Applied Computational Intelligence and Mathematical Methods

This paper deals with macroeconomic convergence at the NUTS2 level for the following six countries: Czechia, Slovakia, Poland, Hungary, Germany and Austria. Prominent spatial dependencies are identified and compared to the Solow-Swan type convergence. The estimation and testing is performed using spatial panel data methodology. At the theoretical and empirical level, properties and performance of spatial panel models are compared with classical cross-sectional and panel (non-spatial) approaches. Given the variety of available frameworks of modeling and estimation of spatial dependencies, significant proportion of this paper is devoted to model specification and robustness analysis issues. Also, topics relevant for appropriate interpretation of the estimated spatio-temporal models are included.

The aim of the paper is to decompose two important economic variables (GDP and unemployment rate in the Czech Republic) into a cyclical and a trend component by applying a state space methodology. An unobserved component model is econometrically estimated by the method of maximum likelihood. The likelihood function is constructed using the square root version of the Kalman filter. The results are economically interpreted and it is found that (1) the cyclical component of output and unemployment rate has already recovered from an initial shock at the beginning of the economic crisis in 2008, (2) there has been a persistently decreased growth of the trend component of output after the outbreak of the economic crisis, (3) the trend component of unemployment rate has been constant during the current crisis which suggests that possible hysteresis effects have not played an important role yet, (4) the growth of the GDP trend component is highly volatile in the Czech Republic.

Business applications, such as market shops, use matrix multiplication to calculate yearly, monthly, or even daily profits based on price and quantity matrices. Matrices comprise large data in computer applications and other fields, which make the efficiency of matrix multiplication a popular research topic. Although the task of computing matrix products is a central operation in many numerical algorithms, it is potentially time consuming, making it one of the most well-studied problems in this field. In this paper, Message Passing Interface (MPI), MapReduce, and Multithreaded methods have been implemented to demonstrate their effectiveness in expediting matrix multiplication in a multi-core system. Simulation results show that the efficiency rates of MPI and MapReduce are 90.11% and 47.94%, respectively, with a multi-core processor on the Market Shop application, indicating better performances compared with those of the multithreaded and sequential methods.

Procedural content generation can be defined as the algorithmical creation of game content with limited or indirect user input. In this paper we present a procedural content generation genetic algorithm for massively multiplayer online games. The incremental generation of content by choosing the most appropriate selection of added blocks allows an efficient progress in the game with a small impact on performance and the consequent ability to deploy such type of game in low performance mobile devices.

Elasticity is one of the essential properties in Cloud Computing that meets changeable needs of customers, and improves resource utilization for providers. In this context, oversubscription is a very powerful technique to increase resource utilization level as much as possible, which leads to a maximization of the profit of cloud providers. In this paper, we propose an autonomous architecture based on the MAPE-K control loop and using memory oversubscription to improve operating performance in a cloud infrastructure. The overload caused by oversubscription is mitigated by the live migration technique of VMs as well as the use of the network memory of the various physical machines of the cluster through the network. This latter technique is usually used as a replacement technique for the swapping disc as it has more performance.

The Cloud has become a new Information Technology(IT) model for delivering resources such as computing and storage to customers on demand, it provides both high flexibility and resources use. However we are gaining these advantages at the cost of high security threats, which presents the major brake for the migration towards Cloud Computing.Cloud Data Storage(CDS) is one of the Cloud services, it allows users to store their data in the Cloud, this service is very useful for companies and individuals, but data security remains the problem which makes customers worried about their data that reside in the Cloud. In this paper, we propose a framework of security to ensure the CDS, which is based on agents, it contains three layers: Cloud Provider layer, Customer layer and Trusted Third Party(TTP) layer.

Characterization of the information collected from the CLINIC CRC SITE – BARRANQUILLA (Headquarter) network infrastructure will serve as a foundation for building the new proposal of technological infrastructure for the new site of the clinic located in Baranoa. It will be designed according to the guidelines of the methodologies of design Top-Down and PPDIOO, which are: analyze requirements, develop a logical design, develop a physical design, test, optimize and document design, testing the network, and monitor performance. Following each of the previous stages 4 scenarios were designed and tested, which simulate the administration and performance of the network services.

