Search

Your search keyword '"Anghel, Andreea"' showing total 160 results
Start Over Author "Anghel, Andreea"

Refine your results

more »

more »

more »

more »
160 results on '"Anghel, Andreea"'

1. Graph Feature Preprocessor: Real-time Extraction of Subgraph-based Features from Transaction Graphs

Author

Blanuša, Jovan, Baraja, Maximo Cravero, Anghel, Andreea, von Niederhäusern, Luc, Altman, Erik, Pozidis, Haris, and Atasu, Kubilay

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence
Abstract

In this paper, we present "Graph Feature Preprocessor", a software library for detecting typical money laundering and fraud patterns in financial transaction graphs in real time. These patterns are used to produce a rich set of transaction features for downstream machine learning training and inference tasks such as money laundering detection. We show that our enriched transaction features dramatically improve the prediction accuracy of gradient-boosting-based machine learning models. Our library exploits multicore parallelism, maintains a dynamic in-memory graph, and efficiently mines subgraph patterns in the incoming transaction stream, which enables it to be operated in a streaming manner. We evaluate our library using highly-imbalanced synthetic anti-money laundering (AML) and real-life Ethereum phishing datasets. In these datasets, the proportion of illicit transactions is very small, which makes the learning process challenging. Our solution, which combines our Graph Feature Preprocessor and gradient-boosting-based machine learning models, is able to detect these illicit transactions with higher minority-class F1 scores than standard graph neural networks. In addition, the end-to-end throughput rate of our solution executed on a multicore CPU outperforms the graph neural network baselines executed on a powerful V100 GPU. Overall, the combination of high accuracy, a high throughput rate, and low latency of our solution demonstrates the practical value of our library in real-world applications. Graph Feature Preprocessor has been integrated into IBM mainframe software products, namely "IBM Cloud Pak for Data on Z" and "AI Toolkit for IBM Z and LinuxONE".

Published
2024

2. Realistic Synthetic Financial Transactions for Anti-Money Laundering Models

Author

Altman, Erik, Blanuša, Jovan, von Niederhäusern, Luc, Egressy, Béni, Anghel, Andreea, and Atasu, Kubilay

Subjects
Computer Science - Artificial Intelligence, Computer Science - Machine Learning, Quantitative Finance - Computational Finance
Abstract

With the widespread digitization of finance and the increasing popularity of cryptocurrencies, the sophistication of fraud schemes devised by cybercriminals is growing. Money laundering -- the movement of illicit funds to conceal their origins -- can cross bank and national boundaries, producing complex transaction patterns. The UN estimates 2-5\% of global GDP or \$0.8 - \$2.0 trillion dollars are laundered globally each year. Unfortunately, real data to train machine learning models to detect laundering is generally not available, and previous synthetic data generators have had significant shortcomings. A realistic, standardized, publicly-available benchmark is needed for comparing models and for the advancement of the area. To this end, this paper contributes a synthetic financial transaction dataset generator and a set of synthetically generated AML (Anti-Money Laundering) datasets. We have calibrated this agent-based generator to match real transactions as closely as possible and made the datasets public. We describe the generator in detail and demonstrate how the datasets generated can help compare different machine learning models in terms of their AML abilities. In a key way, using synthetic data in these comparisons can be even better than using real data: the ground truth labels are complete, whilst many laundering transactions in real data are never detected.

Published
2023

3. Search-based Methods for Multi-Cloud Configuration

Author

Łazuka, Małgorzata, Parnell, Thomas, Anghel, Andreea, and Pozidis, Haralampos

Subjects
Computer Science - Distributed, Parallel, and Cluster Computing, Computer Science - Machine Learning
Abstract

Multi-cloud computing has become increasingly popular with enterprises looking to avoid vendor lock-in. While most cloud providers offer similar functionality, they may differ significantly in terms of performance and/or cost. A customer looking to benefit from such differences will naturally want to solve the multi-cloud configuration problem: given a workload, which cloud provider should be chosen and how should its nodes be configured in order to minimize runtime or cost? In this work, we consider solutions to this optimization problem. We develop and evaluate possible adaptations of state-of-the-art cloud configuration solutions to the multi-cloud domain. Furthermore, we identify an analogy between multi-cloud configuration and the selection-configuration problems commonly studied in the automated machine learning (AutoML) field. Inspired by this connection, we utilize popular optimizers from AutoML to solve multi-cloud configuration. Finally, we propose a new algorithm for solving multi-cloud configuration, CloudBandit (CB). It treats the outer problem of cloud provider selection as a best-arm identification problem, in which each arm pull corresponds to running an arbitrary black-box optimizer on the inner problem of node configuration. Our experiments indicate that (a) many state-of-the-art cloud configuration solutions can be adapted to multi-cloud, with best results obtained for adaptations which utilize the hierarchical structure of the multi-cloud configuration domain, (b) hierarchical methods from AutoML can be used for the multi-cloud configuration task and can outperform state-of-the-art cloud configuration solutions and (c) CB achieves competitive or lower regret relative to other tested algorithms, whilst also identifying configurations that have 65% lower median cost and 20% lower median time in production, compared to choosing a random provider and configuration., Comment: Submitted to IEEE Cloud 2022

Published
2022

4. SnapBoost: A Heterogeneous Boosting Machine

Author

Parnell, Thomas, Anghel, Andreea, Lazuka, Malgorzata, Ioannou, Nikolas, Kurella, Sebastian, Agarwal, Peshal, Papandreou, Nikolaos, and Pozidis, Haralampos

Subjects
Computer Science - Machine Learning, Statistics - Machine Learning
Abstract

Modern gradient boosting software frameworks, such as XGBoost and LightGBM, implement Newton descent in a functional space. At each boosting iteration, their goal is to find the base hypothesis, selected from some base hypothesis class, that is closest to the Newton descent direction in a Euclidean sense. Typically, the base hypothesis class is fixed to be all binary decision trees up to a given depth. In this work, we study a Heterogeneous Newton Boosting Machine (HNBM) in which the base hypothesis class may vary across boosting iterations. Specifically, at each boosting iteration, the base hypothesis class is chosen, from a fixed set of subclasses, by sampling from a probability distribution. We derive a global linear convergence rate for the HNBM under certain assumptions, and show that it agrees with existing rates for Newton's method when the Newton direction can be perfectly fitted by the base hypothesis at each boosting iteration. We then describe a particular realization of a HNBM, SnapBoost, that, at each boosting iteration, randomly selects between either a decision tree of variable depth or a linear regressor with random Fourier features. We describe how SnapBoost is implemented, with a focus on the training complexity. Finally, we present experimental results, using OpenML and Kaggle datasets, that show that SnapBoost is able to achieve better generalization loss than competing boosting frameworks, without taking significantly longer to tune.

