Search results for: MNIST dataset

Authors: Edward Guillén, Jeisson Sánchez, Carlos Omar Ramos

Abstract:

In recent years, a wide variety of applications are developed with Support Vector Machines -SVM- methods and Artificial Neural Networks -ANN-. In general, these methods depend on intrusion knowledge databases such as KDD99, ISCX, and CAIDA among others. New classes of detectors are generated by machine learning techniques, trained and tested over network databases. Thereafter, detectors are employed to detect anomalies in network communication scenarios according to user’s connections behavior. The first detector based on training dataset is deployed in different real-world networks with mobile and non-mobile devices to analyze the performance and accuracy over static detection. The vulnerabilities are based on previous work in telemedicine apps that were developed on the research group. This paper presents the differences on detections results between some network scenarios by applying traditional detectors deployed with artificial neural networks and support vector machines.

Keywords: Anomaly detection, back-propagation neural networks, network intrusion detection systems, support vector machines.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2009

Authors: Marcio Leal, Marta Villamil

Abstract:

Computacional recognition of sign languages aims to allow a greater social and digital inclusion of deaf people through interpretation of their language by computer. This article presents a model of recognition of two of global parameters from sign languages; hand configurations and hand movements. Hand motion is captured through an infrared technology and its joints are built into a virtual three-dimensional space. A Multilayer Perceptron Neural Network (MLP) was used to classify hand configurations and Dynamic Time Warping (DWT) recognizes hand motion. Beyond of the method of sign recognition, we provide a dataset of hand configurations and motion capture built with help of fluent professionals in sign languages. Despite this technology can be used to translate any sign from any signs dictionary, Brazilian Sign Language (Libras) was used as case study. Finally, the model presented in this paper achieved a recognition rate of 80.4%.

Keywords: Sign language recognition, computer vision, infrared, artificial neural network, dynamic time warping.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 878

Authors: Ali Anaissi, Paul J. Kennedy, Madhu Goyal

Abstract:

Analysis and visualization of microarraydata is veryassistantfor biologists and clinicians in the field of diagnosis and treatment of patients. It allows Clinicians to better understand the structure of microarray and facilitates understanding gene expression in cells. However, microarray dataset is a complex data set and has thousands of features and a very small number of observations. This very high dimensional data set often contains some noise, non-useful information and a small number of relevant features for disease or genotype. This paper proposes a non-linear dimensionality reduction algorithm Local Principal Component (LPC) which aims to maps high dimensional data to a lower dimensional space. The reduced data represents the most important variables underlying the original data. Experimental results and comparisons are presented to show the quality of the proposed algorithm. Moreover, experiments also show how this algorithm reduces high dimensional data whilst preserving the neighbourhoods of the points in the low dimensional space as in the high dimensional space.

Keywords: Linear Dimension Reduction, Non-Linear Dimension Reduction, Principal Component Analysis, Biologists.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1574

Authors: S. M. Anowarul Haque, Md. Asiful Islam

Abstract:

Electrical load forecasting is considered to be one of the most indispensable parts of a modern-day electrical power system. To ensure a reliable and efficient supply of electric energy, special emphasis should have been put on the predictive feature of electricity supply. Artificial Neural Network-based approaches have emerged to be a significant area of interest for electric load forecasting research. This paper proposed an Artificial Neural Network model based on the particle swarm optimization algorithm for improved electric load forecasting for Mymensingh, Bangladesh. The forecasting model is developed and simulated on the MATLAB environment with a large number of training datasets. The model is trained based on eight input parameters including historical load and weather data. The predicted load data are then compared with an available dataset for validation. The proposed neural network model is proved to be more reliable in terms of day-wise load forecasting for Mymensingh, Bangladesh.

