Search results for: classifiers accuracy

Authors: Xian Li, Qing-Guo Wang, Jiangshuai Huang, Jidong Liu, Ming Yu, Tan Kok Poh

Abstract:

In power industry, accurate electricity demand forecasting for a certain leading time is important for system operation and control, etc. In this paper, we investigate the modeling and forecasting of Singapore’s electricity demand. Several standard models, such as HWT exponential smoothing model, the ARMA model and the ANNs model have been proposed based on historical demand data. We applied them to Singapore electricity market and proposed three refinements based on simulation to improve the modeling accuracy. Compared with existing models, our refined model can produce better forecasting accuracy. It is demonstrated in the simulation that by adding forecasting error into the forecasting equation, the modeling accuracy could be improved greatly.

Keywords: power industry, electricity demand, modeling, forecasting

Procedia PDF Downloads 641

Authors: Beng Yew Low, Cher Liang Cha, Cheng Yong Teoh

Abstract:

Due to a lack of continual assessment or grade related data, identifying first-year engineering students in a polytechnic education at risk of failing is challenging. Our experience over the years tells us that there is no strong correlation between having good entry grades in Mathematics and the Sciences and excelling in hardcore engineering subjects. Hence, identifying students at risk of failure cannot be on the basis of entry grades in Mathematics and the Sciences alone. These factors compound the difficulty of early identification and intervention. This paper describes the development of a predictive analytics model in the early detection of students at risk of failing and evaluates its effectiveness. Data from continual assessments conducted in term one, supplemented by data of student psychological profiles such as interests and study habits, were used. Three classification techniques, namely Logistic Regression, K Nearest Neighbour, and Random Forest, were used in our predictive model. Based on our findings, Random Forest was determined to be the strongest predictor with an Area Under the Curve (AUC) value of 0.994. Correspondingly, the Accuracy, Precision, Recall, and F-Score were also highest among these three classifiers. Using this Random Forest Classification technique, students at risk of failure could be identified at the end of term one. They could then be assigned to a Learning Support Programme at the beginning of term two. This paper gathers the results of our findings. It also proposes further improvements that can be made to the model.

Keywords: continual assessment, predictive analytics, random forest, student psychological profile

Procedia PDF Downloads 136

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: 6D posture estimation, image recognition, deep learning, AlexNet

Procedia PDF Downloads 157

Authors: Zhaoan Wang

Abstract:

Research in predictive maintenance modeling has improved in the recent years to predict failures and needed maintenance with high accuracy, saving cost and improving manufacturing efficiency. However, classic prediction models provide little valuable insight towards the most important features contributing to the failure. By analyzing and quantifying feature importance in predictive maintenance models, cost saving can be optimized based on business goals. First, multiple classifiers are evaluated with cross-validation to predict the multi-class of failures. Second, predictive performance with features provided by different feature selection algorithms are further analyzed. Third, features selected by different algorithms are ranked and combined based on their predictive power. Finally, linear explainer SHAP (SHapley Additive exPlanations) is applied to interpret classifier behavior and provide further insight towards the specific roles of features in both local predictions and global model behavior. The results of the experiments suggest that certain features play dominant roles in predictive models while others have significantly less impact on the overall performance. Moreover, for multi-class prediction of machine failures, the most important features vary with type of machine failures. The results may lead to improved productivity and cost saving by prioritizing sensor deployment, data collection, and data processing of more important features over less importance features.

Keywords: automated supply chain, intelligent manufacturing, predictive maintenance machine learning, feature engineering, model interpretation

Procedia PDF Downloads 133

Authors: Peter A. L’vov, Roman S. Konovalov, Alexey A. L’vov

Abstract:

The paper presents the advanced digital modification of the conventional current loop circuit for pressure piezoelectric transducers. The optimal DSP algorithms of current loop responses by the maximum likelihood method are applied for diminishing of measurement errors. The loop circuit has some additional advantages such as the possibility to operate with any type of resistance or reactance sensors, and a considerable increase in accuracy and quality of measurements to be compared with AC bridges. The results obtained are dedicated to replace high-accuracy and expensive measuring bridges with current loop circuits.

