Search results for: machine vision
3451 AutoML: Comprehensive Review and Application to Engineering Datasets
Authors: Parsa Mahdavi, M. Amin Hariri-Ardebili
Abstract:
The development of accurate machine learning and deep learning models traditionally demands hands-on expertise and a solid background to fine-tune hyperparameters. With the continuous expansion of datasets in various scientific and engineering domains, researchers increasingly turn to machine learning methods to unveil hidden insights that may elude classic regression techniques. This surge in adoption raises concerns about the adequacy of the resultant meta-models and, consequently, the interpretation of the findings. In response to these challenges, automated machine learning (AutoML) emerges as a promising solution, aiming to construct machine learning models with minimal intervention or guidance from human experts. AutoML encompasses crucial stages such as data preparation, feature engineering, hyperparameter optimization, and neural architecture search. This paper provides a comprehensive overview of the principles underpinning AutoML, surveying several widely-used AutoML platforms. Additionally, the paper offers a glimpse into the application of AutoML on various engineering datasets. By comparing these results with those obtained through classical machine learning methods, the paper quantifies the uncertainties inherent in the application of a single ML model versus the holistic approach provided by AutoML. These examples showcase the efficacy of AutoML in extracting meaningful patterns and insights, emphasizing its potential to revolutionize the way we approach and analyze complex datasets.Keywords: automated machine learning, uncertainty, engineering dataset, regression
Procedia PDF Downloads 553450 Predicting Options Prices Using Machine Learning
Authors: Krishang Surapaneni
Abstract:
The goal of this project is to determine how to predict important aspects of options, including the ask price. We want to compare different machine learning models to learn the best model and the best hyperparameters for that model for this purpose and data set. Option pricing is a relatively new field, and it can be very complicated and intimidating, especially to inexperienced people, so we want to create a machine learning model that can predict important aspects of an option stock, which can aid in future research. We tested multiple different models and experimented with hyperparameter tuning, trying to find some of the best parameters for a machine-learning model. We tested three different models: a Random Forest Regressor, a linear regressor, and an MLP (multi-layer perceptron) regressor. The most important feature in this experiment is the ask price; this is what we were trying to predict. In the field of stock pricing prediction, there is a large potential for error, so we are unable to determine the accuracy of the models based on if they predict the pricing perfectly. Due to this factor, we determined the accuracy of the model by finding the average percentage difference between the predicted and actual values. We tested the accuracy of the machine learning models by comparing the actual results in the testing data and the predictions made by the models. The linear regression model performed worst, with an average percentage error of 17.46%. The MLP regressor had an average percentage error of 11.45%, and the random forest regressor had an average percentage error of 7.42%Keywords: finance, linear regression model, machine learning model, neural network, stock price
Procedia PDF Downloads 723449 Modern Proteomics and the Application of Machine Learning Analyses in Proteomic Studies of Chronic Kidney Disease of Unknown Etiology
Authors: Dulanjali Ranasinghe, Isuru Supasan, Kaushalya Premachandra, Ranjan Dissanayake, Ajith Rajapaksha, Eustace Fernando
Abstract:
Proteomics studies of organisms are considered to be significantly information-rich compared to their genomic counterparts because proteomes of organisms represent the expressed state of all proteins of an organism at a given time. In modern top-down and bottom-up proteomics workflows, the primary analysis methods employed are gel–based methods such as two-dimensional (2D) electrophoresis and mass spectrometry based methods. Machine learning (ML) and artificial intelligence (AI) have been used increasingly in modern biological data analyses. In particular, the fields of genomics, DNA sequencing, and bioinformatics have seen an incremental trend in the usage of ML and AI techniques in recent years. The use of aforesaid techniques in the field of proteomics studies is only beginning to be materialised now. Although there is a wealth of information available in the scientific literature pertaining to proteomics workflows, no comprehensive review addresses various aspects of the combined use of proteomics and machine learning. The objective of this review is to provide a comprehensive outlook on the application of machine learning into the known proteomics workflows in order to extract more meaningful information that could be useful in a plethora of applications such as medicine, agriculture, and biotechnology.Keywords: proteomics, machine learning, gel-based proteomics, mass spectrometry
Procedia PDF Downloads 1463448 Plant Identification Using Convolution Neural Network and Vision Transformer-Based Models
Authors: Virender Singh, Mathew Rees, Simon Hampton, Sivaram Annadurai
Abstract:
Plant identification is a challenging task that aims to identify the family, genus, and species according to plant morphological features. Automated deep learning-based computer vision algorithms are widely used for identifying plants and can help users narrow down the possibilities. However, numerous morphological similarities between and within species render correct classification difficult. In this paper, we tested custom convolution neural network (CNN) and vision transformer (ViT) based models using the PyTorch framework to classify plants. We used a large dataset of 88,000 provided by the Royal Horticultural Society (RHS) and a smaller dataset of 16,000 images from the PlantClef 2015 dataset for classifying plants at genus and species levels, respectively. Our results show that for classifying plants at the genus level, ViT models perform better compared to CNN-based models ResNet50 and ResNet-RS-420 and other state-of-the-art CNN-based models suggested in previous studies on a similar dataset. ViT model achieved top accuracy of 83.3% for classifying plants at the genus level. For classifying plants at the species level, ViT models perform better compared to CNN-based models ResNet50 and ResNet-RS-420, with a top accuracy of 92.5%. We show that the correct set of augmentation techniques plays an important role in classification success. In conclusion, these results could help end users, professionals and the general public alike in identifying plants quicker and with improved accuracy.Keywords: plant identification, CNN, image processing, vision transformer, classification
Procedia PDF Downloads 913447 Deep Learning based Image Classifiers for Detection of CSSVD in Cacao Plants
Authors: Atuhurra Jesse, N'guessan Yves-Roland Douha, Pabitra Lenka
Abstract:
The detection of diseases within plants has attracted a lot of attention from computer vision enthusiasts. Despite the progress made to detect diseases in many plants, there remains a research gap to train image classifiers to detect the cacao swollen shoot virus disease or CSSVD for short, pertinent to cacao plants. This gap has mainly been due to the unavailability of high quality labeled training data. Moreover, institutions have been hesitant to share their data related to CSSVD. To fill these gaps, image classifiers to detect CSSVD-infected cacao plants are presented in this study. The classifiers are based on VGG16, ResNet50 and Vision Transformer (ViT). The image classifiers are evaluated on a recently released and publicly accessible KaraAgroAI Cocoa dataset. The best performing image classifier, based on ResNet50, achieves 95.39\% precision, 93.75\% recall, 94.34\% F1-score and 94\% accuracy on only 20 epochs. There is a +9.75\% improvement in recall when compared to previous works. These results indicate that the image classifiers learn to identify cacao plants infected with CSSVD.Keywords: CSSVD, image classification, ResNet50, vision transformer, KaraAgroAI cocoa dataset
Procedia PDF Downloads 943446 Applications of AI, Machine Learning, and Deep Learning in Cyber Security
Authors: Hailyie Tekleselase
Abstract:
Deep learning is increasingly used as a building block of security systems. However, neural networks are hard to interpret and typically solid to the practitioner. This paper presents a detail survey of computing methods in cyber security, and analyzes the prospects of enhancing the cyber security capabilities by suggests that of accelerating the intelligence of the security systems. There are many AI-based applications used in industrial scenarios such as Internet of Things (IoT), smart grids, and edge computing. Machine learning technologies require a training process which introduces the protection problems in the training data and algorithms. We present machine learning techniques currently applied to the detection of intrusion, malware, and spam. Our conclusions are based on an extensive review of the literature as well as on experiments performed on real enterprise systems and network traffic. We conclude that problems can be solved successfully only when methods of artificial intelligence are being used besides human experts or operators.Keywords: artificial intelligence, machine learning, deep learning, cyber security, big data
Procedia PDF Downloads 1193445 Machine Learning Model Applied for SCM Processes to Efficiently Determine Its Impacts on the Environment
Authors: Elena Puica
Abstract:
This paper aims to investigate the impact of Supply Chain Management (SCM) on the environment by applying a Machine Learning model while pointing out the efficiency of the technology used. The Machine Learning model was used to derive the efficiency and optimization of technology used in SCM and the environmental impact of SCM processes. The model applied is a predictive classification model and was trained firstly to determine which stage of the SCM has more outputs and secondly to demonstrate the efficiency of using advanced technology in SCM instead of recuring to traditional SCM. The outputs are the emissions generated in the environment, the consumption from different steps in the life cycle, the resulting pollutants/wastes emitted, and all the releases to air, land, and water. This manuscript presents an innovative approach to applying advanced technology in SCM and simultaneously studies the efficiency of technology and the SCM's impact on the environment. Identifying the conceptual relationships between SCM practices and their impact on the environment is a new contribution to the research. The authors can take a forward step in developing recent studies in SCM and its effects on the environment by applying technology.Keywords: machine-learning model in SCM, SCM processes, SCM and the environmental impact, technology in SCM
