Search results for: science-informed machine learning

Authors: Angela Meyer

Abstract:

The operation and maintenance cost for wind parks make up a major fraction of the park’s overall lifetime cost. To minimize the cost and risk involved, an optimal operation and maintenance strategy requires continuous monitoring and analysis. In order to facilitate this, we present a decision support system that automatically scans the stream of telemetry sensor data generated from the turbines. By learning decision boundaries and normal reference operating states using machine learning algorithms, the decision support system can detect anomalous operating behavior in individual wind turbines and diagnose the involved turbine sub-systems. Operating personal can be alerted if a normal operating state boundary is exceeded. The presented decision support system and method are applicable for any turbine type and manufacturer providing telemetry data of the turbine operating state. We demonstrate the successful detection and diagnosis of anomalous operating states in a case study at a German onshore wind park comprised of Vestas V112 turbines.

Keywords: anomaly detection, decision support, machine learning, monitoring, performance optimization, wind turbines

Procedia PDF Downloads 167

Authors: Jaya Mathew

Abstract:

Many organizations are faced with the challenge of how to analyze and build Machine Learning models using their sensitive telemetry data. In this paper, we discuss how users can leverage the power of R without having to move their big data around as well as a cloud based solution for organizations willing to host their data in the cloud. By using ScaleR technology to benefit from parallelization and remote computing or R Services on premise or in the cloud, users can leverage the power of R at scale without having to move their data around.

Keywords: predictive maintenance, machine learning, big data, cloud based, on premise solution, R

Procedia PDF Downloads 379

Authors: Aqsa Ali, Aleem Mushtaq, Attaullah Memon, Monna

Abstract:

In this paper, we present a low cost design for a smart glove that can perform sign language recognition to assist the speech impaired people. Specifically, we have designed and developed an Assistive Hand Gesture Interpreter that recognizes hand movements relevant to the American Sign Language (ASL) and translates them into text for display on a Thin-Film-Transistor Liquid Crystal Display (TFT LCD) screen as well as synthetic speech. Linear Bayes Classifiers and Multilayer Neural Networks have been used to classify 11 feature vectors obtained from the sensors on the glove into one of the 27 ASL alphabets and a predefined gesture for space. Three types of features are used; bending using six bend sensors, orientation in three dimensions using accelerometers and contacts at vital points using contact sensors. To gauge the performance of the presented design, the training database was prepared using five volunteers. The accuracy of the current version on the prepared dataset was found to be up to 99.3% for target user. The solution combines electronics, e-textile technology, sensor technology, embedded system and machine learning techniques to build a low cost wearable glove that is scrupulous, elegant and portable.

Keywords: American sign language, assistive hand gesture interpreter, human-machine interface, machine learning, sensing glove

Procedia PDF Downloads 301

Authors: Mitat Uysal, Aynur Uysal

Abstract:

Artificial Intelligence (AI) is fundamentally driven by mathematics, with many of its core algorithms rooted in mathematical principles such as linear algebra, probability theory, calculus, and optimization techniques. This paper explores the deep connection between AI and mathematics, highlighting the role of mathematical concepts in key AI techniques like machine learning, neural networks, and optimization. To demonstrate this connection, a case study involving the implementation of a neural network using Python is presented. This practical example illustrates the essential role that mathematics plays in training a model and solving real-world problems.

Keywords: AI, mathematics, machine learning, optimization techniques, image processing

Procedia PDF Downloads 15

Authors: Anjali Chouksey, Riya Nimje, Jahanvi Saraf

Abstract:

Online transactions have increased dramatically in this new ‘social-distancing’ era. With online transactions, Fraud in online payments has also increased significantly. Frauds are a significant problem in various industries like insurance companies, baking, etc. These frauds include leaking sensitive information related to the credit card, which can be easily misused. Due to the government also pushing online transactions, E-commerce is on a boom. But due to increasing frauds in online payments, these E-commerce industries are suffering a great loss of trust from their customers. These companies are finding credit card fraud to be a big problem. People have started using online payment options and thus are becoming easy targets of credit card fraud. In this research paper, we will be discussing machine learning algorithms. We have used a decision tree, XGBOOST, k-nearest neighbour, logistic-regression, random forest, and SVM on a dataset in which there are transactions done online mode using credit cards. We will test all these algorithms for detecting fraud cases using the confusion matrix, F1 score, and calculating the accuracy score for each model to identify which algorithm can be used in detecting frauds.

