Search results for: Bayesian networks

Authors: Lin Yang, Zhian Mi, Jiacheng Xiao, Rong Li

Abstract:

Swimming with the tide of deep learning, the field of music information retrieval (MIR) experiences parallel development and a sheer variety of feature-learning models has been applied to music classification and tagging tasks. Among those learning techniques, the deep convolutional neural networks (CNNs) have been widespreadly used with better performance than the traditional approach especially in music genre classification and prediction. However, regarding the music recommendation, there is a large semantic gap between the corresponding audio genres and the various aspects of a song that influence user preference. In our study, aiming to bridge the gap, we strive to construct an automatic music aesthetic annotation model with MIDI format for better comparison and measurement of the similarity between music pieces in the way of harmonic analysis. We use the matrix of qualification converted from MIDI files as input to train two different classifiers, support vector machine (SVM) and Decision Tree (DT). Experimental results in performance of a tag prediction task have shown that both learning algorithms are capable of extracting high-level properties in an end-to end manner from music information. The proposed model is helpful to learn the audience taste and then the resulting recommendations are likely to appeal to a niche consumer.

Keywords: Harmonic analysis, machine learning, music classification and tagging, MIDI.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 758

Authors: Luis Fonseca, Filipe Cabral Pinto, Susana Sargento

Abstract:

The increase of the world population, especially in large urban centers, has resulted in new challenges particularly with the control and optimization of public safety. Thus, in the present work, a solution is proposed for the prediction of criminal occurrences in a city based on historical data of incidents and demographic information. The entire research and implementation will be presented start with the data collection from its original source, the treatment and transformations applied to them, choice and the evaluation and implementation of the Machine Learning model up to the application layer. Classification models will be implemented to predict criminal risk for a given time interval and location. Machine Learning algorithms such as Random Forest, Neural Networks, K-Nearest Neighbors and Logistic Regression will be used to predict occurrences, and their performance will be compared according to the data processing and transformation used. The results show that the use of Machine Learning techniques helps to anticipate criminal occurrences, which contributed to the reinforcement of public security. Finally, the models were implemented on a platform that will provide an API to enable other entities to make requests for predictions in real-time. An application will also be presented where it is possible to show criminal predictions visually.

Keywords: Crime prediction, machine learning, public safety, smart city.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1324

Authors: Tania Elena González Alvarado, Ma. Antonieta Martín Granados

Abstract:

The aim of this paper is to explain what a multienterprise tie is, what evidence its analysis provides and how does the cooperation mechanism influence the establishment of a multienterprise tie. The study focuses on businesses of smaller dimension, geographically dispersed and whose businessmen are learning to cooperate in an international environment. The empirical evidence obtained at this moment permits to conclude the following: The tie is not long-lasting, it has an end; opportunism is an opportunity to learn; the multi-enterprise tie is a space to learn about the cooperation mechanism; the local tie permits a businessman to alternate between competition and cooperation strategies; the disappearance of a tie is an experience of learning for a businessman, diminishing the possibility of failure in the next tie; the cooperation mechanism tends to eliminate hierarchical relations; the multienterprise tie diminishes the asymmetries and permits SME-s to have a better position when they negotiate with large companies; the multi-enterprise tie impacts positively on the local system. The collection of empirical evidence was done trough the following instruments: direct observation in a business encounter to which the businesses attended in 2003 (202 Mexican agro industry SME-s), a survey applied in 2004 (129), a questionnaire applied in 2005 (86 businesses), field visits to the businesses during the period 2006-2008 and; a survey applied by telephone in 2008 (55 Mexican agro industry SME-s).

Keywords: Cooperation, multi-enterprise tie, links, networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1273

Authors: A. F. Ali, D. M. Shawky

Abstract:

Discovering new biological knowledge from the highthroughput biological data is a major challenge to bioinformatics today. To address this challenge, we developed a new approach for protein classification. Proteins that are evolutionarily- and thereby functionally- related are said to belong to the same classification. Identifying protein classification is of fundamental importance to document the diversity of the known protein universe. It also provides a means to determine the functional roles of newly discovered protein sequences. Our goal is to predict the functional classification of novel protein sequences based on a set of features extracted from each protein sequence. The proposed technique used datasets extracted from the Structural Classification of Proteins (SCOP) database. A set of spectral domain features based on Fast Fourier Transform (FFT) is used. The proposed classifier uses multilayer back propagation (MLBP) neural network for protein classification. The maximum classification accuracy is about 91% when applying the classifier to the full four levels of the SCOP database. However, it reaches a maximum of 96% when limiting the classification to the family level. The classification results reveal that spectral domain contains information that can be used for classification with high accuracy. In addition, the results emphasize that sequence similarity measures are of great importance especially at the family level.