Keywords: current loop, maximum likelihood method, optimal digital signal processing, precise pressure measurement

Procedia PDF Downloads 529

Authors: Sunita Singh, Rajani Srivastava

Abstract:

For natural resource management and in other applications about earth observation revolutionary remote sensing technology plays a significant role. One of such application in monitoring and classification of crop types at spatial and temporal scale, as it provides latest, most precise and cost-effective information. Present study emphasizes the use of three different vegetation indices of Rapid Eye imagery on crop type classification. It also analyzed the effect of each indices on classification accuracy. Rapid Eye imagery is highly demanded and preferred for agricultural and forestry sectors as it has red-edge and NIR bands. The three indices used in this study were: the Normalized Difference Vegetation Index (NDVI), the Green Normalized Difference Vegetation Index (GNDVI), and the Normalized Difference Red Edge Index (NDRE) and all of these incorporated the Red Edge band. The study area is Varanasi district of Uttar Pradesh, India and Radial Basis Function (RBF) kernel was used here for the Support Vector Machines (SVMs) classification. Classification was performed with these three vegetation indices. The contribution of each indices on image classification accuracy was also tested with single band classification. Highest classification accuracy of 85% was obtained using three vegetation indices. The study concluded that NDRE has the highest contribution on classification accuracy compared to the other vegetation indices and the Rapid Eye imagery can get satisfactory results of classification accuracy without original bands.

Keywords: GNDVI, NDRE, NDVI, rapid eye, vegetation indices

Procedia PDF Downloads 362

Authors: Mohamad R. Moshtagh, Ahmad Bagheri

Abstract:

Accurate fault detection in rotating machinery is of utmost importance to ensure optimal performance and prevent costly downtime in industrial applications. This study presents a robust fault detection system based on vibration data collected from rotating gears under various operating conditions. The considered scenarios include: (1) both gears being healthy, (2) one healthy gear and one faulty gear, and (3) introducing an imbalanced condition to a healthy gear. Vibration data was acquired using a Hentek 1008 device and stored in a CSV file. Python code implemented in the Spider environment was used for data preprocessing and analysis. Winner features were extracted using the Wiener feature selection method. These features were then employed in multiple machine learning algorithms, including Convolutional Neural Networks (CNN), Multilayer Perceptron (MLP), K-Nearest Neighbors (KNN), and Random Forest, to evaluate their performance in detecting and classifying faults in both the training and validation datasets. The comparative analysis of the methods revealed the superior performance of the Wiener-CNN approach. The Wiener-CNN method achieved a remarkable accuracy of 100% for both the two-class (healthy gear and faulty gear) and three-class (healthy gear, faulty gear, and imbalanced) scenarios in the training and validation datasets. In contrast, the other methods exhibited varying levels of accuracy. The Wiener-MLP method attained 100% accuracy for the two-class training dataset and 100% for the validation dataset. For the three-class scenario, the Wiener-MLP method demonstrated 100% accuracy in the training dataset and 95.3% accuracy in the validation dataset. The Wiener-KNN method yielded 96.3% accuracy for the two-class training dataset and 94.5% for the validation dataset. In the three-class scenario, it achieved 85.3% accuracy in the training dataset and 77.2% in the validation dataset. The Wiener-Random Forest method achieved 100% accuracy for the two-class training dataset and 85% for the validation dataset, while in the three-class training dataset, it attained 100% accuracy and 90.8% accuracy for the validation dataset. The exceptional accuracy demonstrated by the Wiener-CNN method underscores its effectiveness in accurately identifying and classifying fault conditions in rotating machinery. The proposed fault detection system utilizes vibration data analysis and advanced machine learning techniques to improve operational reliability and productivity. By adopting the Wiener-CNN method, industrial systems can benefit from enhanced fault detection capabilities, facilitating proactive maintenance and reducing equipment downtime.

Keywords: fault detection, gearbox, machine learning, wiener method

Procedia PDF Downloads 81

Authors: Abobaker Mohammed Qasem Farhan, Shangming Yang, Mohammed Al-Nehari

Abstract:

Pneumonia is a kind of lung disease that creates congestion in the chest. Such pneumonic conditions lead to loss of life of the severity of high congestion. Pneumonic lung disease is caused by viral pneumonia, bacterial pneumonia, or Covidi-19 induced pneumonia. The early prediction and classification of such lung diseases help to reduce the mortality rate. We propose the automatic Computer-Aided Diagnosis (CAD) system in this paper using the deep learning approach. The proposed CAD system takes input from raw computerized tomography (CT) scans of the patient's chest and automatically predicts disease classification. We designed the Hybrid Deep Learning Algorithm (HDLA) to improve accuracy and reduce processing requirements. The raw CT scans have pre-processed first to enhance their quality for further analysis. We then applied a hybrid model that consists of automatic feature extraction and classification. We propose the robust 2D Convolutional Neural Network (CNN) model to extract the automatic features from the pre-processed CT image. This CNN model assures feature learning with extremely effective 1D feature extraction for each input CT image. The outcome of the 2D CNN model is then normalized using the Min-Max technique. The second step of the proposed hybrid model is related to training and classification using different classifiers. The simulation outcomes using the publically available dataset prove the robustness and efficiency of the proposed model compared to state-of-art algorithms.

