Search results for: RSC neural classifier

Authors: Cheng Zeng, George Michailidis, Hitoshi Iyatomi, Leo L. Duan

Abstract:

The conditional density characterizes the distribution of a response variable y given other predictor x and plays a key role in many statistical tasks, including classification and outlier detection. Although there has been abundant work on the problem of Conditional Density Estimation (CDE) for a low-dimensional response in the presence of a high-dimensional predictor, little work has been done for a high-dimensional response such as images. The promising performance of normalizing flow (NF) neural networks in unconditional density estimation acts as a motivating starting point. In this work, the authors extend NF neural networks when external x is present. Specifically, they use the NF to parameterize a one-to-one transform between a high-dimensional y and a latent z that comprises two components [zₚ, zₙ]. The zₚ component is a low-dimensional subvector obtained from the posterior distribution of an elementary predictive model for x, such as logistic/linear regression. The zₙ component is a high-dimensional independent Gaussian vector, which explains the variations in y not or less related to x. Unlike existing CDE methods, the proposed approach coined Augmented Posterior CDE (AP-CDE) only requires a simple modification of the common normalizing flow framework while significantly improving the interpretation of the latent component since zₚ represents a supervised dimension reduction. In image analytics applications, AP-CDE shows good separation of 𝑥-related variations due to factors such as lighting condition and subject id from the other random variations. Further, the experiments show that an unconditional NF neural network based on an unsupervised model of z, such as a Gaussian mixture, fails to generate interpretable results.

Keywords: conditional density estimation, image generation, normalizing flow, supervised dimension reduction

Procedia PDF Downloads 76

Authors: Merouane Boudjerda, Bénina Touaibia, Mustapha K Mihoubi

Abstract:

In Algeria, water resources play a crucial role in economic development. But over the last decades, they are relatively limited and gradually decreasing to the detriment of agriculture. The agricultural irrigation is the primary water consuming sector followed by the domestic and industrial sectors. The research presented in this paper focuses on the optimization of irrigation water demand. Dynamic Programming-Neural Network (DPNN) method is applied to investigate reservoir optimization. The optimal operation rule is formulated to minimize the gap between water release and water irrigation demand. As a case study, Boukerdane dam’s reservoir system in North of Algeria has been selected to examine our proposed optimization model. The application of DPNN method allowed increasing the satisfaction rate (SR) from 34% to 60%. In addition, the operation rule generated showed more reliable and resilience operation for the examined case study.

Keywords: water management, agricultural demand, Boukerdane dam, dynamic programming, artificial neural network

Procedia PDF Downloads 115

Authors: Vijay Pal Singh, Yogesh Chandra Kotiyal

Abstract:

Structures are a combination of various load carrying members which transfer the loads to the foundation from the superstructure safely. At the design stage, the loading of the structure is defined and appropriate material choices are made based upon their properties, mainly related to strength. The strength of materials kept on reducing with time because of many factors like environmental exposure and deformation caused by unpredictable external loads. Hence, to predict the strength of materials used in structures, various techniques are used. Among these techniques, Non-Destructive Techniques (NDT) are the one that can be used to predict the strength without damaging the structure. In the present study, the compressive strength of concrete has been predicted using Artificial Neural Network (ANN). The predicted strength was compared with the experimentally obtained actual compressive strength of concrete and equations were developed for different models. A good co-relation has been obtained between the predicted strength by these models and experimental values. Further, the co-relation has been developed using two NDT techniques for prediction of strength by regression analysis. It was found that the percentage error has been reduced between the predicted strength by using combined techniques in place of single techniques.

Keywords: rebound, ultra-sonic pulse, penetration, ANN, NDT, regression

Procedia PDF Downloads 408

Authors: Mecheri Zeid Belmecheri, Maamar Ahfir, Izzet Kale

Abstract:

Automatic cardiac auscultation is still a subject of research in order to establish an objective diagnosis. Recorded heart sounds as Phonocardiogram signals (PCG) can be used for automatic segmentation into components that have clinical meanings. These are the first sound, S1, the second sound, S2, and the systolic and diastolic components, respectively. In this paper, an automatic method is proposed for the robust segmentation of heart sounds. This method is based on calculating an intermediate sawtooth-shaped signal from the length variation of the recorded Phonocardiogram (PCG) signal in the time domain and, using its positive derivative function that is a binary signal in training a Recurrent Neural Network (RNN). Results obtained in the context of a large database of recorded PCGs with their simultaneously recorded ElectroCardioGrams (ECGs) from different patients in clinical settings, including normal and abnormal subjects, show a segmentation testing performance average of 76 % sensitivity and 94 % specificity.