Keywords: Bioinformatics, Artificial Neural Networks, Protein Sequence Analysis, Feature Extraction.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2359

Authors: Mauro Giacomini, Stefania Bertone, Federico Caneva Soumetz, Carmelina Ruggiero

Abstract:

Environmental micro-organisms include a large number of taxa and some species that are generally considered nonpathogenic, but can represent a risk in certain conditions, especially for elderly people and immunocompromised individuals. Chemotaxonomic identification techniques are powerful tools for environmental micro-organisms, and cellular fatty acid methyl esters (FAME) content is a powerful fingerprinting identification technique. A system based on an unsupervised artificial neural network (ANN) was set up using the fatty acid profiles of standard bacterial strains, obtained by gas-chromatography, used as learning data. We analysed 45 certified strains belonging to Acinetobacter, Aeromonas, Alcaligenes, Aquaspirillum, Arthrobacter, Bacillus, Brevundimonas, Enterobacter, Flavobacterium, Micrococcus, Pseudomonas, Serratia, Shewanella and Vibrio genera. A set of 79 bacteria isolated from a drinking water line (AMGA, the major water supply system in Genoa) were used as an example for identification compared to standard MIDI method. The resulting ANN output map was found to be a very powerful tool to identify these fresh isolates.

Keywords: Cellular fatty acid methyl esters, environmental bacteria, gas-chromatography, unsupervised ANN.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1839

Authors: Muhammad Atif, Cang Yan

Abstract:

The enhancement of low-light images is a significant area of study aimed at enhancing the quality of captured images in challenging lighting environments. Recently, methods based on Convolutional Neural Networks (CNN) have gained prominence as they offer state-of-the-art performance. However, many approaches based on CNN rely on increasing the size and complexity of the neural network. In this study, we propose an alternative method for improving low-light images using an Autoencoders-based multiscale knowledge transfer model. Our method leverages the power of three autoencoders, where the encoders of the first two autoencoders are directly connected to the decoder of the third autoencoder. Additionally, the decoder of the first two autoencoders is connected to the encoder of the third autoencoder. This architecture enables effective knowledge transfer, allowing the third autoencoder to learn and benefit from the enhanced knowledge extracted by the first two autoencoders. We further integrate the proposed model into the Pix-to-Pix GAN framework. By integrating our proposed model as the generator in the GAN framework, we aim to produce enhanced images that not only exhibit improved visual quality but also possess a more authentic and realistic appearance. These experimental results, both qualitative and quantitative, show that our method is better than the state-of-the-art methodologies.

Keywords: Low light image enhancement, deep learning, convolutional neural network, image processing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 31

Authors: Georgios Linardatos, Miltiadis Zamparas, Vlasoula Bekiari, Georgios Bokias, Georgios Hotos

Abstract:

Hydrogels are three-dimensional, hydrophilic, polymeric networks composed of homopolymers or copolymers and are insoluble in water due to the presence of chemical or physical cross-links. When hydrogels come in contact with aqueous solutions, they can effectively sorb and retain the dissolved substances, depending on the nature of the monomeric units comprising the hydrogel. For this reason, hydrogels have been proposed in several studies as water purification agents. At the present work anionic hydrogels bearing negatively charged –COO- groups were prepared and investigated. These gels are based on sodium acrylate (ANa), either homopolymerized (poly(sodiumacrylate), PANa) or copolymerized (P(DMAM-co-ANa)) with N,N Dimethylacrylamide (DMAM). The hydrogels were used to extract some model organic dyes from water. It is found that cationic dyes are strongly sorbed and retained by the hydrogels, while sorption of anionic dyes was negligible. In all cases it was found that both maximum sorption capacity and equilibrium binding constant varied from one dye to the other depending on the chemical structure of the dye, the presence of functional chemical groups and the hydrophobic-hydrophilic balance. Finally, the nonionic hydrogel of the homopolymer poly(N,Ndimethylacrylamide), PDMAM, was also used for reasons of comparison.

