Time filter

Source Type

Pajenga E.,Aleksander Xhuvani University of Elbasan | Rexha T.,University of Tirana | Celiku S.,Oncology Hospital Mother Tereza | Pisha M.,Medical Center
Bosnian Journal of Basic Medical Sciences | Year: 2013

The role of reproductive factors in the aetiology of ovarian cancer had been evaluated in hospital-based case-control study conducted in Albania, providing a total dataset of 283 cases and 1019 controls. Logistic regression models were used to obtain relative risk (OR) estimates. The present results showed that parity had protective effects which increased until the forth birth and the trend in risk was significant (p < 0.01). In each stratum and overall, nulliparous women appeared to be at highly increased risk compared to those who had different number of births (OR=12.5, 95%, CI: 2.4-63.8). Evaluation of early age at menarche and late age at menopause, showed statistically significant increased risk. Furthermore, increased risk was observed between pre-menopausal women and never-married nulliparity women, respectively (OR=1.44 95%, CI: 0.88-2.36; OR=8.98,95%, CI: 1.44 - 56.14), but ovarian cancer risk was reduced for hysterectomized women. These findings suggest that Albanian women have risk factors similar to women in western countries. © 2013 Association of Basic Medical Sciences of FB&H. Source

Grabocka J.,University of Hildesheim | Bedalli E.,Aleksander Xhuvani University of Elbasan | Schmidt-Thieme L.,University of Hildesheim
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2014

Semi-supervised learning is an eminent domain of machine learning focusing on real-life problems where the labeled data instances are scarce. This paper innovatively extends existing factorization models into a supervised nonlinear factorization. The current state of the art methods for semi-supervised regression are based on supervised manifold regularization. In contrast, the latent data constructed by the proposed method jointly reconstructs both the observed predictors and target variables via generative-style nonlinear functions. Dual-form solutions of the nonlinear functions and a stochastic gradient descent technique which learns the low dimensionality data are introduced. The validity of our method is demonstrated in a series of experiments against five state-of-art baselines, clearly improving the prediction accuracy in eleven real-life data sets. © 2014 Springer International Publishing. Source

Grabocka J.,University of Hildesheim | Bedalli E.,University of Tirana | Bedalli E.,Aleksander Xhuvani University of Elbasan | Schmidt-Thieme L.,University of Hildesheim
Advances in Intelligent Systems and Computing | Year: 2013

Time-series classification has gained wide attention within the Machine Learning community, due to its large range of applicability varying from medical diagnosis, financial markets, up to shape and trajectory classification. The current state-of-art methods applied in time-series classification rely on detecting similar instances through neighboring algorithms. Dynamic Time Warping (DTW) is a similarity measure that can identify the similarity of two time-series, through the computation of the optimal warping alignment of time point pairs, therefore DTW is immune towards patterns shifted in time or distorted in size/shape. Unfortunately the classification time complexity of computing the DTW distance of two series is quadratic, subsequently DTW based nearest neighbor classification deteriorates to quartic order of time complexity per test set. The high time complexity order causes the classification of long time series to be practically infeasible. In this study we propose a fast linear classification complexity method. Our method projects the original data to a reduced latent dimensionality using matrix factorization, while the factorization is learned efficiently via stochastic gradient descent with fast convergence rates and early stopping. The latent data dimensionality is set to be as low as the cardinality of the label variable. Finally, Support Vector Machines with polynomial kernels are applied to classify the reduced dimensionality data. Experimentations over long time series datasets from the UCR collection demonstrate the superiority of our method, which is orders of magnitude faster than baselines while being superior even in terms of classification accuracy. © 2013 Springer-Verlag. Source

Kushi E.,Aleksander Xhuvani University of Elbasan
Journal of Environmental Protection and Ecology | Year: 2011

Nowadays the development of tourism must be considered as tightly connected with environment protection, which means the protection of the basic values of the tourism destinations. Since tourism is one of the most important industries in Albania, this paper provides an analysis of the relationship between sustainable tourism and environment protection in this country. It highlights that while tourism provides considerable economic benefits for many regions and communities in Albania, its rapid expansion can also be responsible for adverse environmental impact. The paper also identifies that the main environmental impacts of tourism are the pressure on natural resources, pollution and waste generation and damage to ecosystems. Furthermore, it shows that environmental degradation poses threat to tourism. Therefore, the development of the international tourism market in Albania needs to be cautious, as running for short-term profits, that further damage the environment, exposes a high risk of failure for the tourist activities. This may solely create an image with negative effects in long run that will require time to be reversed. Finally, the analysis suggests the need to promote sustainable tourism development in order to minimise its environmental impacts and to ensure more sustainable management of natural resources. Source

Devolli A.,Aleksander Xhuvani University of Elbasan | Shabani L.,University of Tirana
Journal of Environmental Protection and Ecology | Year: 2011

The hygiene control in a brewery ensures a very good quality of the final product and minimises the risk of contamination. Microbiological control is a regulative system, which function is the prevention and elemination of contaminating microorganisms to protect and to ensure what is the most important for us, the quality of the beer. It is important to detect as soon as possible the dangerous microorganisms that affect the quality, through constant hygienic control, and to optimise the cleaning procedures to improve the economic efficiency. The hygienic control for detection of the presence of harmfull microorganisms always used rapid tests and microbiological cultivation methods. A total control of hygenic conditions in a brewery, from the equipment used in the beer production in boiling of wort, fermentation process, filtration, packing of the product to all environment of the brewery has been done. The present study examines the growth of microorganisms in a brewery. Beer spoilage microorganisms such as lactic acid and acetic acid bacteria, enterobacteria and yeasts were shown to be present on the materials surface of the brewing process. To achieve this goal, are done a lot of microbiological analyses using some standard and selective media, suitable for the examination of contaminanting microorganisms. The results of this study reveal that the hygiene control in a brewery is very important to provide a high-quality beer. Source

Discover hidden collaborations