Keywords: machine learning, fraud detection, artificial intelligence, decision tree, k nearest neighbour, random forest, XGBOOST, logistic regression, support vector machine

Procedia PDF Downloads 148

Authors: Ahmet Karakuş, Akif Can Kilic, Emre Alptekin

Abstract:

A growing number of studies have been conducted to determine how well-being may be predicted using well-designed models. It is necessary to investigate the backgrounds of features in order to construct a viable Subjective Well-Being (SWB) model. We have picked the suitable variables from the literature on SWB that are acceptable for real-world data instructions. The goal of this work is to evaluate the model by feeding it with SWB characteristics and then categorizing the stress levels using machine learning methods to see how well it performs on a real dataset. Despite the fact that it is a multiclass classification issue, we have achieved significant metric scores, which may be taken into account for a specific task.

Keywords: machine learning, multiclassification problem, subjective well-being, perceived stress scale

Procedia PDF Downloads 131

Authors: Ravija Gunawardana, Banuka Athuraliya

Abstract:

Machine learning has emerged as a powerful tool for disease diagnosis and prediction. The use of machine learning algorithms has the potential to improve the accuracy of disease prediction, thereby enabling medical professionals to provide more effective and personalized treatments. This study focuses on developing a machine-learning model for disease prediction using symptoms and X-ray images. The importance of this study lies in its potential to assist medical professionals in accurately diagnosing diseases, thereby improving patient outcomes. Respiratory diseases are a significant cause of morbidity and mortality worldwide, and chest X-rays are commonly used in the diagnosis of these diseases. However, accurately interpreting X-ray images requires significant expertise and can be time-consuming, making it difficult to diagnose respiratory diseases in a timely manner. By incorporating machine learning algorithms, we can significantly enhance disease prediction accuracy, ultimately leading to better patient care. The study utilized the Mask R-CNN algorithm, which is a state-of-the-art method for object detection and segmentation in images, to process chest X-ray images. The model was trained and tested on a large dataset of patient information, which included both symptom data and X-ray images. The performance of the model was evaluated using a range of metrics, including accuracy, precision, recall, and F1-score. The results showed that the model achieved an accuracy rate of over 90%, indicating that it was able to accurately detect and segment regions of interest in the X-ray images. In addition to X-ray images, the study also incorporated symptoms as input data for disease prediction. The study used three different classifiers, namely Random Forest, K-Nearest Neighbor and Support Vector Machine, to predict diseases based on symptoms. These classifiers were trained and tested using the same dataset of patient information as the X-ray model. The results showed promising accuracy rates for predicting diseases using symptoms, with the ensemble learning techniques significantly improving the accuracy of disease prediction. The study's findings indicate that the use of machine learning algorithms can significantly enhance disease prediction accuracy, ultimately leading to better patient care. The model developed in this study has the potential to assist medical professionals in diagnosing respiratory diseases more accurately and efficiently. However, it is important to note that the accuracy of the model can be affected by several factors, including the quality of the X-ray images, the size of the dataset used for training, and the complexity of the disease being diagnosed. In conclusion, the study demonstrated the potential of machine learning algorithms for disease prediction using symptoms and X-ray images. The use of these algorithms can improve the accuracy of disease diagnosis, ultimately leading to better patient care. Further research is needed to validate the model's accuracy and effectiveness in a clinical setting and to expand its application to other diseases.

Keywords: K-nearest neighbor, mask R-CNN, random forest, support vector machine

Procedia PDF Downloads 154

Authors: Kaoutar Ben Azzou, Hanaa Talei

Abstract:

Human resources is an important department in each organization as it manages the life cycle of employees from recruitment training to retirement or termination of contracts. The recruitment process starts with a job opening, followed by a selection of the best-fit candidates from all applicants. Matching the best profile for a job position requires a manual way of looking at many CVs, which requires hours of work that can sometimes lead to choosing not the best profile. The work presented in this paper aims at reducing the workload of HR personnel by automating the preliminary stages of the candidate screening process, thereby fostering a more streamlined recruitment workflow. This tool introduces an automated system designed to help with the recruitment process by scanning candidates' CVs, extracting pertinent features, and employing machine learning algorithms to decide the most fitting job profile for each candidate. Our work employs natural language processing (NLP) techniques to identify and extract key features from unstructured text extracted from a CV, such as education, work experience, and skills. Subsequently, the system utilizes these features to match candidates with job profiles, leveraging the power of classification algorithms.

