The number of design patterns has been rising rapidly, but management and searching facilities appear to be lagging behind. This is why we developed a novel approach, which is used to search for suitable design patterns. It is built in an intelligent component on top of our platform for design patterns` knowledge exchange. In our approach, the developer no longer searches for an appropriate design pattern, but rather the intelligent component asks the developer questions. We did not want to invest extra effort in terms of maintaining a special expert system. Guided dialogues consist of independent questions from different sources and authors, that are combined automatically. The approach is enabled with a novel mathematical model and algorithms. Validation was performed via comparison with human-created expert systems with a decision tree. Experiments were executed in order to verify our approach performance. The control group used a human-created expert system, while others were given a proposing component.
COBISS.SI-ID: 20216342
We have developed a method called allocation, which is intended to classify unbalanced data. The allocation method is a classification ensemble composed of two levels. In the first level there is the allocator, an algorithm of unsupervised learning, which learns to split the original dataset into homogeneous subsets and allocates them to specialised classifiers on the second level. The second level consists of multiple specialised classifiers where each learns on the specific subset of instances allocated to it, and so specialises in a particular type of data. These specialised classifiers return the class of the instances. To test the concept of the allocation method, we developed an allocator with anomaly detection, which uses the one-class SVM classifier. The allocator was tested in combination with the six basic classification methods as the specialised classifiers on the second level. All variants and the combinations of the allocation method were tested on unbalanced and balanced datasets, the latter for the purpose of validation of the allocation as a general classification approach. The results of the allocation method were compared with existing methods for dealing with unbalanced data. The results of the experiments were analysed further with statistical methods, with which we confirmed that the allocation method is an effective alternative to the existing approaches for the classification of unbalanced and balanced data.
COBISS.SI-ID: 19263510
Business Process Modelling is an activity that includes several different roles, e.g. business analysts, technical analysts and software developers. The resulting process diagrams can be either simple or complex. Nonetheless, they must be understandable to everyone, even those without the necessary knowledge of process modelling notations. The goal of our research was to evaluate intuitive understandability of diagrams, modelled in different process modelling notations, with regard to diagram complexity. An empirical research was conducted, including 103 students, with the goal to validate the intuitiveness of the diagrams empirically, modelled in the most commonly used process modelling notations, i.e. Unified Modelling Language Activity Diagram (UML AD), Business Process Model and Notation (BPMN) and Event Driven Process Chain (EPC). In the case of processes with lesser complexity, participants using BPMN diagrams were outperformed significantly by those using either EPC or UML AD ones. When complexity of processes was higher, participants using EPC diagrams performed significantly worse than those using the UML AD and BPMN counterparts. Participants that used UML AD diagrams were not outperformed significantly by users of diagrams in other process modelling notations, regardless of their complexity. UML AD was recognised as being the most versatile notation. Our research can help decide which notation to use when representing processes that have to be understandable by all stakeholders.
COBISS.SI-ID: 19609366
An introduction of an innovative technology such as interactive whiteboard (IWB) in classrooms offers new opportunities for improving educational practices. Every new educational technology has to be adopted by teachers that have to adapt it in a creative way in order to utilise an IWB's potential in instructions fully. The adoption is a result of various factors, whose impact differs across different technology adoption phases. The main objective of this study is to develop and validate an instrument, allowing simultaneous assessment of external factors that affect users' perceptions about performance expectancy and effort expectancy during different technology adoption stages. A moderating variable user type was proposed to understand thedifferences in factors in different adoption stages.Quantitative- qualitative research in the form of an online questionnaire was conducted to test the proposed model. Empirical data gathered from 1,040 teachers were analysed primarily using the structural equation modelling approach. The results of this study showed that user interface quality, personal innovativeness and perceived pedagogical impact are factors that affect teachers' perceptions in all adoption stages. With regard to the proposed moderating variable, this study demonstrated significant differences in several causal effect sizes. A qualitative analysis was conducted to explain further the main reasons for abandoning IWB.
COBISS.SI-ID: 22585608
Harmonising the metadata format alone does not solve the issue of efficient access to relevant information in heterogeneous environments, when different systems use different content, contextual and semantic concepts for certain entities. One such type of these heterogeneous systems is also Current Research Information Systems (CRIS), which store their data primarily in local relational databases, using different formats and various local concepts. In this article, we study the possibilities and propose a new Ontologically Supported Semantic Search Engine (OSSSE) which, in addition to the harmonisation of the metadata format among local CRIS systems, also ensures that the meaning of data and/or concepts that belong to various metadata entities are also harmonised. A special model of ontological infrastructure was designed, and dedicated test ontology was created alongside with a new simplified algorithm for creating ontology, the basis of which is the distinction between new and already existing classes in terms of content. Finally, we evaluated the proposed OSSSE model using a simulation of the search process on the base of 41,113 real searches within SICRIS. The obtained results show that, regardless of the search situation, the proposed OSSSE is always at least as efficient as a search without ontological support in terms of precision, while recall remains the same; the improvement has been shown to be statistically significant with a high confidence interval (p(0.005). The proposed OSSSE model is able to solve the issue of harmonising the data where different heterogeneous systems use different content, contextual and semantic concepts, which is the case in many advanced expert systems. In this manner, the more the search is carried out based on the properties described by the supporting ontology, the more the infrastructure can help a searcher. The proposed concepts, ontological infrastructure and the designed semantic search engine, may well help to improve search precision in several information retrieval systems.
COBISS.SI-ID: 19798806