Keywords: Anionic organic hydrogels, sorption, organic dyes, water purification agents.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2066

Authors: Joseph S. Ojo, Vincent A. Akpan, Oladayo G. Ajileye, Olalekan L, Ojo

Abstract:

In this paper, the intelligent algorithm (IA) that is capable of adapting to dynamical tropical weather conditions is proposed based on fuzzy logic techniques. The IA effectively interacts with the quality of service (QoS) criteria irrespective of the dynamic tropical weather to achieve improvement in the satellite links. To achieve this, an adaptive network-based fuzzy inference system (ANFIS) has been adopted. The algorithm is capable of interacting with the weather fluctuation to generate appropriate improvement to the satellite QoS for efficient services to the customers. 5-year (2012-2016) rainfall rate of one-minute integration time series data has been used to derive fading based on ITU-R P. 618-12 propagation models. The data are obtained from the measurement undertaken by the Communication Research Group (CRG), Physics Department, Federal University of Technology, Akure, Nigeria. The rain attenuation and signal-to-noise ratio (SNR) were derived for frequency between Ku and V-band and propagation angle with respect to different transmitting power. The simulated results show a substantial reduction in SNR especially for application in the area of digital video broadcast-second generation coding modulation satellite networks.

Keywords: Fuzzy logic, intelligent algorithm, Nigeria, QoS, satellite applications, tropical weather.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 818

Authors: Hiroshi Yamakawa, Michiko Yoshida, Motohiro Tsuchiya

Abstract:

The recent development of Information and Communication Technology (ICT) enables new ways of "democratic" decision-making such as a page-ranking system, which estimates the importance of a web page based on indirect trust on that page shared by diverse group of unorganized individuals. These kinds of "democracy" have not been acclaimed yet in the world of real politics. On the other hand, a large amount of data about personal relations including trust, norms of reciprocity, and networks of civic engagement has been accumulated in a computer-readable form by computer systems (e.g., social networking systems). We can use these relations as a new type of social capital to construct a new democratic decision-making system based on a delegation network. In this paper, we propose an effective decision-making support system, which is based on empowering someone's vote whom you trust. For this purpose, we propose two new techniques: the first is for estimating entire vote distribution from a small number of votes, and the second is for estimating active voter choice to promote voting using a delegation network. We show that these techniques could increase the voting ratio and credibility of the whole decision by agent-based simulations.

Keywords: Delegation, network centrality, social network, voting ratio.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1785

Authors: Ahmed Ebian, Mona Shokair, Kamal Awadalla

Abstract:

Mobile WiMAX is a broadband wireless solution that enables convergence of mobile and fixed broadband networks through a common wide area broadband radio access technology and flexible network architecture. It adopts Orthogonal Frequency Division Multiple Access (OFDMA) for improved multi-path performance in Non-Line-Of-Sight (NLOS) environments. Scalable OFDMA (SOFDMA) is introduced in the IEEE 802e[1]. WIMAX system uses one of different types of channel coding but The mandatory channel coding scheme is based on binary nonrecursive Convolutional Coding (CC). There are other several optional channel coding schemes such as block turbo codes, convolutional turbo codes, and low density parity check (LDPC). In this paper a comparison between the performance of WIMAX using turbo code and using convolutional product code (CPC) [2] is made. Also a combination between them had been done. The CPC gives good results at different SNR values compared to both the turbo system, and the combination between them. For example, at BER equal to 10-2 for 128 subcarriers, the amount of improvement in SNR equals approximately 3 dB higher than turbo code and equals approximately 2dB higher than the combination respectively. Several results are obtained at different modulating schemes (16QAM and 64QAM) and different numbers of sub-carriers (128 and 512).

Keywords: Turbo Code, Convolutional Product Code (CPC), Convolutional Product Code (CPC).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3395

Authors: Bassma Aldahlan, Zongming Fei

Abstract:

Recently, there is a great interest in smart parking application. Theses applications are enhanced by a vehicular ad-hoc network, which helps drivers find and reserve satiable packing spaces for a period of time ahead of time. Named Data Networking (NDN) is a future Internet architecture that benefits vehicular ad-hoc networks because of its clean-slate design and pure communication model. In this paper, we proposed an NDN-based frame-work for smart parking that involved a fog computing architecture. The proposed application had two main directions: First, we allowed drivers to query the number of parking spaces in a particular parking lot. Second, we introduced a technique that enabled drivers to make intelligent reservations before their arrival time. We also introduced a “push-based” model supporting the NDN-based framework for smart parking applications. To evaluate the proposed solution’s performance, we analyzed the function for finding parking lots with available parking spaces and the function for reserving a parking space. Our system showed high performance results in terms of response time and push overhead. The proposed reservation application performed better than the baseline approach.