Keywords: automated recruitment, candidate screening, machine learning, human resources management

Procedia PDF Downloads 56

Authors: Kemal Efe Eseller, Göktuğ Yazici

Abstract:

Laser-induced breakdown spectroscopy (LIBS) is a rapid optical atomic emission spectroscopy which is used for material identification and analysis with the advantages of in-situ analysis, elimination of intensive sample preparation, and micro-destructive properties for the material to be tested. LIBS delivers short pulses of laser beams onto the material in order to create plasma by excitation of the material to a certain threshold. The plasma characteristics, which consist of wavelength value and intensity amplitude, depends on the material and the experiment’s environment. In the present work, medicine samples’ spectrum profiles were obtained via LIBS. Medicine samples’ datasets include two different concentrations for both paracetamol based medicines, namely Aferin and Parafon. The spectrum data of the samples were preprocessed via filling outliers based on quartiles, smoothing spectra to eliminate noise and normalizing both wavelength and intensity axis. Statistical information was obtained and principal component analysis (PCA) was incorporated to both the preprocessed and raw datasets. The machine learning models were set based on two different train-test splits, which were 70% training – 30% test and 80% training – 20% test. Cross-validation was preferred to protect the models against overfitting; thus the sample amount is small. The machine learning results of preprocessed and raw datasets were subjected to comparison for both splits. This is the first time that all supervised machine learning classification algorithms; consisting of Decision Trees, Discriminant, naïve Bayes, Support Vector Machines (SVM), k-NN(k-Nearest Neighbor) Ensemble Learning and Neural Network algorithms; were incorporated to LIBS data of paracetamol based pharmaceutical samples, and their different concentrations on preprocessed and raw dataset in order to observe the effect of preprocessing.

Keywords: machine learning, laser-induced breakdown spectroscopy, medicines, principal component analysis, preprocessing

Procedia PDF Downloads 87

Authors: Yuanqing Li, Changyi Lei, Zhaopeng Xue, Zhuo Zheng, Yanbo Long

Abstract:

In agricultural production and breeding, implementing automatic picking robot in orchard farming to reduce human labour and error is challenging. The core function of it is automatic identification based on machine vision. This paper focuses on apple detection and counting in orchards and implements several deep learning methods. Extensive datasets are used and a semi-automatic annotation method is proposed. The proposed deep learning models are in state-of-the-art YOLO family. In view of the essence of the models with various backbones, a multi-dimensional comparison in details is made in terms of counting accuracy, mAP and model memory, laying the foundation for realising automatic precision agriculture.

Keywords: agricultural object detection, deep learning, machine vision, YOLO family

Procedia PDF Downloads 197

Authors: Qing Wu, Fei-Yan Li, Heng-Chang Zhang

Abstract:

Multitask learning support vector machines (SVMs) have recently attracted increasing research attention. Given several related tasks, the single-task learning methods trains each task separately and ignore the inner cross-relationship among tasks. However, multitask learning can capture the correlation information among tasks and achieve better performance by training all tasks simultaneously. In addition, the asymmetric squared loss function can better improve the generalization ability of the models on the most asymmetric distributed data. In this paper, we first make two assumptions on the relatedness among tasks and propose two multitask learning proximal support vector machine algorithms, named MTL-a-PSVM and EMTL-a-PSVM, respectively. MTL-a-PSVM seeks a trade-off between the maximum expectile distance for each task model and the closeness of each task model to the general model. As an extension of the MTL-a-PSVM, EMTL-a-PSVM can select appropriate kernel functions for shared information and private information. Besides, two corresponding special cases named MTL-PSVM and EMTLPSVM are proposed by analyzing the asymmetric squared loss function, which can be easily implemented by solving linear systems. Experimental analysis of three classification datasets demonstrates the effectiveness and superiority of our proposed multitask learning algorithms.

Keywords: multitask learning, asymmetric squared loss, EMTL-a-PSVM, classification

Procedia PDF Downloads 134

Authors: Sun-Young Jang, Sung-Ah Kim, Dongyoun Shin

Abstract:

Current real-estate value estimation, difficult for laymen, usually is performed by specialists. This paper presents an automated estimation process based on big data and machine-learning technology that calculates influences of building conditions on real-estate price measurement. The present study analyzed actual building sales sample data for Nonhyeon-dong, Gangnam-gu, Seoul, Korea, measuring the major influencing factors among the various building conditions. Further to that analysis, a prediction model was established and applied using RapidMiner Studio, a graphical user interface (GUI)-based tool for derivation of machine-learning prototypes. The prediction model is formulated by reference to previous examples. When new examples are applied, it analyses and predicts accordingly. The analysis process discerns the crucial factors effecting price increases by calculation of weighted values. The model was verified, and its accuracy determined, by comparing its predicted values with actual price increases.

