Search results for: supervised learning algorithm

Authors: Eunji Choi, Borang Park, Youngjae Choi, Jinwoo Moon

Abstract:

In this study, a predictive model for estimating the metabolism (MET) of human body was developed for the optimal control of indoor thermal environment. Human body images for indoor activities and human body joint coordinated values were collected as data sets, which are used in predictive model. A deep learning algorithm was used in an initial model, and its number of hidden layers and hidden neurons were optimized. Lastly, the model prediction performance was analyzed after the model being trained through collected data. In conclusion, the possibility of MET prediction was confirmed, and the direction of the future study was proposed as developing various data and the predictive model.

Keywords: deep learning, indoor quality, metabolism, predictive model

Procedia PDF Downloads 251

Authors: Archana Singh, Jyoti Agarwal, Ajay Rana

Abstract:

Data Mining is an efficient technology to discover patterns in large databases. Association Rule Mining techniques are used to find the correlation between the various item sets in a database, and this co-relation between various item sets are used in decision making and pattern analysis. In recent years, the problem of finding association rules from large datasets has been proposed by many researchers. Various research papers on association rule mining (ARM) are studied and analyzed first to understand the existing algorithms. Apriori algorithm is the basic ARM algorithm, but it requires so many database scans. In DIC algorithm, less amount of database scan is needed but complex data structure lattice is used. The main focus of this paper is to propose a new optimized algorithm (Friendly Algorithm) and compare its performance with the existing algorithms A data set is used to find out frequent itemsets and association rules with the help of existing and proposed (Friendly Algorithm) and it has been observed that the proposed algorithm also finds all the frequent itemsets and essential association rules from databases as compared to existing algorithms in less amount of database scan. In the proposed algorithm, an optimized data structure is used i.e. Graph and Adjacency Matrix.

Keywords: association rules, data mining, dynamic item set counting, FP-growth, friendly algorithm, graph

Procedia PDF Downloads 413

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 158

Authors: Ji Eun Shin, Dong Hoon Lim

Abstract:

Data clustering is a common technique used in data analysis and is used in many applications, such as artificial intelligence, pattern recognition, economics, ecology, psychiatry and marketing. K-means clustering is a well-known clustering algorithm aiming to cluster a set of data points to a predefined number of clusters. In this paper, we implement K-means algorithm based on MapReduce framework with RHadoop to make the clustering method applicable to large scale data. RHadoop is a collection of R packages that allow users to manage and analyze data with Hadoop. The main idea is to introduce a combiner as a function of our map output to decrease the amount of data needed to be processed by reducers. The experimental results demonstrated that K-means algorithm using RHadoop can scale well and efficiently process large data sets on commodity hardware. We also showed that our K-means algorithm using RHadoop with combiner was faster than regular algorithm without combiner as the size of data set increases.

Keywords: big data, combiner, K-means clustering, RHadoop

Procedia PDF Downloads 427

Authors: Silmy A. Humaira'

Abstract:

There is an issue about Asian students including Indonesian students that tend to behave passively in the classroom and depend on the teachers’ instruction. Regarding this statement, this study attempts to address the Indonesian high school students’ attitudes on whether they have initiative and be responsible for their learning out of the classroom and if so, why. Therefore, 30 high school students were asked to fill out the questionnaires and interviewed in order to figure out their attitudes towards self-directed learning. The descriptive qualitative research analysis adapted Knowles’s theory (1975) about Self-directed learning (SDL) to analyze the data. The findings show that the students have a potential to possess self-directed learning through ICT, but they have difficulties in choosing appropriate learning strategy, doing self-assessment and conducting self-reflection. Therefore, this study supports the teacher to promote self-directed learning instruction for successful learning by assisting students in dealing with those aforementioned problems. Furthermore, it is expected to be a beneficial reference which gives new insights on the self-directed learning practice in specific context.

