Discovering Critical Factors Affecting RDF Stores Success
197
conducted by Cudré-Mauroux et al. in [7]; they studied the applicability of NoSQL
databases as backend of RDf store, stating that they represent a valid choice when the
workloads are limited. Other studies also surveyed RDF stores in terms of specific
quality attributes [8, 11]. Moreover, the evaluation of RDF store fits nicely into the
study of Delone et al. [9], which evaluates a generic Information System (IS) on
the basis of six success dimensions. According to the model provided by this study,
an IS can be analyzed and evaluated taking into account various characteristics that
affect in turn the subsequent use of the IS and the perceived user satisfaction (such
as the service quality). Even this view is rather abstract, the herein presented work
concur with it, focusing mainly to find various measures for RDF stores that can be
classified under the dimensions of system and information quality defined by Delone
et al.
As the evaluation of an RDF store in terms of its feature can be treated as a
multi-criteria decision problem (MCDM) [36], it can be analyzed in a systematic
way through MCDM, which is an approach widely used to find the best option
choice selected from several and also eventually conflicting alternatives. The MCDM
methods allow to clearly and systematically analyze a specific problem, enabling the
decision makers to analyze a problem with a methodological approach and scale it
according to the evolving requirements of the problem. In this regard, one of the
best-known and most widely proven MCDM techniques is the Analytic Hierarchy
Process (AHP) [29], which allow to bring back a complex decision to a set of pairwise comparisons. From the answers to such comparisons, the AHP method allows
to calculate the weight for each factor in the overall decision problem. Even if the
MCDM approach has been widely used in literature, none of the available research
works, to the best of authors knowledge, have applied it to support systematic analysis
within the RDF stores evaluation.
3 The Methodological Approach
The main objective of this work is to explore the factors affecting the success of an
RDF store. The factors are selected through the analysis of three real self-conducted
case-studies in different application domains, while their criticality is evaluated
through a literature review. Specifically, the proposed approach consists of two main
steps (Fig. 2):
• Analysis of three different self-conducted case studies in the areas of Ambient Assisted Living, Manufacturing and eHealth, in order to elicit the critical
success factors of the RDF stores, selected from functional and non-functional
features.
• Analysis of each of the identified factors in terms of an up-to-date state of the
art based on the literature review.
Each of the steps are described in details in the following sections.
197
conducted by Cudré-Mauroux et al. in [7]; they studied the applicability of NoSQL
databases as backend of RDf store, stating that they represent a valid choice when the
workloads are limited. Other studies also surveyed RDF stores in terms of specific
quality attributes [8, 11]. Moreover, the evaluation of RDF store fits nicely into the
study of Delone et al. [9], which evaluates a generic Information System (IS) on
the basis of six success dimensions. According to the model provided by this study,
an IS can be analyzed and evaluated taking into account various characteristics that
affect in turn the subsequent use of the IS and the perceived user satisfaction (such
as the service quality). Even this view is rather abstract, the herein presented work
concur with it, focusing mainly to find various measures for RDF stores that can be
classified under the dimensions of system and information quality defined by Delone
et al.
As the evaluation of an RDF store in terms of its feature can be treated as a
multi-criteria decision problem (MCDM) [36], it can be analyzed in a systematic
way through MCDM, which is an approach widely used to find the best option
choice selected from several and also eventually conflicting alternatives. The MCDM
methods allow to clearly and systematically analyze a specific problem, enabling the
decision makers to analyze a problem with a methodological approach and scale it
according to the evolving requirements of the problem. In this regard, one of the
best-known and most widely proven MCDM techniques is the Analytic Hierarchy
Process (AHP) [29], which allow to bring back a complex decision to a set of pairwise comparisons. From the answers to such comparisons, the AHP method allows
to calculate the weight for each factor in the overall decision problem. Even if the
MCDM approach has been widely used in literature, none of the available research
works, to the best of authors knowledge, have applied it to support systematic analysis
within the RDF stores evaluation.
3 The Methodological Approach
The main objective of this work is to explore the factors affecting the success of an
RDF store. The factors are selected through the analysis of three real self-conducted
case-studies in different application domains, while their criticality is evaluated
through a literature review. Specifically, the proposed approach consists of two main
steps (Fig. 2):
• Analysis of three different self-conducted case studies in the areas of Ambient Assisted Living, Manufacturing and eHealth, in order to elicit the critical
success factors of the RDF stores, selected from functional and non-functional
features.
• Analysis of each of the identified factors in terms of an up-to-date state of the
art based on the literature review.
Each of the steps are described in details in the following sections.