Published
2020

5. Breadth-first, Depth-next Training of Random Forests

Author

Anghel, Andreea, Ioannou, Nikolas, Parnell, Thomas, Papandreou, Nikolaos, Mendler-Dünner, Celestine, and Pozidis, Haris

Subjects
Computer Science - Machine Learning, Statistics - Machine Learning
Abstract

In this paper we analyze, evaluate, and improve the performance of training Random Forest (RF) models on modern CPU architectures. An exact, state-of-the-art binary decision tree building algorithm is used as the basis of this study. Firstly, we investigate the trade-offs between using different tree building algorithms, namely breadth-first-search (BFS) and depth-search-first (DFS). We design a novel, dynamic, hybrid BFS-DFS algorithm and demonstrate that it performs better than both BFS and DFS, and is more robust in the presence of workloads with different characteristics. Secondly, we identify CPU performance bottlenecks when generating trees using this approach, and propose optimizations to alleviate them. The proposed hybrid tree building algorithm for RF is implemented in the Snap Machine Learning framework, and speeds up the training of RFs by 7.8x on average when compared to state-of-the-art RF solvers (sklearn, H2O, and xgboost) on a range of datasets, RF configurations, and multi-core CPU architectures.

Published
2019

6. A multidisciplinary investigation of the Konopi mansion in Odvos, Romania

Author

Popa Anca, Anghel Andreea, Onescu Iasmina, and Mosoarca Marius

Subjects
Engineering (General). Civil engineering (General), TA1-2040
Abstract

Architectural heritage represents one of the most important assets of a local community, as it embodies their authenticity and spirit of place. This paper takes a look at the nobiliary manor houses in the extra-urban area of the county Arad, which is located in the western part of Romania, close to the border with Hungary. Along the Mures River, there are many manor houses and the interest and curiosity for this type of former residences of the nobility in Romania has increased in the last few years, whether we are talking about entrepreneurs who want to restore them to their former glory or passers-by curious to discover hidden treasures in rural areas. This architecture, that of the extra-urban noble residences, is a particular feature of the Modernity, very valuable in terms of cultural heritage because it was the result of a culture of the European area. It can be said that the noble architecture of the region Banat-Crișana is a result of the regional interpenetration of Central European canons. Early modern noble residences –known by their classical architecture – are characteristic of rural landscape of the Mures Valley, with more than ten such mansions and buildings of high historical value. Of these, the Konopi mansion, whose architecture makes its presence felt in the consciousness of the community, as it marks the silhouette of the Mures valley. The aim of this paper is to study the manor house and the ensemble Konopi and the way in which the residences were thought of and perceived at the time, both from the point of view of their function as a dwelling and as a symbol of the family power in the countryside. The overall objective of this paper is to identify the ways to intervene on the Konopi Castle complex and to generate a strategy for its long-lasting functioning so that the historical monument of regional and local interest can be left for future generations.

Published
2024
Full Text
View/download PDF

7. Sampling Acquisition Functions for Batch Bayesian Optimization

Author

De Palma, Alessandro, Mendler-Dünner, Celestine, Parnell, Thomas, Anghel, Andreea, and Pozidis, Haralampos

Subjects
Computer Science - Machine Learning, Statistics - Machine Learning
Abstract

We present Acquisition Thompson Sampling (ATS), a novel technique for batch Bayesian Optimization (BO) based on the idea of sampling multiple acquisition functions from a stochastic process. We define this process through the dependency of the acquisition functions on a set of model hyper-parameters. ATS is conceptually simple, straightforward to implement and, unlike other batch BO methods, it can be employed to parallelize any sequential acquisition function or to make existing parallel methods scale further. We present experiments on a variety of benchmark functions and on the hyper-parameter optimization of a popular gradient boosting tree algorithm. These demonstrate the advantages of ATS with respect to classical parallel Thompson Sampling for BO, its competitiveness with two state-of-the-art batch BO methods, and its effectiveness if applied to existing parallel BO algorithms., Comment: Presented at BNP@NeurIPS 2018

Published
2019

8. Benchmarking and Optimization of Gradient Boosting Decision Tree Algorithms

Author

Anghel, Andreea, Papandreou, Nikolaos, Parnell, Thomas, De Palma, Alessandro, and Pozidis, Haralampos

Subjects
Computer Science - Machine Learning, Statistics - Machine Learning
Abstract

Gradient boosting decision trees (GBDTs) have seen widespread adoption in academia, industry and competitive data science due to their state-of-the-art performance in many machine learning tasks. One relative downside to these models is the large number of hyper-parameters that they expose to the end-user. To maximize the predictive power of GBDT models, one must either manually tune the hyper-parameters, or utilize automated techniques such as those based on Bayesian optimization. Both of these approaches are time-consuming since they involve repeatably training the model for different sets of hyper-parameters. A number of software GBDT packages have started to offer GPU acceleration which can help to alleviate this problem. In this paper, we consider three such packages: XGBoost, LightGBM and Catboost. Firstly, we evaluate the performance of the GPU acceleration provided by these packages using large-scale datasets with varying shapes, sparsities and learning tasks. Then, we compare the packages in the context of hyper-parameter optimization, both in terms of how quickly each package converges to a good validation score, and in terms of generalization performance., Comment: Workshop on Systems for ML and Open Source Software at NeurIPS 2018, Montreal, Canada

Published
2018

9. Snap ML: A Hierarchical Framework for Machine Learning

Author

Dünner, Celestine, Parnell, Thomas, Sarigiannis, Dimitrios, Ioannou, Nikolas, Anghel, Andreea, Ravi, Gummadi, Kandasamy, Madhusudanan, and Pozidis, Haralampos