In this research a collection of artificial intelligence techniques are combined together to optimize the process of clustering textual transcripts obtained from audio sources. Since clustering techniques have drawbacks that if not taken care of will produce sub optimal clustering solutions, it’s essential to attempt to optimize the clustering algorithms to avoid sub optimal solutions. As an attempt to overcome this problem, different artificial intelligence techniques are applied to avoid clustering problems. The main objectives of this research is to optimize automatic topic clustering of transcribed speech documents, and investigate the impact of applying genetic algorithm optimization and initial centroid selection optimization (ICSO) in combination with K-means clustering algorithm using Chi-Square similarity measure on the accuracy and the sum of square distances (SSD) of the selected clustering algorithm. The evaluation showed that using ICSO with genetic algorithm and K-means clustering algorithm with Chi-square similarity measure achieved the highest accuracy with the least SSD.

The manufacturing cell design problem is part of the cellular manufacturing system and it has been widely studied as an optimization problem. It consists of grouping machines in parts into manufacturing cells in order to minimize the inter-cell movements. In recent years, different approximate methods have been used to solve this problem. In this paper, we propose a new approximate method inspired on the phenomenon of the colonial age, called imperialist competitive algorithm. In the colonial age, the most powerful countries competed to conquer colonies for increasing their power, where the country with highest power was considered the imperialist one. We performed several experiments on a set of 90 instances, where the proposed approach is able to produce optimal values for the whole set of tested instances.

Optical character recognition (OCR) systems play vital role in pattern recognition research. With rapid growth of OCRs for different languages developing OCR for Czech language is looked upon as positive aspect for people speaking Czech language. In this paper, we develop OCR system for Czech language using hierarchical fuzzy convolutional neural networks (HFCNN). We present end-to-end framework that includes pre-processing activities, segments text image, classifies characters and performs recognition. The feature extraction is performed through fuzzy Hough transform. The feature based classification is performed through HFCNN. A comprehensive assessment of proposed method is performed through publicly available Czech language dataset. OCR recognition accuracy is a major concern. There is always an inherent degree of vagueness and impreciseness present in reallife data. Due to this recognition system is treated here through fuzzy sets encompassing indeterminate uncertainty. The simulation studies reveal that deep learning based OCR for Czech language performs consistently better than traditional models. The experimental results demonstrate efficiency of proposed approach.

The binarization of Swarm Intelligence continuous metaheuristics is an area of great interest in operational research. This interest is mainly due to the application of binarized metaheuristics to combinatorial problems. In this article we propose a general binarization algorithm called Percentile Transition Ranking Algorithm (PTRA). PTRA uses the percentile concept as a binarization mechanism. In particular we will apply this mechanism to the Cuckoo Search metaheuristic to solve the set multidimensional Knapsack problem (MKP). We provide necessary experiments to investigate the role of key ingredients of the algorithm. Finally to demonstrate the efficiency of our proposal, we solve Knapsack benchmark instances of the literature. These instances show PTRA competes with the state-of-the-art algorithms.

It is a challenging task to identify sentiment polarity in Arabic journals comments. Algerian daily newspapers interest more and more people in Algeria, and due to this fact they interact with it by comments they post on articles in their websites. In this paper we propose our approach to classify Arabic comments from Algerian Newspapers into positive and negative classes. Publicly-available Arabic datasets are very rare on the Web, which make it very hard to carring out studies in Arabic sentiment analysis. To reduce this gap we have created SIAAC (Sentiment polarity Identification on Arabic Algerian newspaper Comments) a corpus dedicated for this work. Comments are collected from website of well-known Algerian newspaper Echorouk. For experiments two well known supervised learning classifiers Support Vector Machines (SVM) and Naïve Bayes (NB) were used, with a set of different parameters for each one. Recall, Precision and F_measure are computed for each classifier. Best results are obtained in term of precision in both SVM and NB, also the use of bigram increase the results in the two models. Compared with OCA, a well know corpus for Arabic, SIAAC give a competitive results. Obtained results encourage us to continue with others Algerian newspaper to generalize our model.

Ontology is an agreement about a shared conceptualization, which includes frameworks for modeling domain knowledge and agreements about the representation of particular domain theories, often captured in some form of a semantic web formally. However, building ontology is a time consuming task. however, the paper was presented an approach to enrich instances into the exiting ontology and this research presented the technique to extract information from the unstructured text from websites. Support vector machine was used to create model. The results showed that feature reduction and SVM techniques presented the highest precision than SVM approach.

