The increasing average age of the population in most industrialized countries imposes a necessity for developing advanced and practical services using state-of-the-art technologies, dedicated to personal living spaces. In this paper, we introduce hierarchical distributed approach home care systems based on new paradigm known as Internet Things (IoT). proposed generic framework is supported by three-level data management model composed dew computing, fog cloud computing efficient flow IoT...

Smart grid is the process of applying ICT in order to optimize energy consumption and decrease loses. This paper presents a three tier Internet Thing based hierarchical framework for smart home, as reflection present lack intelligent solutions that do not fully use advantages technologies. Our aims extend home microgrid level, integrate all renewable distributed sources from achieve better optimization. As an extension traditional data processing, we define fog computing approach home....

Activity detection is becoming an integral part of many mobile applications. Therefore, the algorithms for this purpose should be lightweight to operate on or other wearable device, but accurate at same time. In paper, we develop a new algorithm activity based Long Short Term Memory networks, which able learn features from raw accelerometer data, completely bypassing process generating hand-crafted features. We evaluate our data collected in controlled setting, as well under field...

Air pollution monitoring and control is becoming a key priority in urban areas due to its substantial effect on human morbidity mortality. This paper presents system architecture for intelligent visualization future prediction by encompassing measurements meteorological parameters. First, model using spatial interpolation built. By adding parameters this further used identify the field evolution position of potential sources air pollution. Using deep learning techniques, provides predictions...

Nowadays, finding the Unmanned Aerial Vehicle (UAV) position in absence of GPS is attractive and challenging problem research community. In this paper, we present a novel algorithm for mini UAV indoor localization based on distance measurements between existing infrastructure consisting WiFi Access Points. Our uses two well-known techniques from literature: Multi-dimensional Scaling (MDS) Weighted Centroid Localization (WCL). Through extensive simulations have shown that our very suitable...

Forecasting air pollution is crucial for understanding the phenomenological and contextual variety of mechanisms underlying in a particular area or region. Analyzing high-dimensional data with spatial temporal dependencies pointed out as major challenge traditional machine learning approaches forecasting. The unprecedented advances deep employed on massive quantities IoT sensor raise high hopes future Drawing past current experiences solutions advanced growing body research topic, we...

Customer churn is one of the main problems in telecommunications industry. Several studies have shown that attracting new customers much more expensive than retaining existing ones. Therefore, companies are focusing on developing accurate and reliable predictive models to identify potential will near future. The aim this paper investigating reasons for telecommunication sector Macedonia. proposed methodology analysis prediction covers several phases: understanding business; selection, data...

Awareness of air pollution is one the key aspects modern smart cities. Policy makers, and other stakeholders, are often ignorant in their immediate surrounding its correlation to local environment micro-climate when making short- or long-term decisions. The Internet Things (IoT) paradigm provides a suitable general framework for monitoring as it incorporates sensor network containing static and/or mobile sensors measuring different pollutants. IoT architectures, although very powerful, have...

Complex networks have recently become the focus of research in many fields. Their structure reveals crucial information for nodes, how they connect and share information. In our work we analyze protein interaction as complex their functional modular later use that annotation proteins within network. We propose several graph representations network, each having different level complexity inclusion graph. aim to explore what benefits drawbacks these proposed graphs are, when are used function...

The purpose of this paper is to explore the linkage between recipe's ingredients and identification a cuisine. This has been tackled as problem cuisine classification. We will examine various approaches (different machine learning algorithms) for recipes classification based on ingredients. output be recommendation methodology, i.e. what kind preprocessing can done improve performance several classifiers dataset we using.

Network-based representations have introduced a revolution in neuroscience, expanding the understanding of brain from activity individual regions to interactions between them. This augmented network view comes at cost high dimensionality, which hinders both our capacity deciphering main mechanisms behind pathologies, and significance any statistical and/or machine learning task used processing this data. A link selection method, allowing remove irrelevant connections given scenario, is an...

One of the greatest issues modern urban environments are facing is poor air quality. It directly affects human health having a long-term negative impact on people's lives and major cause deaths in world. Smart cities combined with advances deep learning provide novel platform for dealing this problem. This paper uses pollution data from smart sensor networks CNN-LSTM architecture to forecast concentration current hour based previous 24-hour several meteorological features hour. Initially...

In this paper, comparative analysis is presented of our three 3D structure-based approaches for the efficient retrieval protein molecules. All rely on structure proteins. first approach, Spherical Trace Transform applied to structures in order produce geometry based descriptors. Additionally, some biological properties are taken, thus forming better integrated descriptor. second modification ray descriptor backbone molecule. third wavelet transformation distance matrix Calpha atoms which...

Traditional teaching, usually based on lectures and tutorials fosters the idea of instruction-driven learning model where students are passive listeners. Besides this approach, Project Based Learning (PBL) as a different paradigm is standing behind constructivism theory, from real-world situations put first place. The purpose paper to present our approach in embedded systems at University. It combination traditional (face-to-face) PBL. Our PBL represents an interdisciplinary project wireless...

In this paper, a 3D structure-based approach is presented for the efficient classification of protein molecules. The method relies on geometric structure proteins. After proper positioning structures, spherical trace transform applied to them produce geometry - based descriptors, which are completely rotation invariant. Additionally, some biological properties taken, and added geometry-based descriptor, thus forming better integrated descriptor. We have used nearest neighbour previously...

This paper addresses the routes planning problem in a scenario where an UAV (Unmanned Aerial Vehicle) and land-based transportation vehicle are used to deliver parcels customer locations. We developed implemented solution based on well-known Bellman-Held-Karp dynamic programming algorithm for Travelling Salesman Problem that finds least-cost both aircraft vehicle. applied this different, randomly selected commercial drones, with different maximum velocity flight range, order select best...

The rapid development of information technology has imposed a high rate change and reconceptualization in the design computer science (CS) courses order to expand access, provide flexibility learning environments, meet expanding needs ICT industry for qualified graduates. In such environment programming skills have become core competence engineering students. issue choosing and/or developing most effective approach teaching computer-programming languages risen greatest importance. This paper...

The proposed protein function prediction methods are mostly based on sequence or structure similarity and do not take into account the semantic extracted from knowledge databases such as Gene Ontology.Many studies have shown that identification of complexes functional modules can be effectively done by clustering interaction network (PIN).A significant number proteins in PIN remain uncharacterized predicting their remains a major challenge system biology.In this paper we present "semantic...