Informatica

The problem of automatic classification of scientific texts is considered. Methods based on statistical analysis of probabilistic distributions of scientific terms in texts are discussed. The procedures for selecting the most informative terms and the method of making use of auxiliary information related to the terms positions are presented. The results of experimental evaluation of proposed algorithms and procedures over real-world data are reported.

Enterprise systems should be assembled out of components and services according to an orchestration schema and taking into account not only functional requirements but also the resulting Quality of Service (QoS). In other words, QoS-aware composition of services and components must be performed. The problem is to find which components or services have to be employed that the resulting system would optimize some QoS attributes while satisfying some other QoS constraints. The paper proposes to use the Constraint Logic Programming approach to solve this problem, that is, we see this problem as a discrete optimization and satisfaction problem.

For many businesses and organizations, the achievement of interoperability has proven to be a highly desirable goal. However, without efficient schema mapping mechanisms or models that allow for the storage and management of information from several distinct systems, the goal of interoperability is impossible to attain. Due to the role of XML as a standard in information exchange, considerable research has been undertaken to find effective methods or algorithms for the conversion from XML to database models. This paper reviews leading research in the field – focusing particularly on three novel approaches taken – and proposes an original schema mapping mechanism using a conceptual model which, due its higher level of abstraction, maximizes the preservation of semantics.

This paper analyses the possibilities of integrating different technological and knowledge representation techniques for the development of a framework for the remote control of multiple agents such as wheelchair-type robots. Large-scale multi-dimensional recognitions of emotional diagnoses of disabled persons often generate a large amount of multi-dimensional data with complex recognition mechanisms, based on the integration of different knowledge representation techniques and complex inference models. The problem is to reveal the main components of a diagnosis as well as to construct flexible decision making models. Sensors can help record primary data for monitoring objects. However the recognition of abnormal situations, clustering of emotional stages and resolutions for certain types of diagnoses is an oncoming issue for bio-robot constructors. The prediction criteria of the diagnosis of the emotional situations of disabled persons are described using knowledge based model of Petri nets. The research results present the development of multi-layered framework architecture with the integration of artificial agents for diagnosis recognition and control of further actions. The method of extension of Petri nets is introduced in the reasoning modules of robots that work in real time. The framework provides movement support for disabled individuals. The fuzzy reasoning is described by using fuzzy logical Petri nets in order to define the physiological state of disabled individuals through recognizing their emotions during their different activities.

In this paper we propose facilitating ontology development by constant evaluation of steps in the process of ontology development. Existing methodologies for ontology development are complex and they require technical knowledge that business users and developers don't poses. By introducing ontology completeness indicator developer is guided throughout the development process and constantly aided by recommendations to progress to next step and improve the quality of ontology. In evaluating the ontology, several aspects are considered; from description, partition, consistency, redundancy and to anomaly. The applicability of the approach was demonstrated on Financial Instruments and Trading Strategies (FITS) ontology with comparison to other approaches.

We present a novel dynamic network interdiction model that accounts for interactions between an interdictor deploying resources on arcs in a digraph and an evader traversing the network from a designated source to a known terminus, wherein the agents may modify strategies in selected subsequent periods according to respective decision and implementation cycles. For the resulting minimax model, we develop a reformulation that facilitates a direct solution procedure using commercial software or via a proposed alternating heuristic. We examine certain related stability and convergence issues, demonstrate special convergence cases, and provide insights into the computational performance of the solution procedures.

This paper presents GSM speech coder indirect identification algorithm based on sending novel identification pilot signals through the GSM speech channel. Each GSM subsystem disturbs identification pilot, while speech coder uniquely changes the tempo-spectral characteristics of the proposed pilot signal. Speech coder identification algorithm identifies speech coder with the usage of robust linear frequency cepstral coefficient (LFCC) feature extraction procedure and fast artificial neural networks. First step of speech coder identification algorithm is the exact position detection of the identification pilot signal using normalized cross correlation approach. Next stage is time-domain windowing of the input signal to convolve each frame of the input speech signal and window spectrum. Consecutive step is a short-time Fast Fourier Transformation to produce the magnitude spectrum of each windowed frame. Further, a noise reduction with spectral subtraction based on spectral smoothing is carried out. In last steps we perform the frequency filtering and Discrete Cosine Transformation to receive 24 uncorrelated cepstral coefficients per frame as a result. Speech coder identification is completed with fast artificial neural network classification using the input feature vector of 24 LFCC coefficients, giving a result of identified speech coder. For GSM speech coder indirect identification evaluation, the standardized GSM ETSI bit-exact implementations were used. Furthermore, a set of custom tools was build. These tools were used to simulate and control various conditions in the GSM system. Final results show that proposed algorithm identifies the GSM-EFR speech coder with the accuracy of 98.85%, the GSM-FR speech coder with 98.71%, and the GSM-HR coder with 98.61%. These scores were achieved at various types of surrounding noises and even at very low SNR conditions.

The paper presents the process of the selection of a potential supplier, which have to be the most appropriate to stakeholders. The selection is based on a set of criteria: Delivery Price, Financial Position, Production Specifications, Standards and Relevant Certificates, Commercial Strength, and the Performance of supplier, etc. The criteria for evaluation and their importance are selected by taking into consideration the interests and goals of the stakeholders. The solution of problem was made by applying a new Additive Ratio ASsessment (ARAS) method with the grey criteria scores – ARAS-G method. The proposed technique could be applied to substantiate the selection of effective alternative of sustainable development, impact on environment, structures, technologies, investments, etc.

Combinatorial problems serve as an important resource for developing practical public key cryptosystems and several combinatorial cryptosystems have been proposed in the cryptographic community. In this paper, a combinatorial public key cryptosystem is proposed. The security of the proposed cryptosystem is dependent on a combinatorial problem involving matrices. The system features fast encryption and decryption. However, the system also suffers from some drawbacks. The ciphertext expansion is relatively large and the key sizes are somewhat larger than that of RSA. The security of the system is carefully examined by illustrating the computational infeasibilities of some attacks on the system.

This paper presents an entire chaos-based biometric remote user authentication scheme on tokens without using passwords. The proposed scheme is based on the chaotic hash function and chaotic pseudo-random number generator to provide secure mutual authentication over an insecure channel between the user and remote server. Compared with the related biometric authentication schemes, the proposed scheme does not require the user password to provide convenience to users. It also does not require time synchronization or delay-time limitations between the user and remote server to resolve time synchronization problems.