Keywords: ICT, learning autonomy, students’ attitudes, self-directed learning

Procedia PDF Downloads 224

Authors: Ibraheem Alzahrani

Abstract:

This paper seeks to discover the relationship between both the social constructivist learning theory and the collaborative learning environment. This relationship can be identified through given an example of the learning environment. Due to wiki characteristics, wiki can be used to understand the relationship between constructivist learning theory and collaborative learning environment. However, several evidences will come in this paper to support the idea of why wiki is the suitable method to explore the relationship between social constructivist theory and the collaborative learning and their role in learning. Moreover, learning activities in wiki classroom will be discussed in this paper to find out the result of the learners' interaction in the classroom groups, which will be through two types of communication; synchronous and asynchronous.

Keywords: social constructivist, collaborative, environment, wiki, activities

Procedia PDF Downloads 498

Authors: Anitha Narasimhan, Aarthy Anandan, Madhan Karky, C. N. Subalalitha

Abstract:

Games can be the excellent tools for teaching a language. There are few e-learning games in Indian languages like word scrabble, cross word, quiz games etc., which were developed mainly for educational purposes. This paper proposes a Tamil word game called, “Porul”, which focuses on education as well as on players’ thinking and decision-making skills. Porul is a multiple choice based quiz game, in which the players attempt to answer questions correctly from the given multiple options that are generated using a unique algorithm called the Option Selection algorithm which explores the semantics of the question in various dimensions namely, synonym, rhyme and Universal Networking Language semantic category. This kind of semantic exploration of the question not only increases the complexity of the game but also makes it more interesting. The paper also proposes a Scoring Algorithm which allots a score based on the popularity score of the question word. The proposed game has been tested using 20,000 Tamil words.

Keywords: Porul game, Tamil word game, option selection, flash card, scoring, algorithm

Procedia PDF Downloads 397

Authors: Beatriz Lafuente Alcázar, Yash Wani, Amit J. Nimunkar

Abstract:

Cardiovascular Diseases (CVDs) are the leading cause of death globally, and around 80% of sudden cardiac deaths are due to arrhythmias or irregular heartbeats. The majority of these pathologies are revealed by either short-term or long-term alterations in the electrocardiogram (ECG) morphology. The ECG is the main diagnostic tool in cardiology. It is a non-invasive, pain free procedure that measures the heart’s electrical activity and that allows the detecting of abnormal rhythms and underlying conditions. A cardiologist can diagnose a wide range of pathologies based on ECG’s form alterations, but the human interpretation is subjective and it is contingent to error. Moreover, ECG records can be quite prolonged in time, which can further complicate visual diagnosis, and deeply retard disease detection. In this context, deep learning methods have risen as a promising strategy to extract relevant features and eliminate individual subjectivity in ECG analysis. They facilitate the computation of large sets of data and can provide early and precise diagnoses. Therefore, the cardiology field is one of the areas that can most benefit from the implementation of deep learning algorithms. In the present study, a deep learning algorithm is trained following a novel approach, using a combination of different databases as the training set. The goal of the algorithm is to achieve the detection of R-peaks in ECG signals. Its performance is further evaluated in ECG signals with different origins and features to test the model’s ability to generalize its outcomes. Performance of the model for detection of R-peaks for clean and noisy ECGs is presented. The model is able to detect R-peaks in the presence of various types of noise, and when presented with data, it has not been trained. It is expected that this approach will increase the effectiveness and capacity of cardiologists to detect divergences in the normal cardiac activity of their patients.

Keywords: arrhythmia, deep learning, electrocardiogram, machine learning, R-peaks

Procedia PDF Downloads 175

Authors: Aadiv Shah, Hari Nair, Vedant Mittal, Alice Cheeran

Abstract:

Present-day agricultural practices such as blanket spraying not only lead to excessive usage of pesticides but also harm the overall crop yield. This paper introduces an algorithm to optimize the traversal of an unmanned aerial vehicle (UAV) in constrained environments. The proposed system focuses on the agricultural application of targeted spraying for locust elimination. Given a satellite image of a farm, target zones that are prone to locust swarm formation are detected through the calculation of the normalized difference vegetation index (NDVI). This is followed by determining the optimal path for traversal of a UAV through these target zones using the proposed algorithm in order to perform pesticide spraying in the most efficient manner possible. Unlike the classic travelling salesman problem involving point-to-point optimization, the proposed algorithm determines an optimal path for multiple regions, independent of its geometry. Finally, the paper explores the idea of implementing reinforcement learning to model complex environmental behaviour and make the path planning mechanism for UAVs agnostic to external environment changes. This system not only presents a solution to the enormous losses incurred due to locust attacks but also an efficient way to automate agricultural practices across the globe in order to improve farmer ergonomics.

Keywords: locust, NDVI, optimization, path planning, reinforcement learning, UAV

Procedia PDF Downloads 245

Authors: J. Kaabi, Y. Harrath

Abstract:

This paper studied the flow shop scheduling problem under machine availability constraints. The machines are subject to flexible preventive maintenance activities. The nonresumable scenario for the jobs was considered. That is, when a job is interrupted by an unavailability period of a machine it should be restarted from the beginning. The objective is to minimize the total tardiness time for the jobs and the advance/tardiness for the maintenance activities. To solve the problem, a genetic algorithm was developed and successfully tested and validated on many problem instances. The computational results showed that the new genetic algorithm outperforms another earlier proposed algorithm.

Keywords: flow shop scheduling, genetic algorithm, maintenance, priority rules

Procedia PDF Downloads 468

Authors: Haya Salah, Srinivas Sharan

Abstract:

Healthcare facilities use appointment systems to schedule their appointments and to manage access to their medical services. With the growing demand for outpatient care, it is now imperative to manage physician's time effectively. However, high variation in consultation duration affects the clinical scheduler's ability to estimate the appointment duration and allocate provider time appropriately. Underestimating consultation times can lead to physician's burnout, misdiagnosis, and patient dissatisfaction. On the other hand, appointment durations that are longer than required lead to doctor idle time and fewer patient visits. Therefore, a good estimation of consultation duration has the potential to improve timely access to care, resource utilization, quality of care, and patient satisfaction. Although the literature on factors influencing consultation length abound, little work has done to predict it using based data-driven approaches. Therefore, this study aims to predict consultation duration using supervised machine learning algorithms (ML), which predicts an outcome variable (e.g., consultation) based on potential features that influence the outcome. In particular, ML algorithms learn from a historical dataset without explicitly being programmed and uncover the relationship between the features and outcome variable. A subset of the data used in this study has been obtained from the electronic medical records (EMR) of four different outpatient clinics located in central Pennsylvania, USA. Also, publicly available information on doctor's characteristics such as gender and experience has been extracted from online sources. This research develops three popular ML algorithms (deep learning, random forest, gradient boosting machine) to predict the treatment time required for a patient and conducts a comparative analysis of these algorithms with respect to predictive performance. The findings of this study indicate that ML algorithms have the potential to predict the provider service time with superior accuracy. While the current approach of experience-based appointment duration estimation adopted by the clinic resulted in a mean absolute percentage error of 25.8%, the Deep learning algorithm developed in this study yielded the best performance with a MAPE of 12.24%, followed by gradient boosting machine (13.26%) and random forests (14.71%). Besides, this research also identified the critical variables affecting consultation duration to be patient type (new vs. established), doctor's experience, zip code, appointment day, and doctor's specialty. Moreover, several practical insights are obtained based on the comparative analysis of the ML algorithms. The machine learning approach presented in this study can serve as a decision support tool and could be integrated into the appointment system for effectively managing patient scheduling.