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Distributed, Parallel, and Cluster Computing
Abstract

We describe a new software framework for fast training of generalized linear models. The framework, named Snap Machine Learning (Snap ML), combines recent advances in machine learning systems and algorithms in a nested manner to reflect the hierarchical architecture of modern computing systems. We prove theoretically that such a hierarchical system can accelerate training in distributed environments where intra-node communication is cheaper than inter-node communication. Additionally, we provide a review of the implementation of Snap ML in terms of GPU acceleration, pipelining, communication patterns and software architecture, highlighting aspects that were critical for achieving high performance. We evaluate the performance of Snap ML in both single-node and multi-node environments, quantifying the benefit of the hierarchical scheme and the data streaming functionality, and comparing with other widely-used machine learning software frameworks. Finally, we present a logistic regression benchmark on the Criteo Terabyte Click Logs dataset and show that Snap ML achieves the same test loss an order of magnitude faster than any of the previously reported results, including those obtained using TensorFlow and scikit-learn., Comment: in Proceedings of the Thirty-Second Conference on Neural Information Processing Systems (NeurIPS 2018)

Published
2018

10. An Insight into Practices Associated with the Control of Internal Parasites in the Dairy Goat Herds of Romania: A Questionnaire Survey.

Author

Potârniche, Adrian-Valentin, Cerbu, Constantin, Olah, Diana, Trif, Emilia, D'Amico, Gianluca, Györke, Adriana, Mickiewicz, Marcin, Nowek, Zofia, Czopowicz, Michał, Nadolu, Dorina, Anghel, Andreea Hortanse, and Kaba, Jarosław

Subjects
ANIMAL herds, EDUCATION of farmers, GOAT diseases, PARASITIC diseases, ANTHELMINTICS, GOATS
Abstract

Simple Summary: Parasitic infections, especially those caused by a particular group of roundworms called gastrointestinal nematodes, are one of the most important diseases of goats worldwide. The widespread and uncontrolled use of deworming drugs (anthelmintics) makes nematodes increasingly resistant, and this phenomenon is called anthelmintic resistance. Romania has one of the highest goat populations in Europe, but little is known about how widespread and resistant gastrointestinal parasites are in this country. That is why we decided to carry out this survey—to find out how Romanian farmers fight parasites in their goats and what their opinion is on the presence of parasites and the effectiveness of deworming drugs in their herds. Our findings are disappointing. Most farmers deworm goats when they think it is necessary, not when laboratory tests show that they really need it. Moreover, they use deworming drugs too often and at too low doses. Such practices are very likely to stimulate parasites to develop anthelmintic resistance, and only intensive education programs for farmers and veterinarians can stop this process. The widespread and uncontrolled use of anthelmintic products has contributed to the emergence of anthelmintic resistance (AR). This phenomenon globally threatens the productivity and welfare of small ruminants. A questionnaire consisting of 34 questions was handed to 234 goat farmers across Romania to gain insight into control practices against internal parasites and the farmers' perception of the parasitic infections present in their herds and the efficacy of anthelmintic treatments. The majority of farmers (88.5%) admitted they had never submitted fecal samples for parasitological laboratory analysis, and 77.4% had treated the animals on their own. In general, the farmers dewormed their goats based on visual body weight estimation. Prophylactic anthelmintic treatment was practiced by more than 85% of the farmers. A traditional control approach based on treating the entire herd at fixed time intervals is widespread among Romanian goat and sheep farmers. The most commonly used anthelmintic drugs in the previous 3 years (2021–2023) were benzimidazoles (85.5%) and macrocyclic lactones (81.6%). Poor anthelmintic efficacy was suspected by 14.5% of farmers, and the minority (18.0%) considered internal parasites as a problem in their herds. Regarding the farmers' perception of the presence of parasites, there was a significant level of uncertainty. This is the first survey carried out in Romanian goat herds, and it provides up-to-date information on practices aimed at controlling internal parasites. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

11. Accelerated ML-Assisted Tumor Detection in High-Resolution Histopathology Images

Author

Ioannou, Nikolas, Stanisavljevic, Milos, Anghel, Andreea, Papandreou, Nikolaos, Andani, Sonali, Rüschoff, Jan Hendrik, Wild, Peter, Gabrani, Maria, Pozidis, Haralampos, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Woeginger, Gerhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Shen, Dinggang, editor, Liu, Tianming, editor, Peters, Terry M., editor, Staib, Lawrence H., editor, Essert, Caroline, editor, Zhou, Sean, editor, Yap, Pew-Thian, editor, and Khan, Ali, editor

Published
2019
Full Text
View/download PDF

12. A Fast and Scalable Pipeline for Stain Normalization of Whole-Slide Images in Histopathology

Author

Stanisavljevic, Milos, Anghel, Andreea, Papandreou, Nikolaos, Andani, Sonali, Pati, Pushpak, Rüschoff, Jan Hendrik, Wild, Peter, Gabrani, Maria, Pozidis, Haralampos, Hutchison, David, Series Editor, Kanade, Takeo, Series Editor, Kittler, Josef, Series Editor, Kleinberg, Jon M., Series Editor, Mattern, Friedemann, Series Editor, Mitchell, John C., Series Editor, Naor, Moni, Series Editor, Pandu Rangan, C., Series Editor, Steffen, Bernhard, Series Editor, Terzopoulos, Demetri, Series Editor, Tygar, Doug, Series Editor, Leal-Taixé, Laura, editor, and Roth, Stefan, editor

Published
2019
Full Text
View/download PDF

13. STUDY ON THE PRODUCTIVITY OF GOATS INCLUDED IN SELECTION BREEDING WORKS IN OUR COUNTRY.

Author

MARINICĂ, Laura, NADOLU, Dorina, ANGHEL, Andreea Hortanse, and PASCAL, Constantin