Keywords: apartment complex, big data, life-cycle building value analysis, machine learning

Procedia PDF Downloads 374

Authors: Alvaro Figueira, Bruno Cabral

Abstract:

Nowadays, students are used to be assessed through an online platform. Educators have stepped up from a period in which they endured the transition from paper to digital. The use of a diversified set of question types that range from quizzes to open questions is currently common in most university courses. In many courses, today, the evaluation methodology also fosters the students’ online participation in forums, the download, and upload of modified files, or even the participation in group activities. At the same time, new pedagogy theories that promote the active participation of students in the learning process, and the systematic use of problem-based learning, are being adopted using an eLearning system for that purpose. However, although there can be a lot of feedback from these activities to student’s, usually it is restricted to the assessments of online well-defined tasks. In this article, we propose an automatic system that informs students of abnormal deviations of a 'correct' learning path in the course. Our approach is based on the fact that by obtaining this information earlier in the semester, may provide students and educators an opportunity to resolve an eventual problem regarding the student’s current online actions towards the course. Our goal is to prevent situations that have a significant probability to lead to a poor grade and, eventually, to failing. In the major learning management systems (LMS) currently available, the interaction between the students and the system itself is registered in log files in the form of registers that mark beginning of actions performed by the user. Our proposed system uses that logged information to derive new one: the time each student spends on each activity, the time and order of the resources used by the student and, finally, the online resource usage pattern. Then, using the grades assigned to the students in previous years, we built a learning dataset that is used to feed a machine learning meta classifier. The produced classification model is then used to predict the grades a learning path is heading to, in the current year. Not only this approach serves the teacher, but also the student to receive automatic feedback on her current situation, having past years as a perspective. Our system can be applied to online courses that integrate the use of an online platform that stores user actions in a log file, and that has access to other student’s evaluations. The system is based on a data mining process on the log files and on a self-feedback machine learning algorithm that works paired with the Moodle LMS.

Keywords: data mining, e-learning, grade prediction, machine learning, student learning path

Procedia PDF Downloads 122

Authors: Hexiong Liu

Abstract:

Powder metallurgy is the optimal method for the consolidation and preparation of W(Mo) alloys, which exhibit excellent application prospects at high temperatures. The properties of W(Mo) alloys are closely related to the sintered density. However, controlling the sintered density and porosity of these alloys is still challenging. In the past, the regulation methods mainly focused on time-consuming and costly trial-and-error experiments. In this study, the sintering data for more than a dozen W(Mo) alloys constituted a small-scale dataset, including both solid and liquid phases of sintering. Furthermore, simple descriptors were used to predict the sintered density of W(Mo) alloys based on the descriptor selection strategy and machine learning method (ML), where the ML algorithm included the least absolute shrinkage and selection operator (Lasso) regression, k-nearest neighbor (k-NN), random forest (RF), and multi-layer perceptron (MLP). The results showed that the interpretable descriptors extracted by our proposed selection strategy and the MLP neural network achieved a high prediction accuracy (R>0.950). By further predicting the sintered density of W(Mo) alloys using different sintering processes, the error between the predicted and experimental values was less than 0.063, confirming the application potential of the model.

Keywords: sintered density, machine learning, interpretable descriptors, W(Mo) alloy

Procedia PDF Downloads 82

Authors: Marian Sorin Ionescu, Olivia Negoita, Cosmin Dobrin

Abstract:

This scientific communication reports and discusses learning models adaptable to modern business problems and models specific to digital concepts and paradigms. In the PAC (probably approximately correct) learning model approach, in which the learning process begins by receiving a batch of learning examples, the set of learning processes is used to acquire a hypothesis, and when the learning process is fully used, this hypothesis is used in the prediction of new operational examples. For complex business models, a lot of models should be introduced and evaluated to estimate the induced results so that the totality of the results are used to develop a predictive rule, which anticipates the choice of new models. In opposition, for online learning-type processes, there is no separation between the learning (training) and predictive phase. Every time a business model is approached, a test example is considered from the beginning until the prediction of the appearance of a model considered correct from the point of view of the business decision. After choosing choice a part of the business model, the label with the logical value "true" is known. Some of the business models are used as examples of learning (training), which helps to improve the prediction mechanisms for future business models.