Procedia PDF Downloads 1103444 A Comparative Study of Malware Detection Techniques Using Machine Learning Methods
Authors: Cristina Vatamanu, Doina Cosovan, Dragos Gavrilut, Henri Luchian
Abstract:
In the past few years, the amount of malicious software increased exponentially and, therefore, machine learning algorithms became instrumental in identifying clean and malware files through semi-automated classification. When working with very large datasets, the major challenge is to reach both a very high malware detection rate and a very low false positive rate. Another challenge is to minimize the time needed for the machine learning algorithm to do so. This paper presents a comparative study between different machine learning techniques such as linear classifiers, ensembles, decision trees or various hybrids thereof. The training dataset consists of approximately 2 million clean files and 200.000 infected files, which is a realistic quantitative mixture. The paper investigates the above mentioned methods with respect to both their performance (detection rate and false positive rate) and their practicability.Keywords: ensembles, false positives, feature selection, one side class algorithm
Procedia PDF Downloads 2863443 Mathematics Vision of the Companies' Growth with Educational Technologies
Authors: Valencia P. L. Rodrigo, Morita A. Adelina, Vargas V. Martin
Abstract:
This proposal consists of an analysis of macro concepts involved within an organization growth using educational technologies, which will relate each concept, in a mathematical way with a vision of harmonic work. Working collaboratively, competitively and cooperatively so that this growth is harmonious and homogenous, coining a new term, Harmonic Work. The Harmonic Work ensures that the organization grows in all business directions, allowing managers to project a much more accurate growth, making clear the contribution of each department, resulting in an algorithm that analyzes each of the variables both endogenous and exogenous, establishing different performance indicators in its process of growth.Keywords: business projection, collaboration, competitiveness, educational technology, harmonious growth
Procedia PDF Downloads 3133442 Rapid Soil Classification Using Computer Vision, Electrical Resistivity and Soil Strength
Authors: Eugene Y. J. Aw, J. W. Koh, S. H. Chew, K. E. Chua, Lionel L. J. Ang, Algernon C. S. Hong, Danette S. E. Tan, Grace H. B. Foo, K. Q. Hong, L. M. Cheng, M. L. Leong
Abstract:
This paper presents a novel rapid soil classification technique that combines computer vision with four-probe soil electrical resistivity method and cone penetration test (CPT), to improve the accuracy and productivity of on-site classification of excavated soil. In Singapore, excavated soils from local construction projects are transported to Staging Grounds (SGs) to be reused as fill material for land reclamation. Excavated soils are mainly categorized into two groups (“Good Earth” and “Soft Clay”) based on particle size distribution (PSD) and water content (w) from soil investigation reports and on-site visual survey, such that proper treatment and usage can be exercised. However, this process is time-consuming and labour-intensive. Thus, a rapid classification method is needed at the SGs. Computer vision, four-probe soil electrical resistivity and CPT were combined into an innovative non-destructive and instantaneous classification method for this purpose. The computer vision technique comprises soil image acquisition using industrial grade camera; image processing and analysis via calculation of Grey Level Co-occurrence Matrix (GLCM) textural parameters; and decision-making using an Artificial Neural Network (ANN). Complementing the computer vision technique, the apparent electrical resistivity of soil (ρ) is measured using a set of four probes arranged in Wenner’s array. It was found from the previous study that the ANN model coupled with ρ can classify soils into “Good Earth” and “Soft Clay” in less than a minute, with an accuracy of 85% based on selected representative soil images. To further improve the technique, the soil strength is measured using a modified mini cone penetrometer, and w is measured using a set of time-domain reflectometry (TDR) probes. Laboratory proof-of-concept was conducted through a series of seven tests with three types of soils – “Good Earth”, “Soft Clay” and an even mix of the two. Validation was performed against the PSD and w of each soil type obtained from conventional laboratory tests. The results show that ρ, w and CPT measurements can be collectively analyzed to classify soils into “Good Earth” or “Soft Clay”. It is also found that these parameters can be integrated with the computer vision technique on-site to complete the rapid soil classification in less than three minutes.Keywords: Computer vision technique, cone penetration test, electrical resistivity, rapid and non-destructive, soil classification
Procedia PDF Downloads 2053441 Investigating the Viability of Ultra-Low Parameter Count Networks for Real-Time Football Detection
Authors: Tim Farrelly
Abstract:
In recent years, AI-powered object detection systems have opened the doors for innovative new applications and products, especially those operating in the real world or ‘on edge’ – namely, in sport. This paper investigates the viability of an ultra-low parameter convolutional neural network specially designed for the detection of footballs on ‘on the edge’ devices. The main contribution of this paper is the exploration of integrating new design features (depth-wise separable convolutional blocks and squeezed and excitation modules) into an ultra-low parameter network and demonstrating subsequent improvements in performance. The results show that tracking the ball from Full HD images with negligibly high accu-racy is possible in real-time.Keywords: deep learning, object detection, machine vision applications, sport, network design
Procedia PDF Downloads 1403440 Optimization for the Hydraulic Clamping System of an Internal Circulation Two-Platen Injection Molding Machine
Authors: Jian Wang, Lu Yang, Jiong Peng
Abstract:
Internal circulation two-platen clamping system for injection molding machine (IMM) has many potential advantages on energy-saving. In order to estimate its properties, experiments in this paper were carried out. Displacement and pressure of the components were measured. In comparison, the model of hydraulic clamping system was established by using AMESim. The related parameters as well as the energy consumption could be calculated. According to the analysis, the hydraulic system was optimized in order to reduce the energy consumption.Keywords: AMESim, energy-saving, injection molding machine, internal circulation
Procedia PDF Downloads 5433439 Using Artificial Intelligence Technology to Build the User-Oriented Platform for Integrated Archival Service
Authors: Lai Wenfang
Abstract:
Tthis study will describe how to use artificial intelligence (AI) technology to build the user-oriented platform for integrated archival service. The platform will be launched in 2020 by the National Archives Administration (NAA) in Taiwan. With the progression of information communication technology (ICT) the NAA has built many systems to provide archival service. In order to cope with new challenges, such as new ICT, artificial intelligence or blockchain etc. the NAA will try to use the natural language processing (NLP) and machine learning (ML) skill to build a training model and propose suggestions based on the data sent to the platform. NAA expects the platform not only can automatically inform the sending agencies’ staffs which records catalogues are against the transfer or destroy rules, but also can use the model to find the details hidden in the catalogues and suggest NAA’s staff whether the records should be or not to be, to shorten the auditing time. The platform keeps all the users’ browse trails; so that the platform can predict what kinds of archives user could be interested and recommend the search terms by visualization, moreover, inform them the new coming archives. In addition, according to the Archives Act, the NAA’s staff must spend a lot of time to mark or remove the personal data, classified data, etc. before archives provided. To upgrade the archives access service process, the platform will use some text recognition pattern to black out automatically, the staff only need to adjust the error and upload the correct one, when the platform has learned the accuracy will be getting higher. In short, the purpose of the platform is to deduct the government digital transformation and implement the vision of a service-oriented smart government.Keywords: artificial intelligence, natural language processing, machine learning, visualization
Procedia PDF Downloads 1663438 Comparison of Visio-spatial Intelligence Between Amateur Rugby and Netball Players Using a Hand-Eye Coordination Specific Visual Test Battery
Authors: Lourens Millard, Gerrit Jan Breukelman, Nonkululeko Mathe
Abstract:
Aim: The research aims to investigate the differences in visio-spatial skills (VSS) between athletes and non-athletes, as well as variations across sports, presenting conflicting findings. Therefore, the objective of this study was to determine if there exist significant differences in visio-spatial intelligence skills between rugby players and netball players, and whether such disparities are present when comparing both groups to non-athletes. Methods: Participants underwent an optometric assessment, followed by an evaluation of VSS using six established tests: the Hart Near Far Rock, saccadic eye movement, evasion, accumulator, flash memory, and ball wall toss tests. Results: The results revealed that rugby players significantly outperformed netball players in speed of recognition, peripheral awareness, and hand-eye coordination (p=.000). Moreover, both rugby players and netball players performed significantly better than non-athletes in five of the six tests (p=.000), with the exception being the visual memory test (p=.809). Conclusion: This discrepancy in performance suggests that certain VSS are superior in athletes compared to non-athletes, highlighting potential implications for theories of vision, test selection, and the development of sport-specific VSS testing batteries. Furthermore, the use of a hand-eye coordination-specific VSS test battery effectively differentiated between different sports. However, this pattern was not consistent across all VSS tests, indicating that further research should explore the training methods employed by both sports, as these factors may contribute to the observed differences.Keywords: visio-spatial intelligence (VSI), rugby vision, netball vision, visual skills, sport vision.
