Search results for: Clusterization and classification algorithms
1750 Prediction of Writer Using Tamil Handwritten Document Image Based on Pooled Features
Authors: T. Thendral, M. S. Vijaya, S. Karpagavalli
Abstract:
Tamil handwritten document is taken as a key source of data to identify the writer. Tamil is a classical language which has 247 characters include compound characters, consonants, vowels and special character. Most characters of Tamil are multifaceted in nature. Handwriting is a unique feature of an individual. Writer may change their handwritings according to their frame of mind and this place a risky challenge in identifying the writer. A new discriminative model with pooled features of handwriting is proposed and implemented using support vector machine. It has been reported on 100% of prediction accuracy by RBF and polynomial kernel based classification model.Keywords: Classification, Feature extraction, Support vector machine, Training, Writer.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16901749 IMLFQ Scheduling Algorithm with Combinational Fault Tolerant Method
Authors: MohammadReza EffatParvar, Akbar Bemana, Mehdi EffatParvar
Abstract:
Scheduling algorithms are used in operating systems to optimize the usage of processors. One of the most efficient algorithms for scheduling is Multi-Layer Feedback Queue (MLFQ) algorithm which uses several queues with different quanta. The most important weakness of this method is the inability to define the optimized the number of the queues and quantum of each queue. This weakness has been improved in IMLFQ scheduling algorithm. Number of the queues and quantum of each queue affect the response time directly. In this paper, we review the IMLFQ algorithm for solving these problems and minimizing the response time. In this algorithm Recurrent Neural Network has been utilized to find both the number of queues and the optimized quantum of each queue. Also in order to prevent any probable faults in processes' response time computation, a new fault tolerant approach has been presented. In this approach we use combinational software redundancy to prevent the any probable faults. The experimental results show that using the IMLFQ algorithm results in better response time in comparison with other scheduling algorithms also by using fault tolerant mechanism we improve IMLFQ performance.Keywords: IMLFQ, Fault Tolerant, Scheduling, Queue, Recurrent Neural Network.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15211748 On-line Image Mosaicing of Live Stem Cells
Authors: Alessandro Bevilacqua, Alessandro Gherardi, Filippo Piccinini
Abstract:
Image mosaicing is a technique that permits to enlarge the field of view of a camera. For instance, it is employed to achieve panoramas with common cameras or even in scientific applications, to achieve the image of a whole culture in microscopical imaging. Usually, a mosaic of cell cultures is achieved through using automated microscopes. However, this is often performed in batch, through CPU intensive minimization algorithms. In addition, live stem cells are studied in phase contrast, showing a low contrast that cannot be improved further. We present a method to study the flat field from live stem cells images even in case of 100% confluence, this permitting to build accurate mosaics on-line using high performance algorithms.
Keywords: Microscopy, image mosaicing, stem cells.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14931747 An Expert System Designed to Be Used with MOEAs for Efficient Portfolio Selection
Authors: K. Metaxiotis, K. Liagkouras
Abstract:
This study presents an Expert System specially designed to be used with Multiobjective Evolutionary Algorithms (MOEAs) for the solution of the portfolio selection problem. The validation of the proposed hybrid System is done by using data sets from Hang Seng 31 in Hong Kong, DAX 100 in Germany and FTSE 100 in UK. The performance of the proposed system is assessed in comparison with the Non-dominated Sorting Genetic Algorithm II (NSGAII). The evaluation of the performance is based on different performance metrics that evaluate both the proximity of the solutions to the Pareto front and their dispersion on it. The results show that the proposed hybrid system is efficient for the solution of this kind of problems.
Keywords: Expert Systems, Multiobjective optimization, Evolutionary Algorithms, Portfolio Selection.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 17481746 Mimicking Morphogenesis for Robust Behaviour of Cellular Architectures
Authors: David Jones, Richard McWilliam, Alan Purvis
Abstract:
Morphogenesis is the process that underpins the selforganised development and regeneration of biological systems. The ability to mimick morphogenesis in artificial systems has great potential for many engineering applications, including production of biological tissue, design of robust electronic systems and the co-ordination of parallel computing. Previous attempts to mimick these complex dynamics within artificial systems have relied upon the use of evolutionary algorithms that have limited their size and complexity. This paper will present some insight into the underlying dynamics of morphogenesis, then show how to, without the assistance of evolutionary algorithms, design cellular architectures that converge to complex patterns.