Keywords: Cloud Computing, Vehicular Named Data Networking, Smart Parking Applications, Fog Computing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 226

Authors: Meng-Lung Lin, Chien-Min Chu, Chung-Hung Tsai, Chih-Cheng Chen, Chen-Yuan Chen

Abstract:

The study of tourist activities and the mapping of their routes in space and time has become an important issue in tourism management. Here we represent space-time paths for the tourism industry by visualizing individual tourist activities and the paths followed using a 3D Geographic Information System (GIS). Considerable attention has been devoted to the measurement of accessibility to shopping, eating, walking and other services at the tourist destination. I turns out that GIS is a useful tool for studying the spatial behaviors of tourists in the area. The value of GIS is especially advantageous for space-time potential path area measures, especially for the accurate visualization of possible paths through existing city road networks. This study seeks to apply space-time concepts with a detailed street network map obtained from Google Maps to measure tourist paths both spatially and temporally. These paths are further determined based on data obtained from map questionnaires regarding the trip activities of 40 individuals. The analysis of the data makes it possible to determining the locations of the more popular paths. The results can be visualized using 3D GIS to show the areas and potential activity opportunities accessible to tourists during their travel time.

Keywords: Tourist activity analysis, space-time path, GIS, geovisualization, activity-travel pattern.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2492

Authors: Priyanka Paliwal, N.P. Patidar

Abstract:

Distributed Power generation has gained a lot of attention in recent times due to constraints associated with conventional power generation and new advancements in DG technologies .The need to operate the power system economically and with optimum levels of reliability has further led to an increase in interest in Distributed Generation. However it is important to place Distributed Generator on an optimum location so that the purpose of loss minimization and voltage regulation is dully served on the feeder. This paper investigates the impact of DG units installation on electric losses, reliability and voltage profile of distribution networks. In this paper, our aim would be to find optimal distributed generation allocation for loss reduction subjected to constraint of voltage regulation in distribution network. The system is further analyzed for increased levels of Reliability. Distributed Generator offers the additional advantage of increase in reliability levels as suggested by the improvements in various reliability indices such as SAIDI, CAIDI and AENS. Comparative studies are performed and related results are addressed. An analytical technique is used in order to find the optimal location of Distributed Generator. The suggested technique is programmed under MATLAB software. The results clearly indicate that DG can reduce the electrical line loss while simultaneously improving the reliability of the system.

Keywords: AENS, CAIDI, Distributed Generation, lossreduction, Reliability, SAIDI

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3100

Authors: S. Mohideen Badhusha, K. Duraiswamy

Abstract:

Most of the existing video streaming protocols provide video services without considering security aspects in decentralized mobile ad-hoc networks. The security policies adapted to the currently existing non-streaming protocols, do not comply with the live video streaming protocols resulting in considerable vulnerability, high bandwidth consumption and unreliability which cause severe security threats, low bandwidth and error prone transmission respectively in video streaming applications. Therefore a synergized methodology is required to reduce vulnerability and bandwidth consumption, and enhance reliability in the video streaming applications in MANET. To ensure the security measures with reduced bandwidth consumption and improve reliability of the video streaming applications, a Secure Low-bandwidth Video Streaming through Reliable Multipath Propagation (SLVRMP) protocol architecture has been proposed by incorporating the two algorithms namely Secure Low-bandwidth Video Streaming Algorithm and Reliable Secure Multipath Propagation Algorithm using Layered Video Coding in non-overlapping zone routing network topology. The performances of the proposed system are compared to those of the other existing secure multipath protocols Sec-MR, SPREAD using NS 2.34 and the simulation results show that the performances of the proposed system get considerably improved.

Keywords: Bandwidth consumption, layered video coding, multipath propagation, reliability, security threats, video streaming applications, vulnerability.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1882

Authors: Pema Choejey, David Murray, Chun Che Fung

Abstract:

Bhutan is becoming increasingly dependent on Information and Communications Technologies (ICTs), especially the Internet for performing the daily activities of governments, businesses, and individuals. Consequently, information systems and networks are becoming more exposed and vulnerable to cybersecurity threats. This paper highlights the findings of the survey study carried out to understand the perceptions of cybersecurity implementation among government organizations in Bhutan. About 280 ICT personnel were surveyed about the effectiveness of cybersecurity implementation in their organizations. A questionnaire based on a 5 point Likert scale was used to assess the perceptions of respondents. The questions were asked on cybersecurity practices such as cybersecurity policies, awareness and training, and risk management. The survey results show that less than 50% of respondents believe that the cybersecurity implementation is effective: cybersecurity policy (40%), risk management (23%), training and awareness (28%), system development life cycle (34%); incident management (26%), and communications and operational management (40%). The findings suggest that many of the cybersecurity practices are inadequately implemented and therefore, there exist a gap in achieving a required cybersecurity posture. This study recommends government organizations to establish a comprehensive cybersecurity program with emphasis on cybersecurity policy, risk management, and awareness and training. In addition, the research study has practical implications to both government and private organizations for implementing and managing cybersecurity.