Cat swarm optimization (CSO) is a novel metaheuristic based on swarm intelligence, presented in 2006 has demonstrated great potential generating good results and excellent performances simulating the behavior of domestic cats using two behavior: seeking and tracing mode, this mode are classified using a mixture rate (MR), this parameter finally defines the number of individuals who work by exploring and exploiting. This work presents an improvement structure of a binary cat swarm optimization using a total reboot of the population when loss diversity it is detected.

This research aims to develop a model for analysis of student behavior through e-Learning based on data mining technique in case of Suan Sunandha Rajabhat University. The student data set was composed of 5392 personal records and, to compare the effective of algorithm, the model was created under decision tree and Bayesian networks techniques. The result found that showed that Bayesian networks technique showed higher performance and the percentage of prediction is accurate 91.32%.

There are existing implementations of Latent Dirichlet Allocation (LDA) algorithm as a semantic classifier to arrange the data for efficient retrieval. However, the problem of learning or inferencing the posterior distribution of the algorithm is trivial. Inferencing directly the prior distribution could lead to time taken to increase exponentially. It is due to the coupling of the hyperparameters. Several inference algorithms have been implemented together with LDA to solve this issue. The inference algorithm used in this research work is Gibbs sampling. Research using Gibbs sampling shows promising results in comparison to other inference algorithms, especially in the performance of the algorithm. It still takes a long time to compute the topic distribution of the data. There are still room for improvement in the time taken for the algorithm to complete the topic distribution. Using two datasets, an evaluation of the performance of the algorithm has been conducted. Results show that Gibbs sampling as the inference algorithm provides a better prediction on the optimal number of topic of the data in comparison to Variational Expectation Maximization (VEM).

This paper aims to present a hypothetic theory of intelligent security system. In society the threat of cyber-attacks is getting louder and the use of computers, criminal activity has also changed from physical to cybernetic intrusion. There had been many cyber security solutions used to counteract these attacks, however we highlight the importance of self-protected systems in defense and in a correct analysis of cyber attacks. The internet is vulnerable to cyber-attacks as well as the information found in data systems and through a form of recognition and extraction of relevant information, we can represent data as shared data and integrated to intelligent system. What was used us a static firewall is now intended to be dynamic and self-critical. By techniques of data analysis, statistics, machine learning, data mining, the cybersecurity and privacy challenges are within our reach. This paper examines data mining techniques in order to predict pathways of Internet security and which considerations are involved in the theoretical solutions presented for the privacy systems such as the e-Learning environments.

Associative classification is recognized by its high accuracy and strong flexibility in managing unstructured data. However, the performance is still induced by low quality dataset which comprises of noised and distorted data during data collection. The noisy data affected support value of an itemset and so it influenced the performance of an associative classification. The performance of associative classification is relied on the classification where the classification is worked based on the class association rules which generated from frequent rule mining process. To optimize the frequent itemsets based on the support value, in this research, we proposed a new optimization pruning technique to prune decision tree according to the correlation of each decision tree branches using genetic algorithm.

The vehicle routing problem and its variants such as the multi-depot vehicle routing problem are well-known NP-hard combinatorial optimization problems with wide engineering and theoretical background. In this paper a new hybrid technique based on intelligent water drop algorithm and simulated annealing is proposed to solve the multi-depot vehicle routing problem. The intelligent water drop algorithm is a stochastic population based metaheuristic optimization algorithm that uses a constructive approach to find optimal solutions of a given problem. Simulated annealing is a popular local search meta-heuristic approach with the key features of being able to provide a means to escape local optima by allowing hill-climbing moves with the hope of finding a global optimum. The performance of the hybrid algorithm is evaluated on a set of 23 benchmark instances and the results obtained compared with the best known solutions. The computational results show that the proposed method can produce good solutions, indicating that it is a good alternative algorithm for solving the multi-depot vehicle routing problem.

It is possible that a technology affects other technologies. In this paper, we explored the possibility to reveal the performance of improved information retrieval through the extraction method of social network. Any extracted social network structurally is not a complete graph so it is possible to build the star social networks as the optimal form of graph, which guides to model the implication of information retrieval: the formulation of recall and precision, by using a sample, it show better performance on average over $$90\%$$ and $$58\%$$, respectively.