Keywords: heart sounds, PCG segmentation, event detection, recurrent neural networks, PCG curve length

Procedia PDF Downloads 165

Authors: Amal Khalifa, Nicolas Vana Santos

Abstract:

Steganography has been known for centuries as an efficient approach for covert communication. Due to its popularity and ease of access, image steganography has attracted researchers to find secure techniques for hiding information within an innocent looking cover image. In this research, we propose a novel deep-learning approach to digital image steganography. The proposed method, DeepWaveletFusion, uses convolutional neural networks (CNN) to hide a secret image into a cover image of the same size. Two CNNs are trained back-to-back to merge the Discrete Wavelet Transform (DWT) of both colored images and eventually be able to blindly extract the hidden image. Based on two different image similarity metrics, a weighted gain function is used to guide the learning process and maximize the quality of the retrieved secret image and yet maintaining acceptable imperceptibility. Experimental results verified the high recoverability of DeepWaveletFusion which outperformed similar deep-learning-based methods.

Keywords: deep learning, steganography, image, discrete wavelet transform, fusion

Procedia PDF Downloads 62

Authors: Anoop T. R., Otman Basir, Robert F. Hess, Eileen E. Birch, Brooke A. Koritala, Reed M. Jost, Becky Luu, David Stager, Ben Thompson

Abstract:

Strabismus refers to a misalignment of the eyes. Early detection and treatment of strabismus in childhood can prevent the development of permanent vision loss due to abnormal development of visual brain areas. We developed a two-stage method for strabismus detection and classification based on photographs of the face. The first stage detects the presence or absence of strabismus, and the second stage classifies the type of strabismus. The first stage comprises face detection using Haar cascade, facial landmark estimation, face alignment, aligned face landmark detection, segmentation of the eye region, and detection of strabismus using VGG 16 convolution neural networks. Face alignment transforms the face to a canonical pose to ensure consistency in subsequent analysis. Using facial landmarks, the eye region is segmented from the aligned face and fed into a VGG 16 CNN model, which has been trained to classify strabismus. The CNN determines whether strabismus is present and classifies the type of strabismus (exotropia, esotropia, and vertical deviation). If stage 1 detects strabismus, the eye region image is fed into stage 2, which starts with the estimation of pupil center coordinates using mask R-CNN deep neural networks. Then, the distance between the pupil coordinates and eye landmarks is calculated along with the angle that the pupil coordinates make with the horizontal and vertical axis. The distance and angle information is used to characterize the degree and direction of the strabismic eye misalignment. This model was tested on 100 clinically labeled images of children with (n = 50) and without (n = 50) strabismus. The True Positive Rate (TPR) and False Positive Rate (FPR) of the first stage were 94% and 6% respectively. The classification stage has produced a TPR of 94.73%, 94.44%, and 100% for esotropia, exotropia, and vertical deviations, respectively. This method also had an FPR of 5.26%, 5.55%, and 0% for esotropia, exotropia, and vertical deviation, respectively. The addition of one more feature related to the location of corneal light reflections may reduce the FPR, which was primarily due to children with pseudo-strabismus (the appearance of strabismus due to a wide nasal bridge or skin folds on the nasal side of the eyes).

Keywords: strabismus, deep neural networks, face detection, facial landmarks, face alignment, segmentation, VGG 16, mask R-CNN, pupil coordinates, angle deviation, horizontal and vertical deviation

Procedia PDF Downloads 64

Authors: Li Hui, Riyadh Hindi

Abstract:

In the United States, bridge construction often employs overhang brackets to support the deck overhang, the weight of fresh concrete, and loads from construction equipment. This approach, however, can lead to significant torsional moments on the exterior girders, potentially causing excessive girder rotation. Such rotations can result in various safety and maintenance issues, including thinning of the deck, reduced concrete cover, and cracking during service. Traditionally, these issues are addressed by installing temporary lateral bracing systems and conducting comprehensive torsional analysis through detailed finite element analysis for the construction of bridge deck overhang. However, this process is often intricate and time-intensive, with the spacing between temporary lateral bracing systems usually relying on the field engineers’ expertise. In this study, a deep neural network model is introduced to limit exterior girder rotation during bridge deck construction. The model predicts the optimal spacing between temporary bracing systems. To train this model, over 10,000 finite element models were generated in SAP2000, incorporating varying parameters such as girder dimensions, span length, and types and spacing of lateral bracing systems. The findings demonstrate that the deep neural network provides an effective and efficient alternative for limiting the exterior girder rotation for bridge deck construction. By reducing dependence on extensive finite element analyses, this approach stands out as a significant advancement in improving safety and maintenance effectiveness in the construction of bridge decks.

Keywords: bridge deck construction, exterior girder rotation, deep learning, finite element analysis

Procedia PDF Downloads 44

Authors: Ming-Jen Huang, Chun-Fang Huang, Chiching Wei

Abstract:

With the recent advance of the deep neural network, we observe new applications of NLP (natural language processing) and CV (computer vision) powered by deep neural networks for processing business documents. However, creating a real-world document processing system needs to integrate several NLP and CV tasks, rather than treating them separately. There is a need to have a unified approach for processing documents containing textual and graphical elements with rich formats, diverse layout arrangement, and distinct semantics. In this paper, a framework that fulfills this unified approach is presented. The framework includes a representation model definition for holding the information generated by various tasks and specifications defining the coordination between these tasks. The framework is a blueprint for building a system that can process documents with rich formats, styles, and multiple types of elements. The flexible and lightweight design of the framework can help build a system for diverse business scenarios, such as contract monitoring and reviewing.

