developer and other users or scientists to replicate, validate, and further expand original methods. However, most workflows are not reproduced for a variety of reasons, which is referred to as workflow decay. There are several factors for the irreproducibility of geoprocessing workflows. Among these, one of the key problems for the irreproducibility of workflows is the lack of sufficient metadata description of the workflow. As a result, in this research, we developed a way to improve the semantics of workflows to enhance workflow reproducibility. This research's main goal is to develop a method to enhance the reproducibility of geoprocessing workflows using semantic annotation. The implementation of the proof of concept required the development and integration of several components. An ontology has been developed in the Living Textbook (LTB), which helps model and store the semantic description used to embed the workflows. This research also studied the added values of concept mapping tools like LTB for ontology development and semantic enrichment. To automate the semantic annotation process, the JavaScript application has been developed. The prototype application has been designed for two user groups – workflow creator and workflow consumer. This study has proven that, embedding additional semantic descriptions to the workflow increases the users understandability of a given workflow. This is attested by usability testing.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/89090
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page