Subjects
GOAT breeds, GOATS, GOAT milk, ANIMAL herds, DAIRY processing
Abstract

The growing interest in goat milk in our country has required the expansion of research in this species, covering various areas such as nutrition, reproduction, exploitation conditions, milk and meat processing methods, and others. The purpose of this study is to analyze the distribution of goats included in selection works in relation to the total number of goats at national level reported by the National Institute of Statistics, with the following aspects being evidenced: the proportion of goats included in selection works, the distribution by breeds of the number of goats in the genealogical register and their average milk/lactation/breed/region production. The study is of a statistical nature and focuses exclusively on the herds in the Genealogical Register on 31.08.2023, and the results show that the South East Region stands out with an impressive herd of goats, representing 36.16% compared to the Centre Region where, in the selection work, only 2.88% goats are included. The conclusions presented provide a detailed numerical perspective and the potential of each region within the national goat milk scene. These consolidating insights are essential pillars for guiding future strategies and developments in this area. [ABSTRACT FROM AUTHOR]

Published
2024

14. Acceleration of Decision-Tree Ensemble Models on the IBM Telum Processor

Author

Papandreou, Nikolaos, primary, van Lunteren, Jan, additional, Anghel, Andreea, additional, Parnell, Thomas, additional, Petermann, Martin, additional, Stanisavljevic, Milos, additional, Lichtenau, Cedric, additional, Sica, Andrew, additional, Röhm, Dominic, additional, Tzortzatos, Elpida, additional, and Pozidis, Haralampos, additional

Published
2023
Full Text
View/download PDF

15. Synthetic Traffic Model of the Graph500 Communications

Author

Fuentes, Pablo, Vallejo, Enrique, Bosque, José Luis, Beivide, Ramón, Anghel, Andreea, Rodríguez, Germán, Gusat, Mitch, Minkenberg, Cyriel, Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Weikum, Gerhard, Series editor, Carretero, Jesus, editor, Garcia-Blas, Javier, editor, Ko, Ryan K.L., editor, Mueller, Peter, editor, and Nakano, Koji, editor

Published
2016
Full Text
View/download PDF

17. Quantifying Communication in Graph Analytics

Author

Anghel, Andreea, Rodriguez, German, Prisacari, Bogdan, Minkenberg, Cyriel, Dittmann, Gero, Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Weikum, Gerhard, Series editor, Kunkel, Julian M., editor, and Ludwig, Thomas, editor

Published
2015
Full Text
View/download PDF

18. Polymorphisms in autophagy genes and active pulmonary tuberculosis susceptibility in Romania

Author

Cucu Mihai Gabriel, Streața Ioana, Riza Anca Lelia, Cimpoeru Alina Liliana, Șerban-Șoșoi Simona, Ciocoiu Adela, Pleșea Răzvan Mihail, Popescu Elena Leocadia, Dorobanțu Ștefania, Anghel Andreea, Stroe Aida Maria, Ștefan Andreea Nicoleta, Cioboată Ramona, Băzăvan Ileana, Ciontea Marius Sorin, Căpitănescu Iulia, Olteanu Mihai, Nițu Mimi, Burada Florin, Tătaru Tiberiu, Netea Mihai, van Crevel Reinout, Olaru Marian, Mixich Francisc, and Ioana Mihai

Subjects
tuberculosis, gene polymorphism, autophagy, innate immune system, allele, Medicine
Abstract

Autophagy, a homeostatic process involved in nutrient regeneration and immune responses, may be involved in intracellular killing of M. tuberculosis. Several studies linked variation in autophagy genes with susceptibility to pulmonary tuberculosis, but others did not confirm these findings.

Published
2017
Full Text
View/download PDF

19. Scalable High Resolution Traffic Heatmaps: Coherent Queue Visualization for Datacenters

Author

Anghel, Andreea, Birke, Robert, Gusat, Mitch, Hutchison, David, editor, Kanade, Takeo, editor, Kittler, Josef, editor, Kleinberg, Jon M., editor, Kobsa, Alfred, editor, Mattern, Friedemann, editor, Mitchell, John C., editor, Naor, Moni, editor, Nierstrasz, Oscar, editor, Pandu Rangan, C., editor, Steffen, Bernhard, editor, Terzopoulos, Demetri, editor, Tygar, Doug, editor, Weikum, Gerhard, editor, Dainotti, Alberto, editor, Mahanti, Anirban, editor, and Uhlig, Steve, editor

Published
2014
Full Text
View/download PDF

24. EFFECT OF BREED, AGE AND FOOD ON REPRODUCTIVE EFFICIENCY OF FEMALE SHEEP OF THE TSIGAI - RUSTY VARIETY PUREBREED AND THEIR CROSS WITH THE SUFFOLK AND GERMAN BLACKFACE.

Author

ILIȘIU, Cristian-Vasile, ILIȘIU, Elena, ANGHEL, Andreea-Hortensa, ILIȘIU, Vasile-Călin, NADOLU, Dorina, and Zoia-ZAMFIR, Camelia

Subjects
SHEEP, ANIMAL breeding, SHEEP milk, ANIMAL weaning, ANIMAL breeds, HAY as feed, FEMALES
Abstract

Fertility, prolificacy and weaning rate were compared for Tsigai - rusty variety (TIRU) purebreed ewes and their Suffolk (50%) × German Blackface (37.5%) x Tsigai - rusty variety (12.5%) (S x BF x TIRU) contemporaries, mated at 8 and 18 months of age. The fecundity, prolificacy and weaning rate for animals mated at 8 months of age and fed in the shelter with hay and corn-barley based concentrate was 78.43%, 105.0% and 78.57% to females from TIRU and 80.0%, 100.0% and 54.17% for animals of the S x BF x TIRU genotype, respectively. For animals of the two breeds mated at 18 months of age and grazing on pastures, the fecundity, prolificacy and weaning rate was 64.71%, 100.0% and 87.88% to females from TIRU and 73.33%, 104.55% and 78.26% for animals of the S x BF x TIRU genotype, respectively. 88.24% females from TIRU and 90.0% from S x BF x TIRU genotype, mated at 8 months, have lambed once up to age of 23-24 months, and 56.86% females from TIRU and 56.66% from the animals from S x BF x TIRU genotype have lambed twice. [ABSTRACT FROM AUTHOR]

Published
2023

27. Correlations between Biochemical Markers in Thyroid Pathology with those in Carbohydrates Metabolism

Author

Anghel Andreea, Rosoiu Natalia, Voiculescu Anne-Lise Mihaela, Gurgas Leonard, and Petcu Lucian