Keywords: Load forecasting, artificial neural network, particle swarm optimization.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 686

Authors: Fatemeh Valizadeh Gamchi, Edward L. Boone

Abstract:

This research focuses on Lyme disease, a widespread infectious condition in the United States caused by the bacterium Borrelia burgdorferi sensu stricto. It is critical to identify environmental and economic elements that are contributing to the spread of the disease. This study examined data from Virginia to identify a subset of explanatory variables significant for Lyme disease case numbers. To identify relevant variables and avoid overfitting, linear poisson, and regularization regression methods such as ridge, lasso, and elastic net penalty were employed. Cross-validation was performed to acquire tuning parameters. The methods proposed can automatically identify relevant disease count covariates. The efficacy of the techniques was assessed using four criteria on three simulated datasets. Finally, using the Virginia Department of Health’s Lyme disease dataset, the study successfully identified key factors, and the results were consistent with previous studies.

Keywords: Lyme disease, Poisson generalized linear model, Ridge regression, Lasso Regression, elastic net regression.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 123

Authors: Elham Serkani, Hossein Gharaee Garakani, Naser Mohammadzadeh, Elaheh Vaezpour

Abstract:

Intrusion detection systems (IDS) are the main components of network security. These systems analyze the network events for intrusion detection. The design of an IDS is through the training of normal traffic data or attack. The methods of machine learning are the best ways to design IDSs. In the method presented in this article, the pruning algorithm of C5.0 decision tree is being used to reduce the features of traffic data used and training IDS by the least square vector algorithm (LS-SVM). Then, the remaining features are arranged according to the predictor importance criterion. The least important features are eliminated in the order. The remaining features of this stage, which have created the highest level of accuracy in LS-SVM, are selected as the final features. The features obtained, compared to other similar articles which have examined the selected features in the least squared support vector machine model, are better in the accuracy, true positive rate, and false positive. The results are tested by the UNSW-NB15 dataset.

Keywords: Intrusion detection system, decision tree, support vector machine, feature selection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1240

Authors: S. K. Aruna, S. Chitra

Abstract:

This research paper presents a framework for classifying Magnetic Resonance Imaging (MRI) images for Dementia. Dementia, an age-related cognitive decline is indicated by degeneration of cortical and sub-cortical structures. Characterizing morphological changes helps understand disease development and contributes to early prediction and prevention of the disease. Modelling, that captures the brain’s structural variability and which is valid in disease classification and interpretation is very challenging. Features are extracted using Gabor filter with 0, 30, 60, 90 orientations and Gray Level Co-occurrence Matrix (GLCM). It is proposed to normalize and fuse the features. Independent Component Analysis (ICA) selects features. Support Vector Machine (SVM) classifier with different kernels is evaluated, for efficiency to classify dementia. This study evaluates the presented framework using MRI images from OASIS dataset for identifying dementia. Results showed that the proposed feature fusion classifier achieves higher classification accuracy.

Keywords: Magnetic resonance imaging, dementia, Gabor filter, gray level co-occurrence matrix, support vector machine.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2116

Authors: Long Van Hoang, Mitsuyasu Yabe

Abstract:

Environmental factors affect agriculture production productivity and efficiency resulted in changing of profit efficiency. This paper attempts to estimate the impacts of environmental factors to profitability of rice farmers in the Red River Delta of Vietnam. The dataset was extracted from 349 rice farmers using personal interviews. Both OLS and MLE trans-log profit functions were used in this study. Five production inputs and four environmental factors were included in these functions. The estimation of the stochastic profit frontier with a two-stage approach was used to measure profitability. The results showed that the profit efficiency was about 75% on the average and environmental factors change profit efficiency significantly beside farm specific characteristics. Plant disease, soil fertility, irrigation apply and water pollution were the four environmental factors cause profit loss in rice production. The result indicated that farmers should reduce household size, farm plots, apply row seeding technique and improve environmental factors to obtain high profit efficiency with special consideration is given for irrigation water quality improvement.

