Ludwik Kuzniarz

Guidelines for conducting systematic mapping studies in software engineering: An update

Abstract Context Systematic mapping studies are used to structure a research area, while systematic reviews are focused on gathering and synthesizing evidence. The most recent guidelines for systematic mapping are from 2008. Since that time, many suggestions have been made of how to improve systematic literature reviews (SLRs). There is a need to evaluate how researchers conduct the process of systematic mapping and identify how the guidelines should be updated based on the lessons learned from the existing systematic maps and SLR guidelines. Objective To identify how the systematic mapping process is conducted (including search, study selection, analysis and presentation of data, etc.); to identify improvement potentials in conducting the systematic mapping process and updating the guidelines accordingly. Method We conducted a systematic mapping study of systematic maps, considering some practices of systematic review guidelines as well (in particular in relation to defining the search and to conduct a quality assessment). Results In a large number of studies multiple guidelines are used and combined, which leads to different ways in conducting mapping studies. The reason for combining guidelines was that they differed in the recommendations given. Conclusion The most frequently followed guidelines are not sufficient alone. Hence, there was a need to provide an update of how to conduct systematic mapping studies. New guidelines have been proposed consolidating existing findings.

Consistency problems in UML-based software development

This survey of the workshop series Consistency Problems in UML-based Software Development aims to help readers to find the guidelines of the papers. First, general considerations about consistency and related problems are discussed. Next, the approaches proposed in the workshop papers to handle the problems are categorized and summarized. The last section includes extended abstracts of the papers from the current workshop.

An empirical study on using stereotypes to improve understanding of UML models

Stereotypes were introduced into the Unified Modeling Language (UML) to provide means of customizing this visual, general purpose, object-oriented modeling language, for its usage in specific application domains. The primary purpose of stereotypes is to brand an existing model element with a specific semantics. In addition, stereotypes can also be used as notational shorthand. The paper elaborates on this role of stereotypes from the perspective of UML, clarifies the role and describes a controlled experiment aimed at evaluation of the role - in the context of model understanding. The results of the experiment support the claim that stereotypes with graphical icons for their representation play a significant role in comprehension of models and show the size of the improvement.

Empirical assessment of using stereotypes to improve comprehension of UML models: A set of experiments

Stereotypes were introduced into the Unified Modeling Language to provide means of customizing this general purpose modeling language for its usage in specific application domains. The primary role of stereotypes is to brand an existing model element with specific semantics, but stereotypes can also be used to provide means of a secondary classification of modeling elements. This paper elaborates on the influence of stereotypes on the comprehension of models. The paper describes a set of controlled experiments performed in academia and industry which were aimed at evaluating the role of stereotypes in improving comprehension of UML models. The results of the experiments show that stereotypes play a significant role in the comprehension of models and the improvement achieved both by students and industry professionals.

An empirical assessment of using stereotypes to improve reading techniques in software inspections

Stereotypes were introduced into the Unified Modeling Language (UML) in order to provide a means of customizing the language for particular needs. The stereotypes can increase the comprehension of UML diagrams and therefor influence reading techniques used for inspections of software artefacts. In this paper we evaluate how the usage of stereotypes in UML designs influences outcomes of three reading techniques used for verification and validation of UML models. The study presented in this paper is done in the context of the UML domain modeling, but the results can be generalized to other kinds of models expressed in UML. The results show that the presence of stereotypes improves the efficiency and effectiveness of the studied methods and shows the magnitude of the improvements. We also investigate which of the reading techniques are the most efficient and effective for analysis of UML designs with stereotypes.

Teaching Modeling - An Initial Classification of Related Issues

Modeling is an important skill needed in both science and engineering domains. In software engineering, in particular, models are ubiquitous artefacts. The development, manipulation and understanding of models is therefore an important learning objective. The paper describes motivations and an outline of a research project aimed at classifying issues related to the teaching and learning of modeling together with some initial research results.

Empirical extension of a classification framework for addressing consistency in model based development

Context: Consistency constitutes an important aspect in practical realization of modeling ideas in the process of software development and in the related research which is diverse. A classification framework has been developed, in order to aid the model based software construction by categorizing research problems related to consistency. However, the framework does not include information on the importance of classification elements. Objective: The aim was to extend the classification framework with information about the relative importance of the elements constituting the classification. The research question was how to express and obtain this information. Method: A survey was conducted on a sample of 24 stakeholders from academia and industry, with different roles, who answered a quantitative questionnaire. Specifically, the respondents prioritized perspectives and issues using an extended hierarchical voting scheme based on the hundred dollar test. The numerical data obtained were first weighted and normalized and then they were analyzed by descriptive statistics and bar charts. Results: The detailed analysis of the data revealed the relative importance of consistency perspectives and issues under different views, allowing for the desired extension of the classification framework with empirical information. The most highly valued issues come from the pragmatics perspective. These issues are the most important for tool builders and practitioners from industry, while for the responders from academia theory group some issues from the concepts perspective are equally important. Conclusion: The method of using empirical data from a hierarchical cumulative voting scheme for extending existing research classification framework is useful for including information regarding the importance of the classification elements.

Properties of stereotypes from the perspective of their role in designs

Stereotypes in object-oriented software development can be perceived in various ways and they can be used for various purposes. As a consequence of these variations, assessing quality of stereotypes needs to be purpose-specific. In this paper we identify eight types of stereotypes and provide a set of criteria for assessing quality of stereotypes. The criteria for each type are formed by a set of properties that characterizes its stereotypes. The identified types are based on the purpose of each stereotype (its role in designs) and its expressiveness. We identified the types of stereotypes and their properties in an empirical way by investigating stereotypes from UML profiles used in industrial software development. The properties are intended to be used in our further research for developing guidelines for creating and using stereotypes in a more efficient way.

Best practices for teaching UML based software development

Software development is the core activity performed within software engineering. The Unified Modelling Language (UML) has become a de facto standard language for expressing artefacts used and produced within software development process. As a result of that there is an expanding activity related to teaching the process of software development and the language at both higher education institutions and in software industry. The aim of the education in UML is to prepare software developers to effectively use UML and modelling at their work. Therefore guidelines on a proper way of teaching and learning processes and notations are foreseen. The paper elaborates on experiences and lessons learned from the course on UML based software development delivered as part of software engineering curriculum. In the paper we identify a set of best practices for teaching modelling and UML. The best practices are based on the experiences gathered from multiple course deliveries over a period of five years.

Generating Domain Models from Ontologies

The paper presents and elaborates on the idea of automatic acquisition of knowledge about domain structure from ontologies into an object-oriented software development process. The information required to be included in the domain model produced during the development process is identified. The existence of the knowledge in ontologies is investigated. Requirements for ontology description languages are formulated followed by brief evaluation of existing languages against these requirements. A schema for domain knowledge acquisition is outlined. A realization of the schema is sketched in the paper while the implementation details can be found in the technical report.