Keywords: CT scan, Covid-19, deep learning, image processing, lung disease classification

Procedia PDF Downloads 157

Authors: Rikson Gultom

Abstract:

Hate Speech and Abusive language on social media is difficult to detect, usually, it is detected after it becomes viral in cyberspace, of course, it is too late for prevention. An early detection system that has a fairly good accuracy is needed so that it can reduce conflicts that occur in society caused by postings on social media that attack individuals, groups, and governments in Indonesia. The purpose of this study is to find an early detection model on Twitter social media using machine learning that has high accuracy from several machine learning methods studied. In this study, the support vector machine (SVM), Naïve Bayes (NB), and Random Forest Decision Tree (RFDT) methods were compared with the Support Vector machine with genetic algorithm (SVM-GA), Nave Bayes with genetic algorithm (NB-GA), and Random Forest Decision Tree with Genetic Algorithm (RFDT-GA). The study produced a comparison table for the accuracy of the hate speech and abusive language detection model, and presented it in the form of a graph of the accuracy of the six algorithms developed based on the Indonesian-language Twitter dataset, and concluded the best model with the highest accuracy.

Keywords: abusive language, hate speech, machine learning, optimization, social media

Procedia PDF Downloads 129

Authors: Ashlesha Khanapure, Harsh Kashyap, Abhinav Anand, Sanjana Habib, Anupama Bidargaddi

Abstract:

Technologies that are developing quickly are being developed daily in a variety of disciplines, particularly the medical field. For the purpose of detecting bone fractures in X-ray pictures of different body segments, our work compares the ResNet-50 and MobileNetV3 architectures. It evaluates accuracy and computing efficiency with X-rays of the elbow, hand, and shoulder from the MURA dataset. Through training and validation, the models are evaluated on normal and fractured images. While ResNet-50 showcases superior accuracy in fracture identification, MobileNetV3 showcases superior speed and resource optimization. Despite ResNet-50’s accuracy, MobileNetV3’s swifter inference makes it a viable choice for real-time clinical applications, emphasizing the importance of balancing computational efficiency and accuracy in medical imaging. We created a graphical user interface (GUI) for MobileNet V3 model bone fracture detection. This research underscores MobileNetV3’s potential to streamline bone fracture diagnoses, potentially revolutionizing orthopedic medical procedures and enhancing patient care.

Keywords: CNN, MobileNet V3, ResNet-50, healthcare, MURA, X-ray, fracture detection

Procedia PDF Downloads 68

Authors: Ibrahim Elnour Abdelrahman Eltayeb

Abstract:

Computational modelling is a method used to investigate joint forces during a movement. It can get high accuracy in the joint forces via subject-specific models. However, the construction of subject-specific models remains time-consuming and expensive. The purpose of this paper was to identify what alterations we can make to generic computational models to get a better estimation of the joint forces. It appraised the impact of these alterations on the accuracy of the estimated joint forces. It found different strategies of alterations: joint model, muscle model, and an optimisation problem. All these alterations affected joint contact force accuracy, so showing the potential for improving the model predictions without involving costly and time-consuming medical images.

Keywords: joint force, joint model, optimisation problem, validation

Procedia PDF Downloads 170

Authors: Marrone Silverio Melo Dantas Pedro Henrique Dreyer, Gabriel Fonseca Reis de Souza, Daniel Bezerra, Ricardo Souza, Silvia Lins, Judith Kelner, Djamel Fawzi Hadj Sadok

Abstract:

The creation of a dataset is time-consuming and often discourages researchers from pursuing their goals. To overcome this problem, we present and discuss two solutions adopted for the automation of this process. Both optimize valuable user time and resources and support video object segmentation with object tracking and 3D projection. In our scenario, we acquire images from a moving robotic arm and, for each approach, generate distinct annotated datasets. We evaluated the precision of the annotations by comparing these with a manually annotated dataset, as well as the efficiency in the context of detection and classification problems. For detection support, we used YOLO and obtained for the projection dataset an F1-Score, accuracy, and mAP values of 0.846, 0.924, and 0.875, respectively. Concerning the tracking dataset, we achieved an F1-Score of 0.861, an accuracy of 0.932, whereas mAP reached 0.894. In order to evaluate the quality of the annotated images used for classification problems, we employed deep learning architectures. We adopted metrics accuracy and F1-Score, for VGG, DenseNet, MobileNet, Inception, and ResNet. The VGG architecture outperformed the others for both projection and tracking datasets. It reached an accuracy and F1-score of 0.997 and 0.993, respectively. Similarly, for the tracking dataset, it achieved an accuracy of 0.991 and an F1-Score of 0.981.

Keywords: RJ45, automatic annotation, object tracking, 3D projection

Procedia PDF Downloads 170

Authors: Robert Long

Abstract:

The results of a one-year study on the evolution of complexity, accuracy, and fluency (CAF) in the compositions of Japanese L2 university students throughout a semester are presented in this study. One goal was to determine if any improvement in writing abilities over this academic term had occurred, while another was to examine methods of editing. Participants had 30 minutes to write each essay with an additional 10 minutes allotted for editing. As for editing, participants were divided into two groups, one of which utilized an online grammar checker, while the other half self-edited their initial manuscripts. From the three different institutions, there was a total of 159 students. Research questions focused on determining if the CAF had evolved over the previous year, identifying potential variations in editing techniques, and describing the connections between the CAF dimensions. According to the findings, there was some improvement in accuracy (fewer errors) in all three of the measures), whereas there was a marked decline in complexity and fluency. As for the second research aim relating to the interaction among the three dimensions (CAF) and of possible increases in fluency being offset by decreases in grammatical accuracy, results showed (there is a logical high correlation with clauses and word counts, and mean length of T-unit (MLT) and (coordinate phrase of T-unit (CP/T) as well as MLT and clause per T-unit (C/T); furthermore, word counts and error/100 ratio correlated highly with error-free clause totals (EFCT). Issues of syntactical complexity had a negative correlation with EFCT, indicating that more syntactical complexity relates to decreased accuracy. Concerning a difference in error correction between those who self-edited and those who used an online grammar correction tool, results indicated that the variable of errors-free clause ratios (EFCR) had the greatest difference regarding accuracy, with fewer errors noted with writers using an online grammar checker. As for possible differences between the first and second (edited) drafts regarding CAF, results indicated there were positive changes in accuracy, the most significant change seen in complexity (CP/T and MLT), while there were relatively insignificant changes in fluency. Results also indicated significant differences among the three institutions, with Fujian University of Technology having the most fluency and accuracy. These findings suggest that to raise students' awareness of their overall writing development, teachers should support them in developing more complex syntactic structures, improving their fluency, and making more effective use of online grammar checkers.

Keywords: complexity, accuracy, fluency, writing

Procedia PDF Downloads 42

Authors: Hamidreza Hoshyarmanesh, Benjamin Durante, Alex Irwin, Sanju Lama, Kourosh Zareinia, Garnette R. Sutherland

Abstract:

This study reports on a proposed method to evaluate the accuracy and precision of Excalibur I, a neurosurgery-specific haptic hand controller, designed and developed at Project neuroArm. Having an efficient and successful robot-assisted telesurgery is considerably contingent on how accurate and precise a haptic hand controller (master/local robot) would be able to interpret the kinematic indices of motion, i.e., position and orientation, from the surgeon’s upper limp to the slave/remote robot. A proposed test rig is designed and manufactured according to standard ASTM F2554-10 to determine the accuracy and precision range of Excalibur I at four different locations within its workspace: central workspace, extreme forward, far left and far right. The test rig is metrologically characterized by a coordinate measuring machine (accuracy and repeatability < ± 5 µm). Only the serial linkage of the haptic device is examined due to the use of the Structural Length Index (SLI). The results indicate that accuracy decreases by moving from the workspace central area towards the borders of the workspace. In a comparative study, Excalibur I performs on par with the PHANToM PremiumTM 3.0 and more accurate/precise than the PHANToM PremiumTM 1.5. The error in Cartesian coordinate system shows a dominant component in one direction (δx, δy or δz) for the movements on horizontal, vertical and inclined surfaces. The average error magnitude of three attempts is recorded, considering all three error components. This research is the first promising step to quantify the kinematic performance of Excalibur I.