Keywords: Profit efficiency; Profit function; Environmental factors; OLS and MLE estimations; Rice Production; Vietnam

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3554

Authors: Yishu Gong, Liangliang Yang, Jianyu Zhang, Zhengyu Chen, Sihong He, Xusheng Zhang, Wei Zhang

Abstract:

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that affects millions of people worldwide and is characterized by cognitive decline and behavioral changes. People living with Alzheimer’s disease often find it hard to complete routine tasks. However, there are limited objective assessments that aim to quantify the difficulty of certain tasks for AD patients compared to non-AD people. In this study, we propose to use speech emotion recognition (SER), especially the frustration level as a potential biomarker for quantifying the difficulty patients experience when describing a picture. We build an SER model using data from the IEMOCAP dataset and apply the model to the DementiaBank data to detect the AD/non-AD group difference and perform longitudinal analysis to track the AD disease progression. Our results show that the frustration level detected from the SER model can possibly be used as a cost-effective tool for objective tracking of AD progression in addition to the Mini-Mental State Examination (MMSE) score.

Keywords: Alzheimer’s disease, Speech Emotion Recognition, longitudinal biomarker, machine learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 274

Authors: Sajjad Ahmed, Knut Hinkelmann, Flavio Corradini

Abstract:

Fake news detection research is still in the early stage as this is a relatively new phenomenon in the interest raised by society. Machine learning helps to solve complex problems and to build AI systems nowadays and especially in those cases where we have tacit knowledge or the knowledge that is not known. We used machine learning algorithms and for identification of fake news; we applied three classifiers; Passive Aggressive, Naïve Bayes, and Support Vector Machine. Simple classification is not completely correct in fake news detection because classification methods are not specialized for fake news. With the integration of machine learning and text-based processing, we can detect fake news and build classifiers that can classify the news data. Text classification mainly focuses on extracting various features of text and after that incorporating those features into classification. The big challenge in this area is the lack of an efficient way to differentiate between fake and non-fake due to the unavailability of corpora. We applied three different machine learning classifiers on two publicly available datasets. Experimental analysis based on the existing dataset indicates a very encouraging and improved performance.

Keywords: Fake news detection, types of fake news, machine learning, natural language processing, classification techniques.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1512

Authors: Omer Nebil Yaveroglu, Tolga Can

Abstract:

In this study, a high accuracy protein-protein interaction prediction method is developed. The importance of the proposed method is that it only uses sequence information of proteins while predicting interaction. The method extracts phylogenetic profiles of proteins by using their sequence information. Combining the phylogenetic profiles of two proteins by checking existence of homologs in different species and fitting this combined profile into a statistical model, it is possible to make predictions about the interaction status of two proteins. For this purpose, we apply a collection of pattern recognition techniques on the dataset of combined phylogenetic profiles of protein pairs. Support Vector Machines, Feature Extraction using ReliefF, Naive Bayes Classification, K-Nearest Neighborhood Classification, Decision Trees, and Random Forest Classification are the methods we applied for finding the classification method that best predicts the interaction status of protein pairs. Random Forest Classification outperformed all other methods with a prediction accuracy of 76.93%

Keywords: Protein Interaction Prediction, Phylogenetic Profile, SVM , ReliefF, Decision Trees, Random Forest Classification

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1613

Authors: Jatin Kumar Choudhary, Naren Rayala, Abbas Eslami Kiasari, Fahimeh Jafari

Abstract:

Great Britain has one of the safest road networks in the world. However, the consequences of any death or serious injury are devastating for loved ones, as well as for those who help the severely injured. This paper aims to analyse Great Britain's road safety situation and show the response measures for areas where the total damage caused by accidents can be significantly and quickly reduced. For the past 30 years, the UK has had a good record in reducing fatalities over the past 30 years, there is still a considerable number of road deaths. The government continues to scale back road deaths empowering responsible road users by identifying and prosecuting the parameters that make the roads less safe. This study represents an exploratory analysis with deep insights which could provide policy makers with invaluable insights into how accidents happen and how they can be mitigated. We use STATS19 data published by the UK government. Since we need more information about locations which is not provided in STATA19, we first expand the features of the dataset using OpenStreetMap and Visual Crossing. This paper also provides a discussion regarding new road safety methods.