Keywords: Morphogenesis, regeneration, robustness, convergence, cellular automata.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14811745 Extraction of Significant Phrases from Text
Authors: Yuan J. Lui
Abstract:
Prospective readers can quickly determine whether a document is relevant to their information need if the significant phrases (or keyphrases) in this document are provided. Although keyphrases are useful, not many documents have keyphrases assigned to them, and manually assigning keyphrases to existing documents is costly. Therefore, there is a need for automatic keyphrase extraction. This paper introduces a new domain independent keyphrase extraction algorithm. The algorithm approaches the problem of keyphrase extraction as a classification task, and uses a combination of statistical and computational linguistics techniques, a new set of attributes, and a new machine learning method to distinguish keyphrases from non-keyphrases. The experiments indicate that this algorithm performs better than other keyphrase extraction tools and that it significantly outperforms Microsoft Word 2000-s AutoSummarize feature. The domain independence of this algorithm has also been confirmed in our experiments.
Keywords: classification, keyphrase extraction, machine learning, summarization
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 20391744 A New Heuristic Approach for the Large-Scale Generalized Assignment Problem
Authors: S. Raja Balachandar, K.Kannan
Abstract:
This paper presents a heuristic approach to solve the Generalized Assignment Problem (GAP) which is NP-hard. It is worth mentioning that many researches used to develop algorithms for identifying the redundant constraints and variables in linear programming model. Some of the algorithms are presented using intercept matrix of the constraints to identify redundant constraints and variables prior to the start of the solution process. Here a new heuristic approach based on the dominance property of the intercept matrix to find optimal or near optimal solution of the GAP is proposed. In this heuristic, redundant variables of the GAP are identified by applying the dominance property of the intercept matrix repeatedly. This heuristic approach is tested for 90 benchmark problems of sizes upto 4000, taken from OR-library and the results are compared with optimum solutions. Computational complexity is proved to be O(mn2) of solving GAP using this approach. The performance of our heuristic is compared with the best state-ofthe- art heuristic algorithms with respect to both the quality of the solutions. The encouraging results especially for relatively large size test problems indicate that this heuristic approach can successfully be used for finding good solutions for highly constrained NP-hard problems.
Keywords: Combinatorial Optimization Problem, Generalized Assignment Problem, Intercept Matrix, Heuristic, Computational Complexity, NP-Hard Problems.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 23361743 Energy Detection Based Sensing and Primary User Traffic Classification for Cognitive Radio
Authors: Urvee B. Trivedi, U. D. Dalal
Abstract:
As wireless communication services grow quickly; the seriousness of spectrum utilization has been on the rise gradually. An emerging technology, cognitive radio has come out to solve today’s spectrum scarcity problem. To support the spectrum reuse functionality, secondary users are required to sense the radio frequency environment, and once the primary users are found to be active, the secondary users are required to vacate the channel within a certain amount of time. Therefore, spectrum sensing is of significant importance. Once sensing is done, different prediction rules apply to classify the traffic pattern of primary user. Primary user follows two types of traffic patterns: periodic and stochastic ON-OFF patterns. A cognitive radio can learn the patterns in different channels over time. Two types of classification methods are discussed in this paper, by considering edge detection and by using autocorrelation function. Edge detection method has a high accuracy but it cannot tolerate sensing errors. Autocorrelation-based classification is applicable in the real environment as it can tolerate some amount of sensing errors.Keywords: Cognitive radio (CR), probability of detection (PD), probability of false alarm (PF), primary User (PU), secondary user (SU), Fast Fourier transform (FFT), signal to noise ratio (SNR).
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14561742 Calcification Classification in Mammograms Using Decision Trees
Authors: S. Usha, S. Arumugam
Abstract:
Cancer affects people globally with breast cancer being a leading killer. Breast cancer is due to the uncontrollable multiplication of cells resulting in a tumour or neoplasm. Tumours are called ‘benign’ when cancerous cells do not ravage other body tissues and ‘malignant’ if they do so. As mammography is an effective breast cancer detection tool at an early stage which is the most treatable stage it is the primary imaging modality for screening and diagnosis of this cancer type. This paper presents an automatic mammogram classification technique using wavelet and Gabor filter. Correlation feature selection is used to reduce the feature set and selected features are classified using different decision trees.