Strategic decisions are those that have a far reaching effect on the environment and on the organization itself, however, they need a strategic diagnosis. Decision makers, for lack of knowledge of the literature and without a more efficient methodology, sometimes delay and/or make decisions that might not have been the best. This work aims to mitigate this problem, proposing a method that can assist the decision maker. The solution is based on two tools in the field of business strategy: the SWOT matrix (strengths and weaknesses in internal analysis and opportunities and threats in external analysis) and the map of strategic postures (survival, maintenance, growth and development), with support of the Multi-Attribute Global Inference of Quality (MAGIQ) multicriteria analysis tool [1]. Thus, depending on the situation of the strategic diagnosis in which the organization is located, the solution would indicate what possible strategic decisions the organization should adopt. A methodology for feeding, filtering, calculating, positioning and selecting strategies, including inbound and outbound reports, is proposed.

The amount of unstructured documents daily produced has dramatically increased in the last few years. As a result, automatic structuring of these contents has become an urgent need: it constitutes a prerequisite to any further automatic processing in term of annotation, indexing, information retrieval, etc. Nevertheless, a lack of automatic structuring methods for the Arabic normative texts is perceived. In this context, a method for automatic structuring of Arabic normative texts is presented in this paper. A standardized structure of Arabic normative texts is defined: two levels of granularity are identified: thematic and logic. A semantic annotation rule base is also developed to automatically structure documents according to these levels of granularity. Obtained results are very promising: the overall performance reached 94.53% for Precision, 91.21% for Recall and 92.84% for F-score.

This article describes computer aided analysis using computational intelligence methods for analysis and simulation research of a crane system during sequential movements. A parametric solid model has been specified, designed with a CAD/CAE environment, which allows to evaluate its stability for selected configurations and conditions of operation. Neural-network-supported analysis of varying contact forces exerted by the outriggers onto the ground, stabilizing and overturning torques, mass centre during handling allowed to specify trajectories ensuring stability of the crane. The results of the simulation research have been presented as changes of stability conditions depending on: angular position of the column with its telescopic arms and booms, the position of the telescopic arms themselves, the mass of individual components of the load system, as well as the load value applied onto it.

According to the adaptive behavior of organizational innovation, this paper explores the impact of organizational inertia on incremental innovation and radical innovation from the dimensions of structural inertia and cognitive inertia by analyzing the root of technological innovation network inertia, and taking the network embeddedness as adjustment variable. The research is carried out with the industry with high R & D intensity as the object, and the multiple regression model is used to do the empirical test, it is proved that organizational inertia has significant positive effect on incremental innovation, adaptive behavior helps to reveal the organization of technological innovation network, to enhance organizational innovation ability, and has important significance to maintain the network stability.

Econometric analysis is a non-trivial discipline applied to different areas of an enterprise or economy, in order to express economic reality and anticipate economic phenomena. This requires a great deal of econometric knowledge using a number of sophisticated methods and their good capabilities for correct and high quality interpretation of results. Currently, the intelligent system is a solution that is capable of performing highly complex tasks in the same way as people approach these tasks. In the context of ambient intelligence, it is possible to use personalized, contextual awareness and adaptive attributes for the design of an intelligent econometric system. In our work, we focused on the concept of an intelligent econometric system together with the application of OLAP technology for the creation of interactive analytical outputs. This new concept of the system is presented in an example of a data analysis to derive the forecast from the econometric model by designing a multidimensional view of the data.

In this paper, we discuss the results of research on the optimization modeling of ground motion attenuation in the subduction zone of the model Youngs et al. [1] using two methods: the Levenberg-Marquard and Bruce-Force method. This modeling is particularly important in the case of seismicity. Given that it takes a good model for predicting the strength of earthquakes in order to reduce the risk of the impact of natural disasters. Two major contributions of this study are ground motion attenuation model specific to the subduction zone that has been optimized with the Levenberg-Marquard method and Bruce-Force uses a model Youngs et al. [1] and a proof that the Levenberg-Marquard method for optimization model is better than Bruce-Force method. The Levenberg-Marquardt method has been proven to provide more accurate results on the modeling of ground motion attenuation which is indicated by a very small deviation between the values of PGA predictable results with the PGA actual values.

This paper will serve as a basic introduction to the observability of diffusion process as an example of distributed parameter systems. The aim of this research is to reconstruct initial state not well known $$x_{0}$$, which is known in certain subregions and unknown in others, and to give important results related to internal pointwise sensor in different geometrical situations. Many applications are investigated whether in one-dimensional case or two-dimensional one.

The research has application-oriented character and is aimed at solving practical task of forecasting additional manpower needs in the regional economy of the Murmansk region. The authors propose to integrate the existing methods for determining manpower needs with simulation modeling. Models of system dynamics (simulation models) are used to design tools to determine a number of employed people and job vacancies for each type of economy activity. The peculiarity of the developed tools is the possibility to consider large regional investment projects.