Keywords: Road safety, data analysis, OpenStreetMap, feature expanding.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 348

Authors: Yasaswi Palagummi, Sareh Rowlands

Abstract:

Generalised Zero-Shot Learning, often known as GZSL, is an advanced variant of zero-shot learning in which the samples in the unseen category may be either seen or unseen. GZSL methods typically have a bias towards the seen classes because they learn a model to perform recognition for both the seen and unseen classes using data samples from the seen classes. This frequently leads to the misclassification of data from the unseen classes into the seen classes, making the task of GZSL more challenging. In this work, we propose an approach leveraging the Shifted Window based Self-Attention in the Swin Transformer (Swin-GZSL) to work in the inductive GZSL problem setting. We run experiments on three popular benchmark datasets: CUB, SUN, and AWA2, which are specifically used for ZSL and its other variants. The results show that our model based on Swin Transformer has achieved state-of-the-art harmonic mean for two datasets - AWA2 and SUN and near-state-of-the-art for the other dataset - CUB. More importantly, this technique has a linear computational complexity, which reduces training time significantly. We have also observed less bias than most of the existing GZSL models.

Keywords: Generalised Zero-shot Learning, Inductive Learning, Shifted-Window Attention, Swin Transformer, Vision Transformer.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 222

Authors: Jedol Dayou, Jackson Hian Wui Chang, Rubena Yusoff, Ag. Sufiyan Abd. Hamid, Fauziah Sulaiman, Justin Sentian

Abstract:

Aerosols are small particles suspended in air that have wide varying spatial and temporal distributions. The concentration of aerosol in total columnar atmosphere is normally measured using aerosol optical depth (AOD). In long-term monitoring stations, accurate AOD retrieval is often difficult due to the lack of frequent calibration. To overcome this problem, a near-sea-level Langley calibration algorithm is developed using the combination of clear-sky detection model and statistical filter. It attempts to produce a dataset that consists of only homogenous and stable atmospheric condition for the Langley calibration purposes. In this paper, a radiance-based validation method is performed to further investigate the feasibility and consistency of the proposed algorithm at different location, day, and time. The algorithm is validated using SMARTS model based n DNI value. The overall results confirmed that the proposed calibration algorithm feasible and consistent for measurements taken at different sites and weather conditions.

Keywords: Aerosol optical depth, direct normal irradiance, Langley calibration, radiance-based validation, SMARTS.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1808

Authors: Chiung-Ju Huang, Yuan-Hong Ho

Abstract:

This study utilizes the International Monetary Fund (IMF) Fiscal Rules Dataset focusing on four specific fiscal rules such as expenditure rule, revenue rule, budget balance rule, and debt rule and five main characteristics of each fiscal rule those are monitoring, enforcement, coverage, legal basis, and escape clause to construct the Fiscal Rule Index for nine countries in the Asia-Pacific region from 1996 to 2015. After constructing the fiscal rule index for each country, we utilize the Panel Generalized Method of Moments (Panel GMM) by using the constructed fiscal rule index to examine the effectiveness of fiscal rules in reducing procyclicality. Empirical results show that national fiscal rules have a significantly negative impact on procyclicality of government expenditure. Additionally, stricter fiscal rules combined with high government effectiveness are effective in reducing procyclicality of government expenditure. Results of this study indicate that for nine Asia-Pacific countries, policymakers’ use of fiscal rules and government effectiveness to reducing procyclicality of fiscal policy are effective.