Keywords: document processing, framework, formal definition, machine learning

Procedia PDF Downloads 197

Authors: Piyangkun Kukutapan, Siridech Boonsang

Abstract:

The article proposes maximum power point tracking without mechanical sensor using Multilayer Perceptron Neural Network (MLPNN). The aim of article is to reduce the cost and complexity but still retain efficiency. The experimental is that duty cycle is generated maximum power, if it has suitable qualification. The measured data from DC generator, voltage (V), current (I), power (P), turnover rate of power (dP), and turnover rate of voltage (dV) are used as input for MLPNN model. The output of this model is duty cycle for driving the converter. The experiment implemented using Arduino Uno board. This diagram is compared to MPPT using MLPNN and P&O control (Perturbation and Observation control). The experimental results show that the proposed MLPNN based approach is more efficiency than P&O algorithm for this application.

Keywords: maximum power point tracking, multilayer perceptron netural network, optimal duty cycle, DC generator

Procedia PDF Downloads 311

Authors: Somesh Kumar Sharma, Sunanad Gupta

Abstract:

The contribution of aviation to society and economy is undisputedly significant. The aviation industry drives economic and social progress by contributing prominently to tourism, commerce and improved quality of life. Identifying the amount of fuel consumed by an aircraft while moving in both airspace and ground networks is critical to air transport economics. Aviation fuel is a major operating cost parameter of the aviation industry and at the same time it is prone to various constraints. This article aims to develop a model for fuel consumption of aviation product. The paper tailors the information for the fuel consumption optimization in terms of information development, information evaluation and information refinement. The information is evaluated and refined using statistical package R and Factor Analysis which is further validated with neural networking. The study explores three primary dimensions which are finally summarized into 23 influencing variables in contrast to 96 variables available in literature. The 23 variables explored in this study should be considered as highly influencing variables for fuel consumption which will contribute significantly towards fuel optimization.

Keywords: fuel consumption, civil aviation industry, neural networking, optimization

Procedia PDF Downloads 317

Authors: Souhir Ben Amor, Heni Boubaker, Lotfi Belkacem

Abstract:

This aims of this paper is to forecast the electricity spot prices. First, we focus on modeling the conditional mean of the series so we adopt a generalized fractional -factor Gegenbauer process (k-factor GARMA). Secondly, the residual from the -factor GARMA model has used as a proxy for the conditional variance; these residuals were predicted using two different approaches. In the first approach, a local linear wavelet neural network model (LLWNN) has developed to predict the conditional variance using the Back Propagation learning algorithms. In the second approach, the Gegenbauer generalized autoregressive conditional heteroscedasticity process (G-GARCH) has adopted, and the parameters of the k-factor GARMA-G-GARCH model has estimated using the wavelet methodology based on the discrete wavelet packet transform (DWPT) approach. The empirical results have shown that the k-factor GARMA-G-GARCH model outperform the hybrid k-factor GARMA-LLWNN model, and find it is more appropriate for forecasts.

Keywords: electricity price, k-factor GARMA, LLWNN, G-GARCH, forecasting

Procedia PDF Downloads 215

Authors: I. Alhajri, G. Zahedi, R. Alazmi, A. Akbari

Abstract:

In this paper, an Artificial Neural Network (ANN) was developed to predict Asphaltene Precipitation (AP) during the injection of carbon dioxide into crude oil reservoirs. In this study, the experimental data from six different oil fields were collected. Seventy percent of the data was used to develop the ANN model, and different ANN architectures were examined. A network with the Trainlm training algorithm was found to be the best network to estimate the AP. To check the validity of the proposed model, the model was used to predict the AP for the thirty percent of the data that was unevaluated. The Mean Square Error (MSE) of the prediction was 0.0018, which confirms the excellent prediction capability of the proposed model. In the second part of this study, the ANN model predictions were compared with modified Hirschberg model predictions. The ANN was found to provide more accurate estimates compared to the modified Hirschberg model. Finally, the proposed model was employed to examine the effect of different operating parameters during gas injection on the AP. It was found that the AP is mostly sensitive to the reservoir temperature. Furthermore, the carbon dioxide concentration in liquid phase increases the AP.

Keywords: artificial neural network, asphaltene, CO2 injection, Hirschberg model, oil reservoirs

Procedia PDF Downloads 353

Authors: Tesfaye Fenta Boka, Niu Zhendong

Abstract:

Recommender systems play a crucial role in connecting individuals with the items they require, as is particularly evident in the rapid growth of online grocery shopping platforms. These systems predominantly rely on user-centered recommendations, where items are suggested based on individual preferences, garnering considerable attention and adoption. However, our focus lies on the item-centered recommendation task within the grocery shopping context. In the reverse next item period recommendation (RNPR) task, we are presented with a specific item and challenged to identify potential users who are likely to consume it in the upcoming period. Despite the ever-expanding inventory of products on online grocery platforms, the cold start item problem persists, posing a substantial hurdle in delivering personalized and accurate recommendations for new or niche grocery items. To address this challenge, we propose a Graph Neural Network (GNN)-based approach. By capitalizing on the inherent relationships among grocery items and leveraging users' historical interactions, our model aims to provide reliable and context-aware recommendations for cold-start items. This integration of GNN technology holds the promise of enhancing recommendation accuracy and catering to users' individual preferences. This research contributes to the advancement of personalized recommendations in the online grocery shopping domain. By harnessing the potential of GNNs and exploring item-centered recommendation strategies, we aim to improve the overall shopping experience and satisfaction of users on these platforms.