Keywords: Breast Cancer, Mammogram, Symlet Wavelets, Gabor Filters, Decision Trees
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 17361741 Classifier Based Text Mining for Neural Network
Authors: M. Govindarajan, R. M. Chandrasekaran
Abstract:
Text Mining is around applying knowledge discovery techniques to unstructured text is termed knowledge discovery in text (KDT), or Text data mining or Text Mining. In Neural Network that address classification problems, training set, testing set, learning rate are considered as key tasks. That is collection of input/output patterns that are used to train the network and used to assess the network performance, set the rate of adjustments. This paper describes a proposed back propagation neural net classifier that performs cross validation for original Neural Network. In order to reduce the optimization of classification accuracy, training time. The feasibility the benefits of the proposed approach are demonstrated by means of five data sets like contact-lenses, cpu, weather symbolic, Weather, labor-nega-data. It is shown that , compared to exiting neural network, the training time is reduced by more than 10 times faster when the dataset is larger than CPU or the network has many hidden units while accuracy ('percent correct') was the same for all datasets but contact-lences, which is the only one with missing attributes. For contact-lences the accuracy with Proposed Neural Network was in average around 0.3 % less than with the original Neural Network. This algorithm is independent of specify data sets so that many ideas and solutions can be transferred to other classifier paradigms.Keywords: Back propagation, classification accuracy, textmining, time complexity.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 42011740 Soft Real-Time Fuzzy Task Scheduling for Multiprocessor Systems
Authors: Mahdi Hamzeh, Sied Mehdi Fakhraie, Caro Lucas
Abstract:
All practical real-time scheduling algorithms in multiprocessor systems present a trade-off between their computational complexity and performance. In real-time systems, tasks have to be performed correctly and timely. Finding minimal schedule in multiprocessor systems with real-time constraints is shown to be NP-hard. Although some optimal algorithms have been employed in uni-processor systems, they fail when they are applied in multiprocessor systems. The practical scheduling algorithms in real-time systems have not deterministic response time. Deterministic timing behavior is an important parameter for system robustness analysis. The intrinsic uncertainty in dynamic real-time systems increases the difficulties of scheduling problem. To alleviate these difficulties, we have proposed a fuzzy scheduling approach to arrange real-time periodic and non-periodic tasks in multiprocessor systems. Static and dynamic optimal scheduling algorithms fail with non-critical overload. In contrast, our approach balances task loads of the processors successfully while consider starvation prevention and fairness which cause higher priority tasks have higher running probability. A simulation is conducted to evaluate the performance of the proposed approach. Experimental results have shown that the proposed fuzzy scheduler creates feasible schedules for homogeneous and heterogeneous tasks. It also and considers tasks priorities which cause higher system utilization and lowers deadline miss time. According to the results, it performs very close to optimal schedule of uni-processor systems.Keywords: Computational complexity, Deadline, Feasible scheduling, Fuzzy scheduling, Priority, Real-time multiprocessor systems, Robustness, System utilization.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 21151739 Parameters Extraction for Pseudomorphic HEMTs Using Genetic Algorithms
Authors: Mazhar B. Tayel, Amr H. Yassin
Abstract:
A proposed small-signal model parameters for a pseudomorphic high electron mobility transistor (PHEMT) is presented. Both extrinsic and intrinsic circuit elements of a smallsignal model are determined using genetic algorithm (GA) as a stochastic global search and optimization tool. The parameters extraction of the small-signal model is performed on 200-μm gate width AlGaAs/InGaAs PHEMT. The equivalent circuit elements for a proposed 18 elements model are determined directly from the measured S- parameters. The GA is used to extract the parameters of the proposed small-signal model from 0.5 up to 18 GHz.