Keywords: Counter-cyclical policy, fiscal rules, government effectiveness, procyclical policy.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 825

Authors: A. Shukla, A. Tarsauliya, R. Tiwari, S. Sharma

Abstract:

Cancer classification to their corresponding cohorts has been key area of research in bioinformatics aiming better prognosis of the disease. High dimensionality of gene data has been makes it a complex task and requires significance data identification technique in order to reducing the dimensionality and identification of significant information. In this paper, we have proposed a novel approach for classification of oral cancer into metastasis positive and negative patients. We have used significance analysis of microarrays (SAM) for identifying significant genes which constitutes gene signature. 3 different gene signatures were identified using SAM from 3 different combination of training datasets and their classification accuracy was calculated on corresponding testing datasets using k-Nearest Neighbour (kNN), Fuzzy C-Means Clustering (FCM), Support Vector Machine (SVM) and Backpropagation Neural Network (BPNN). A final gene signature of only 9 genes was obtained from above 3 individual gene signatures. 9 gene signature-s classification capability was compared using same classifiers on same testing datasets. Results obtained from experimentation shows that 9 gene signature classified all samples in testing dataset accurately while individual genes could not classify all accurately.

Keywords: Cancer, Gene Signature, SAM, Classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2076

Authors: Liwen Hou

Abstract:

The booming mobile business has been penetrating the taxi industry worldwide with P2P (peer to peer) taxi services, as an emerging business model, transforming the industry. Parallel with other mobile businesses, member recommendations and online reviews are believed to be very effective with regard to acquiring new users for P2P taxi services. Based on an empirical dataset of the taxi industry in China, this study aims to reveal which factors influence users’ recommendations and review-writing behaviors. Differing from the existing literature, this paper takes the taxi driver’s perspective into consideration and hence selects a group of variables related to the drivers. We built two models to reflect the factors that influence the number of recommendations and reviews posted on the platform (i.e., the app). Our models show that all factors, except the driver’s score, significantly influence the recommendation behavior. Likewise, only one factor, passengers’ bad reviews, is insignificant in generating more drivers’ reviews. In the conclusion, we summarize the findings and limitations of the research.

Keywords: Online recommendation, P2P taxi service, review-writing, word of mouth.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1379

Authors: S. Piramu Kailasam

Abstract:

This study is expected to compress true color image with compression algorithms in color spaces to provide high compression rates. The need of high compression ratio is to improve storage space. Alternative aim is to rank compression algorithms in a suitable color space. The dataset is sequence of true color images with size 128 x 128. HAAR Wavelet is one of the famous wavelet transforms, has great potential and maintains image quality of color images. HAAR wavelet Transform using Set Partitioning in Hierarchical Trees (SPIHT) algorithm with different color spaces framework is applied to compress sequence of images with angles. Embedded Zerotrees of Wavelet (EZW) is a powerful standard method to sequence data. Hence the proposed compression frame work of HAAR wavelet, xyz color space, morphological gradient and applied image with EZW compression, obtained improvement to other methods, in terms of Compression Ratio, Mean Square Error, Peak Signal Noise Ratio and Bits Per Pixel quality measures.

Keywords: Color Spaces, HAAR Wavelet, Morphological Gradient, Embedded Zerotrees Wavelet Compression.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 518

Authors: A. Borji, M. Hamidi

Abstract:

A new method for color image segmentation using fuzzy logic is proposed in this paper. Our aim here is to automatically produce a fuzzy system for color classification and image segmentation with least number of rules and minimum error rate. Particle swarm optimization is a sub class of evolutionary algorithms that has been inspired from social behavior of fishes, bees, birds, etc, that live together in colonies. We use comprehensive learning particle swarm optimization (CLPSO) technique to find optimal fuzzy rules and membership functions because it discourages premature convergence. Here each particle of the swarm codes a set of fuzzy rules. During evolution, a population member tries to maximize a fitness criterion which is here high classification rate and small number of rules. Finally, particle with the highest fitness value is selected as the best set of fuzzy rules for image segmentation. Our results, using this method for soccer field image segmentation in Robocop contests shows 89% performance. Less computational load is needed when using this method compared with other methods like ANFIS, because it generates a smaller number of fuzzy rules. Large train dataset and its variety, makes the proposed method invariant to illumination noise

Keywords: Comprehensive learning Particle Swarmoptimization, fuzzy classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1956

Authors: Umar Albalawi, Sang C. Suh, Jinoh Kim

Abstract:

As internet continues to expand its usage with an  enormous number of applications, cyber-threats have significantly  increased accordingly. Thus, accurate detection of malicious traffic in  a timely manner is a critical concern in today’s Internet for security.  One approach for intrusion detection is to use Machine Learning (ML)  techniques. Several methods based on ML algorithms have been  introduced over the past years, but they are largely limited in terms of  detection accuracy and/or time and space complexity to run. In this  work, we present a novel method for intrusion detection that  incorporates a set of supervised learning algorithms. The proposed  technique provides high accuracy and outperforms existing techniques  that simply utilizes a single learning method. In addition, our  technique relies on partial flow information (rather than full  information) for detection, and thus, it is light-weight and desirable for  online operations with the property of early identification. With the  mid-Atlantic CCDC intrusion dataset publicly available, we show that  our proposed technique yields a high degree of detection rate over 99%  with a very low false alarm rate (0.4%). 

 

Keywords: Intrusion Detection, Supervised Learning, Traffic Classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2035

Authors: Yakin Hajlaoui, Richard Labib, Jean-Franc¸ois Plante, Michel Gamache

Abstract:

This study presents the Multi-Layer Inverse Distance Weighting Model (ML-IDW), inspired by the mathematical formulation of both multi-layer neural networks (ML-NNs) and Inverse Distance Weighting model (IDW). ML-IDW leverages ML-NNs’ processing capabilities, characterized by compositions of learnable non-linear functions applied to input features, and incorporates IDW’s ability to learn anisotropic spatial dependencies, presenting a promising solution for nonlinear spatial interpolation and learning from complex spatial data. We employ gradient descent and backpropagation to train ML-IDW. The performance of the proposed model is compared against conventional spatial interpolation models such as Kriging and standard IDW on regression and classification tasks using simulated spatial datasets of varying complexity. Our results highlight the efficacy of ML-IDW, particularly in handling complex spatial dataset, exhibiting lower mean square error in regression and higher F1 score in classification.

Keywords: Deep Learning, Multi-Layer Neural Networks, Gradient Descent, Spatial Interpolation, Inverse Distance Weighting.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 39

Authors: Rozilawati Binti Dollah, Masaki Aono

Abstract:

Classifying biomedical literature is a difficult and challenging task, especially when a large number of biomedical articles should be organized into a hierarchical structure. In this paper, we present an approach for classifying a collection of biomedical text abstracts downloaded from Medline database with the help of ontology alignment. To accomplish our goal, we construct two types of hierarchies, the OHSUMED disease hierarchy and the Medline abstract disease hierarchies from the OHSUMED dataset and the Medline abstracts, respectively. Then, we enrich the OHSUMED disease hierarchy before adapting it to ontology alignment process for finding probable concepts or categories. Subsequently, we compute the cosine similarity between the vector in probable concepts (in the “enriched" OHSUMED disease hierarchy) and the vector in Medline abstract disease hierarchies. Finally, we assign category to the new Medline abstracts based on the similarity score. The results obtained from the experiments show the performance of our proposed approach for hierarchical classification is slightly better than the performance of the multi-class flat classification.

Keywords: Biomedical literature, hierarchical text classification, ontology alignment, text mining.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2011

Authors: Alexander Goncharenko, Andrey Denisov, Sergey Alyamkin, Evgeny Terentev

Abstract:

The neural network quantization is highly desired procedure to perform before running neural networks on mobile devices. Quantization without fine-tuning leads to accuracy drop of the model, whereas commonly used training with quantization is done on the full set of the labeled data and therefore is both time- and resource-consuming. Real life applications require simplification and acceleration of quantization procedure that will maintain accuracy of full-precision neural network, especially for modern mobile neural network architectures like Mobilenet-v1, MobileNet-v2 and MNAS. Here we present a method to significantly optimize training with quantization procedure by introducing the trained scale factors for discretization thresholds that are separate for each filter. Using the proposed technique, we quantize the modern mobile architectures of neural networks with the set of train data of only ∼ 10% of the total ImageNet 2012 sample. Such reduction of train dataset size and small number of trainable parameters allow to fine-tune the network for several hours while maintaining the high accuracy of quantized model (accuracy drop was less than 0.5%). Ready-for-use models and code are available in the GitHub repository.