Keywords: clinical decision support system, machine learning algorithms, patient scheduling, prediction models, provider service time

Procedia PDF Downloads 119

Authors: Sai Sankalp Vemavarapu

Abstract:

This paper discusses the application of machine learning in the field of transportation engineering for predicting engineering properties of pavement more accurately and efficiently. Predicting the elastic properties aid us in assessing the current road conditions and taking appropriate measures to avoid any inconvenience to commuters. This improves the longevity and sustainability of the pavement layer while reducing its overall life-cycle cost. As an example, we have implemented differential evolution (DE) in the back-calculation of the elastic modulus of multi-layered pavement. The proposed DE global optimization back-calculation approach is integrated with a forward response model. This approach treats back-calculation as a global optimization problem where the cost function to be minimized is defined as the root mean square error in measured and computed deflections. The optimal solution which is elastic modulus, in this case, is searched for in the solution space by the DE algorithm. The best DE parameter combinations and the most optimum value is predicted so that the results are reproducible whenever the need arises. The algorithm’s performance in varied scenarios was analyzed by changing the input parameters. The prediction was well within the permissible error, establishing the supremacy of DE.

Keywords: cost function, differential evolution, falling weight deflectometer, genetic algorithm, global optimization, metaheuristic algorithm, multilayered pavement, pavement condition assessment, pavement layer moduli back calculation

Procedia PDF Downloads 159

Authors: Omar Dib, Alexandre Caminada, Marie-Ange Manier

Abstract:

The purpose of this study is to introduce a novel approach to solve the one-to-one shortest path problem. A directed connected graph is assumed in which all edges’ weights are positive. Our method is based on a memetic algorithm in which we combine a genetic algorithm (GA) and a variable neighborhood search method (VNS). We compare our approximate method with two exact algorithms Dijkstra and Integer Programming (IP). We made experimentations using random generated, complete and real graph instances. In most case studies, numerical results show that our method outperforms exact methods with 5% average gap to the optimality. Our algorithm’s average speed is 20-times faster than Dijkstra and more than 1000-times compared to IP. The details of the experimental results are also discussed and presented in the paper.

Keywords: shortest path problem, Dijkstra’s algorithm, integer programming, memetic algorithm

Procedia PDF Downloads 462

Authors: Bruce X. B. Yu, Yan Liu, Keith C. C. Chan

Abstract:

We propose to record Activities of Daily Living (ADLs) of elderly people using a vision-based system so as to provide better assistive and personalization technologies. Current ADL-related research is based on data collected with help from non-elderly subjects in laboratory environments and the activities performed are predetermined for the sole purpose of data collection. To obtain more realistic datasets for the application, we recorded ADLs for the elderly with data collected from real-world environment involving real elderly subjects. Motivated by the need to collect data for more effective research related to elderly care, we chose to collect data in the room of an elderly person. Specifically, we installed Kinect, a vision-based sensor on the ceiling, to capture the activities that the elderly subject performs in the morning every day. Based on the data, we identified 12 morning activities that the elderly person performs daily. To recognize these activities, we created a HARELCARE framework to investigate into the effectiveness of existing Human Activity Recognition (HAR) algorithms and propose the use of a transfer learning algorithm for HAR. We compared the performance, in terms of accuracy, and training progress. Although the collected dataset is relatively small, the proposed algorithm has a good potential to be applied to all daily routine activities for healthcare purposes such as evidence-based diagnosis and treatment.

Keywords: daily activity recognition, healthcare, IoT sensors, transfer learning

Procedia PDF Downloads 126

Authors: Farouk Lawan Gambo

Abstract:

Data compression shrinks files into fewer bits then their original presentation. It has more advantage on internet because the smaller a file, the faster it can be transferred but learning most of the concepts in data compression are abstract in nature therefore making them difficult to digest by some students (Engineers in particular). To determine the best approach toward learning data compression technique, this paper first study the learning preference of engineering students who tend to have strong active, sensing, visual and sequential learning preferences, the paper also study the advantage that mobility of learning have experienced; Learning at the point of interest, efficiency, connection, and many more. A survey is carried out with some reasonable number of students, through random sampling to see whether considering the learning preference and advantages in mobility of learning will give a promising improvement over the traditional way of learning. Evidence from data analysis using Ms-Excel as a point of concern for error-free findings shows that there is significance different in the students after using learning content provided on smart phone, also the result of the findings presented in, bar charts and pie charts interpret that mobile learning has to be promising feature of learning.