Subjects
endocrine system, 030222 orthopedics, glycosylated haemoglobin, endocrine system diseases, Thyroid pathology, thyroid dysfunction, business.industry, Physiology, Carbohydrate metabolism, tsh - thyroid stimulating hormone, General Biochemistry, Genetics and Molecular Biology, 03 medical and health sciences, 0302 clinical medicine, diabete, Medicine, 030212 general & internal medicine, business, Biochemical markers
Abstract

Thyroid dysfunction is increasingly found in patients with diabetes. Diabetes can affect thyroid function to a variable extent and unrecognized thyroid dysfunction, not only worsens metabolic control, but also prevents the management of diabetes. The aim of this study was to screen the incidence of thyroid disease in Constanta County, Romania in patients diagnosed with type II diabetes, in order to assess the tendency to associate thyroid hormone dysfunction with the diabetic process by correlating glycemic parameters and thyroid profile in serum. The study included a number of 153 patients who presented within 3 months for blood tests, in compliance with GDPR rules. The following parameters were processed from these patients: FT3 - free triiodothyronine, FT4 - free thyroxine, TSH - thyroid stimulating hormone, Anti-TPO - thyroid peroxidase antibodies (anti-thyroid peroxidase), Serum glucose, Glycosylated haemoglobin - HbA1C. In patients of both sexes studied, the mean TSH was 3.15 IU / ml, males’ TSH mean being 2.82 IU / ml and higher, but not above the maximum limit. Blood glucose ranged from an average of 92.11 mg / dl in the 153 patients, 92.41 mg / dl in male patients and 93.82 mg / dl in female patients. Glycated haemoglobin (%) ranged from 6.79% in female patients to 6.21% in male patients, with a mean of 6.5%. Pathology derived from disorders of carbohydrate metabolism may be present in patients with thyroid changes.

Published
2020

28. The Influence of Thermal Processing on Antioxidant Capacity of Tomatoes Based Products

Author

Margaritti Doina, Tomos Sirma, Schroder Verginica, Anghel Andreea, and Buşuricu Florica

Subjects
0106 biological sciences, business.industry, antioxidant activity, 04 agricultural and veterinary sciences, lycopene, 040401 food science, 01 natural sciences, General Biochemistry, Genetics and Molecular Biology, Antioxidant capacity, 0404 agricultural biotechnology, 010608 biotechnology, Medicine, thermal processing, Food science, tomatoes, business
Abstract

We intend to study the influence of thermal processing on the antioxidant capacity of tomato products. The products were analyzed for lycopene content and antioxidant activity. Lycopene was extracted using a mixture of hexane: ethyl alcohol: acetone, in ration of 1:25 (w: v). The antioxidant capacity was evaluated by the DPPH method. The obtained results show a higher quantity of lycopene in the processed products (8, 4800 ±0, 48079 - 11,8217 ±0. 37010 mg/100g), compared to the fresh ones (2.9400±0.1510-4.0933±0.16773 mg/100g) and the antioxidant activity registers higher values in the aqueous extracts (21-45%). The thermal processing concentrates the content of lycopene and preserves the antioxidant activity, so that for the optimal nutritional intake of antioxidant phytonutrients, the consumption of processed tomatoes is indicated, to the detriment of the fresh tomatoes.

Published
2019

29. The Influence of Sheep Age at First Calving on the Productive Performance of Lambs from Tsigai breed.

Author

Ilișiu, Elena, Anghel, Andreea-Hortanse, Ilișiu, Vasile-Călin, Zamfir, Camelia-Zoia, Ilișiu, Cristian-Vasile, Mare, Daniela-Rodica, Nadolu, Dorina, and Chirteș, Ion-Dumitru

Subjects
*LAMB (Meat), *LAMBS, *SHEEP, *MEAT quality, *ANIMAL weaning, *SHEEP breeding, *BREEDING
Abstract

The aim of this study was to determine the effects of age on reproductive indices and on growth performances, carcass and meat quality of Tsigai lambs coming from young females (8 months age), in order to improve meat quality and meat sensory characteristics of lambs. 222 adult ewes and 51 young females were pursued in the breeding and calving season 2020-2021. The fecundity, prolificacy and weaning rate was 78.43%, 105% and 78.57% to young females and 94.60%, 105.24% and 91.40% to adult ewes, respectively. Thirty lambs (L1, n = 14 heads coming from young females and L2, n = 16 heads from adult ewes) were used in the experiment from birth up to end of intensive fattening of 100 days. No significant differences (p > 0.05) were found between the two lots with regard at final weight and average daily gain during fattening period, but significant differences (p < 0.001) were found for average daily gain from birth to the end of fattening. Significant differences (p <0.001) were also recorded between meat from the two groups with regard to juiciness and overall difference. The meat of lambs from L1 has showed improved eating qualities, resulting in a more juicy and tender meat, in which the specific lamb taste was attenuated. [ABSTRACT FROM AUTHOR]

Published
2022

30. DIGITAL VIRUSES VS BIOLOGICAL VIRUSES - A LITERATURE REVIEW.

Author

Anghel, Cătălin, Iacobescu, Paul, and Anghel, Andreea Alexandra

Subjects
LITERATURE reviews, ARTIFICIAL intelligence
Abstract

Computer viruses and biological viruses are bugs that make our lives difficult. This paper contains a brief review of the literature aimed to identify the relevant pieces of knowledge, differences and similarities between computer viruses and biological viruses. Also point out the potential intelligent behavior of both biological and computer viruses. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

31. QUANTITATIVE AND QUALITATIVE VARIATION OF SAANEN GOAT MILK KEEPED IN EXTENDED LACTATION FOR TWO YEARS.

Author

NADOLU, Dorina, ZAMFIR, Camelia Zoia, ANGHEL, Andreea Hortanse, and ILIȘIU, Elena

Subjects
GOAT milk, GOATS, LACTATION, MILK yield, GOAT farming, DOMESTIC markets
Abstract