Keywords: Awareness and training, cybersecurity, cybersecurity policy, risk management, security risks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1560

Authors: Banu Karaşah, Derya Sarı

Abstract:

Nature constantly changes as a result of human necessities. This change mostly feels in natural water sources which are reconstructed with an effect of dams and motorways. In other respects, visual quality of the landscape gets a new and different character during and after the construction of dams and motorways. Changing and specialization new landscapes will be very important to protection-usage balance to explore sustainable usage facilities. The main cause of the selection of Artvin city is that it has very important geographical location and one of the most attraction points in the World with its biodiversity, conservation areas and natural landscape characteristics. Many hydroelectric station and 7 dams are situated, 3 of them have already been built on the Çoruh River in the province of Artvin. As a result of dams, motorways route were reshaped and the ways which have already changed because of elevation is directly affected several of natural destruction. In contrast, many different reservoirs in Coruh Basin provide new vista point that has high visual quality. In this study, we would like to evaluate with sustainable landscape design in 76 km river corridor, which is mainly based on Deriner, Borçka and Muratlı Dams and determination of their basin-lakes recreational potential and opportunities. Lastly, we are going to give some suggestion about the potential of the corridor.

Keywords: Artvin, dam reservoirs, landscape assessment, river corridor, visual quality.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1963

Authors: Jessica Lòpez Espejel, Mahaman Sanoussi Yahaya Alassan, Walid Dahhane, El Hassane Ettifouri

Abstract:

Pretrained transformer-based models have shown high performance in natural language generation task. However, a new wave of interest has surged: automatic programming language generation. This task consists of translating natural language instructions to a programming code. Despite the fact that well-known pretrained models on language generation have achieved good performance in learning programming languages, effort is still needed in automatic code generation. In this paper, we introduce JaCoText, a model based on Transformers neural network. It aims to generate java source code from natural language text. JaCoText leverages advantages of both natural language and code generation models. More specifically, we study some findings from the state of the art and use them to (1) initialize our model from powerful pretrained models, (2) explore additional pretraining on our java dataset, (3) carry out experiments combining the unimodal and bimodal data in the training, and (4) scale the input and output length during the fine-tuning of the model. Conducted experiments on CONCODE dataset show that JaCoText achieves new state-of-the-art results.

Keywords: Java code generation, Natural Language Processing, Sequence-to-sequence Models, Transformers Neural Networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 855

Authors: S. Chiali, Z.Eberrichi, M.Malki

Abstract:

The number of framework conceived for e-learning constantly increase, unfortunately the creators of learning materials and educational institutions engaged in e-formation adopt a “proprietor" approach, where the developed products (courses, activities, exercises, etc.) can be exploited only in the framework where they were conceived, their uses in the other learning environments requires a greedy adaptation in terms of time and effort. Each one proposes courses whose organization, contents, modes of interaction and presentations are unique for all learners, unfortunately the latter are heterogeneous and are not interested by the same information, but only by services or documents adapted to their needs. Currently the new tendency for the framework conceived for e-learning, is the interoperability of learning materials, several standards exist (DCMI (Dublin Core Metadata Initiative)[2], LOM (Learning Objects Meta data)[1], SCORM (Shareable Content Object Reference Model)[6][7][8], ARIADNE (Alliance of Remote Instructional Authoring and Distribution Networks for Europe)[9], CANCORE (Canadian Core Learning Resource Metadata Application Profiles)[3]), they converge all to the idea of learning objects. They are also interested in the adaptation of the learning materials according to the learners- profile. This article proposes an approach for the composition of courses adapted to the various profiles (knowledge, preferences, objectives) of learners, based on two ontologies (domain to teach and educational) and the learning objects.

Keywords: Adaptive educational hypermedia systems (AEHS), E-learning, Learner's model, Learning objects, Metadata, Ontology.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1959

Authors: P. Priakanth, P. Thangaraj

Abstract:

Due to the complex network architecture, the mobile adhoc network-s multihop feature gives additional problems to the users. When the traffic load at each node gets increased, the additional contention due its traffic pattern might cause the nodes which are close to destination to starve the nodes more away from the destination and also the capacity of network is unable to satisfy the total user-s demand which results in an unfairness problem. In this paper, we propose to create an algorithm to compute the optimal MAC-layer bandwidth assigned to each flow in the network. The bottleneck links contention area determines the fair time share which is necessary to calculate the maximum allowed transmission rate used by each flow. To completely utilize the network resources, we compute two optimal rates namely, the maximum fair share and minimum fair share. We use the maximum fair share achieved in order to limit the input rate of those flows which crosses the bottleneck links contention area when the flows that are not allocated to the optimal transmission rate and calculate the following highest fair share. Through simulation results, we show that the proposed protocol achieves improved fair share and throughput with reduced delay.

