Informatica

The way that forensic examiners compare fingerprints highly differs from the behaviour of current automatic fingerprint identification algorithms. Experts usually use all the information in the fingerprint, not only minutiae, while automatic algorithms don’t. Partial (especially latent) fingerprint matching algorithms still report low accuracy values in comparison to those achieved by experts. This difference is mainly due to the features used in each case. In this work, a novel approach for matching partial fingerprints is presented. We introduce a new fingerprint feature, named Distinctive Ridge Point (DRP), combined with an improved triangle-based representation which also uses minutiae. The new feature describes the neighbouring ridges of minutiae in a novel way. A modified version of a fingerprint matching algorithm presented in a previous work is used for matching two triangular representations of minutiae and DRPs. The experiments conducted on NIST27 database with a background added of 29000 tenprint impressions from NIST14 and NIST4 databases showed the benefits of this approach. The results show that using the proposal we achieved an accuracy of 70.9% in rank-1, improving in an 11% the accuracy obtained using minutiae and the reference point. This result is comparable with the best accuracy reached in the state of the art while the amount of features is reduced.

The paper deals with the causality driven modelling method applied for the domain deep knowledge elicitation. This method is suitable for discovering causal relationships in domains that are characterized by internal circular causality, e.g. control and management, regulatory processes, self-regulation and renewal. Such domains are organizational systems (i.e. enterprise) or cyber-social systems, also biological systems, ecological systems, and other complex systems. Subject domain may be of different nature: real-world activities or documented content. A causality driven approach is applied here for the learning content analysis and normalization of the knowledge structures. This method was used in the field of education, and a case study of learning content renewal is provided. The domain here is a real world area – a learning content is about. The paper is on how to align the existing learning content and current (new) knowledge of the domain using the same causality driven viewpoint and the described models (frameworks). Two levels of the domain causal modelling are obtained. The first level is the discovery of the causality of the domain using the Management Transaction (MT) framework. Secondly, a deep knowledge structure of MT is revealed through a more detailed framework called the Elementary Management Cycle (EMC). The algorithms for updating the LO content in two steps are presented. Traceability matrix indicates the mismatch of the LO content (old knowledge) and new domain knowledge. Classification of the content discrepancies and an example of the study program content analysis is presented. The main outcome of the causality driven modelling approach is the effectiveness of discovering the deep knowledge when the relevant domain causality frameworks are applicable.

This paper presents a model which integrates inbound and outbound logistics with a crossdocking system. This model integrates the problem of routing inbound vehicles between suppliers and cross-docks and outbound vehicles between cross-docks and retailers, considering logistics costs and the products properties. This model aims to minimize the total cost by optimizing assignment of products to suppliers and retailers and operations of inbound and outbound vehicles. We developed an endosymbiotic evolutionary algorithm, which yields good performance in concurrent searches for the solutions of multiple subproblems and validate the performance using several numerical examples.

In cases where the balance problem of an assembly line with the aim to distribute the work loads among the stations as equal as possible, the concept of entropy function can be used. In this paper, a typical assembly line balancing problem with different objective functions such as entropy-based objective function plus two more objective functions like equipment purchasing cost and worker time-dependent wage is formulated. The non-linear entropy-based objective function is approximated as a linear function using the bounded variable method of linear programming. A new hybrid fuzzy programming approach is proposed to solve the proposed multi-objective formulation efficiently. The extensive computational experiments on some test problems proves the efficiency of the proposed solution approach comparing to the available approaches of the literature.

A standard problem in certain applications requires one to find a reconstruction of an analogue signal f from a sequence of its samples $f{({t_{k}})_{k}}$. The great success of such a reconstruction consists, under additional assumptions, in the fact that an analogue signal f of a real variable $t\in \mathbb{R}$ can be represented equivalently by a sequence of complex numbers $f{({t_{k}})_{k}}$, i.e. by a digital signal. In the sequel, this digital signal can be processed and filtered very efficiently, for example, on digital computers. The sampling theory is one of the theoretical foundations of the conversion from analog to digital signals. There is a long list of impressive research results in this area starting with the classical work of Shannon. Note that the well known Shannon sampling theory is mainly for one variable signals. In this paper, we concern with bandlimited signals of several variables, whose restriction to Euclidean space ${\mathbb{R}^{n}}$ has finite p-energy. We present sampling series, where signals are sampled at Nyquist rate. These series involve digital samples of signals and also samples of their partial derivatives. It is important that our reconstruction is stable in the sense that sampling series converge absolutely and uniformly on the whole ${\mathbb{R}^{n}}$. Therefore, having a stable reconstruction process, it is possible to bound the approximation error, which is made by using only of the partial sum with finitely many samples.