Keywords: accuracy, advanced metrology, hand controller, precision, robot-assisted surgery, tele-operation, workspace

Procedia PDF Downloads 336

Authors: Watchareeporn Chaimongkol

Abstract:

In this paper, we compare the statistical measures accuracy of composite forecasting model to estimate automobile customer demand in Thailand. A modified simple exponential smoothing and autoregressive integrate moving average (ARIMA) forecasting model is built to estimate customer demand of passenger cars, instead of using information of historical sales data. Our model takes into account special characteristic of the Thai automobile market such as sales promotion, advertising and publicity, petrol price, and interest rate for loan. We evaluate our forecasting model by comparing forecasts with actual data using six accuracy measurements, mean absolute percentage error (MAPE), geometric mean absolute error (GMAE), symmetric mean absolute percentage error (sMAPE), mean absolute scaled error (MASE), median relative absolute error (MdRAE), and geometric mean relative absolute error (GMRAE).

Keywords: composite forecasting, simple exponential smoothing model, autoregressive integrate moving average model selection, accuracy measurements

Procedia PDF Downloads 362

Authors: Wing Kam Fung, Kexin Yu

Abstract:

Child trafficking has been a serious problem in modern China. The Chinese government has established a national anti-trafficking DNA database to help reunite missing children with their families. The database collects DNA information from missing children's parents, trafficked and homeless children, then conducts paternity tests to find matched pairs. This paper considers the matching accuracy in such cases by looking into the exclusion probability in paternity testing. First, the situation of child trafficking in China is introduced. Next, derivations of the exclusion probability for both one-parent and two-parents cases are given, followed by extension to allow for 1 or 2 mutations. The accuracy of paternity testing of child trafficking cases is then assessed using the exclusion probabilities and available data. Finally, the number of loci that should be used to ensure a correct match is investigated.

Keywords: child trafficking, DNA database, exclusion probability, paternity testing

Procedia PDF Downloads 459

Authors: Afaneen Anwer, Samara M. Kamil

Abstract:

Load Forecast required prediction accuracy based on optimal operation and maintenance. A good accuracy is the basis of economic dispatch, unit commitment, and system reliability. A good load forecasting system fulfilled fast speed, automatic bad data detection, and ability to access the system automatically to get the needed data. In this paper, the formulation of the load forecasting is discussed and the solution is obtained by using artificial neural network method. A MATLAB environment has been used to solve the load forecasting schedule of Iraqi super grid network considering the daily load for three years. The obtained results showed a good accuracy in predicting the forecasted load.

Keywords: load forecasting, neural network, back-propagation algorithm, Iraqi power system

Procedia PDF Downloads 583

Authors: Auwal Abubakar, Ahmed Ahidjo, Shazril Imran Shaukat, Noor Khairiah A. Karim, Gokula Kumar Appalanaido, Hafiz Mohd Zin

Abstract:

Background: The use of optical surface imaging systems (OSISs) is increasingly becoming popular in radiotherapy practice, especially during breast cancer treatment. This study reviews the accuracy of the available commercial OSISs for breast radiotherapy. Method: A literature search was conducted and identified the available commercial OSISs from different manufacturers that are integrated into radiotherapy practice for setup verification during breast radiotherapy. Studies that evaluated the accuracy of the OSISs during breast radiotherapy using cone beam computed tomography (CBCT) as a reference were retrieved and analyzed. The physics and working principles of the systems from each manufacturer were discussed together with their respective strength and limitations. Results: A total of five (5) different commercially available OSISs from four (4) manufacturers were identified, each with a different working principle. Six (6) studies were found to evaluate the accuracy of the systems during breast radiotherapy in conjunction with CBCT as a goal standard. The studies revealed that the accuracy of the system in terms of mean difference ranges from 0.1 to 2.1 mm. The correlation between CBCT and OSIS ranges between 0.4 and 0.9. The limit of agreements obtained using bland Altman analysis in the studies was also within an acceptable range. Conclusion: The OSISs have an acceptable level of accuracy and could be used safely during breast radiotherapy. The systems are non-invasive, ionizing radiation-free, and provide real-time imaging of the target surface at no extra concomitant imaging dose. However, the system should only be used to complement rather than replace x-ray-based image guidance techniques such as CBCT.