Saanen goats, specialized in milk production, improved in terms of quantity and quality (protein and fat content), are normally milked for 270-300 days with an annual production of 650-1050 l of milk / lactation and an average protein content of 2.9% and fat content of 3.2%. This study was conducted in the first goat farm in Romania that initiated prolonged lactation of goats. The research was carried out on a batch of primiparous goats, maintained in an intensive exploitation system, which after the first calving were subdivided according to milk production measured for a week. Two batches of 128 goats with a production of 3-4 liters/day and 94 goats with a production of over 4 liters/day were established. By modulating the diet and the light regime, the goats were milked continuously for 690 to 742 days, monitoring the amount of milk, protein and lipid levels on a monthly basis. Along with controlled breeding programs, prolonged lactation contributes to ensuring on the domestic market a continuity in the supply of goat's milk, throughout the year. [ABSTRACT FROM AUTHOR]

Published
2022

34. REPRODUCTIVE PERFORMANCE IN ALPINE GOATS ACCORDING TO THE APPLICATION OF A SIMPLIFIED PROTOCOL FOR INDUCING ESTROUS INTO THE REPRODUCTIVE OUT OF SEASON.

Author

NADOLU, Dorina, ANGHEL, Andreea Hortanse, ZAMFIR, Camelia Zoia, and ILIȘIU, Elena

Subjects
*GOATS, *GOAT breeds, *FROZEN semen, *ARTIFICIAL insemination, *ABORTION, *ESTRUS
Abstract

Studies in the field of goat breeding have shown that the most widely used and effective way to induce and synchronize estrus in the breeding season is the hormonal method with Chronogest sponges and the administration of Folligon and a prostaglandin. In the present study, we looked at the reproductive performance of a batch of 148 Alpina goats synchronized in April with Chronogest sponges maintained intravaginally for 11 days and the injection of 400 IU Folligon without prostaglandin administration. Artificial insemination with frozen semen was performed 43 +/- 2 hours after the removal of the sponges and the following reproductive aspects were established: the degree of the cervix opening at the time of artificial insemination, ultrasound diagnosis of 50 days gestation after artificial insemination with the identification of cases of pseud-gestation (4.05%), the rate of calving (56.76%) and prolificacy (310.71%). In farm conditions, specific to our country, the exclusion of prostaglandin administration decreases the risk of abortions by lysis of luteal bodies in possibly pregnant goats at the time of treatment and insemination. [ABSTRACT FROM AUTHOR]

Published
2022

35. Nutritional quality of some non-alcoholic beverages from the Romanian market.

Author

Florica, Buşuricu, Verginica, Schroder, Doina, Margaritti, Hortensa, Anghel Andreea, and Sirma, Tomoş

Subjects
NUTRITIONAL value, NON-alcoholic beverages, HEALTH, FRUIT extracts
Abstract

In the recent years, young consumers choose to consume from the category of soft drinks, without acknowledging that their health may be endangered. To this end, we aim to contribute to the real scientific estimation of the nutrient intake from a number of 20 waters with vitamins and minerals having a reduced addition of fruit extract (1-3%), compared to juices that have 10-30 % fruits extract, from those sold in Romania. We analyzed the presentation of the nutritional declaration, the quality of the additives and the profiling of nutrients through the SENS system. The obtained results classify the products in the category of those with low energy value and in the category “significant sources of vitamins and minerals”; there are from food classes 2 and 4, as a nutritional profile. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

36. THE EFFECT OF DIET ON GROWTH PERFORMANCES, CARCASS AND MEAT QUALITY CHARACTERISTICS OF LAMBS FROM TSIGAI BREED.

Author

ZAMFIR, Camelia Zoia, ILIȘIU, Elena, CHIRTEȘ, Ion, ILIȘIU, Vasile Călin, ANGHEL, Andreea Hortensa, ILIȘIU, Cristian Vasile, and NADOLU, Dorina

Subjects
MEAT quality, LAMB (Meat), DIET, FISH growth
Abstract

The aim of this study was to determine the effects of diet on growth performances, carcass and meat quality of Tsigai lambs - rusty variety fed with different diets, in order to improve meat quality and meat sensory characteristics of lambs; forty male lambs (L1 and L2, n = 20 heads/group) were used in experiment from birth up to 5 months. No significant differences (p > 0.05) were found between the two lots with regard at final weight and average daily gain. The diet had a significant effect between L1 and L2 groups in terms of hot carcass weight and hot slaughter yield (p < 0.001), cold carcass weight and cold slaughter yield (p < 0.01), as well for commercial yield (p < 0.05). The juiciness and overall difference were strongly influenced by the diet administered in the present study, significant differences (p <0.001) being recorded between meat from the two groups. The diet administered to L1 influenced positively the eating qualities of lamb meat, resulting in a more juicy and tender meat, in which the specific lamb taste was attenuated. [ABSTRACT FROM AUTHOR]

Published
2022

39. Artificial Insemination of Carpathian Goats with Semen Preserved in Different Forms

Author

Anghel Andreea, Sonea Cristinel, Dorina Nadolu, and Anghelescu Claudiu

Subjects
Plant science, Ecology (disciplines), Fresh semen, Artificial insemination, medicine.medical_treatment, medicine, Zoology, Semen, General Medicine, Biology
Abstract

The artificial insemination (AI) technique is an essential tool in goat breeding programs, since it increases the efficiency of sire genetic evaluation and the extension of genetic improvements; at the same time, it enables control of parturition dates with a view to meeting market demands. The objective of this study was to determine the fecundity rate according to the mode of semen preservation before artificial insemination. The percentage of kidding was influenced by the duration of storage: 3 h = 88%; 14 h =821%.

Published
2016

40. Analytical Model for the Electrodynamical Forces inside Electrical Contacts of Commutation Devices

Author

Flaviu Mihai Frigura-Iliasa, Emil Lazarescu, Hannelore Filipescu, Alexandra Popescu, Lia Dolga, and Anghel Andreea

Subjects
Physics, 020209 energy, 020208 electrical & electronic engineering, High voltage, 02 engineering and technology, Mechanics, Electrical contacts, 0202 electrical engineering, electronic engineering, information engineering, Equipotential, Power engineering, Contact area, Low voltage, Voltage, Contactor
Abstract

This paper aims to present a new analytical model for the contact area of all switching devices, at all voltage and current levels. It was proposed in order to recalculate all the computational formulas related to the electrodynamical forces which inevitably appear in a mostly symmetric contact, with a different shape, other than a sharpen or pointed one. When applying this model, the very simplifying hypothesis of the constant current density on equipotential stiction surfaces for the crossing current is completely abandoned. These new expressions could experimentally be verified by means of a simple device. By using this model, we can study the “attraction”, electrodynamical forces that may appear in an electrical contact belonging to any switching device used in electrical and power engineering technique. This research could improve both design and maintenance of new and existing switching equipment like contactors, low voltage switches, oil submersed switches or even SF 6 high voltage ones.

