Most recognition system of human facial emotions are assessed solely on accuracy, even if other performance criteria are also thought to be important in the evaluation process such as sensitivity, precision, F-measure, and G-mean. Moreover, the most common problem that must be resolved in face emotion recognition systems is the feature extraction methods, which is comparable to traditional manual feature extraction methods. This traditional method is not able to extract features efficiently. In other words, there are redundant amount of features which are considered not significant, which affect the classification performance. In this work, a new system to recognize human facial emotions from images is proposed. The HOG (Histograms of Or

      The transition of customers from one telecom operator to another has a direct impact on the company's growth and revenue. Traditional classification algorithms fail to predict churn effectively. This research introduces a deep learning model for predicting customers planning to leave to another operator. The model works on a high-dimensional large-scale data set. The performance of the model was measured against other classification algorithms, such as Gaussian NB, Random Forrest, and Decision Tree in predicting churn. The evaluation was performed based on accuracy, precision, recall, F-measure, Area Under Curve (AUC), and Receiver Operating Characteristic (ROC) Curve. The proposed deep learning model performs better than othe

Home New Trends in Information and Communications Technology Applications Conference paper Audio Compression Using Transform Coding with LZW and Double Shift Coding Zainab J. Ahmed & Loay E. George Conference paper First Online: 11 January 2022 126 Accesses Part of the Communications in Computer and Information Science book series (CCIS,volume 1511) Abstract The need for audio compression is still a vital issue, because of its significance in reducing the data size of one of the most common digital media that is exchanged between distant parties. In this paper, the efficiencies of two audio compression modules were investigated; the first module is based on discrete cosine transform and the second module is based on discrete wavelet tr

A three-stage learning algorithm for deep multilayer perceptron (DMLP) with effective weight initialisation based on sparse auto-encoder is proposed in this paper, which aims to overcome difficulties in training deep neural networks with limited training data in high-dimensional feature space. At the first stage, unsupervised learning is adopted using sparse auto-encoder to obtain the initial weights of the feature extraction layers of the DMLP. At the second stage, error back-propagation is used to train the DMLP by fixing the weights obtained at the first stage for its feature extraction layers. At the third stage, all the weights of the DMLP obtained at the second stage are refined by error back-propagation. Network structures an

In this study, genetic algorithm was used to predict the reaction kinetics of Iraqi heavy naphtha catalytic reforming process located in Al-Doura refinery in Baghdad.  One-dimensional steady state model was derived to describe commercial catalytic reforming unit consisting of four catalytic reforming reactors in series process.

The experimental information (Reformate composition and output temperature) for each four reactors collected at different operating conditions was used to predict the parameters of the proposed kinetic model. The kinetic model involving 24 components, 1 to 11 carbon atoms for paraffins and 6 to 11 carbon atom for naphthenes and aromatics with 71 reactions. The pre-exponential Arrhenius constants and a

Detection of early clinical keratoconus (KCN) is a challenging task, even for expert clinicians. In this study, we propose a deep learning (DL) model to address this challenge. We first used Xception and InceptionResNetV2 DL architectures to extract features from three different corneal maps collected from 1371 eyes examined in an eye clinic in Egypt. We then fused features using Xception and InceptionResNetV2 to detect subclinical forms of KCN more accurately and robustly. We obtained an area under the receiver operating characteristic curves (AUC) of 0.99 and an accuracy range of 97–100% to distinguish normal eyes from eyes with subclinical and established KCN. We further validated the model based on an independent dataset with

Voice Activity Detection (VAD) is considered as an important pre-processing step in speech processing systems such as speech enhancement, speech recognition, gender and age identification. VAD helps in reducing the time required to process speech data and to improve final system accuracy by focusing the work on the voiced part of the speech. An automatic technique for VAD using Fuzzy-Neuro technique (FN-AVAD) is presented in this paper. The aim of this work is to alleviate the problem of choosing the best threshold value in traditional VAD methods and achieves automaticity by combining fuzzy clustering and machine learning techniques. Four features are extracted from each speech segment, which are short term energy, zero-crossing rate, auto

CO2 Laser (10600nm) is the recent method in the management of challenging skin scar resulting from trauma, burn and surgical wound. The aim of this study was to evaluate the efficacy & safety of fractional CO2 laser (10600nm) in treatment of skin scar. Materials and Methods:Twenty patients with different types of scars treated with fractional CO2 (10600nm) laser, (10 patients) were given additional intralesional Triamcinolone. Results: All of the twenty patients included in this study showed some sort of improvements in scar texture, height and pliability and all of the ten patients who received intralesional Triamcinolone after laser show complete satisfaction. Conclusion:Fractional CO2 (10600nm) laser can be used as alternative, ef

    The problem statement discussed in this paper is a new technique for the presentation of painterly rendering that uses a K-mean segmentation to divide the input image into a set of regions (depending on the grayscale of the regions). Segmenting the input image helps users use different brush strokes and easily change the strokes' shape, size, or orientation for different regions. Every region is painted using different brush kinds. The properties of the brush strokes are chosen depending on the region's details. The brush stroke properties, such as size, color, shape, location, and orientation, are extracted from the source image using statistical tools. The number of regions is set up manually and depends on the input image. This