Keywords: Distillation, machine learning, neural networks, quantization.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 732

Authors: Adriano Z. Zambom, Preethi Ravikumar

Abstract:

One of the biggest challenges in nonparametric regression is the curse of dimensionality. Additive models are known to overcome this problem by estimating only the individual additive effects of each covariate. However, if the model is misspecified, the accuracy of the estimator compared to the fully nonparametric one is unknown. In this work the efficiency of completely nonparametric regression estimators such as the Loess is compared to the estimators that assume additivity in several situations, including additive and non-additive regression scenarios. The comparison is done by computing the oracle mean square error of the estimators with regards to the true nonparametric regression function. Then, a backward elimination selection procedure based on the Akaike Information Criteria is proposed, which is computed from either the additive or the nonparametric model. Simulations show that if the additive model is misspecified, the percentage of time it fails to select important variables can be higher than that of the fully nonparametric approach. A dimension reduction step is included when nonparametric estimator cannot be computed due to the curse of dimensionality. Finally, the Boston housing dataset is analyzed using the proposed backward elimination procedure and the selected variables are identified.

Keywords: Additive models, local polynomial regression, residuals, mean square error, variable selection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1011

Authors: Sunita Jahirabadkar, Parag Kulkarni

Abstract:

Many real-world data sets consist of a very high dimensional feature space. Most clustering techniques use the distance or similarity between objects as a measure to build clusters. But in high dimensional spaces, distances between points become relatively uniform. In such cases, density based approaches may give better results. Subspace Clustering algorithms automatically identify lower dimensional subspaces of the higher dimensional feature space in which clusters exist. In this paper, we propose a new clustering algorithm, ISC – Intelligent Subspace Clustering, which tries to overcome three major limitations of the existing state-of-art techniques. ISC determines the input parameter such as є – distance at various levels of Subspace Clustering which helps in finding meaningful clusters. The uniform parameters approach is not suitable for different kind of databases. ISC implements dynamic and adaptive determination of Meaningful clustering parameters based on hierarchical filtering approach. Third and most important feature of ISC is the ability of incremental learning and dynamic inclusion and exclusions of subspaces which lead to better cluster formation.

Keywords: Density based clustering, high dimensional data, subspace clustering, dynamic parameter setting.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2018

Authors: Parvinder S. Sandhu, Urvashi Malhotra, E. Ardil

Abstract:

There is lot of work done in prediction of the fault proneness of the software systems. But, it is the severity of the faults that is more important than number of faults existing in the developed system as the major faults matters most for a developer and those major faults needs immediate attention. In this paper, we tried to predict the level of impact of the existing faults in software systems. Neuro-Fuzzy based predictor models is applied NASA-s public domain defect dataset coded in C programming language. As Correlation-based Feature Selection (CFS) evaluates the worth of a subset of attributes by considering the individual predictive ability of each feature along with the degree of redundancy between them. So, CFS is used for the selecting the best metrics that have highly correlated with level of severity of faults. The results are compared with the prediction results of Logistic Models (LMT) that was earlier quoted as the best technique in [17]. The results are recorded in terms of Accuracy, Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). The results show that Neuro-fuzzy based model provide a relatively better prediction accuracy as compared to other models and hence, can be used for the modeling of the level of impact of faults in function based systems.

Keywords: Software Metrics, Fuzzy, Neuro-Fuzzy, Software Faults, Accuracy, MAE, RMSE.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1356

Authors: Alireza Keshmiri, Shahriar Bagheri, Nan Wu

Abstract:

This research presents an analytical model for the development of an energy harvester with piezoelectric rings stacked at the boundary of the structure based on the Adomian decomposition method. The model is applied to geometrically non-uniform beams to derive the steady-state dynamic response of the structure subjected to base motion excitation and efficiently harvest the subsequent vibrational energy. The in-plane polarization of the piezoelectric rings is employed to enhance the electrical power output. A parametric study for the proposed energy harvester with various design parameters is done to prepare the dataset required for optimization. Finally, simulation-based optimization technique helps to find the optimum structural design with maximum efficiency. To solve the optimization problem, an artificial neural network is first trained to replace the simulation model, and then, a genetic algorithm is employed to find the optimized design variables. Higher geometrical non-uniformity and length of the beam lowers the structure natural frequency and generates a larger power output.