Keywords: recommender systems, cold start item recommendations, online grocery shopping platforms, graph neural networks

Procedia PDF Downloads 69

Authors: Abbas Ahmadi, Bijan Raie, Mohammad Reza Neyshabouri, Mohammad Ali Ghorbani, Farrokh Asadzadeh

Abstract:

In recent years, vast areas of Urmia Lake in Dasht-e-Tabriz has dried up leading to saline sediments exposure on the surface lake coastal areas being highly susceptible to wind erosion. This study was conducted to investigate wind erosion and its relevance to soil physicochemical properties and also modeling of wind erodibility (WE) using artificial intelligence techniques. For this purpose, 96 soil samples were collected from 0-5 cm depth in 414000 hectares using stratified random sampling method. To measure the WE, all samples (<8 mm) were exposed to 5 different wind velocities (9.5, 11, 12.5, 14.1 and 15 m s-1 at the height of 20 cm) in wind tunnel and its relationship with soil physicochemical properties was evaluated. According to the results, WE varied within the range of 76.69-9.98 (g m-2 min-1)/(m s-1) with a mean of 10.21 and coefficient of variation of 94.5% showing a relatively high variation in the studied area. WE was significantly (P<0.01) affected by soil physical properties, including mean weight diameter, erodible fraction (secondary particles smaller than 0.85 mm) and percentage of the secondary particle size classes 2-4.75, 1.7-2 and 0.1-0.25 mm. Results showed that the mean weight diameter, erodible fraction and percentage of size class 0.1-0.25 mm demonstrated stronger relationship with WE (coefficients of determination were 0.69, 0.67 and 0.68, respectively). This study also compared efficiency of multiple linear regression (MLR), gene expression programming (GEP), artificial neural network (MLP), artificial neural network based on genetic algorithm (MLP-GA) and artificial neural network based on whale optimization algorithm (MLP-WOA) in predicting of soil wind erodibility in Dasht-e-Tabriz. Among 32 measured soil variable, percentages of fine sand, size classes of 1.7-2.0 and 0.1-0.25 mm (secondary particles) and organic carbon were selected as the model inputs by step-wise regression. Findings showed MLP-WOA as the most powerful artificial intelligence techniques (R2=0.87, NSE=0.87, ME=0.11 and RMSE=2.9) to predict soil wind erodibility in the study area; followed by MLP-GA, MLP, GEP and MLR and the difference between these methods were significant according to the MGN test. Based on the above finding MLP-WOA may be used as a promising method to predict soil wind erodibility in the study area.

Keywords: wind erosion, erodible fraction, gene expression programming, artificial neural network

Procedia PDF Downloads 49

Authors: Vipul M. Patel, Hemantkumar B. Mehta

Abstract:

Technological innovations in electronic world demand novel, compact, simple in design, less costly and effective heat transfer devices. Closed Loop Pulsating Heat Pipe (CLPHP) is a passive phase change heat transfer device and has potential to transfer heat quickly and efficiently from source to sink. Thermal performance of a CLPHP is governed by various parameters such as number of U-turns, orientations, input heat, working fluids and filling ratio. The present paper is an attempt to predict the thermal performance of a CLPHP using Artificial Neural Network (ANN). Filling ratio and heat input are considered as input parameters while thermal resistance is set as target parameter. Types of neural networks considered in the present paper are radial basis, generalized regression, linear layer, cascade forward back propagation, feed forward back propagation; feed forward distributed time delay, layer recurrent and Elman back propagation. Linear, logistic sigmoid, tangent sigmoid and Radial Basis Gaussian Function are used as transfer functions. Prediction accuracy is measured based on the experimental data reported by the researchers in open literature as a function of Mean Absolute Relative Deviation (MARD). The prediction of a generalized regression ANN model with spread constant of 4.8 is found in agreement with the experimental data for MARD in the range of ±1.81%.