Keywords: machine learning, business models, convex analysis, online learning

Procedia PDF Downloads 141

Authors: Natascha C. D'Amico, Enzo Grossi, Giovanni Valbusa, Ala Malasevschi, Gianpiero Cardone, Sergio Papa

Abstract:

Purpose: To characterize, through a radiomic approach, the nature of areas classified PI-RADS (Prostate Imaging Reporting and Data System) 3/5, recognized in multiparametric prostate magnetic resonance with T2-weighted (T2w), diffusion and perfusion sequences with paramagnetic contrast. Methods and Materials: 24 cases undergoing multiparametric prostate MR and biopsy were admitted to this pilot study. Clinical outcome of the PI-RADS 3/5 was found through biopsy, finding 8 malignant tumours. The analysed images were acquired with a Philips achieva 1.5T machine with a CE- T2-weighted sequence in the axial plane. Semi-automatic tumour segmentation was carried out on MR images using 3DSlicer image analysis software. 45 shape-based, intensity-based and texture-based features were extracted and represented the input for preprocessing. An evolutionary algorithm (a TWIST system based on KNN algorithm) was used to subdivide the dataset into training and testing set and select features yielding the maximal amount of information. After this pre-processing 20 input variables were selected and different machine learning systems were used to develop a predictive model based on a training testing crossover procedure. Results: The best machine learning system (three-layers feed-forward neural network) obtained a global accuracy of 90% ( 80 % sensitivity and 100% specificity ) with a ROC of 0.82. Conclusion: Machine learning systems coupled with radiomics show a promising potential in distinguishing benign from malign tumours in PI-RADS 3/5 areas.

Keywords: machine learning, MR prostate, PI-Rads 3, radiomics

Procedia PDF Downloads 188

Authors: Kamel Belammi, Houria Fatrim

Abstract:

imbalanced data set, a problem often found in real world application, can cause seriously negative effect on classification performance of machine learning algorithms. There have been many attempts at dealing with classification of imbalanced data sets. In medical diagnosis classification, we often face the imbalanced number of data samples between the classes in which there are not enough samples in rare classes. In this paper, we proposed a learning method based on a cost sensitive extension of Least Mean Square (LMS) algorithm that penalizes errors of different samples with different weight and some rules of thumb to determine those weights. After the balancing phase, we applythe different classifiers (support vector machine (SVM), k- nearest neighbor (KNN) and multilayer neuronal networks (MNN)) for balanced data set. We have also compared the obtained results before and after balancing method.

Keywords: multilayer neural networks, k- nearest neighbor, support vector machine, imbalanced medical data, least mean square algorithm, diabetes

Procedia PDF Downloads 532

Authors: Jonathan Heng, Yoong Cheah Huei

Abstract:

A strategic model that does not trigger any false alarms to detect anomalies in Secure Water Treatment (SWaT) test bed is presented. This model uses machine learning invariants formulated from streamlining the general form of Auto-Regressive models with eXogenous input. A creative generalized CUSUM algorithm to integrate the invariants and the detection strategy technique is successfully developed and tested in the SWaT Programmable Logic Controllers (PLCs). Three steps to fine-tune parameters, b and τ in the generalized algorithm are stated and an example used to demonstrate the tuning process is discussed. This approach can swiftly and effectively detect various scopes of cyber-attacks such as multiple points single stage and multiple points multiple stages in SWaT. This technique can be applied in water treatment plants and other cyber physical systems like power and gas plants too.

Keywords: machine learning invariants, generalized CUSUM algorithm with invariants and detection strategy, scope of cyber attacks, strategic model, tuning parameters

Procedia PDF Downloads 181

Authors: Danica Maljkovic, Igor Balen, Bojana Dalbelo Basic

Abstract:

Following the regulation of EU Directive on Energy Efficiency, specifically Article 9, individual metering in district heating systems has to be introduced by the end of 2016. These directions have been implemented in member state’s legal framework, Croatia is one of these states. The directive allows installation of both heat metering devices and heat cost allocators. Mainly due to bad communication and PR, the general public false image was created that the heat cost allocators are devices that save energy. Although this notion is wrong, the aim of this work is to develop a model that would precisely express the influence of installation heat cost allocators on potential energy savings in each unit within multifamily buildings. At the same time, in recent years, a science of machine learning has gain larger application in various fields, as it is proven to give good results in cases where large amounts of data are to be processed with an aim to recognize a pattern and correlation of each of the relevant parameter as well as in the cases where the problem is too complex for a human intelligence to solve. A special method of machine learning, decision tree method, has proven an accuracy of over 92% in prediction general building consumption. In this paper, a machine learning algorithms will be used to isolate the sole impact of installation of heat cost allocators on a single building in multifamily houses connected to district heating systems. Special emphasises will be given regression analysis, logistic regression, support vector machines, decision trees and random forest method.

Keywords: district heating, heat cost allocator, energy efficiency, machine learning, decision tree model, regression analysis, logistic regression, support vector machines, decision trees and random forest method

Procedia PDF Downloads 249

Authors: Alhusein Almahjoub, Dongyu Qiu

Abstract:

This paper deals with the random access procedure in next-generation networks and presents the solution to reduce total service time (TST) which is one of the most important performance metrics in current and future internet of things (IoT) based networks. The proposed solution focuses on the calculation of optimal transmission probability which maximizes the success probability and reduces TST. It uses the information of several idle preambles in every time slot, and based on it, it estimates the number of backlogged IoT devices using Naïve Bayes estimation which is a type of supervised learning in the machine learning domain. The estimation of backlogged devices is necessary since optimal transmission probability depends on it and the eNodeB does not have information about it. The simulations are carried out in MATLAB which verify that the proposed solution gives excellent performance.

Keywords: random access, LTE/LTE-A, 5G, machine learning, Naïve Bayes estimation

Procedia PDF Downloads 145

Authors: Saeed Eskandari, Seyed Rasoul Mehdikhani

Abstract:

Radiation sources have been used in many industries, such as gamma sources in medical imaging. These waves have destructive effects on humans and the environment. It is very important to detect and find the source of these waves because these sources cannot be seen by the eye. A portable robot has been designed and built with the purpose of revealing radiation sources that are able to scan the place from 5 to 20 meters away and shows the location of the sources according to the intensity of the waves on a two-dimensional digital image. The operation of the robot is done by measuring the pixels separately. By increasing the image measurement resolution, we will have a more accurate scan of the environment, and more points will be detected. But this causes a lot of time to be spent on scanning. In this paper, to overcome this challenge, we designed a method that can optimize this time. In this method, a small number of important points of the environment are measured. Hence the remaining pixels are predicted and estimated by regression algorithms in machine learning. The research method is based on comparing the actual values of all pixels. These steps have been repeated with several other radiation sources. The obtained results of the study show that the values estimated by the regression method are very close to the real values.

Keywords: regression, machine learning, scan radiation, robot

Procedia PDF Downloads 79

Authors: Ahmad Saleem

Abstract:

Ethereum is the second largest crypto currency in blockchain ecosystem. Along with standard transactions, it supports smart contracts and NFT’s. Current research trends are focused on analyzing the overall structure of the network its growth and behavior. Ethereum addresses are anonymous and can be created on fly. The nature of Ethereum network and addresses make it hard to predict their behavior. The activity period of an ethereum address is not much analyzed. Using machine learning we can make early prediction about the disposability of the address. In this paper we analyzed the lifetime of the addresses. We also identified and predicted the disposable addresses using machine learning models and compared the results.

Keywords: blockchain, Ethereum, cryptocurrency, prediction

Procedia PDF Downloads 97

Authors: Szymon Rutkowski, Tomasz Korbak

Abstract:

We present a reinforcement-learning-based method of training neural machine translation models without parallel corpora. The standard encoder-decoder approach to machine translation suffers from two problems we aim to address. First, it needs parallel corpora, which are scarce, especially for low-resource languages. Second, it lacks psychological plausibility of learning procedure: learning a foreign language is about learning to communicate useful information, not merely learning to transduce from one language’s 'encoding' to another. We instead pose the problem of learning to translate as learning a policy in a communication game between two agents: the translator and the classifier. The classifier is trained beforehand on a natural language inference task (determining the entailment relation between a premise and a hypothesis) in the target language. The translator produces a sequence of actions that correspond to generating translations of both the hypothesis and premise, which are then passed to the classifier. The translator is rewarded for classifier’s performance on determining entailment between sentences translated by the translator to disciple’s native language. Translator’s performance thus reflects its ability to communicate useful information to the classifier. In effect, we train a machine translation model without the need for parallel corpora altogether. While similar reinforcement learning formulations for zero-shot translation were proposed before, there is a number of improvements we introduce. While prior research aimed at grounding the translation task in the physical world by evaluating agents on an image captioning task, we found that using a linguistic task is more sample-efficient. Natural language inference (also known as recognizing textual entailment) captures semantic properties of sentence pairs that are poorly correlated with semantic similarity, thus enforcing basic understanding of the role played by compositionality. It has been shown that models trained recognizing textual entailment produce high-quality general-purpose sentence embeddings transferrable to other tasks. We use stanford natural language inference (SNLI) dataset as well as its analogous datasets for French (XNLI) and Polish (CDSCorpus). Textual entailment corpora can be obtained relatively easily for any language, which makes our approach more extensible to low-resource languages than traditional approaches based on parallel corpora. We evaluated a number of reinforcement learning algorithms (including policy gradients and actor-critic) to solve the problem of translator’s policy optimization and found that our attempts yield some promising improvements over previous approaches to reinforcement-learning based zero-shot machine translation.

Keywords: agent-based language learning, low-resource translation, natural language inference, neural machine translation, reinforcement learning

Procedia PDF Downloads 128

Authors: Ava Dodd, Jonathan Adams

Abstract:

The purpose of this study is to explore the characteristics of developing a machine learning application using synthetic data. The study is structured to develop the application for the purpose of deploying the computer vision model. The findings discuss the realities of attempting to develop a computer vision model for practical purpose, and detail the processes, tools, and techniques that were used to meet accuracy requirements. The research reveals that synthetic data represents another variable that can be adjusted to improve the performance of a computer vision model. Further, a suite of tools and tuning recommendations are provided.

Keywords: computer vision, machine learning, synthetic data, YOLOv4

Procedia PDF Downloads 225

Authors: Shaan Khosla, Jon Krohn

Abstract:

In a recruiting context, machine learning models are valuable for recommendations: to predict the best candidates for a vacancy, to match the best vacancies for a candidate, and compile a set of similar candidates for any given candidate. While useful to create these models, validating their accuracy in a recommendation context is difficult due to a sparsity of data. In this report, we use network graph data to generate useful representations for candidates and vacancies. We use candidates and vacancies as network nodes and designate a bi-directional link between them based on the candidate interviewing for the vacancy. After using node2vec, the embeddings are used to construct a validation dataset with a ranked order, which will help validate new recommender systems.

Keywords: AI, machine learning, NLP, recruiting

Procedia PDF Downloads 84

Authors: Rohini Hariharan, Yazhini R, Bhamidipati Naga Shrikarti

Abstract:

In the realm of cricket, particularly within the context of the Indian Premier League (IPL), the ability to predict team scores accurately holds significant importance for both cricket enthusiasts and stakeholders alike. This paper presents a comprehensive study on IPL score prediction utilizing various machine learning algorithms, including Support Vector Machines (SVM), XGBoost, Multiple Regression, Linear Regression, K-nearest neighbors (KNN), and Random Forest. Through meticulous data preprocessing, feature engineering, and model selection, we aimed to develop a robust predictive framework capable of forecasting team scores with high precision. Our experimentation involved the analysis of historical IPL match data encompassing diverse match and player statistics. Leveraging this data, we employed state-of-the-art machine learning techniques to train and evaluate the performance of each model. Notably, Multiple Regression emerged as the top-performing algorithm, achieving an impressive accuracy of 77.19% and a precision of 54.05% (within a threshold of +/- 10 runs). This research contributes to the advancement of sports analytics by demonstrating the efficacy of machine learning in predicting IPL team scores. The findings underscore the potential of advanced predictive modeling techniques to provide valuable insights for cricket enthusiasts, team management, and betting agencies. Additionally, this study serves as a benchmark for future research endeavors aimed at enhancing the accuracy and interpretability of IPL score prediction models.