Published
2019

41. Preliminary Results on Lambs Performances of Tsigai Breed Fed with Different Diets.

Author

Ilișiu, Elena, Anghel, Andreea-Hortanse, Ilișiu, Cristian-Vasile, Zoia-Zamfir, Camelia, Ilișiu, Vasile - Călin, and Nadolu, Dorina

Subjects
*LAMBS, *ULTRASONIC measurement, *EYE muscles, *LUMBAR vertebrae, *BIRTH weight, *SHEEP breeding
Abstract

The present study was conducted to determine the effect of ram, sex and diet on growth performances of Tsigai lambs - rusty variety from birth up to 5 months. The birth weight, weight at 28 days and weaning weight were observed on 212 lambs coming from 10 rams' groups. At 5 months age, the depth, eye muscle area of Longissimus dorsi, muscle perimeter and the backfat thickness covering this muscle at the cross of sectional area between the 12th and 13th ribs, and between 3-4 lumbar vertebrae was determined using ultrasonic measurements from 80 lambs (40 male from two lots - L1 and L2 - 20 heads/lot and 40 female - L3) fed with different diets. At this time, the weights of lambs at 5 months were also recorded. The overall men weight of lambs at 5 months was 42.23 kg at L1, 40.18 kg at L2 and 34.58 kg at L3. The weight at 5 months was found to be significantly affected by sex. The diet had no significant effect on growth performances of the lambs. The means for ultrasonic measurement was significantly affected by sex. The variation observed between lots means were significant for eye muscle depth at 3-4 lumbar vertebrae and for eye muscle area, but not significant for backfat thickness, and muscle perimeter. Regarding the phenotypic correlations between ultrasound measurements with weight at 5 months, from a total of 45 traits couples 57.78% are small correlation (0.00-0.30), 20.00% are medium to high correlations (0.31-0.60) and high correlations recorded 22.22 % (0.61-1.00). Important information on body composition of lambs can be obtained practically by ultrasonic measurements on live animals. When combined in a breeding program with lamb market weights, these measurements will provide a way to increase both meat yield and the quality of Tsigai lambs. [ABSTRACT FROM AUTHOR]

Published
2021

42. Snap ML: A Hierarchical Framework for Machine Learning

Author

D��nner, Celestine, Parnell, Thomas, Sarigiannis, Dimitrios, Ioannou, Nikolas, Anghel, Andreea, Ravi, Gummadi, Kandasamy, Madhusudanan, and Pozidis, Haralampos

Subjects
FOS: Computer and information sciences, Computer Science - Machine Learning, Artificial Intelligence (cs.AI), Computer Science - Distributed, Parallel, and Cluster Computing, Computer Science - Artificial Intelligence, Distributed, Parallel, and Cluster Computing (cs.DC), Machine Learning (cs.LG)
Abstract

We describe a new software framework for fast training of generalized linear models. The framework, named Snap Machine Learning (Snap ML), combines recent advances in machine learning systems and algorithms in a nested manner to reflect the hierarchical architecture of modern computing systems. We prove theoretically that such a hierarchical system can accelerate training in distributed environments where intra-node communication is cheaper than inter-node communication. Additionally, we provide a review of the implementation of Snap ML in terms of GPU acceleration, pipelining, communication patterns and software architecture, highlighting aspects that were critical for achieving high performance. We evaluate the performance of Snap ML in both single-node and multi-node environments, quantifying the benefit of the hierarchical scheme and the data streaming functionality, and comparing with other widely-used machine learning software frameworks. Finally, we present a logistic regression benchmark on the Criteo Terabyte Click Logs dataset and show that Snap ML achieves the same test loss an order of magnitude faster than any of the previously reported results, including those obtained using TensorFlow and scikit-learn., in Proceedings of the Thirty-Second Conference on Neural Information Processing Systems (NeurIPS 2018)

Published
2018

45. PRELIMINARY RESULTS OF ARTIFICIAL INSEMINATION WITH FRESH DILUTED SEMEN DURING NATURAL ESTROUS AT EWES.

Author

Nadolu, Dorina, Anghel, Andreea Hortanse, Tămâianu, B., Ilisiu, Elena, and Nacu, G.

Subjects
*EWES, *ARTIFICIAL insemination of sheep, *SEMEN analysis
Abstract

Based on the context of the development and extension of the ewe’s exploitation system and on the application of the European legislation en force regarding the conditions for reproduction and certification from genealogical registry, the main concern of the sheep breeders became the assisted breeding with rams of known origin and genetic value. Thus, the interest for artificial insemination has increased among sheep breeders who are facing a shortage of rams with known genetic value, pure breed, in the application of selection and breeding programs. By intensive use of 5 rams in the natural breeding of 195 Palas Merinos, the male exhaustion and a rate of return to estrous cycle of 53.84% were obtained. After a 5 days of stimulating feeding diet and sexual rest of the rams, the semen was collected and qualitatively analyzed. 105 ewes at the 2nd or 3rd estrous cycle were artificially inseminated. The insemination was performed during natural estrous, on the day of the detection of estrous, with diluted raw semen, with a motility of over 80%. The rate of non-returning to the estrous at 15-19 days from artificial insemination was of 84.76%. Diagnosis confirmation was made by transabdominal ultrasound at 45 - 50 days after artificial insemination and it indicated a gestation rate of 74.28%. [ABSTRACT FROM AUTHOR]

Published
2021

46. SEMEN QUALITY OF MATURE CARPATHIAN BUCKS DURING NON-BREADING SEASON.

Author

Anghel, Andreea Hortanse, Nadolu, Dorina, Tămâianu, B., Anghel, Fl., and Nacu, G.