Procedia PDF Downloads 283437 Support Vector Regression with Weighted Least Absolute Deviations
Authors: Kang-Mo Jung
Abstract:
Least squares support vector machine (LS-SVM) is a penalized regression which considers both fitting and generalization ability of a model. However, the squared loss function is very sensitive to even single outlier. We proposed a weighted absolute deviation loss function for the robustness of the estimates in least absolute deviation support vector machine. The proposed estimates can be obtained by a quadratic programming algorithm. Numerical experiments on simulated datasets show that the proposed algorithm is competitive in view of robustness to outliers.Keywords: least absolute deviation, quadratic programming, robustness, support vector machine, weight
Procedia PDF Downloads 5193436 A Study of Permission-Based Malware Detection Using Machine Learning
Authors: Ratun Rahman, Rafid Islam, Akin Ahmed, Kamrul Hasan, Hasan Mahmud
Abstract:
Malware is becoming more prevalent, and several threat categories have risen dramatically in recent years. This paper provides a bird's-eye view of the world of malware analysis. The efficiency of five different machine learning methods (Naive Bayes, K-Nearest Neighbor, Decision Tree, Random Forest, and TensorFlow Decision Forest) combined with features picked from the retrieval of Android permissions to categorize applications as harmful or benign is investigated in this study. The test set consists of 1,168 samples (among these android applications, 602 are malware and 566 are benign applications), each consisting of 948 features (permissions). Using the permission-based dataset, the machine learning algorithms then produce accuracy rates above 80%, except the Naive Bayes Algorithm with 65% accuracy. Of the considered algorithms TensorFlow Decision Forest performed the best with an accuracy of 90%.Keywords: android malware detection, machine learning, malware, malware analysis
Procedia PDF Downloads 1573435 Application of Machine Learning Techniques in Forest Cover-Type Prediction
Authors: Saba Ebrahimi, Hedieh Ashrafi
Abstract:
Predicting the cover type of forests is a challenge for natural resource managers. In this project, we aim to perform a comprehensive comparative study of two well-known classification methods, support vector machine (SVM) and decision tree (DT). The comparison is first performed among different types of each classifier, and then the best of each classifier will be compared by considering different evaluation metrics. The effect of boosting and bagging for decision trees is also explored. Furthermore, the effect of principal component analysis (PCA) and feature selection is also investigated. During the project, the forest cover-type dataset from the remote sensing and GIS program is used in all computations.Keywords: classification methods, support vector machine, decision tree, forest cover-type dataset
Procedia PDF Downloads 2083434 A Molding Surface Auto-inspection System
Authors: Ssu-Han Chen, Der-Baau Perng
Abstract:
Molding process in IC manufacturing secures chips against the harms done by hot, moisture or other external forces. While a chip was being molded, defects like cracks, dilapidation, or voids may be embedding on the molding surface. The molding surfaces the study poises to treat and the ones on the market, though, differ in the surface where texture similar to defects is everywhere. Manual inspection usually passes over low-contrast cracks or voids; hence an automatic optical inspection system for molding surface is necessary. The proposed system is consisted of a CCD, a coaxial light, a back light as well as a motion control unit. Based on the property of statistical textures of the molding surface, a series of digital image processing and classification procedure is carried out. After training of the parameter associated with above algorithm, result of the experiment suggests that the accuracy rate is up to 93.75%, contributing to the inspection quality of IC molding surface.Keywords: molding surface, machine vision, statistical texture, discrete Fourier transformation
Procedia PDF Downloads 4253433 Machine Learning-Enabled Classification of Climbing Using Small Data
Authors: Nicholas Milburn, Yu Liang, Dalei Wu
Abstract:
Athlete performance scoring within the climbing do-main presents interesting challenges as the sport does not have an objective way to assign skill. Assessing skill levels within any sport is valuable as it can be used to mark progress while training, and it can help an athlete choose appropriate climbs to attempt. Machine learning-based methods are popular for complex problems like this. The dataset available was composed of dynamic force data recorded during climbing; however, this dataset came with challenges such as data scarcity, imbalance, and it was temporally heterogeneous. Investigated solutions to these challenges include data augmentation, temporal normalization, conversion of time series to the spectral domain, and cross validation strategies. The investigated solutions to the classification problem included light weight machine classifiers KNN and SVM as well as the deep learning with CNN. The best performing model had an 80% accuracy. In conclusion, there seems to be enough information within climbing force data to accurately categorize climbers by skill.Keywords: classification, climbing, data imbalance, data scarcity, machine learning, time sequence
Procedia PDF Downloads 1383432 Hate Speech Detection Using Deep Learning and Machine Learning Models
Authors: Nabil Shawkat, Jamil Saquer
Abstract:
Social media has accelerated our ability to engage with others and eliminated many communication barriers. On the other hand, the widespread use of social media resulted in an increase in online hate speech. This has drastic impacts on vulnerable individuals and societies. Therefore, it is critical to detect hate speech to prevent innocent users and vulnerable communities from becoming victims of hate speech. We investigate the performance of different deep learning and machine learning algorithms on three different datasets. Our results show that the BERT model gives the best performance among all the models by achieving an F1-score of 90.6% on one of the datasets and F1-scores of 89.7% and 88.2% on the other two datasets.Keywords: hate speech, machine learning, deep learning, abusive words, social media, text classification
Procedia PDF Downloads 1273431 A Neural Network Approach for an Automatic Detection and Localization of an Open Phase Circuit of a Five-Phase Induction Machine Used in a Drivetrain of an Electric Vehicle
Authors: Saad Chahba, Rabia Sehab, Ahmad Akrad, Cristina Morel
Abstract:
Nowadays, the electric machines used in urban electric vehicles are, in most cases, three-phase electric machines with or without a magnet in the rotor. Permanent Magnet Synchronous Machine (PMSM) and Induction Machine (IM) are the main components of drive trains of electric and hybrid vehicles. These machines have very good performance in healthy operation mode, but they are not redundant to ensure safety in faulty operation mode. Faced with the continued growth in the demand for electric vehicles in the automotive market, improving the reliability of electric vehicles is necessary over the lifecycle of the electric vehicle. Multiphase electric machines respond well to this constraint because, on the one hand, they have better robustness in the event of a breakdown (opening of a phase, opening of an arm of the power stage, intern-turn short circuit) and, on the other hand, better power density. In this work, a diagnosis approach using a neural network for an open circuit fault or more of a five-phase induction machine is developed. Validation on the simulator of the vehicle drivetrain, at reduced power, is carried out, creating one and more open circuit stator phases showing the efficiency and the reliability of the new approach to detect and to locate on-line one or more open phases of a five-induction machine.Keywords: electric vehicle drivetrain, multiphase drives, induction machine, control, open circuit (OC) fault diagnosis, artificial neural network
Procedia PDF Downloads 1993430 Highly Accurate Tennis Ball Throwing Machine with Intelligent Control
Authors: Ferenc Kovács, Gábor Hosszú
Abstract:
The paper presents an advanced control system for tennis ball throwing machines to improve their accuracy according to the ball impact points. A further advantage of the system is the much easier calibration process involving the intelligent solution of the automatic adjustment of the stroking parameters according to the ball elasticity, the self-calibration, the use of the safety margin at very flat strokes and the possibility to placing the machine to any position of the half court. The system applies mathematical methods to determine the exact ball trajectories and special approximating processes to access all points on the aimed half court.Keywords: control system, robot programming, robot control, sports equipment, throwing machine
Procedia PDF Downloads 3883429 Constructing a Physics Guided Machine Learning Neural Network to Predict Tonal Noise Emitted by a Propeller
Authors: Arthur D. Wiedemann, Christopher Fuller, Kyle A. Pascioni
Abstract:
With the introduction of electric motors, small unmanned aerial vehicle designers have to consider trade-offs between acoustic noise and thrust generated. Currently, there are few low-computational tools available for predicting acoustic noise emitted by a propeller into the far-field. Artificial neural networks offer a highly non-linear and adaptive model for predicting isolated and interactive tonal noise. But neural networks require large data sets, exceeding practical considerations in modeling experimental results. A methodology known as physics guided machine learning has been applied in this study to reduce the required data set to train the network. After building and evaluating several neural networks, the best model is investigated to determine how the network successfully predicts the acoustic waveform. Lastly, a post-network transfer function is developed to remove discontinuity from the predicted waveform. Overall, methodologies from physics guided machine learning show a notable improvement in prediction performance, but additional loss functions are necessary for constructing predictive networks on small datasets.Keywords: aeroacoustics, machine learning, propeller, rotor, neural network, physics guided machine learning
Procedia PDF Downloads 2163428 Machine Learning Automatic Detection on Twitter Cyberbullying
Authors: Raghad A. Altowairgi
Abstract:
With the wide spread of social media platforms, young people tend to use them extensively as the first means of communication due to their ease and modernity. But these platforms often create a fertile ground for bullies to practice their aggressive behavior against their victims. Platform usage cannot be reduced, but intelligent mechanisms can be implemented to reduce the abuse. This is where machine learning comes in. Understanding and classifying text can be helpful in order to minimize the act of cyberbullying. Artificial intelligence techniques have expanded to formulate an applied tool to address the phenomenon of cyberbullying. In this research, machine learning models are built to classify text into two classes; cyberbullying and non-cyberbullying. After preprocessing the data in 4 stages; removing characters that do not provide meaningful information to the models, tokenization, removing stop words, and lowering text. BoW and TF-IDF are used as the main features for the five classifiers, which are; logistic regression, Naïve Bayes, Random Forest, XGboost, and Catboost classifiers. Each of them scores 92%, 90%, 92%, 91%, 86% respectively.Keywords: cyberbullying, machine learning, Bag-of-Words, term frequency-inverse document frequency, natural language processing, Catboost
Procedia PDF Downloads 1243427 Software Transactional Memory in a Dynamic Programming Language at Virtual Machine Level
Authors: Szu-Kai Hsu, Po-Ching Lin
Abstract:
As more and more multi-core processors emerge, traditional sequential programming paradigm no longer suffice. Yet only few modern dynamic programming languages can leverage such advantage. Ruby, for example, despite its wide adoption, only includes threads as a simple parallel primitive. The global virtual machine lock of official Ruby runtime makes it impossible to exploit full parallelism. Though various alternative Ruby implementations do eliminate the global virtual machine lock, they only provide developers dated locking mechanism for data synchronization. However, traditional locking mechanism error-prone by nature. Software Transactional Memory is one of the promising alternatives among others. This paper introduces a new virtual machine: GobiesVM to provide a native software transactional memory based solution for dynamic programming languages to exploit parallelism. We also proposed a simplified variation of Transactional Locking II algorithm. The empirical results of our experiments show that support of STM at virtual machine level enables developers to write straightforward code without compromising parallelism or sacrificing thread safety. Existing source code only requires minimal or even none modi cation, which allows developers to easily switch their legacy codebase to a parallel environment. The performance evaluations of GobiesVM also indicate the difference between sequential and parallel execution is significant.Keywords: global interpreter lock, ruby, software transactional memory, virtual machine
Procedia PDF Downloads 2773426 Housing Price Prediction Using Machine Learning Algorithms: The Case of Melbourne City, Australia
Authors: The Danh Phan
Abstract:
House price forecasting is a main topic in the real estate market research. Effective house price prediction models could not only allow home buyers and real estate agents to make better data-driven decisions but may also be beneficial for the property policymaking process. This study investigates the housing market by using machine learning techniques to analyze real historical house sale transactions in Australia. It seeks useful models which could be deployed as an application for house buyers and sellers. Data analytics show a high discrepancy between the house price in the most expensive suburbs and the most affordable suburbs in the city of Melbourne. In addition, experiments demonstrate that the combination of Stepwise and Support Vector Machine (SVM), based on the Mean Squared Error (MSE) measurement, consistently outperforms other models in terms of prediction accuracy.Keywords: house price prediction, regression trees, neural network, support vector machine, stepwise
Procedia PDF Downloads 2173425 The Effect of Postural Sway and Technical Parameters of 8 Weeks Technical Training Performed with Restrict of Visual Input on the 10-12 Ages Soccer Players
Authors: Nurtekin Erkmen, Turgut Kaplan, Halil Taskin, Ahmet Sanioglu, Gokhan Ipekoglu
Abstract:
The aim of this study was to determine the effects of an 8 week soccerspecific technical training with limited vision perception on postural control and technical parameters in 10-12 aged soccer players. Subjects in this study were 24 male young soccer players (age: 11.00 ± 0.56 years, height: 150.5 ± 4.23 cm, body weight: 41.49 ± 7.56 kg). Subjects were randomly divided as two groups: Training and control. Balance performance was measured by Biodex Balance System (BBS). Short pass, speed dribbling, 20 m speed with ball, ball control, juggling tests were used to measure soccer players’ technical performances with a ball. Subjects performed soccer training 3 times per week for 8 weeks. In each session, training group with limited vision perception and control group with normal vision perception committed soccer-specific technical drills for 20 min. Data analyzed with t-test for independent samples and Mann-Whitney U between groups and paired t-test and Wilcoxon test between pre-posttests. No significant difference was found balance scores and with eyes open and eyes closed and LOS test between training and control groups after training (p>0.05). After eight week of training there are no significant difference in balance score with eyes open for both training and control groups (p>0.05). Balance scores decreased in training and control groups after the training (p<0.05). The completion time of LOS test shortened in both training and control groups after training (p<0.05). The training developed speed dribbling performance of training group (p<0.05). On the other hand, soccer players’ performance in training and control groups increased in 20 m speed with a ball after eight week training (p<0.05). In conclusion; the results of this study indicate that soccer-specific training with limited vision perception may not improves balance performance in 10-12 aged soccer players, but it develops speed dribbling performance.Keywords: Young soccer players, vision perception, postural control, technical
