Data Science

Data Science

Scientific areas of our expertise

Data Science today becomes an area of research present in various domains of society. Our experts for more than 10 years actively involved in research activities in the fields of machine learning, data mining and artificial intelligence. Evidence of this is a large number of published scientific articles in various reputable scientific journals and databases.

  • Machine learning and data mining,
  • Data clustering, attribute selection, data regression,
  • Data prediction/forecasting,
  • Predictive modelling on customer demand,
  • Deep learning & Neural Network,
  • Fuzzy learning and systems.

Some of the published scientific articles of our experts

A decision support system for prediction of wastewater pumping station failure based on CBR continuous learning model

Engineering, Technology and Applied Science Research 9(5):4745-4749 · October 2019
Abstract — Nowadays the communities are facing the problem of waste and wastewater. While wastewater systems have become more complex, the need for development of sustainable solution for wastewater management emerged. Therefore, the development of a Decision Support System (DSS) for wastewater disposal management became necessary. This paper presents a new DSS for predicting the failure of wastewater pumping stations, the system architecture and its implementation. The prediction model is based on the Case Based Reasoning (CBR) classification method. The standard CBR classification technique has been upgraded with an algorithm for continuous learning. The paper describes the system structure, its connection to the wastewater system, the internal processes involved in the prediction process and the implemented algorithm for continuous learning. Furthermore, the features used in the prediction are indicated along with the achieved results and the method of results evaluation. The test and obtained results indicate that the proposed DSS is efficient and capable of providing very good results in the prediction process.


Proceedings of the 29th International DAAAM Symposium 2018 (pp.0486-0494)
Abstract — This paper proposes the framework and implemented algorithm for text skew angle detection and auto correction of scanned documents. The process of scanning the documents might cause their displacement and rotation, which later affects the difficulty in text recognition process. The paper presents a developed application with the algorithm adapted specially for the framework. The implemented algorithm is based on Fast Fourier Transform. The framework enables text skew angle measurement and according to the archived results performs rotation and correction of the scanned document. Evaluation of the framework was focused on determining accuracy in skew angle detection, on determining the implemented algorithm sensitivity based on the resolution and quality of scanned documents. The evaluation showed that the proposed framework achieved good results which provide guidance for the further development of similar frameworks.

Adaptable web prediction framework for disease prediction based on the hybrid Case Base Reasoning mode

Engineering, Technology & Applied Science Research (ETASR), Vol. 6, No. 6, 2016, pp. 1212-1216. EOS ASSOC, GASTOUNI, GREECE
ISSN: 2241-4487
Abstract — Nowadays, we are witnessing the rapid development of medicine and various methods that are used for early detection of diseases. In order to make quality decisions in diagnosis and prevention of disease, various decision support systems based on machine learning methods are introduced in the medical domain. Such systems play an increasingly important role in medical practice. This paper presents a new web framework concept for disease prediction. The proposed framework is object-oriented and enables online prediction of various diseases. The framework enables online creation of different autonomous prediction models depending on the characteristics of diseases. Prediction process in the framework is based on a hybrid Case Based Reasoning classifier. The framework was evaluated on a few disease datasets from public repositories. Experimental evaluation shows that the proposed framework achieved high diagnosis accuracy.

Web Prediction Framework for College Selection Based on the Hybrid Case Based Reasoning Model and Expert’s Knowledge

International Journal of Hybrid Intelligent Systems, vol. 13, no. 3-4, pp. 161-171, 2016 Published by IOS Press, The Netherlands
Abstract — Higher education today represents the basis of any successful society. Every day we are witnessing an increase in the number of HEI, an increase in the number of students but also an increase in the number of dropouts. This paper presents a new concept of the prediction framework which enables the selection of future college students based on their socio-demographic characteristics. The framework enables college autonomy in creating their own predictive models based on the characteristics of its students. In the prediction process, the framework has the ability of dynamic adjustment according to specific characteristics of each college. The framework is object-oriented and enables the performance of an online prediction process. The proposed framework uses a hybrid Case Based Reasoning (CBR) model and expert’s knowledge. The hybrid CBR model has integrated several methods of machine learning: Information Gain, K-means and Case-based reasoning. The study used datasets collected from several colleges, a part of the Croatian Information System for Higher Education (ISVU). The case study demonstrates that our proposed web prediction framework is efficient and capable of providing very good results in the process of prediction. The achieved results provide guidelines for the future development of the prediction framework.

Case-Based Reasoning: A Hybrid Classification Model Improved with an Expert's Knowledge for High-Dimensional Problems

Conference: ICCBR 2016 :18th International Conference on Case-Based Reasoning, June 13 - 14, 2016, Venice, Italy International Journal of Computer, Electrical, Automation, Control and Information Engineering Vol:10, No:6, 2016 (World Academy of Science, Engineering and Technology)
Abstract — Data mining and classification of objects is the process of data analysis, using various machine learning techniques, which is used today in various fields of research. This paper presents a concept of hybrid classification model improved with the expert knowledge. The hybrid model in its algorithm has integrated several machine learning techniques (Information Gain, K-means, and Case-Based Reasoning) and the expert’s knowledge into one. The knowledge of experts is used to determine the importance of features. The paper presents the model algorithm and the results of the case study in which the emphasis was put on achieving the maximum classification accuracy without reducing the number of features.

Business bankruptcy prediction based on hybrid CBR mode

21st European Concurrent Engineering Conference 2015, (ECEC), 27-29 April 2015 in Lisbon, Portugal ISBN: 9789077381885
Abstract — Predicting bankruptcy and early identification of the financial crisis nowadays has become a field of particular interest in which various studies have been conducted. The consequences of business bankruptcy have a negative impact on the whole society. Signs of financial distress can be detected much earlier before bankruptcy occurs. For this reason, a variety of scientific methods have been developed for timely detection of the difficulties in the business. The main purpose of this paper is to analyze quality prediction of business bankruptcy using the novel hybrid Case-based reasoning (CBR) model with a new approach of data classification. The hybrid model in its algorithm has integrated several methods of machine learning: Information Gain, K-means and Case-based reasoning.

Evaluation of Croatian Development Strategies using SWOT Analyses with Fuzzy TOPSIS Method and K-means methods

Journal: Journal of Economics, Business and Management
Abstract — The purpose of this paper is to analyse Croatian national development strategy and Croatian local/regional development strategies. Upon defining the purpose we can define the main goal of this paper and which is creation of a tool based on funny logics which will give results on whether the chosen areas of analysed strategies are correspondent or not. For that purposes linguistic variables for areas of importance of national and regional development strategies have been defined with corresponding weight impact factors. Defined areas have been of various development priorities and have been sorted according to SWOT analysis components. Upon definition of variables and weight impact factors the analysis has been conducted.