Keywords: MAC-layer, MANETs, Multihop, optimal rate, Transmission.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1507

Authors: Sofien Chtourou, Mohamed Chtourou, Omar Hammami

Abstract:

Embedded systems need to respect stringent real time constraints. Various hardware components included in such systems such as cache memories exhibit variability and therefore affect execution time. Indeed, a cache memory access from an embedded microprocessor might result in a cache hit where the data is available or a cache miss and the data need to be fetched with an additional delay from an external memory. It is therefore highly desirable to predict future memory accesses during execution in order to appropriately prefetch data without incurring delays. In this paper, we evaluate the potential of several artificial neural networks for the prediction of instruction memory addresses. Neural network have the potential to tackle the nonlinear behavior observed in memory accesses during program execution and their demonstrated numerous hardware implementation emphasize this choice over traditional forecasting techniques for their inclusion in embedded systems. However, embedded applications execute millions of instructions and therefore millions of addresses to be predicted. This very challenging problem of neural network based prediction of large time series is approached in this paper by evaluating various neural network architectures based on the recurrent neural network paradigm with pre-processing based on the Self Organizing Map (SOM) classification technique.

Keywords: Address, data set, memory, prediction, recurrentneural network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1674

Authors: Vasudha Hegde, Narendra Chaulagain, H. M. Ravikumar, Sonu Mishra, Siva Yellampalli

Abstract:

Acoustic sensors are extensively used in recent days not only for sensing and condition monitoring applications but also for small scale energy harvesting applications to power wireless sensor networks (WSN) due to their inherent advantages. The natural frequency of the structure plays a major role in energy harvesting applications since the sensor key element has to operate at resonant frequency. In this paper, circular diaphragm based MEMS acoustic sensor is modelled by Lumped Element Model (LEM) and the natural frequency is compared with the simulated model using Finite Element Method (FEM) tool COMSOL Multiphysics. The sensor has the circular diaphragm of 3000 µm radius and thickness of 30 µm to withstand the high SPL (Sound Pressure Level) and also to withstand the various fabrication steps. A Piezoelectric ZnO layer of thickness of 1 µm sandwiched between two aluminium electrodes of thickness 0.5 µm and is coated on the diaphragm. Further, a channel with radius 3000 µm radius and length 270 µm is connected at the bottom of the diaphragm. The natural frequency of the structure by LEM method is approximately 16.6 kHz which is closely matching with that of simulated structure with suitable approximations.

Keywords: Acoustic sensor, diaphragm based, lumped element modeling, natural frequency, piezoelectric.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1028

Authors: R.V Pawar, P.P.Kajave, S.N.Mali

Abstract:

The speech signal conveys information about the identity of the speaker. The area of speaker identification is concerned with extracting the identity of the person speaking the utterance. As speech interaction with computers becomes more pervasive in activities such as the telephone, financial transactions and information retrieval from speech databases, the utility of automatically identifying a speaker is based solely on vocal characteristic. This paper emphasizes on text dependent speaker identification, which deals with detecting a particular speaker from a known population. The system prompts the user to provide speech utterance. System identifies the user by comparing the codebook of speech utterance with those of the stored in the database and lists, which contain the most likely speakers, could have given that speech utterance. The speech signal is recorded for N speakers further the features are extracted. Feature extraction is done by means of LPC coefficients, calculating AMDF, and DFT. The neural network is trained by applying these features as input parameters. The features are stored in templates for further comparison. The features for the speaker who has to be identified are extracted and compared with the stored templates using Back Propogation Algorithm. Here, the trained network corresponds to the output; the input is the extracted features of the speaker to be identified. The network does the weight adjustment and the best match is found to identify the speaker. The number of epochs required to get the target decides the network performance.