Keywords: ANN models, CLPHP, filling ratio, generalized regression, spread constant

Procedia PDF Downloads 274

Authors: Shima Soltanzadeh, Mohammad Hosain Fazel Zarandi, Mojtaba Barzegar Astanjin

Abstract:

Diagnosis of male infertility by the laboratory tests is expensive and, sometimes it is intolerable for patients. Filling out the questionnaire and then using classification method can be the first step in decision-making process, so only in the cases with a high probability of infertility we can use the laboratory tests. In this paper, we evaluated the performance of four classification methods including naive Bayesian, neural network, logistic regression and fuzzy c-means clustering as a classification, in the diagnosis of male infertility due to environmental factors. Since the data are unbalanced, the ROC curves are most suitable method for the comparison. In this paper, we also have selected the more important features using a filtering method and examined the impact of this feature reduction on the performance of each methods; generally, most of the methods had better performance after applying the filter. We have showed that using fuzzy c-means clustering as a classification has a good performance according to the ROC curves and its performance is comparable to other classification methods like logistic regression.

Keywords: classification, fuzzy c-means, logistic regression, Naive Bayesian, neural network, ROC curve

Procedia PDF Downloads 319

Authors: Alexander Hudson, Shaogang Gong

Abstract:

Structure determination is key to understanding protein function at a molecular level. Whilst significant advances have been made in predicting structure and function from amino acid sequence, researchers must still rely on expensive, time-consuming analytical methods to visualise detailed protein conformation. In this study, we demonstrate that it is possible to make accurate (≥80%) predictions of protein class and architecture from structures determined at low (>3A) resolution, using a deep convolutional neural network trained on high-resolution (≤3A) structures represented as 2D matrices. Thus, we provide proof of concept for high-speed, low-cost protein structure classification at low resolution, and a basis for extension to prediction of function. We investigate the impact of the input representation on classification performance, showing that side-chain information may not be necessary for fine-grained structure predictions. Finally, we confirm that high resolution, low-resolution and NMR-determined structures inhabit a common feature space, and thus provide a theoretical foundation for boosting with single-image super-resolution.

Keywords: transfer learning, protein distance maps, protein structure classification, neural networks

Procedia PDF Downloads 115

Authors: Vijaya Yuvaram Singh V M, Kameshwar Rao J V

Abstract:

The challenge with development of neural network based methods for medical is the availability of data. Anatomical landmark detection in the medical domain is a process to find points on the x-ray scan report of the patient. Most of the time this task is done manually by trained professionals as it requires precision and domain knowledge. Traditionally object detection based methods are used for landmark detection. Here, we utilize reinforcement learning and query based method to train a single agent capable of detecting multiple landmarks. A deep Q network agent is trained to detect single and multiple landmarks present on hip and shoulder from x-ray scan of a patient. Here a single agent is trained to find multiple landmark making it superior to having individual agents per landmark. For the initial study, five images of different patients are used as the environment and tested the agents performance on two unseen images.

Keywords: reinforcement learning, medical landmark detection, multi target detection, deep neural network

Procedia PDF Downloads 129

Authors: Mannan Malik Muhammad Naeem, Jeong Myung Yung

Abstract:

Functional near infrared spectroscopy (fNIRS) has a favorable position in non-invasive brain imaging techniques. The concentration change of oxygenated hemoglobin and de-oxygenated hemoglobin during particular cognitive activity is the basis for this neuro-imaging modality. Two wavelengths of near-infrared light can be used with modified Beer-Lambert law to explain the indirect status of neuronal activity inside brain. The temporal resolution of fNIRS is very good for real-time brain computer-interface applications. The portability, low cost and an acceptable temporal resolution of fNIRS put it on a better position in neuro-imaging modalities. In this study, an optimization model for impulse response function has been used to estimate/predict initial dip using fNIRS data. In addition, the activity strength parameter related to motor based cognitive task has been analyzed. We found an initial dip that remains around 200-300 millisecond and better localize neural activity.

Keywords: fNIRS, brain-computer interface, optimization algorithm, adaptive signal processing

Procedia PDF Downloads 208

Authors: Jonathan Gong

Abstract:

Purpose AI-driven solutions are at the forefront of many pathology and medical imaging methods. Using algorithms designed to better the experience of medical professionals within their respective fields, the efficiency and accuracy of diagnosis can improve. In particular, X-rays are a fast and relatively inexpensive test that can diagnose diseases. In recent years, X-rays have not been widely used to detect and diagnose COVID-19. The under use of Xrays is mainly due to the low diagnostic accuracy and confounding with pneumonia, another respiratory disease. However, research in this field has expressed a possibility that artificial neural networks can successfully diagnose COVID-19 with high accuracy. Models and Data The dataset used is the COVID-19 Radiography Database. This dataset includes images and masks of chest X-rays under the labels of COVID-19, normal, and pneumonia. The classification model developed uses an autoencoder and a pre-trained convolutional neural network (DenseNet201) to provide transfer learning to the model. The model then uses a deep neural network to finalize the feature extraction and predict the diagnosis for the input image. This model was trained on 4035 images and validated on 807 separate images from the ones used for training. The images used to train the classification model include an important feature: the pictures are cropped beforehand to eliminate distractions when training the model. The image segmentation model uses an improved U-Net architecture. This model is used to extract the lung mask from the chest X-ray image. The model is trained on 8577 images and validated on a validation split of 20%. These models are calculated using the external dataset for validation. The models’ accuracy, precision, recall, f1-score, IOU, and loss are calculated. Results The classification model achieved an accuracy of 97.65% and a loss of 0.1234 when differentiating COVID19-infected, pneumonia-infected, and normal lung X-rays. The segmentation model achieved an accuracy of 97.31% and an IOU of 0.928. Conclusion The models proposed can detect COVID-19, pneumonia, and normal lungs with high accuracy and derive the lung mask from a chest X-ray with similarly high accuracy. The hope is for these models to elevate the experience of medical professionals and provide insight into the future of the methods used.