Keywords: optical surface imaging system, Cone beam computed tomography (CBCT), surface guided radiotherapy, Breast radiotherapy

Procedia PDF Downloads 66

Authors: Amin Amini

Abstract:

A Kinect sensor has been utilized as a cheap and accurate alternative to 3D laser scanners and electronic distance measurement (EDM) systems. This research presents an inexpensive and easy-to-setup system that utilizes the Microsoft Kinect v2 sensor as a surveying and measurement tool and investigates the possibility of using such a device as a replacement for conventional theodolite systems. The system was tested in an indoor environment where its accuracy in distance and angle measurements was tested using virtual markers in a 3D Cartesian environment. The system has shown an average accuracy of 97.94 % in measuring distances and 99.11 % and 98.84 % accuracy for area and perimeter, respectively, within the Kinect’s surveying range of 1.5 to 6 meters. The research also tested the system competency for relative angle determination between two objects.

Keywords: kinect v2, 3D measurement, depth map, ToF

Procedia PDF Downloads 69

Authors: Faiza Babur Khan, Sohail Asghar

Abstract:

With an emergence of distributed environment, cloud computing is proving to be the most stimulating computing paradigm shift in computer technology, resulting in spectacular expansion in IT industry. Many companies have augmented their technical infrastructure by adopting cloud resource sharing architecture. Cloud computing has opened doors to unlimited opportunities from application to platform availability, expandable storage and provision of computing environment. However, from a security viewpoint, an added risk level is introduced from clouds, weakening the protection mechanisms, and hardening the availability of privacy, data security and on demand service. Issues of trust, confidentiality, and integrity are elevated due to multitenant resource sharing architecture of cloud. Trust or reliability of cloud refers to its capability of providing the needed services precisely and unfailingly. Confidentiality is the ability of the architecture to ensure authorization of the relevant party to access its private data. It also guarantees integrity to protect the data from being fabricated by an unauthorized user. So in order to assure provision of secured cloud, a roadmap or model is obligatory to analyze a security problem, design mitigation strategies, and evaluate solutions. The aim of the paper is twofold; first to enlighten the factors which make cloud security critical along with alleviation strategies and secondly to propose an intrusion detection model that identifies the attackers in a preventive way using machine learning Random Forest classifier with an accuracy of 99.8%. This model uses less number of features. A comparison with other classifiers is also presented.

Keywords: cloud security, threats, machine learning, random forest, classification

Procedia PDF Downloads 320

Authors: Kemka Ihemelandu, Chukwuemeka Ihemelandu

Abstract:

Melanoma remains a public health crisis, with incidence rates increasing rapidly in the past decades. Improving diagnostic accuracy to decrease misdiagnosis using Artificial intelligence (AI) continues to be documented. Unfortunately, unintended racially biased outcomes, a product of lack of diversity in the dataset used, with a noted class imbalance favoring lighter vs. darker skin tone, have increasingly been recognized as a problem.Resulting in noted limitations of the accuracy of the Convolutional neural network (CNN)models. CNN models are prone to biased output due to biases in the dataset used to train them. Our aim in this study was the optimization of convolutional neural network algorithms to mitigate bias in the automated diagnosis of melanoma. We hypothesized that our proposed training algorithms based on a data augmentation method to optimize the diagnostic accuracy of a CNN classifier by generating new training samples from the original ones will reduce bias in the automated diagnosis of melanoma. We applied geometric transformation, including; rotations, translations, scale change, flipping, and shearing. Resulting in a CNN model that provided a modifiedinput data making for a model that could learn subtle racial features. Optimal selection of the momentum and batch hyperparameter increased our model accuracy. We show that our augmented model reduces bias while maintaining accuracy in the automated diagnosis of melanoma.

Keywords: bias, augmentation, melanoma, convolutional neural network

Procedia PDF Downloads 213

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

Abstract:

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

Keywords: distillation, machine learning, neural networks, quantization

Procedia PDF Downloads 328

Authors: Shaukat Ali Shahee, Usha Ananthakumar

Abstract:

A data set exhibits class imbalance problem when one class has very few examples compared to the other class, and this is also referred to as between class imbalance. The traditional classifiers fail to classify the minority class examples correctly due to its bias towards the majority class. Apart from between-class imbalance, imbalance within classes where classes are composed of a different number of sub-clusters with these sub-clusters containing different number of examples also deteriorates the performance of the classifier. Previously, many methods have been proposed for handling imbalanced dataset problem. These methods can be classified into four categories: data preprocessing, algorithmic based, cost-based methods and ensemble of classifier. Data preprocessing techniques have shown great potential as they attempt to improve data distribution rather than the classifier. Data preprocessing technique handles class imbalance either by increasing the minority class examples or by decreasing the majority class examples. Decreasing the majority class examples lead to loss of information and also when minority class has an absolute rarity, removing the majority class examples is generally not recommended. Existing methods available for handling class imbalance do not address both between-class imbalance and within-class imbalance simultaneously. In this paper, we propose a method that handles between class imbalance and within class imbalance simultaneously for binary classification problem. Removing between class imbalance and within class imbalance simultaneously eliminates the biases of the classifier towards bigger sub-clusters by minimizing the error domination of bigger sub-clusters in total error. The proposed method uses model-based clustering to find the presence of sub-clusters or sub-concepts in the dataset. The number of examples oversampled among the sub-clusters is determined based on the complexity of sub-clusters. The method also takes into consideration the scatter of the data in the feature space and also adaptively copes up with unseen test data using Lowner-John ellipsoid for increasing the accuracy of the classifier. In this study, neural network is being used as this is one such classifier where the total error is minimized and removing the between-class imbalance and within class imbalance simultaneously help the classifier in giving equal weight to all the sub-clusters irrespective of the classes. The proposed method is validated on 9 publicly available data sets and compared with three existing oversampling techniques that rely on the spatial location of minority class examples in the euclidean feature space. The experimental results show the proposed method to be statistically significantly superior to other methods in terms of various accuracy measures. Thus the proposed method can serve as a good alternative to handle various problem domains like credit scoring, customer churn prediction, financial distress, etc., that typically involve imbalanced data sets.

Keywords: classification, imbalanced dataset, Lowner-John ellipsoid, model based clustering, oversampling

Procedia PDF Downloads 418

Authors: Amr Abdel Azim Ali, G. A. Elsheikh, Moutaz Hegazy

Abstract:

This paper presents verification of a modeling and simulation for a Spacecraft (SC) attitude and orbit control system. Detailed formulation of coupled SC orbital and attitude equations of motion is performed in order to achieve accepted accuracy to meet the requirements of multitargets tracking and orbit correction complex modes. Correction of the target parameter based on the estimated state vector during shooting time to enhance pointing accuracy is considered. Time-optimal nonlinear feedback control technique was used in order to take full advantage of the maximum torques that the controller can deliver. This simulation provides options for visualizing SC trajectory and attitude in a 3D environment by including an interface with V-Realm Builder and VR Sink in Simulink/MATLAB. Verification data confirms the simulation results, ensuring that the model and the proposed control law can be used successfully for large and fast tracking and is robust enough to keep the pointing accuracy within the desired limits with considerable uncertainty in inertia and control torque.

Keywords: attitude and orbit control, time-optimal nonlinear feedback control, modeling and simulation, pointing accuracy, maximum torques

Procedia PDF Downloads 332

Authors: Sayna Bahraini, Wendy Gifford, Ian Graham, Liquaa Wazni, Suzettee Bremault-Phillips, Rebekah Hackbusch, Catrine Demers, Mary Egan

Abstract:

Objective: Guidelines for palliative and spiritual care emphasize the importance of screening patients for spiritual suffering. The aim of this review was to synthesize the research evidence on the accuracy of measures used to screen adults for spiritual suffering. Methods: A systematic review has been conducted. We searched five scientific databases to identify relevant articles. Two independent reviewers screened extracted data and assessed study methodological quality. Results: We identified five articles that yielded information on 24 spiritual screening measures. Among all identified measures, the 2-item Meaning/Joy & Self-Described Struggle has the highest sensitivity (82-87%), and the revised Rush protocol has the highest specificity (81-90%). The methodological quality of all included studies was low. Significance of Results: While most of the identified spiritual screening measures are brief (comprise 1 to 12 number of items), few have sufficient accuracy to effectively screen patients for spiritual suffering. We advise clinicians to use their critical appraisal skills and clinical judgment when selecting and using any of the identified measures to screen for spiritual suffering.

Keywords: screening, suffering, spirituality, diagnostic test accuracy, systematic review

Procedia PDF Downloads 143

Authors: Ahmed A. D. Sarhan, Mum Wai Yip, M. Sayuti, Lim Siew Fen

Abstract:

Ti6Al4V alloy is highly used in the automotive and aerospace industry due to its good machining characteristics. Micro EDM drilling is commonly used to drill micro hole on extremely hard material with very high depth to diameter ratio. In this study, the parameters of micro-electrical discharge machining (EDM) in drilling of Ti6Al4V alloy is optimized for higher machining accuracy with less hole-dilation and hole taper ratio. The micro-EDM machining parameters includes, peak current and pulse on time. Fuzzy analysis was developed to evaluate the machining accuracy. The analysis shows that hole-dilation and hole-taper ratio are increased with the increasing of peak current and pulse on time. However, the surface quality deteriorates as the peak current and pulse on time increase. The combination that gives the optimum result for hole dilation is medium peak current and short pulse on time. Meanwhile, the optimum result for hole taper ratio is low peak current and short pulse on time.

