Papapesios, Nikolaos; (2023) The Importance of Provenance in Increasing Usefulness and Usability in Information Derived from Geospatial Sources. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Papapesios_10163519_Thesis.pdf Download (8MB) | Preview

Abstract

Information derived from geospatial sources is used in decision-making in various sectors, such as in defence (Franklin et al., 2013; Roy et al., 2017), government (central and local) (Harding, 2006; Scott & Rajabifard, 2017; Sutanta et al., 2016), and non-government organisations (NGOs) (Crooks & Wise, 2013; Quill, 2018). However, decision-makers, when interacting with digital interfaces, do not always have an easy way to decide whether to make use of the information provided to support their decisions (and if so, how much they can rely on it). A factor that may influence decision-making is the well-documented provenance of the information (Ma et al., 2014). Provenance can answer questions such as who created the information, when it was created, why it was created, when it was updated, and who owns it. Thus, understanding whether presenting the provenance alongside the delivered information, in a useful and usable way can assist decision-makers in making better use of the derived information (i.e., be able to rely to the information more during their decision-making processes) is important. This PhD thesis investigates how provenance can be integrated by evaluating decision-makers perceptions in three relevant fields. Firstly, a human-computer interaction (HCI) methodology was implemented to understand how decision-makers currently interact with analytical products derived from geospatial data and how provenance can potentially be presented as part of this interaction, thus making them to rely more on the data. Interviews and questionnaires were conducted to further understand the decision-makers’ perspectives on this challenge and their actual needs. Secondly, a prototype was developed based on decision-makers needs to examine through usability testing if, with provenance, their decisions would be enriched. The prototype includes preferred visualisation techniques presented in a usable way that can increase decision-makers’ confidence in the given geospatial data. Finally, the prototype was evaluated, increasing the understanding of decision-makers’ needs. The findings of the semi-structured interviews identified four important provenance elements - date produced, currency (how up to date), data source, name of the organisation that produced the data - and highlight that the participants did not want to receive extensive provenance information. The questionnaires also confirmed these results. In addition, completeness was added as another valuable element for decision-makers. Both studies also demonstrated that provenance elements should be presented simply and understandably, such as a combination of table formats with visual effects. The selected methodological framework demonstrates how to approach end-users with relatively low-technical data expertise; understand their environment, duties, and needs; and design usable interfaces that can facilitate their decisions and hence to rely more in the provided geospatial data. HCI methods to address these critical gaps between producer and user sides should be noticeably referred to, since they provide the proper support to collect additional information and design further user requirements. Finally, although this holistic approach is implemented in three sectors where provenance is deemed significant, it also provides the required knowledge to apply it in more sectors when those gaps are identified.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item