Keywords: artificial intelligence, convolutional neural networks, deep learning, image processing, machine learning

Procedia PDF Downloads 112

Authors: Ben Soltane Cheima, Ittansa Yonas Kelbesa

Abstract:

Speaker Identification (SI) is the task of establishing identity of an individual based on his/her voice characteristics. The SI task is typically achieved by two-stage signal processing: training and testing. The training process calculates speaker specific feature parameters from the speech and generates speaker models accordingly. In the testing phase, speech samples from unknown speakers are compared with the models and classified. Even though performance of speaker identification systems has improved due to recent advances in speech processing techniques, there is still need of improvement. In this paper, a Closed-Set Tex-Independent Speaker Identification System (CISI) based on a Multiple Classifier System (MCS) is proposed, using Mel Frequency Cepstrum Coefficient (MFCC) as feature extraction and suitable combination of vector quantization (VQ) and Gaussian Mixture Model (GMM) together with Expectation Maximization algorithm (EM) for speaker modeling. The use of Voice Activity Detector (VAD) with a hybrid approach based on Short Time Energy (STE) and Statistical Modeling of Background Noise in the pre-processing step of the feature extraction yields a better and more robust automatic speaker identification system. Also investigation of Linde-Buzo-Gray (LBG) clustering algorithm for initialization of GMM, for estimating the underlying parameters, in the EM step improved the convergence rate and systems performance. It also uses relative index as confidence measures in case of contradiction in identification process by GMM and VQ as well. Simulation results carried out on voxforge.org speech database using MATLAB highlight the efficacy of the proposed method compared to earlier work.

Keywords: feature extraction, speaker modeling, feature matching, Mel frequency cepstrum coefficient (MFCC), Gaussian mixture model (GMM), vector quantization (VQ), Linde-Buzo-Gray (LBG), expectation maximization (EM), pre-processing, voice activity detection (VAD), short time energy (STE), background noise statistical modeling, closed-set tex-independent speaker identification system (CISI)

Procedia PDF Downloads 288

Authors: Irene Yi

Abstract:

Machine learning, natural language processing, and neural network models of language are becoming more and more prevalent in the fields of technology and linguistics today. Training data for machines are at best, large corpora of human literature and at worst, a reflection of the ugliness in society. Machines have been trained on millions of human books, only to find that in the course of human history, derogatory and sexist adjectives are used significantly more frequently when describing females in history and literature than when describing males. This is extremely problematic, both as training data, and as the outcome of natural language processing. As machines start to handle more responsibilities, it is crucial to ensure that they do not take with them historical sexist and misogynistic notions. This paper gathers data and algorithms from neural network models of language having to deal with syntax, semantics, sociolinguistics, and text classification. Results are significant in showing the existing intentional and unintentional misogynistic notions used to train machines, as well as in developing better technologies that take into account the semantics and syntax of text to be more mindful and reflect gender equality. Further, this paper deals with the idea of non-binary gender pronouns and how machines can process these pronouns correctly, given its semantic and syntactic context. This paper also delves into the implications of gendered grammar and its effect, cross-linguistically, on natural language processing. Languages such as French or Spanish not only have rigid gendered grammar rules, but also historically patriarchal societies. The progression of society comes hand in hand with not only its language, but how machines process those natural languages. These ideas are all extremely vital to the development of natural language models in technology, and they must be taken into account immediately.

Keywords: gendered grammar, misogynistic language, natural language processing, neural networks

Procedia PDF Downloads 100

Authors: Muhammad Ali Sarwar, Muhammad Farooq, Nayab Hassan, Hammad Hassan

Abstract:

Pakistan is highly recognized for its agriculture and is well known for producing substantial amounts of wheat, cotton, and sugarcane. However, some factors contribute to a decline in crop quality and a reduction in overall output. One of the main factors contributing to this decline is the presence of weed and its late detection. This process of detection is manual and demands a detailed inspection to be done by the farmer itself. But by the time detection of weed, the farmer will be able to save its cost and can increase the overall production. The focus of this research is to identify and classify the four main types of weeds (Small-Flowered Cranesbill, Chick Weed, Prickly Acacia, and Black-Grass) that are prevalent in our region’s major crops. In this work, we implemented three different deep learning techniques: YOLO-v5, Inception-v3, and Deep CNN on the same Dataset, and have concluded that deep convolutions neural network performed better with an accuracy of 97.45% for such classification. In relative to the state of the art, our proposed approach yields 2% better results. We devised the architecture in an efficient way such that it can be used in real-time.