Keywords: data analysis, compression techniques, learning content, traditional learning approach

Procedia PDF Downloads 343

Authors: Arman S. Kussainov, Altynbek K. Beisekov

Abstract:

This work focuses on building an efficient universal procedure to construct a single density of states from the multiple pieces of data provided by the parallel implementation of the Wang Landau Monte Carlo based algorithm. The Ising and Pott models were used as the examples of the two-dimensional spin lattices to construct their densities of states. Sampled energy space was distributed between the individual walkers with certain overlaps. This was made to include the latest development of the algorithm as the density of states replica exchange technique. Several factors of immediate importance for the seamless stitching process have being considered. These include but not limited to the speed and universality of the initial parallel algorithm implementation as well as the data post-processing to produce the expected smooth density of states.

Keywords: density of states, Monte Carlo, parallel algorithm, Wang Landau algorithm

Procedia PDF Downloads 405

Authors: Tayfun Çay, Yasar İnceyol, Abdurrahman Özbeyaz

Abstract:

Land reallocation is one of the most important steps in land consolidation projects. Many different models were proposed for land reallocation in the literature such as Fuzzy Logic, block priority based land reallocation and Spatial Decision Support Systems. A model including four parts is considered for automatic block reallocation with genetic algorithm method in land consolidation projects. These stages are preparing data tables for a project land, determining conditions and constraints of land reallocation, designing command steps and logical flow chart of reallocation algorithm and finally writing program codes of Genetic Algorithm respectively. In this study, we designed the first three steps of the considered model comprising four steps.

Keywords: land consolidation, landholding, land reallocation, optimization, genetic algorithm

Procedia PDF Downloads 424

Authors: Dimitri Golenko-Ginzburg

Abstract:

A CPM network project with deterministic activity durations, in which activities require homogenous resources with fixed capacities, is considered. The problem is to determine the optimal schedule of starting times for all network activities within their maximal allowable limits (in order not to exceed the network's critical time) to minimize the maximum required resources for the project at any point in time. In case when a non-critical activity may start only at discrete moments with the pregiven time span, the problem becomes NP-complete and an optimal solution may be obtained via a look-over algorithm. For the case when a look-over requires much computational time an approximate algorithm is suggested. The algorithm's performance ratio, i.e., the relative accuracy error, is determined. Experimentation has been undertaken to verify the suggested algorithm.

Keywords: resource smoothing problem, CPM network, lookover algorithm, lexicographical order, approximate algorithm, accuracy estimate

Procedia PDF Downloads 298

Authors: Jozef Jurcik, Miroslav Gutten, Milan Sebok, Daniel Korenciak, Jerzy Roj

Abstract:

The paper presents the research results of electronic fuel injection system, which can be used for diagnostics of automotive systems. In the paper is described the construction and operation of a typical fuel injection system and analyzed its electronic part. It has also been proposed method for the detection of the injector malfunction, based on the analysis of differential current or voltage characteristics. In order to detect the fault state, it is needed to use self-learning process, by the use of an appropriate self-learning algorithm.

Keywords: electronic fuel injector, diagnostics, measurement, testing device

Procedia PDF Downloads 547

Authors: Xabier Olaz Moratinos

Abstract:

Drone navigation is critical, particularly during the initial phases, such as the initial ascension, where pilots may fail due to strong external interferences that could potentially lead to a crash. In this ongoing work, a drone has been successfully trained to perform an ascent of up to 6 meters at speeds with external disturbances pushing it up to 24 mph, with the DQN algorithm managing external forces affecting the system. It has been demonstrated that the system can control its height, position, and stability in all three axes (roll, pitch, and yaw) throughout the process. The learning process is carried out in the Gazebo simulator, which emulates interferences, while ROS is used to communicate with the agent.