Keywords: PHEMT, Genetic Algorithms, small signal modeling, optimization.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 22531738 An Empirical Evaluation of Performance of Machine Learning Techniques on Imbalanced Software Quality Data
Authors: Ruchika Malhotra, Megha Khanna
Abstract:
The development of change prediction models can help the software practitioners in planning testing and inspection resources at early phases of software development. However, a major challenge faced during the training process of any classification model is the imbalanced nature of the software quality data. A data with very few minority outcome categories leads to inefficient learning process and a classification model developed from the imbalanced data generally does not predict these minority categories correctly. Thus, for a given dataset, a minority of classes may be change prone whereas a majority of classes may be non-change prone. This study explores various alternatives for adeptly handling the imbalanced software quality data using different sampling methods and effective MetaCost learners. The study also analyzes and justifies the use of different performance metrics while dealing with the imbalanced data. In order to empirically validate different alternatives, the study uses change data from three application packages of open-source Android data set and evaluates the performance of six different machine learning techniques. The results of the study indicate extensive improvement in the performance of the classification models when using resampling method and robust performance measures.Keywords: Change proneness, empirical validation, imbalanced learning, machine learning techniques, object-oriented metrics.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15071737 Control of an Asymmetrical Design of a Pneumatically Actuated Ambidextrous Robot Hand
Authors: Emre Akyürek, Anthony Huynh, Tatiana Kalganova
Abstract:
The Ambidextrous Robot Hand is a robotic device with the purpose to mimic either the gestures of a right or a left hand. The symmetrical behavior of its fingers allows them to bend in one way or another keeping a compliant and anthropomorphic shape. However, in addition to gestures they can reproduce on both sides, an asymmetrical mechanical design with a three tendons routing has been engineered to reduce the number of actuators. As a consequence, control algorithms must be adapted to drive efficiently the ambidextrous fingers from one position to another and to include grasping features. These movements are controlled by pneumatic muscles, which are nonlinear actuators. As their elasticity constantly varies when they are under actuation, the length of pneumatic muscles and the force they provide may differ for a same value of pressurized air. The control algorithms introduced in this paper take both the fingers asymmetrical design and the pneumatic muscles nonlinearity into account to permit an accurate control of the Ambidextrous Robot Hand. The finger motion is achieved by combining a classic PID controller with a phase plane switching control that turns the gain constants into dynamic values. The grasping ability is made possible because of a sliding mode control that makes the fingers adapt to the shape of an object before strengthening their positions.
Keywords: Ambidextrous hand, intelligent algorithms, nonlinear actuators, pneumatic muscles, robotics, sliding control.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 22321736 Sentiment Analysis of Fake Health News Using Naive Bayes Classification Models
Authors: Danielle Shackley, Yetunde Folajimi
Abstract:
As more people turn to the internet seeking health related information, there is more risk of finding false, inaccurate, or dangerous information. Sentiment analysis is a natural language processing technique that assigns polarity scores of text, ranging from positive, neutral and negative. In this research, we evaluate the weight of a sentiment analysis feature added to fake health news classification models. The dataset consists of existing reliably labeled health article headlines that were supplemented with health information collected about COVID-19 from social media sources. We started with data preprocessing, tested out various vectorization methods such as Count and TFIDF vectorization. We implemented 3 Naive Bayes classifier models, including Bernoulli, Multinomial and Complement. To test the weight of the sentiment analysis feature on the dataset, we created benchmark Naive Bayes classification models without sentiment analysis, and those same models were reproduced and the feature was added. We evaluated using the precision and accuracy scores. The Bernoulli initial model performed with 90% precision and 75.2% accuracy, while the model supplemented with sentiment labels performed with 90.4% precision and stayed constant at 75.2% accuracy. Our results show that the addition of sentiment analysis did not improve model precision by a wide margin; while there was no evidence of improvement in accuracy, we had a 1.9% improvement margin of the precision score with the Complement model. Future expansion of this work could include replicating the experiment process, and substituting the Naive Bayes for a deep learning neural network model.
Keywords: Sentiment analysis, Naive Bayes model, natural language processing, topic analysis, fake health news classification model.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4391735 Comparative Analysis of Classical and Parallel Inpainting Algorithms Based on Affine Combinations of Projections on Convex Sets
Authors: Irina Maria Artinescu, Costin Radu Boldea, Eduard-Ionut Matei
Abstract:
The paper is a comparative study of two classical vari-ants of parallel projection methods for solving the convex feasibility problem with their equivalents that involve variable weights in the construction of the solutions. We used a graphical representation of these methods for inpainting a convex area of an image in order to investigate their effectiveness in image reconstruction applications. We also presented a numerical analysis of the convergence of these four algorithms in terms of the average number of steps and execution time, in classical CPU and, alternativaly, in parallel GPU implementation.