Keywords: deep convolution networks, Yolo, machine learning, agriculture

Procedia PDF Downloads 91

Authors: Martins Y. Otache, John J. Musa, Abayomi I. Kuti, Mustapha Mohammed

Abstract:

The performance of an artificial neural network (ANN) is contingent on a host of factors, for instance, the network optimisation scheme. In view of this, the study examined the general implications of the ANN training optimisation algorithm on its forecast performance. To this end, the Bayesian regularisation (Br), Levenberg-Marquardt (LM), and the adaptive learning gradient descent: GDM (with momentum) algorithms were employed under different ANN structural configurations: (1) single-hidden layer, and (2) double-hidden layer feedforward back propagation network. Results obtained revealed generally that the gradient descent with momentum (GDM) optimisation algorithm, with its adaptive learning capability, used a relatively shorter time in both training and validation phases as compared to the Levenberg- Marquardt (LM) and Bayesian Regularisation (Br) algorithms though learning may not be consummated; i.e., in all instances considering also the prediction of extreme flow conditions for 1-day and 5-day ahead, respectively especially using the ANN model. In specific statistical terms on the average, model performance efficiency using the coefficient of efficiency (CE) statistic were Br: 98%, 94%; LM: 98 %, 95 %, and GDM: 96 %, 96% respectively for training and validation phases. However, on the basis of relative error distribution statistics (MAE, MAPE, and MSRE), GDM performed better than the others overall. Based on the findings, it is imperative to state that the adoption of ANN for real-time forecasting should employ training algorithms that do not have computational overhead like the case of LM that requires the computation of the Hessian matrix, protracted time, and sensitivity to initial conditions; to this end, Br and other forms of the gradient descent with momentum should be adopted considering overall time expenditure and quality of the forecast as well as mitigation of network overfitting. On the whole, it is recommended that evaluation should consider implications of (i) data quality and quantity and (ii) transfer functions on the overall network forecast performance.

Keywords: streamflow, neural network, optimisation, algorithm

Procedia PDF Downloads 139

Authors: Phakamani Xaba, Robert Huberts, Bilainu Oboirien

Abstract:

The present study was aimed at developing & comparing two neural-network-based predictive models to predict shelf-life/product age of South African margarine using free fatty acid (FFA), water droplet size (D3.3), water droplet distribution (e-sigma), moisture content, peroxide value (PV), anisidine valve (AnV) and total oxidation (totox) value as input variables to the model. Brick margarine products which had varying ages ranging from fresh i.e. week 0 to week 47 were sourced. The brick margarine products which had been stored at 10 & 25 °C and were characterized. JMP and MATLAB models to predict shelf-life/ margarine age were developed and their performances were compared. The key performance indicators to evaluate the model performances were correlation coefficient (CC), root mean square error (RMSE), and mean absolute percentage error (MAPE) relative to the actual data. The MATLAB-developed model showed a better performance in all three performance indicators. The correlation coefficient of the MATLAB model was 99.86% versus 99.74% for the JMP model, the RMSE was 0.720 compared to 1.005 and the MAPE was 7.4% compared to 8.571%. The MATLAB model was selected to be the most accurate, and then, the number of hidden neurons/ nodes was optimized to develop a single predictive model. The optimized MATLAB with 10 neurons showed a better performance compared to the models with 1 & 5 hidden neurons. The developed models can be used by margarine manufacturers, food research institutions, researchers etc, to predict shelf-life/ margarine product age, optimize addition of antioxidants, extend shelf-life of products and proactively troubleshoot for problems related to changes which have an impact on shelf-life of margarine without conducting expensive trials.

Keywords: margarine shelf-life, predictive modelling, neural networks, oil oxidation

Procedia PDF Downloads 181

Authors: Racha El Kadiri, Mohamed Sultan, Henrique Momm, Zachary Blair, Rachel Schultz, Tamer Al-Bayoumi

Abstract:

The Mediterranean Basin is a very diverse region of nationalities and climate zones, with a strong dependence on agricultural activities. Predicting long term (with a lead of 1 to 12 months) rainfall, and future droughts could contribute in a sustainable management of water resources and economical activities. In this study, an integrated approach was adopted to construct predictive tools with lead times of 0 to 12 months to forecast rainfall amounts over nine subzones of the Mediterranean Basin region. The following steps were conducted: (1) acquire, assess and intercorrelate temporal remote sensing-based rainfall products (e.g. The CPC Merged Analysis of Precipitation [CMAP]) throughout the investigation period (1979 to 2016), (2) acquire and assess monthly values for all of the climatic indices influencing the regional and global climatic patterns (e.g., Northern Atlantic Oscillation [NOI], Southern Oscillation Index [SOI], and Tropical North Atlantic Index [TNA]); (3) delineate homogenous climatic regions and select nine pilot study zones, (4) apply data mining methods (e.g. neural networks, principal component analyses) to extract relationships between the observed rainfall and the controlling factors (i.e. climatic indices with multiple lead-time periods) and (5) use the constructed predictive tools to forecast monthly rainfall and dry and wet periods. Preliminary results indicate that rainfall and dry/wet periods were successfully predicted with lead zones of 0 to 12 months using the adopted methodology, and that the approach is more accurately applicable in the southern Mediterranean region.