Subjects
*SEMEN analysis, *ESTRUS, *ARTIFICIAL insemination
Abstract

The recognition of the differences specific to each breed and those caused by the estrous season in terms of the breeding performance of the bucks, leads to the improvement of the management of individual semen samples used in the artificial insemination programs. The aim of the study was to evaluate the quality of the semen collected during non-breeding season, from 3 Carpathian bucks, 3 years of age. The semen was collected by artificial vagina between June and July 2019. Two methods of semen dilution, with and without seminal plasma were compared, in order to highlight the effect of seminal plasma in the refrigeration process. Tris-based extender was used, with and without egg yolk. The quality of the semen was evaluated immediately after collection by assessing the volume ejaculated, concentration, pH and motility and viability immediately after collection and also at 24 and 48 hours, after refrigeration at a temperature of 4° C. The results showed a decrease of the cytological parameters (motility, viability) by 15-20% every 24 hours of refrigeration in the semen samples from both methods. In conclusion, for insemination during non-breeding season with refrigerated semen, within 48 hours after collection, there is no need for centrifugation in order to remove the seminal plasma during semen’s preparation process. [ABSTRACT FROM AUTHOR]

Published
2021

47. MONITORING THE QUALITATIVE PARAMETERS OF THE REFRIGERATED RAM SEMEN DURING NON-BREADING SEASON.

Author

Tămâianu, B., Anghel, Andreea Hortanse, Nadolu, Dorina, Ilișiu, Elena, and Nacu, G.

Subjects
*ARTIFICIAL insemination, *FROZEN semen, *RAMS
Abstract

The association of the biotechnology of artificial insemination with the conservation of the semen allows the use of a small number of males with which a large number of females can be inseminated. Artificial insemination gives numerous genetic and economic benefits for animal production, being the safest method to bring into the herd the superior genes, of valuable individuals. The aim of this study was to evaluate the effect of Tris-based extender containing 2% glycerol and 20% egg yolk on the motility and viability of rams' semen collected in non-breeding season and refrigerated at 4°C. During 30 days, a total of 16 ejaculates from 6 Texel rams were collected. The semen was collected by artificial vagina during May-June 2019. The main qualitative parameters: motility and viability were evaluated immediately after collection and also at 24, 48, 72, 96 and 120 hours. The results were statistically processed and the extender’s influence and storage time were analyzed. In conclusion, the Tris-based extender containing glycerol and egg yolk showed a satisfactory protective effect on rams' semen collected in non-breeding season and refrigerated at 4°C. Storage time (P <0.05) has significantly affected the qualitative parameters of the semen recording a 10-15% decrease of the qualitative parameters every 24 hours. [ABSTRACT FROM AUTHOR]

Published
2021

49. CHEMICAL COMPOSITION OF MILK AT CARPATHIAN GOATS UNDER SEMI-INTENSIVE FARMING CONDITIONS.

Author

Anghel, Andreea, Nadolu, Dorina, and Busuricu, Florica

Subjects
*GOAT milk, *AGRICULTURAL intensification, *PROTEIN content of milk
Abstract

The studies were performed on milk samples collected from 178 Carpathian females belonging to a farm from southeastern Romania. The females were grazed daily, benefiting from natural pasture, specific to the Dobrogea area. Before each milking, 100 g of special granules for goat's milk were administered. After weaning (April), the goats were milked twice a day for 5 months. Milk production control was carried out monthly, after the weaning of the kids, for three consecutive months, starting with May. For this, the amount of milk milked in the morning and in the evening was weighed individually. The computation of the total individual production was calculated by applying the Fleischmann formula in the case of milking the goat after a period of lactation, when the first milking is done after the flat-rate date of the end of lactation. For primiparous, the total individual milk production was between 124.35 kg and 199.76 kg with an average of 164.93 kg/head/day. At multiparous, the average total individual milk production was 168.32Kg. Biochemical parameters (fat, protein, lactose) were analyzed by ultrasonic method in a standardized analyzer for goat's milk. At primiparous, the fat concentration reaches a maximum of 3.12 g% in July, and at multiparous 3.36g% in June. The protein concentration had values between 2.87g and 3.08g% at primiparous and between 2.90g% and 3.22g% at multiparous. The results of our analyzes recorded a maximum of 4.36% for lactose in June for primiparous and 4.58g% for multiparous. The percentage of mineral substances varies between 0.65 g% and 0.67 g% at primiparous and between 0.64 and 0.65 g% at multiparous. [ABSTRACT FROM AUTHOR]

Published
2020

50. RESEARCH REGARDING THE SLAUGHTERING OUTPUT, TISSUE STRUCTURE AND THE QUALITY OF THE CARCASSES AT THE HYBRID LAMBS FOR MEAT COMPARATIVELY TO THEIR CONTEMPORARIES FROM THE PARENTAL BREEDS.

Author

NEGRE, Cristina Stefania, VICOVAN, Gabriel Petru, ANGHEL, Andreea, ENCIU, Ana, ZAMFIR, Camelia Zoia, STANCIU, Maria, and NICOLESCU, Alina Narcisa

Subjects
LAMB (Meat), ERECTOR spinae muscles, SLAUGHTERING, BREEDING, TISSUES
Abstract

The present work aims to present the superiority of the hybrid lambs for meat regarding the slaughtering output, the tissue structure and the quality of the carcasses, comparatively to their with contemporaries from the parental breeds. The research was made on lambs fattened until the living weight 40 kg/head from the breeds of Palas Merino, Palas Meat Breed and F1 hybrids of Palas Meat Breed x Palas Merino and F1 Suffolk x Palas Merino. The highest slaughtering output was that of the F1 hybrids Suffolk x Palas Merino, out of them R1 having the value of 51,21 % and R2 of 55,94 % besides 44,25 % and respectively 49,54 % at the contemporaries of Palas Merino. Also, these hybrids had also the best carcasses, being categorized by conformation in U class (very good carcasses) and by fattening degree in the 3rd class 3 (medium fat carcasses). [ABSTRACT FROM AUTHOR]

Published
2020

Catalog

Books, media, physical & digital resources
See catalog results