Keywords: indian premier league (IPL), cricket, score prediction, machine learning, support vector machines (SVM), xgboost, multiple regression, linear regression, k-nearest neighbors (KNN), random forest, sports analytics

Procedia PDF Downloads 53

Authors: Violeta Damjanovic-Behrendt

Abstract:

This paper presents an approach for optimal cyber security decisions to protect instances of a federated Internet of Things (IoT) platform in the cloud. The presented solution implements the repeated Stackelberg Security Game (SSG) and a model called Stochastic Human behaviour model with AttRactiveness and Probability weighting (SHARP). SHARP employs the Subjective Utility Quantal Response (SUQR) for formulating a subjective utility function, which is based on the evaluations of alternative solutions during decision-making. We augment the repeated SSG (including SHARP and SUQR) with a reinforced learning algorithm called Naïve Q-Learning. Naïve Q-Learning belongs to the category of active and model-free Machine Learning (ML) techniques in which the agent (either the defender or the attacker) attempts to find an optimal security solution. In this way, we combine GT and ML algorithms for discovering optimal cyber security policies. The proposed security optimization components will be validated in a collaborative cloud platform that is based on the Industrial Internet Reference Architecture (IIRA) and its recently published security model.

Keywords: security, internet of things, cloud computing, stackelberg game, machine learning, naive q-learning

Procedia PDF Downloads 354

Authors: Shatha Ghareeb, Rawaa Al-Jumeily, Thar Baker

Abstract:

In Abu Dhabi, there are many different education curriculums where sector of private schools and quality assurance is supervising many private schools in Abu Dhabi for many nationalities. As there are many different education curriculums in Abu Dhabi to meet expats’ needs, there are different requirements for registration and success. In addition, there are different age groups for starting education in each curriculum. In fact, each curriculum has a different number of years, assessment techniques, reassessment rules, and exam boards. Currently, students that transfer curriculums are not being placed in the right year group due to different start and end dates of each academic year and their date of birth for each year group is different for each curriculum and as a result, we find students that are either younger or older for that year group which therefore creates gaps in their learning and performance. In addition, there is not a way of storing student data throughout their academic journey so that schools can track the student learning process. In this paper, we propose to develop a computational framework applicable in multicultural countries such as UAE in which multi-education systems are implemented. The ultimate goal is to use cloud and fog computing technology integrated with Artificial Intelligence techniques of Machine Learning to aid in a smooth transition when assigning students to their year groups, and provide leveling and differentiation information of students who relocate from a particular education curriculum to another, whilst also having the ability to store and access student data from anywhere throughout their academic journey.

Keywords: admissions, algorithms, cloud computing, differentiation, fog computing, levelling, machine learning

Procedia PDF Downloads 142

Authors: Kuei-Ling Sun, Emily Chia-Yu Su

Abstract:

Allergy is a hypersensitive overreaction of the immune system to environmental stimuli, and a major health problem. These overreactions include rashes, sneezing, fever, food allergies, anaphylaxis, asthmatic, shock, or other abnormal conditions. Allergies can be caused by food, insect stings, pollen, animal wool, and other allergens. Their development of allergies is due to both genetic and environmental factors. Allergies involve immunoglobulin E antibodies, a part of the body’s immune system. Immunoglobulin E antibodies will bind to an allergen and then transfer to a receptor on mast cells or basophils triggering the release of inflammatory chemicals such as histamine. Based on the increasingly serious problem of environmental change, changes in lifestyle, air pollution problem, and other factors, in this study, we both collect allergens and non-allergens from several databases and use several machine learning methods for classification, including logistic regression (LR), stepwise regression, decision tree (DT) and neural networks (NN) to do the model comparison and determine the permutations of allergen amino acid’s sequence.

Keywords: allergy, classification, decision tree, logistic regression, machine learning

Procedia PDF Downloads 303

Authors: Emanuel Koseos

Abstract:

Data-Driven Decision Making (DDDM) refers to making decisions that are based on historical data in order to inform practice, develop strategies and implement policies that benefit organizational settings. In educational technology, DDDM facilitates the implementation of differential educational learning approaches such as Educational Data Mining (EDM) and Competency-Based Education (CBE), which commonly target university classrooms. There is a current need for DDDM models applied to middle and secondary schools from a concern for assessing the needs, progress and performance of students and educators with respect to regional standards, policies and evolution of curriculums. To address these concerns, we propose a DDDM reference model developed using educational key process initiatives as inputs to a machine learning framework implemented with statistical software (SAS, R) to provide a best-practices, complex-free and automated approach for educators at their regional level. We assessed the efficiency of the model over a six-year period using data from 45 schools and grades K-12 in the Langley, BC, Canada regional school district. We concluded that the model has wider appeal, such as business learning systems.

Keywords: competency-based learning, data-driven decision making, machine learning, secondary schools

Procedia PDF Downloads 174