Keywords: convex feasibility problem, convergence analysis, ınpainting, parallel projection methods
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4221734 A New Approach for Classifying Large Number of Mixed Variables
Authors: Hashibah Hamid
Abstract:
The issue of classifying objects into one of predefined groups when the measured variables are mixed with different types of variables has been part of interest among statisticians in many years. Some methods for dealing with such situation have been introduced that include parametric, semi-parametric and nonparametric approaches. This paper attempts to discuss on a problem in classifying a data when the number of measured mixed variables is larger than the size of the sample. A propose idea that integrates a dimensionality reduction technique via principal component analysis and a discriminant function based on the location model is discussed. The study aims in offering practitioners another potential tool in a classification problem that is possible to be considered when the observed variables are mixed and too large.Keywords: classification, location model, mixed variables, principal component analysis.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15421733 Gradual Shot Boundary Detection and Classification Based on Fractal Analysis
Authors: Zeinab Zeinalpour-Tabrizi, Faeze Asdaghi, Mahmooh Fathy, Mohammad Reza Jahed-Motlagh
Abstract:
Shot boundary detection is a fundamental step for the organization of large video data. In this paper, we propose a new method for video gradual shots detection and classification, using advantages of fractal analysis and AIS-based classifier. Proposed features are “vertical intercept" and “fractal dimension" of each frame of videos which are computed using Fourier transform coefficients. We also used a classifier based on Clonal Selection Algorithm. We have carried out our solution and assessed it according to the TRECVID2006 benchmark dataset.
Keywords: shot boundary detection, gradual shots, fractal analysis, artificial immune system, choose Clooney.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19101732 A Study on Neural Network Training Algorithm for Multiface Detection in Static Images
Authors: Zulhadi Zakaria, Nor Ashidi Mat Isa, Shahrel A. Suandi
Abstract:
This paper reports the study results on neural network training algorithm of numerical optimization techniques multiface detection in static images. The training algorithms involved are scale gradient conjugate backpropagation, conjugate gradient backpropagation with Polak-Riebre updates, conjugate gradient backpropagation with Fletcher-Reeves updates, one secant backpropagation and resilent backpropagation. The final result of each training algorithms for multiface detection application will also be discussed and compared.Keywords: training algorithm, multiface, static image, neural network
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 25621731 A Fast Block-based Evolutional Algorithm for Combinatorial Problems
Authors: Huang, Wei-Hsiu Chang, Pei-Chann, Wang, Lien-Chun
Abstract:
The problems with high complexity had been the challenge in combinatorial problems. Due to the none-determined and polynomial characteristics, these problems usually face to unreasonable searching budget. Hence combinatorial optimizations attracted numerous researchers to develop better algorithms. In recent academic researches, most focus on developing to enhance the conventional evolutional algorithms and facilitate the local heuristics, such as VNS, 2-opt and 3-opt. Despite the performances of the introduction of the local strategies are significant, however, these improvement cannot improve the performance for solving the different problems. Therefore, this research proposes a meta-heuristic evolutional algorithm which can be applied to solve several types of problems. The performance validates BBEA has the ability to solve the problems even without the design of local strategies.