Keywords: machine learning, DQN, gazebo, navigation

Procedia PDF Downloads 101

Authors: Muhamamd Shahid Farooq, Maazan Zafar, Rizawana Akhtar

Abstract:

E-learning technologies are the blessings of advancements in science and technology. These facilitate the learners to get information at any place and any time by improving their self-confidence, self-efficacy and effectiveness in teaching learning process. E-learning provides an individualized learning experience for learners and remove barriers faced by students during new and creative ways of gaining information. It provides a wide range of facilities to enable the teachers and students for effective and purposeful learning. This study was conducted to explore the attitudes of university students and teachers towards e-learning working in a metropolitan university of Pakistan. The personal, institutional and technological characteristics of the teachers and students of higher education institution effect the adoption of e-learning. For this descriptive study 449 students and 35 university teachers were surveyed by using a Likert scale type questionnaire consisting of 52 statements relating to six factors "perceived usefulness, intention to adopt e-learning, ease of e-learning use, availability resources, e-learning stressors, and pressure to use e-learning". Data were analyzed by making comparisons on the basis of different demographic factors. The findings of the study show that both type of respondents have positive attitude towards e-learning. However, the male and female respondents differ in their opinion for e-learning implementation.

Keywords: e-learning, ICT, e-sources of learning, questionnaire

Procedia PDF Downloads 524

Authors: Ashish James, Anusha James

Abstract:

The impressive predictive performance of deep learning techniques on a wide range of tasks has led to its widespread use. Estimating the confidence of these predictions is paramount for improving the safety and reliability of such systems. However, the uncertainty estimates provided by neural networks (NNs) tend to be overconfident and unreasonable. Ensemble of NNs typically produce good predictions but uncertainty estimates tend to be inconsistent. Inspired by these, this paper presents a framework that can quantitatively estimate the uncertainties by leveraging the advances in transfer learning through slight modification to the existing training pipelines. This promising algorithm is developed with an intention of deployment in real world problems which already boast a good predictive performance by reusing those pretrained models. The idea is to capture the behavior of the trained NNs for the base task by augmenting it with the uncertainty estimates from a supplementary network. A series of experiments with known and unknown distributions show that the proposed approach produces well calibrated uncertainty estimates with high quality predictions.

Keywords: uncertainty estimation, neural networks, transfer learning, regression

Procedia PDF Downloads 126

Authors: Abdelaziz Moujane, Abedelali Boulli, Abdellah Ouigmane

Abstract:

The present work focuses on the assessment of forestlandscape changes in the region of ZaouitAhansal, usingmultitemporal satellite images at high spatial resolution.Severalremotesensingmethodswereappliednamely: The supervised classification algorithm and NDVI whichwerecombined in a GIS environment to quantify the extent and change in density of forest stands (holmoak, juniper, thya, Aleppo pine, crops, and others).The resultsobtainedshowedthat the forest of ZaouitAhansal has undergonesignificantdegradationresulting in a decrease in the area of juniper, cedar, and zeenoak, as well as an increase in the area of baresoil and agricultural land. The remotesensing data providedsatisfactoryresults for identifying and quantifying changes in forestcover. In addition, thisstudycould serve as a reference for the development of management strategies and restoration programs.

Keywords: remote sensing, GIS, satellite image, NDVI, deforestation, zaouit ahansal

Procedia PDF Downloads 148

Authors: Aishik Deb, Rituparna Sinha

Abstract:

We have developed an algorithm to detect the abnormal segment/"structural variation in the genome across a number of samples. We have worked on simulated as well as real data from the BAM Files and have designed a segmentation algorithm where abnormal segments are detected. This algorithm aims to improve the accuracy and performance of the existing CNV-CH algorithm. The next-generation sequencing (NGS) approach is very fast and can generate large sequences in a reasonable time. So the huge volume of sequence information gives rise to the need for Big Data and parallel approaches of segmentation. Therefore, we have designed a map-reduce approach for the existing CNV-CH algorithm where a large amount of sequence data can be segmented and structural variations in the human genome can be detected. We have compared the efficiency of the traditional and map-reduce algorithms with respect to precision, sensitivity, and F-Score. The advantages of using our algorithm are that it is fast and has better accuracy. This algorithm can be applied to detect structural variations within a genome, which in turn can be used to detect various genetic disorders such as cancer, etc. The defects may be caused by new mutations or changes to the DNA and generally result in abnormally high or low base coverage and quantification values.