Keywords: Piezoelectricity, energy harvesting, simulation-based optimization, artificial neural network, genetic algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 845

Authors: Lina Galinskaitė, Gytautas Ignatavičius

Abstract:

Animal vehicle collisions (AVCs) affect human safety, cause property damage and wildlife welfare. The number of AVCs are increasing and creating serious implications for the animal conservation and management. Roe deer (Capreolus capreolus) and other large ungulates (moose, wild boar, red deer) are the most frequently collided ungulate with vehicles in Europe. Therefore, we analyzed temporal patterns of roe deer vehicle collisions (RDVC) occurring in Lithuania. Using a comprehensive dataset, consisting of 15,891 data points, we examined the influence of different time units (i.e. time of the day, day of week, month, and season) on RDVC. We identified accident periods within the analyzed time units. Highest frequencies of RDVC occurred on Fridays. Highest frequencies of roe deer-vehicle accidents occurred in May, November and December. Regarding diurnal patterns, most of RDVC occur after sunset and before sunset (during dark hours). Since vehicle collisions with animals showed temporal variation, these should be taken into consideration in developing statistical models of spatial AVC patterns, and also in planning strategies to reduce accident risk.

Keywords: Animal vehicle collision, diurnal patterns, road safety, roe deer, statistical analysis.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 497

Authors: Kuo-Yang Cheng, Yi-Lun Pan, Chang-Hsing Wu, His-En Yu, Hui-Shan Chen, Weicheng Huang

Abstract:

The main mission of Ezilla is to provide a friendly interface to access the virtual machine and quickly deploy the high performance computing environment. Ezilla has been developed by Pervasive Computing Team at National Center for High-performance Computing (NCHC). Ezilla integrates the Cloud middleware, virtualization technology, and Web-based Operating System (WebOS) to form a virtual computer in distributed computing environment. In order to upgrade the dataset and speedup, we proposed the sensor observation system to deal with a huge amount of data in the Cassandra database. The sensor observation system is based on the Ezilla to store sensor raw data into distributed database. We adopt the Ezilla Cloud service to create virtual machines and login into virtual machine to deploy the sensor observation system. Integrating the sensor observation system with Ezilla is to quickly deploy experiment environment and access a huge amount of data with distributed database that support the replication mechanism to protect the data security.

Keywords: Cloud, Virtualization, Cassandra, WebOS

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1869

Authors: S. Bader, N. Essoukri Ben Amara

Abstract:

With the technological development and rise of virtual worlds, these spaces are becoming more and more attractive for cybercriminals, hidden behind avatars and fictitious identities. Since access to these spaces is not restricted or controlled, some impostors take advantage of gaining unauthorized access and practicing cyber criminality. This paper proposes an identity management approach for securing access to virtual worlds. The major purpose of the suggested solution is to install a strong security mechanism to protect virtual identities represented by avatars. Thus, only legitimate users, through their corresponding avatars, are allowed to access the platform resources. Access is controlled by integrating an authentication process based on biometrics. In the request process for registration, a user fingerprint is enrolled and then encrypted into a watermark utilizing a cancelable and non-invertible algorithm for its protection. After a user personalizes their representative character, the biometric mark is embedded into the avatar through a watermarking procedure. The authenticity of the avatar identity is verified when it requests authorization for access. We have evaluated the proposed approach on a dataset of avatars from various virtual worlds, and we have registered promising performance results in terms of authentication accuracy, acceptation and rejection rates.

Keywords: Identity management, security, biometrics authentication and authorization, avatar, virtual world.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1657