Procedia PDF Downloads 4643424 Rapid Soil Classification Using Computer Vision with Electrical Resistivity and Soil Strength
Authors: Eugene Y. J. Aw, J. W. Koh, S. H. Chew, K. E. Chua, P. L. Goh, Grace H. B. Foo, M. L. Leong
Abstract:
This paper presents the evaluation of various soil testing methods such as the four-probe soil electrical resistivity method and cone penetration test (CPT) that can complement a newly developed novel rapid soil classification scheme using computer vision, to improve the accuracy and productivity of on-site classification of excavated soil. In Singapore, excavated soils from the local construction industry are transported to Staging Grounds (SGs) to be reused as fill material for land reclamation. Excavated soils are mainly categorized into two groups (“Good Earth” and “Soft Clay”) based on particle size distribution (PSD) and water content (w) from soil investigation reports and on-site visual survey, such that proper treatment and usage can be exercised. However, this process is time-consuming and labor-intensive. Thus, a rapid classification method is needed at the SGs. Four-probe soil electrical resistivity and CPT were evaluated for their feasibility as suitable additions to the computer vision system to further develop this innovative non-destructive and instantaneous classification method. The computer vision technique comprises soil image acquisition using an industrial-grade camera; image processing and analysis via calculation of Grey Level Co-occurrence Matrix (GLCM) textural parameters; and decision-making using an Artificial Neural Network (ANN). It was found from the previous study that the ANN model coupled with ρ can classify soils into “Good Earth” and “Soft Clay” in less than a minute, with an accuracy of 85% based on selected representative soil images. To further improve the technique, the following three items were targeted to be added onto the computer vision scheme: the apparent electrical resistivity of soil (ρ) measured using a set of four probes arranged in Wenner’s array, the soil strength measured using a modified mini cone penetrometer, and w measured using a set of time-domain reflectometry (TDR) probes. Laboratory proof-of-concept was conducted through a series of seven tests with three types of soils – “Good Earth”, “Soft Clay,” and a mix of the two. Validation was performed against the PSD and w of each soil type obtained from conventional laboratory tests. The results show that ρ, w and CPT measurements can be collectively analyzed to classify soils into “Good Earth” or “Soft Clay” and are feasible as complementing methods to the computer vision system.Keywords: computer vision technique, cone penetration test, electrical resistivity, rapid and non-destructive, soil classification
Procedia PDF Downloads 2283423 An Application for Risk of Crime Prediction Using Machine Learning
Authors: Luis Fonseca, Filipe Cabral Pinto, Susana Sargento
Abstract:
The increase of the world population, especially in large urban centers, has resulted in new challenges particularly with the control and optimization of public safety. Thus, in the present work, a solution is proposed for the prediction of criminal occurrences in a city based on historical data of incidents and demographic information. The entire research and implementation will be presented start with the data collection from its original source, the treatment and transformations applied to them, choice and the evaluation and implementation of the Machine Learning model up to the application layer. Classification models will be implemented to predict criminal risk for a given time interval and location. Machine Learning algorithms such as Random Forest, Neural Networks, K-Nearest Neighbors and Logistic Regression will be used to predict occurrences, and their performance will be compared according to the data processing and transformation used. The results show that the use of Machine Learning techniques helps to anticipate criminal occurrences, which contributed to the reinforcement of public security. Finally, the models were implemented on a platform that will provide an API to enable other entities to make requests for predictions in real-time. An application will also be presented where it is possible to show criminal predictions visually.Keywords: crime prediction, machine learning, public safety, smart city
Procedia PDF Downloads 1043422 TDApplied: An R Package for Machine Learning and Inference with Persistence Diagrams
Authors: Shael Brown, Reza Farivar
Abstract:
Persistence diagrams capture valuable topological features of datasets that other methods cannot uncover. Still, their adoption in data pipelines has been limited due to the lack of publicly available tools in R (and python) for analyzing groups of them with machine learning and statistical inference. In an easy-to-use and scalable R package called TDApplied, we implement several applied analysis methods tailored to groups of persistence diagrams. The two main contributions of our package are comprehensiveness (most functions do not have implementations elsewhere) and speed (shown through benchmarking against other R packages). We demonstrate applications of the tools on simulated data to illustrate how easily practical analyses of any dataset can be enhanced with topological information.Keywords: machine learning, persistence diagrams, R, statistical inference
Procedia PDF Downloads 78