Keywords: Combinatorial problems, Artificial Chromosomes, Blocks Mining, Block Recombination
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14061730 Optimal Economic Load Dispatch Using Genetic Algorithms
Authors: Vijay Kumar, Jagdev Singh, Yaduvir Singh, Sanjay Sood
Abstract:
In a practical power system, the power plants are not located at the same distance from the center of loads and their fuel costs are different. Also, under normal operating conditions, the generation capacity is more than the total load demand and losses. Thus, there are many options for scheduling generation. In an interconnected power system, the objective is to find the real and reactive power scheduling of each power plant in such a way as to minimize the operating cost. This means that the generator’s real and reactive powers are allowed to vary within certain limits so as to meet a particular load demand with minimum fuel cost. This is called optimal power flow problem. In this paper, Economic Load Dispatch (ELD) of real power generation is considered. Economic Load Dispatch (ELD) is the scheduling of generators to minimize total operating cost of generator units subjected to equality constraint of power balance within the minimum and maximum operating limits of the generating units. In this paper, genetic algorithms are considered. ELD solutions are found by solving the conventional load flow equations while at the same time minimizing the fuel costs.Keywords: ELD, Equality constraints, Genetic algorithms, Strings.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 38451729 An Engineering Approach to Forecast Volatility of Financial Indices
Authors: Irwin Ma, Tony Wong, Thiagas Sankar
Abstract:
By systematically applying different engineering methods, difficult financial problems become approachable. Using a combination of theory and techniques such as wavelet transform, time series data mining, Markov chain based discrete stochastic optimization, and evolutionary algorithms, this work formulated a strategy to characterize and forecast non-linear time series. It attempted to extract typical features from the volatility data sets of S&P100 and S&P500 indices that include abrupt drops, jumps and other non-linearity. As a result, accuracy of forecasting has reached an average of over 75% surpassing any other publicly available results on the forecast of any financial index.Keywords: Discrete stochastic optimization, genetic algorithms, genetic programming, volatility forecast
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16201728 Multi-Layer Perceptron and Radial Basis Function Neural Network Models for Classification of Diabetic Retinopathy Disease Using Video-Oculography Signals
Authors: Ceren Kaya, Okan Erkaymaz, Orhan Ayar, Mahmut Özer
Abstract:
Diabetes Mellitus (Diabetes) is a disease based on insulin hormone disorders and causes high blood glucose. Clinical findings determine that diabetes can be diagnosed by electrophysiological signals obtained from the vital organs. 'Diabetic Retinopathy' is one of the most common eye diseases resulting on diabetes and it is the leading cause of vision loss due to structural alteration of the retinal layer vessels. In this study, features of horizontal and vertical Video-Oculography (VOG) signals have been used to classify non-proliferative and proliferative diabetic retinopathy disease. Twenty-five features are acquired by using discrete wavelet transform with VOG signals which are taken from 21 subjects. Two models, based on multi-layer perceptron and radial basis function, are recommended in the diagnosis of Diabetic Retinopathy. The proposed models also can detect level of the disease. We show comparative classification performance of the proposed models. Our results show that proposed the RBF model (100%) results in better classification performance than the MLP model (94%).
Keywords: Diabetic retinopathy, discrete wavelet transform, multi-layer perceptron, radial basis function, video-oculography.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 13301727 Curvelet Features with Mouth and Face Edge Ratios for Facial Expression Identification
Authors: S. Kherchaoui, A. Houacine
Abstract:
This paper presents a facial expression recognition system. It performs identification and classification of the seven basic expressions; happy, surprise, fear, disgust, sadness, anger, and neutral states. It consists of three main parts. The first one is the detection of a face and the corresponding facial features to extract the most expressive portion of the face, followed by a normalization of the region of interest. Then calculus of curvelet coefficients is performed with dimensionality reduction through principal component analysis. The resulting coefficients are combined with two ratios; mouth ratio and face edge ratio to constitute the whole feature vector. The third step is the classification of the emotional state using the SVM method in the feature space.
Keywords: Facial expression identification, curvelet coefficients, support vector machine (SVM).
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 18341726 6D Posture Estimation of Road Vehicles from Color Images
Authors: Yoshimoto Kurihara, Tad Gonsalves
Abstract:
Currently, in the field of object posture estimation, there is research on estimating the position and angle of an object by storing a 3D model of the object to be estimated in advance in a computer and matching it with the model. However, in this research, we have succeeded in creating a module that is much simpler, smaller in scale, and faster in operation. Our 6D pose estimation model consists of two different networks – a classification network and a regression network. From a single RGB image, the trained model estimates the class of the object in the image, the coordinates of the object, and its rotation angle in 3D space. In addition, we compared the estimation accuracy of each camera position, i.e., the angle from which the object was captured. The highest accuracy was recorded when the camera position was 75°, the accuracy of the classification was about 87.3%, and that of regression was about 98.9%.