Keywords: Average Mean Distance function, Backpropogation, Linear Predictive Coding, MultilayeredPerceptron,

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1892

Authors: Ashraf M. A. Ahmad

Abstract:

Delivering streaming video over wireless is an important component of many interactive multimedia applications running on personal wireless handset devices. Such personal devices have to be inexpensive, compact, and lightweight. But wireless channels have a high channel bit error rate and limited bandwidth. Delay variation of packets due to network congestion and the high bit error rate greatly degrades the quality of video at the handheld device. Therefore, mobile access to multimedia contents requires video transcoding functionality at the edge of the mobile network for interworking with heterogeneous networks and services. Therefore, to guarantee quality of service (QoS) delivered to the mobile user, a robust and efficient transcoding scheme should be deployed in mobile multimedia transporting network. Hence, this paper examines the challenges and limitations that the video transcoding schemes in mobile multimedia transporting network face. Then handheld resources, network conditions and content based mobile and wireless video transcoding is proposed to provide high QoS applications. Exceptional performance is demonstrated in the experiment results. These experiments were designed to verify and prove the robustness of the proposed approach. Extensive experiments have been conducted, and the results of various video clips with different bit rate and frame rate have been provided.

Keywords: Content, Object detection, Transcoding, Texture, Temporal, Video.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1346

Authors: Pariwat Somnuek

Abstract:

This study focused on tourism logistic services in the border areas of Thailand by an analysis and comparison of the opinions of tourists, villagers, and entrepreneurs of these services. Sample representatives of this study were a total of 600 villagers and 15 entrepreneurs in the three border areas consisting of Chong Anma, Chong Sa-Ngam, and Chong Jom checkpoints. For methodology, survey questionnaires, situation analysis, TOWS matrix, and focus group discussions were used for data collection, as well as descriptive analysis and statistics such as arithmetic means and standard deviations, were employed for data analysis. The findings revealed that business potential was at the medium level and entrepreneurs were satisfied with their turnovers. However, perspectives of transportation and tourism services provided for tourists need to be immediately improved. Recommendations for the potential development included promotion of border tourism destinations and foreign investments into accommodation, restaurants, and transport, as well as the establishment of business networks between Thailand and Cambodia, along with the introduction of new tourism destinations by co-operation between entrepreneurs in both countries. These initiatives may lead to increased visitors, collaboration of security offices, and an improved image of tourism security.

Keywords: Business potential, potential development, tourism logistics, services.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1046

Authors: Cameron Hamilton, Walter Potter, Gerrit Hoogenboom, Ronald McClendon, Will Hobbs

Abstract:

A model was constructed to predict the amount of solar radiation that will make contact with the surface of the earth in a given location an hour into the future. This project was supported by the Southern Company to determine at what specific times during a given day of the year solar panels could be relied upon to produce energy in sufficient quantities. Due to their ability as universal function approximators, an artificial neural network was used to estimate the nonlinear pattern of solar radiation, which utilized measurements of weather conditions collected at the Griffin, Georgia weather station as inputs. A number of network configurations and training strategies were utilized, though a multilayer perceptron with a variety of hidden nodes trained with the resilient propagation algorithm consistently yielded the most accurate predictions. In addition, a modeled direct normal irradiance field and adjacent weather station data were used to bolster prediction accuracy. In later trials, the solar radiation field was preprocessed with a discrete wavelet transform with the aim of removing noise from the measurements. The current model provides predictions of solar radiation with a mean square error of 0.0042, though ongoing efforts are being made to further improve the model’s accuracy.

Keywords: Artificial Neural Networks, Resilient Propagation, Solar Radiation, Time Series Forecasting.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2761

Authors: Qian Liu, Steve Furber

Abstract:

To explore how the brain may recognise objects in its general,accurate and energy-efficient manner, this paper proposes the use of a neuromorphic hardware system formed from a Dynamic Video Sensor (DVS) silicon retina in concert with the SpiNNaker real-time Spiking Neural Network (SNN) simulator. As a first step in the exploration on this platform a recognition system for dynamic hand postures is developed, enabling the study of the methods used in the visual pathways of the brain. Inspired by the behaviours of the primary visual cortex, Convolutional Neural Networks (CNNs) are modelled using both linear perceptrons and spiking Leaky Integrate-and-Fire (LIF) neurons. In this study’s largest configuration using these approaches, a network of 74,210 neurons and 15,216,512 synapses is created and operated in real-time using 290 SpiNNaker processor cores in parallel and with 93.0% accuracy. A smaller network using only 1/10th of the resources is also created, again operating in real-time, and it is able to recognise the postures with an accuracy of around 86.4% - only 6.6% lower than the much larger system. The recognition rate of the smaller network developed on this neuromorphic system is sufficient for a successful hand posture recognition system, and demonstrates a much improved cost to performance trade-off in its approach.