The simplest hypothesis of DNA strand symmetry states that proportions of nucleotides of the same base pair are approximately equal within single DNA strands. Results of extensive empirical studies using asymmetry measures and various visualization tools show that for long DNA sequences (approximate) strand symmetry generally holds with rather rare exceptions. In the paper, a formal definition of DNA strand local symmetry is presented, characterized in terms of generalized logits and tested for the longest non-coding sequences of bacterial genomes. Validity of a special regression-type probabilistic structure of the data is supposed. This structure is compatible with probability distribution of random nucleotide sequences at a steady state of a context-dependent reversible Markov evolutionary process. The null hypothesis of strand local symmetry is rejected in majority of bacterial genomes suggesting that even neutral mutations are skewed with respect to leading and lagging strands.

Conventional large vocabulary automatic speech recognition (ASR) systems require a mapping from words into sub-word units to generalize over the words that were absent in the training data and to enable the robust estimation of acoustic model parameters. This paper surveys the research done during the last 15 years on the topic of word to sub-word mappings for Lithuanian ASR systems. It also compares various phoneme and grapheme based mappings across a broad range of acoustic modelling techniques including monophone and triphone based Hidden Markov models (HMM), speaker adaptively trained HMMs, subspace gaussian mixture models (SGMM), feed-forward time delay neural network (TDNN), and state-of-the-art low frame rate bidirectional long short term memory (LFR BLSTM) recurrent deep neural network. Experimental comparisons are based on a 50-hour speech corpus. This paper shows that the best phone-based mapping significantly outperforms a grapheme-based mapping. It also shows that the lowest phone error rate of an ASR system is achieved by the phoneme-based lexicon that explicitly models syllable stress and represents diphthongs as single phonetic units.

Certificate-based cryptography (CB-PKC) is an attractive public key setting, which reduces the complexity of public key infrastructure in traditional public key settings and resolves the key escrow problem in ID-based public key settings. In the past, a large number of certificate-based signature and encryption schemes were proposed. Nevertheless, the security assumptions of these schemes are mainly relied on the difficulties of the discrete logarithm and factorization problems. Unfortunately, both problems will be resolved when quantum computers come true in the future. Public key cryptography from lattices is one of the important candidates for post-quantum cryptography. However, there is little work on certificate-based cryptography from lattices. In the paper, we propose a new and efficient certificate-based signature (CBS) scheme from lattices. Under the short integer solution (SIS) assumption from lattices, the proposed CBS scheme is shown to be existential unforgeability against adaptive chosen message attacks. Performance comparisons are made to demonstrate that the proposed CBS scheme from lattices is better than the previous lattice-based CBS scheme in terms of private key size and signature size.

Fuzzy c-means (FCM) is a well-known and widely applied fuzzy clustering method. Although there have been considerable studies which focused on the selection of better fuzzifier values in FCM, there is still not one widely accepted criterion. Also, in practical applications, the distributions of many data sets are not uniform. Hence, it is necessary to understand the impact of cluster size distribution on the selection of fuzzifier value. In this paper, the coefficient of variation (CV) is used to measure the variation of cluster sizes in a data set, and the difference of coefficient of variation (DCV) is the change of variation in cluster sizes after FCM clustering. Then, considering that the fuzzifier value with which FCM clustering produces minor change in cluster variation is better, a criterion for fuzzifier selection in FCM is presented from cluster size distribution perspective, followed by a fuzzifier selection algorithm called CSD-m (cluster size distribution for fuzzifier selection) algorithm. Also, we developed an indicator called Influence Coefficient of Fuzzifier ($\mathit{ICF}$) to measure the influence of fuzzifier values on FCM clustering results. Finally, experimental results on 8 synthetic data sets and 4 real-world data sets illustrate the effectiveness of the proposed criterion and CSD-m algorithm. The results also demonstrate that the widely used fuzzifier value $m=2$ is not optimal for many data sets with large variation in cluster sizes. Based on the relationship between ${\mathit{CV}_{0}}$ and $\mathit{ICF}$, we further found that there is a linear correlation between the extent of fuzzifier value influence and the original cluster size distributions.