Keywords: Micro EDM, Ti-6Al-4V alloy, fuzzy logic based analysis, optimization, machining accuracy

Procedia PDF Downloads 496

Authors: S. F. Wong, W. I. Ho, B. Lin, Q. Huang

Abstract:

ERP has been used in many enterprises for management, the accuracy of the production forecasting module is vital to the decision making of the enterprise, and the profit is affected directly. Therefore, enhancing the accuracy of the production forecasting module can also increase the efficiency and profitability. To deal with a lot of data, a suitable, reliable and accurate statistics model is necessary. LSSVM and Grey System are two main models to be studied in this paper, and a case study is used to demonstrate how the combination model is effective to the result of forecasting.

Keywords: ERP, grey system, LSSVM, production forecasting

Procedia PDF Downloads 464

Authors: Goran Begović

Abstract:

In recent years, many researchers are focusing on non-intrusive measuring methods when it comes to human biosignals. These methods provide solutions for everyday use, whether it’s health monitoring or finessing the workout routine. One of the biggest issues with these solutions is that the sensors’ accuracy is highly variable due to many factors, such as ambiental light, skin color diversity, etc. That is why we wanted to explore different outcomes under those kinds of circumstances in order to find the most optimal algorithm(s) for extracting heart rate (HR) information. The optimization of such algorithms can benefit the wider, cheaper, and safer application of home health monitoring, without having to visit medical professionals as often when it comes to observing heart irregularities. In this study, we explored the accuracy of infrared (IR), red, and green LED sensorics in a controlled environment and compared the results with a medically accurate ECG monitoring device.

Keywords: data science, ECG, heart rate, holter monitor, LED sensors

Procedia PDF Downloads 128

Authors: Monika Rawat, Rahul Kumar

Abstract:

Input output Buffer Information Specification (IBIS) models are used for describing the analog behavior of the Input Output (I/O) buffers of a digital device. They are widely used to perform signal integrity analysis. Advantages of using IBIS models include simple structure, IP protection and fast simulation time with reasonable accuracy. As design complexity of driver and receiver increases, capturing exact behavior from transistor level model into IBIS model becomes an essential task to achieve better accuracy. In this paper, an improvement in existing methodology of generating IBIS model for complex I/O interfaces such as Inter-Integrated Circuit (I2C) and Low Voltage Differential Signaling (LVDS) is proposed. Furthermore, the accuracy and computational performance of standard method and proposed approach with respect to SPICE are presented. The investigations will be useful to further improve the accuracy of IBIS models and to enhance their wider acceptance.

Keywords: IBIS, signal integrity, open-drain buffer, low voltage differential signaling, behavior modelling, transient simulation

Procedia PDF Downloads 197

Authors: Konstantinos Perifanos, Eirini Florou, Dionysis Goutsos

Abstract:

This paper presents and benchmarks a number of end-to-end Deep Learning based models for metaphor detection in Greek. We combine Convolutional Neural Networks and Recurrent Neural Networks with representation learning to bear on the metaphor detection problem for the Greek language. The models presented achieve exceptional accuracy scores, significantly improving the previous state-of-the-art results, which had already achieved accuracy 0.82. Furthermore, no special preprocessing, feature engineering or linguistic knowledge is used in this work. The methods presented achieve accuracy of 0.92 and F-score 0.92 with Convolutional Neural Networks (CNNs) and bidirectional Long Short Term Memory networks (LSTMs). Comparable results of 0.91 accuracy and 0.91 F-score are also achieved with bidirectional Gated Recurrent Units (GRUs) and Convolutional Recurrent Neural Nets (CRNNs). The models are trained and evaluated only on the basis of training tuples, the related sentences and their labels. The outcome is a state-of-the-art collection of metaphor detection models, trained on limited labelled resources, which can be extended to other languages and similar tasks.

Keywords: metaphor detection, deep learning, representation learning, embeddings

Procedia PDF Downloads 154