Keywords: Spiking neural network (SNN), convolutional neural network (CNN), posture recognition, neuromorphic system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2053

Authors: M. Bouguerra, I. Habi

Abstract:

The continuity in the electric supply of the electric installations is becoming one of the main requirements of the electric supply network (generation, transmission, and distribution of the electric energy). The achievement of this requirement depends from one side on the structure of the electric network and on the other side on the avaibility of the reserve source provided to maintain the supply in case of failure of the principal one. The avaibility of supply does not only depends on the reliability parameters of the both sources (principal and reserve) but it also depends on the reliability of the circuit breaker which plays the role of interlocking the reserve source in case of failure of the principal one. In addition, the principal source being under operation, its control can be ideal and sure, however, for the reserve source being in stop, a preventive maintenances which proceed on time intervals (periodicity) and for well defined lengths of time are envisaged, so that this source will always available in case of the principal source failure. The choice of the periodicity of preventive maintenance of the source of reserve influences directly the reliability of the electric feeder system In this work and on the basis of the semi- markovian's processes, the influence of the time of interlocking the reserve source upon the reliability of an industrial electric network is studied and is given the optimal time of interlocking the reserve source in case of failure the principal one, also the influence of the periodicity of the preventive maintenance of the source of reserve is studied and is given the optimal periodicity.

Keywords: Semi-Markovians processes, reliability, optimization, industrial electric network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1270

Authors: A. Faro, D. Giordano, F. Maiorana

Abstract:

Data mining uses a variety of techniques each of which is useful for some particular task. It is important to have a deep understanding of each technique and be able to perform sophisticated analysis. In this article we describe a tool built to simulate a variation of the Kohonen network to perform unsupervised clustering and support the entire data mining process up to results visualization. A graphical representation helps the user to find out a strategy to optimize classification by adding, moving or delete a neuron in order to change the number of classes. The tool is able to automatically suggest a strategy to optimize the number of classes optimization, but also support both tree classifications and semi-lattice organizations of the classes to give to the users the possibility of passing from one class to the ones with which it has some aspects in common. Examples of using tree and semi-lattice classifications are given to illustrate advantages and problems. The tool is applied to classify macroeconomic data that report the most developed countries- import and export. It is possible to classify the countries based on their economic behaviour and use the tool to characterize the commercial behaviour of a country in a selected class from the analysis of positive and negative features that contribute to classes formation. Possible interrelationships between the classes and their meaning are also discussed.

Keywords: Unsupervised classification, Kohonen networks, macroeconomics, Visual data mining, Cluster interpretation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1541

Authors: Rajine Swetha R, B. Shekar Babu, Sumithra Devi K.A

Abstract:

Electronic Systems are the core of everyday lives. They form an integral part in financial networks, mass transit, telephone systems, power plants and personal computers. Electronic systems are increasingly based on complex VLSI (Very Large Scale Integration) integrated circuits. Initial electronic design automation is concerned with the design and production of VLSI systems. The next important step in creating a VLSI circuit is Physical Design. The input to the physical design is a logical representation of the system under design. The output of this step is the layout of a physical package that optimally or near optimally realizes the logical representation. Physical design problems are combinatorial in nature and of large problem sizes. Darwin observed that, as variations are introduced into a population with each new generation, the less-fit individuals tend to extinct in the competition of basic necessities. This survival of fittest principle leads to evolution in species. The objective of the Genetic Algorithms (GA) is to find an optimal solution to a problem .Since GA-s are heuristic procedures that can function as optimizers, they are not guaranteed to find the optimum, but are able to find acceptable solutions for a wide range of problems. This survey paper aims at a study on Efficient Algorithms for VLSI Physical design and observes the common traits of the superior contributions.

Keywords: Genetic Algorithms, Physical Design, VLSI.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1739

Authors: K. Chandra Sekaran

Abstract:

This paper describes a paradigmatic approach to develop architecture of secure systems by describing the requirements from four different points of view: that of the owner, the administrator, the user, and the network. Deriving requirements and developing architecture implies the joint elicitation and describing the problem and the structure of the solution. The view points proposed in this paper are those we consider as requirements towards their contributions as major parties in the design, implementation, usage and maintenance of secure systems. The dramatic growth of the technology of Internet and the applications deployed in World Wide Web have lead to the situation where the security has become a very important concern in the development of secure systems. Many security approaches are currently being used in organizations. In spite of the widespread use of many different security solutions, the security remains a problem. It is argued that the approach that is described in this paper for the development of secure architecture is practical by all means. The models representing these multiple points of view are termed the requirements model (views of owner and administrator) and the operations model (views of user and network). In this paper, this multiple view paradigm is explained by first describing the specific requirements and or characteristics of secure systems (particularly in the domain of networks) and the secure architecture / system development methodology.

Keywords: Multiple view paradigms, requirements model, operations model, secure system, owner, administrator, user, network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1368