Keywords: rainfall, neural networks, climatic indices, Mediterranean

Procedia PDF Downloads 295

Authors: R. J. Kuo, W. C. Cheng, W. C. Lien, T. J. Yang

Abstract:

Taiwan is a hyper endemic area for the Hepatitis B virus (HBV). The estimated total number of HBsAg carriers in the general population who are more than 20 years old is more than 3 million. Therefore, a case record review is conducted from January 2003 to June 2007 for all patients with a diagnosis of acute hepatitis who were admitted to the Emergency Department (ED) of a well-known teaching hospital. The cost for the use of medical resources is defined as the total medical fee. In this study, principal component analysis (PCA) is firstly employed to reduce the number of dimensions. Support vector regression (SVR) and artificial neural network (ANN) are then used to develop the forecasting model. A total of 117 patients meet the inclusion criteria. 61% patients involved in this study are hepatitis B related. The computational result shows that the proposed PCA-SVR model has superior performance than other compared algorithms. In conclusion, the Child-Pugh score and echogram can both be used to predict the cost of medical resources for patients with acute hepatitis in the ED.

Keywords: acute hepatitis, medical resource cost, artificial neural network, support vector regression

Procedia PDF Downloads 409

Authors: Galal Omer, Onisimo Mutanga, Elfatih M. Abdel-Rahman, Elhadi Adam

Abstract:

Threatened tree species (TTS) play a significant role in ecosystem functioning and services, land use dynamics, and other socio-economic aspects. Such aspects include ecological, economic, livelihood, security-based, and well-being benefits. The development of techniques for mapping and monitoring TTS is thus critical for understanding the functioning of ecosystems. The advent of advanced imaging systems and supervised learning algorithms has provided an opportunity to classify TTS over fragmenting landscape. Recently, vegetation maps have been produced using advanced imaging systems such as WorldView-2 (WV-2) and robust classification algorithms such as support vectors machines (SVM) and artificial neural network (ANN). However, delineation of TTS in a fragmenting landscape using high resolution imagery has widely remained elusive due to the complexity of the species structure and their distribution. Therefore, the objective of the current study was to examine the utility of the advanced WV-2 data for mapping TTS in the fragmenting Dukuduku indigenous forest of South Africa using SVM and ANN classification algorithms. The results showed the robustness of the two machine learning algorithms with an overall accuracy (OA) of 77.00% (total disagreement = 23.00%) for SVM and 75.00% (total disagreement = 25.00%) for ANN using all eight bands of WV-2 (8B). This study concludes that SVM and ANN classification algorithms with WV-2 8B have the potential to classify TTS in the Dukuduku indigenous forest. This study offers relatively accurate information that is important for forest managers to make informed decisions regarding management and conservation protocols of TTS.

Keywords: artificial neural network, threatened tree species, indigenous forest, support vector machines

Procedia PDF Downloads 496

Authors: Zhaoxin Luo, Michael Zhu

Abstract:

In natural languages, there are always complex semantic hierarchies. Obtaining the feature representation based on these complex semantic hierarchies becomes the key to the success of the model. Several RNN models have recently been proposed to use latent indicators to obtain the hierarchical structure of documents. However, the model that only uses a single-layer latent indicator cannot achieve the true hierarchical structure of the language, especially a complex language like Chinese. In this paper, we propose a deep layered model that stacks arbitrarily many RNN layers equipped with latent indicators. After using EM and training it hierarchically, our model solves the computational problem of stacking RNN layers and makes it possible to stack arbitrarily many RNN layers. Our deep hierarchical model not only achieves comparable results to large pre-trained models on the Chinese short text classification problem but also achieves state of art results on the Chinese long text classification problem.

Keywords: nature language processing, recurrent neural network, hierarchical structure, document classification, Chinese

Procedia PDF Downloads 48

Authors: Greeshma Sharma, Sushil Chandra, Vijander Singh, Alok Prakash Mittal

Abstract:

Navigational ability requires spatial representation, planning, and memory. It covers three interdependent domains, i.e. cognitive and perceptual factors, neural information processing, and variability in brain microstructure. Many attempts have been made to see the role of spatial representation in the navigational ability, and the individual differences have been identified in the neural substrate. But, there is also a need to address the influence of planning, memory on navigational ability. The present study aims to evaluate relations of aforementioned factors in the navigational ability. Total 30 participants volunteered in the study of a virtual shopping complex and subsequently were classified into good and bad navigators based on their performances. The result showed that planning ability was the most correlated factor for the navigational ability and also the discriminating factor between the good and bad navigators. There was also found the correlations between spatial memory recall and navigational ability. However, non-verbal episodic memory and spatial memory recall were also found to be correlated with the learning variable. This study attempts to identify differences between people with more and less navigational ability on the basis of planning and memory.

Keywords: memory, planning navigational ability, virtual reality

Procedia PDF Downloads 313