Marche Polytechnic University

About: Marche Polytechnic University is a education organization based out in Ancona, Italy. It is known for research contribution in the topics: Population & Cancer. The organization has 5905 authors who have published 15769 publications receiving 382286 citations. The organization is also known as: Universitá Politecnica delle Marche & Universita Politecnica delle Marche.

Non-Saccharomyces wine yeasts have a promising role in biotechnological approaches to winemaking

Abstract: Biotechnology as applied to winemaking includes several aspects of the fermentation industry, such as monitoring of microbial populations, use of selected starter cultures, and control of undesired yeasts. Over the last few decades, the control of microorganisms using biotechnological approaches has become of increasing importance in the winemaking field. The profusion of selected starter strains has allowed more extensive use of inoculated fermentations, with a consequent improvement in the control of fermentation combined with the use of new biotechnological processes in winemaking. As a consequence of this re-evaluation of the role of non-Saccharomyces yeasts in winemaking over the last few years, several studies have evaluated the use of controlled mixed fermentations using Saccharomyces and different, non-Saccharomyces, yeast species that are a part of the winemaking environment. In this context, mixed fermentations using controlled inoculations of Saccharomyces cerevisiae starter cultures and non-Saccharomyces yeasts represent a practical way towards improving wine complexity and enhancing specific characteristics of a wine. Another trait in the use of non-Saccharomyces yeasts in winemaking relates to the control of spoilage microorganisms. Indeed, more strict control of undesirable yeasts is required during the various phases of wine fermentation. Moreover, there is now increasing interest in the use of natural antimicrobial agents in foods and, in this context, killer yeasts might have important roles in the control of spontaneous and/or spoilage microflora. Thus, killer toxins appear to represent an attractive solution for use as antimicrobial agents, to partially, or even completely, substitution for chemical agent use even if application costs could limit their use in winemaking.

Historical changes of the Mediterranean Sea ecosystem: modelling the role and impact of primary productivity and fisheries changes over time.

Abstract: The Mediterranean Sea has been defined "under siege" because of intense pressures from multiple human activities; yet there is still insufficient information on the cumulative impact of these stressors on the ecosystem and its resources. We evaluate how the historical (1950-2011) trends of various ecosystems groups/species have been impacted by changes in primary productivity (PP) combined with fishing pressure. We investigate the whole Mediterranean Sea using a food web modelling approach. Results indicate that both changes in PP and fishing pressure played an important role in driving species dynamics. Yet, PP was the strongest driver upon the Mediterranean Sea ecosystem. This highlights the importance of bottom-up processes in controlling the biological characteristics of the region. We observe a reduction in abundance of important fish species (~34%, including commercial and non-commercial) and top predators (~41%), and increases of the organisms at the bottom of the food web (~23%). Ecological indicators, such as community biomass, trophic levels, catch and diversity indicators, reflect such changes and show overall ecosystem degradation over time. Since climate change and fishing pressure are expected to intensify in the Mediterranean Sea, this study constitutes a baseline reference for stepping forward in assessing the future management of the basin.

Unsupervised electric motor fault detection by using deep autoencoders

Abstract: Fault diagnosis of electric motors is a fundamental task for production line testing, and it is usually performed by experienced human operators. In the recent years, several methods have been proposed in the literature for detecting faults automatically. Deep neural networks have been successfully employed for this task, but, up to the authors ʼ knowledge, they have never been used in an unsupervised scenario. This paper proposes an unsupervised method for diagnosing faults of electric motors by using a novelty detection approach based on deep autoencoders. In the proposed method, vibration signals are acquired by using accelerometers and processed to extract Log-Mel coefficients as features. Autoencoders are trained by using normal data only, i.e., data that do not contain faults. Three different autoencoders architectures have been evaluated: the multi-layer perceptron ( MLP ) autoencoder, the convolutional neural network autoencoder, and the recurrent autoencoder composed of long short-term memory ( LSTM ) units. The experiments have been conducted by using a dataset created by the authors, and the proposed approaches have been compared to the one-class support vector machine ( OC-SVM ) algorithm. The performance has been evaluated in terms area under curve ( AUC ) of the receiver operating characteristic curve, and the results showed that all the autoencoder-based approaches outperform the OC-SVM algorithm. Moreover, the MLP autoencoder is the most performing architecture, achieving an AUC equal to 99.11 %.