Keywords: AlexNet, Deep learning, image recognition, 6D posture estimation.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 5671725 An Improved QRS Complex Detection for Online Medical Diagnosis
Authors: I. L. Ahmad, M. Mohamed, N. A. Ab. Ghani
Abstract:
This paper presents the work of signal discrimination specifically for Electrocardiogram (ECG) waveform. ECG signal is comprised of P, QRS, and T waves in each normal heart beat to describe the pattern of heart rhythms corresponds to a specific individual. Further medical diagnosis could be done to determine any heart related disease using ECG information. The emphasis on QRS Complex classification is further discussed to illustrate the importance of it. Pan-Tompkins Algorithm, a widely known technique has been adapted to realize the QRS Complex classification process. There are eight steps involved namely sampling, normalization, low pass filter, high pass filter (build a band pass filter), derivation, squaring, averaging and lastly is the QRS detection. The simulation results obtained is represented in a Graphical User Interface (GUI) developed using MATLAB.Keywords: ECG, Pan Tompkins Algorithm, QRS Complex, Simulation
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 25621724 Efficiency of Floristic and Molecular Markers to Determine Diversity in Iranian Populations of T. boeoticum
Authors: M. R. Naghavi, M. Maleki, S. F. Tabatabaei
Abstract:
In order to study floristic and molecular classification of common wild wheat (Triticum boeoticum Boiss.), an analysis was conducted on populations of the Triticum boeoticum collected from different regions of Iran. Considering all floristic compositions of habitats, six floristic groups (syntaxa) within the populations were identified. A high level of variation of T. boeoticum also detected using SSR markers. Our results showed that molecular method confirmed the grouping of floristic method. In other word, the results from our study indicate that floristic classification are still useful, efficient, and economic tools for characterizing the amount and distribution of genetic variation in natural populations of T. boeoticum. Nevertheless, molecular markers appear as useful and complementary techniques for identification and for evaluation of genetic diversity in studied populations.Keywords: T. boeoticum, diversity, floristic, SSRs.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 13351723 A Serial Hierarchical Support Vector Machine and 2D Feature Sets Act for Brain DTI Segmentation
Authors: Mohammad Javadi
Abstract:
Serial hierarchical support vector machine (SHSVM) is proposed to discriminate three brain tissues which are white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF). SHSVM has novel classification approach by repeating the hierarchical classification on data set iteratively. It used Radial Basis Function (rbf) Kernel with different tuning to obtain accurate results. Also as the second approach, segmentation performed with DAGSVM method. In this article eight univariate features from the raw DTI data are extracted and all the possible 2D feature sets are examined within the segmentation process. SHSVM succeed to obtain DSI values higher than 0.95 accuracy for all the three tissues, which are higher than DAGSVM results.
Keywords: Brain segmentation, DTI, hierarchical, SVM.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 18441722 Analysis of Sonogram Images of Thyroid Gland Based on Wavelet Transform
Authors: M. Bastanfard, B. Jalaeian, S. Jafari
Abstract:
Sonogram images of normal and lymphocyte thyroid tissues have considerable overlap which makes it difficult to interpret and distinguish. Classification from sonogram images of thyroid gland is tackled in semiautomatic way. While making manual diagnosis from images, some relevant information need not to be recognized by human visual system. Quantitative image analysis could be helpful to manual diagnostic process so far done by physician. Two classes are considered: normal tissue and chronic lymphocyte thyroid (Hashimoto's Thyroid). Data structure is analyzed using K-nearest-neighbors classification. This paper is mentioned that unlike the wavelet sub bands' energy, histograms and Haralick features are not appropriate to distinguish between normal tissue and Hashimoto's thyroid.Keywords: Sonogram, thyroid, Haralick feature, wavelet.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 13091721 Design and Implementation of a Software Platform Based on Artificial Intelligence for Product Recommendation
Authors: G. Settanni, A. Panarese, R. Vaira, A. Galiano
Abstract:
Nowadays, artificial intelligence is used successfully in the field of e-commerce for its ability to learn from a large amount of data. In this research study, a prototype software platform was designed and implemented in order to suggest to users the most suitable products for their needs. The platform includes a recommender system based on artificial intelligence algorithms that provide suggestions and decision support to the customer. Specifically, support vector machine algorithms have been implemented combined with natural language processing techniques that allow the user to interact with the system, express their requests and receive suggestions. The interested user can access the web platform on the internet using a computer, tablet or mobile phone, register, provide the necessary information and view the products that the system deems them the most appropriate. The platform also integrates a dashboard that allows the use of the various functions, which the platform is equipped with, in an intuitive and simple way. Also, Long Short-Term Memory algorithms have been implemented and trained on historical data in order to predict customer scores of the different items. Items with the highest scores are recommended to customers.
Keywords: Deep Learning, Long Short-Term Memory, Machine Learning, Recommender Systems, Support Vector Machine.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 290