Peak ground acceleration (PGA) is a measure of earthquake acceleration in the ground. The prediction information about PGA is important to minimize the effect of earthquake. The method for prediction is Ordinary Kriging. It is geostatistic method used to predict data in certain locations which have autocorrelation. The sample data used in this research are PGA in Meuraxa, Banda Aceh 2006. The steps of research methodology consist of autocorrelations identify by Moran’s I and LISA, build semivariograms, and prediction by Ordinary Kriging. The results is Ordinary Kriging can be applied to predict PGA. It was shown by evaluate of mean and MSE value. According to mean value of three prediction, all models (Gaussian, Spherical, and Exponential) have mean 0,3534; 0,3584; and 0,3555 which approaches the actual PGA mean 0.34. According to MSE value, it can be seen that all models have small MSE or relatively closed to zero.

Technology has become a vital component for organizations. Thus, it is necessary to ensure quality and efficient IT solutions to meet the expectations of the business areas. In this scenario, we realize the need to optimize decision-making in the IT infrastructure problem management process, thus ensuring greater availability of IT solutions that support business. The objective of this work is to propose a model for selection and prioritization of IT infrastructure problems in the Multiple Criteria Decision Analysis. We present the context of the decision of the problem and by applying cognitive mapping we identify the most relevant criteria in order to detect the problems that generate the most impacts in the business.

We propose an agent-based model to simulate internal migration of workers. The model is based on mathematical equations describing the socio-economic characteristics of the agents and their behaviors. The main assumption of the model is the objective of individuals to find decent (formal) employment.The model simulates the migration of agents from a rural-type region of origin to a destination region with more important economic activities. The model consists of two main stages: before and after migration. In the first stage, potential migrants and migrants are determined. The second stage involves job search and demographic and socio-economic updates. The model is divided into several modules: (i) Initiation and migration, (ii) Job search module, and (iii) Module of economic and demographic transitions.

In this paper the application capabilities of the heuristic method of mathematical statistics - classifications trees - are considered. Based on the analysis of sociological research results of the political tolerance of students in the Murmansk region, an algorithm for identifying and forecasting risk groups for intolerant behavior is described.

Honeypots are unconventional tools to study methods, tools, and goals of attackers. In addition to IP addresses, these tools collect also timestamps. Therefore, time series analysis of data collected by honeypots can bring different view for prediction of attacks. In the paper, we focus on the model AR(1) and bootstrap based on AR(1) model to predict attacks against honeynet. For this purpose, we used data collected in CZ.NIC honeynet consists of Kippo honeypots in medium-interaction mode. The prediction of attacks is based on 75 weeks data and it has been verified by five weeks data. In the paper, we have shown that prediction model AR(1) and bootstrap based on AR(1) model are suitable for prediction of attacks.

The important contemporary issue of VLSI design verification is its time-consuming. The hardware model, written, for instance, with VHDL, is verified by formal and dynamic verification approaches. Dynamic verification (simulation) is widely-used due to the possibility of full automation of the process, but takes too much time due to its redundancy. The concept of the coverage-directed test generation is to redirect the test generator such as to reach uncovered metric points. There are several approaches for this, including genetic algorithms using, Bayesian network, Data mining, etc.The new cognitive approach to the coverage-directed test generation (CA CDG) is proposed within this paper. It is based on a cognitive map usage. The CA CDG is described, some simulation results are given. Also the future work areas are outlined.

We present an agent-based model for studying the dynamics of Aedes vexans and Culex poicilipes mosquito populations taking into account interactions with animal herds, water ponds and climate factors in the Ferlo region (Senegal). The main objective of this work is to show the impact of intra-annual season’s variability on the Aedes vexans and Culex poicilipes mosquito populations dynamics. We have designed a UML class diagram representing interactions between animal hosts, mosquitoes, and water ponds and used this diagram to create an agent based model which helps us to carry out the sensitivities analysis on dry spells. The obtained results show that there is a growth of Aedes vexans mosquito populations at the end of each dry spell following by rainfall and the appearance of Culex poicilipes mosquito populations at the end of August coinciding with the disappearance of Culex poicilipes mosquito populations. The developed model provides a tool for understanding and predicting the dynamics of Aedes vexans and Culex poicilipes mosquito populations in the short and long term. It can also be used to study the sensitivities analysis of daily rainfalls of other types of vectors.