Keywords: cancer detection, convex hull segmentation, map reduce, next generation sequencing

Procedia PDF Downloads 128

Authors: Güzi̇n Ti̇rkeş, Özge Teki̇n, Kerem Kurtuluş, Y. Yekta Yurtseven, Murat Baran

Abstract:

A scanning electron microscope (SEM) is a type of electron microscope mainly used in nanoscience and nanotechnology areas. Automatic image recognition and classification are among the general areas of application concerning SEM. In line with these usages, the present paper proposes a deep learning algorithm that classifies SEM images into nine categories by means of an online application to simplify the process. The NFFA-EUROPE - 100% SEM data set, containing approximately 21,000 images, was used to train and test the algorithm at 80% and 20%, respectively. Validation was carried out using a separate data set obtained from the Middle East Technical University (METU) in Turkey. To increase the accuracy in the results, the Inception ResNet-V2 model was used in view of the Fine-Tuning approach. By using a confusion matrix, it was observed that the coated-surface category has a negative effect on the accuracy of the results since it contains other categories in the data set, thereby confusing the model when detecting category-specific patterns. For this reason, the coated-surface category was removed from the train data set, hence increasing accuracy by up to 96.5%.

Keywords: convolutional neural networks, deep learning, image classification, scanning electron microscope

Procedia PDF Downloads 118

Authors: Shuge Lei, Haonan Hu, Dasheng Sun, Huabin Zhang, Kehong Yuan, Jian Dai, Yan Tong

Abstract:

It remains challenging to measure reliability from classification results from different machine learning models. This paper proposes a reliable soup optimization algorithm based on the model weight fusion algorithm Model Soup, aiming to improve reliability by using dual-channel reliability as the objective function to fuse a series of weights in the breast ultrasound classification models. Experimental results on breast ultrasound clinical datasets demonstrate that reliable soup significantly enhances the reliability of breast ultrasound image classification tasks. The effectiveness of the proposed approach was verified via multicenter trials. The results from five centers indicate that the reliability optimization algorithm can enhance the reliability of the breast ultrasound image classification model and exhibit low multicenter correlation.

Keywords: breast ultrasound image classification, feature attribution, reliability assessment, reliability optimization

Procedia PDF Downloads 78

Authors: Y. Alrubyli

Abstract:

Unmanned Arial Vehicles (UAVs), and more specifically, quadcopters need to be stable during their flights. Altitude stability is usually achieved by using a PID controller that is built into the flight controller software. Furthermore, the PID controller has gains that need to be tuned to reach optimal altitude stabilization during the quadcopter’s flight. For that, control system engineers need to tune those gains by using extensive modeling of the environment, which might change from one environment and condition to another. As quadcopters penetrate more sectors, from the military to the consumer sectors, they have been put into complex and challenging environments more than ever before. Hence, intelligent self-stabilizing quadcopters are needed to maneuver through those complex environments and situations. Here we show that by using online reinforcement learning with minimal background knowledge, the altitude stability of the quadcopter can be achieved using a model-free approach. We found that by using background knowledge instead of letting the online reinforcement learning algorithm wander for a while to tune the PID gains, altitude stabilization can be achieved faster. In addition, using this approach will accelerate development by avoiding extensive simulations before applying the PID gains to the real-world quadcopter. Our results demonstrate the possibility of using the trial and error approach of reinforcement learning combined with background knowledge to achieve faster quadcopter altitude stabilization in different environments and conditions.

Keywords: reinforcement learning, Q-leanring, online learning, PID tuning, unmanned aerial vehicle, quadcopter

Procedia PDF Downloads 166

Authors: Sulaiman Almutairy, Trevor Davies, Yota Dimitriadi

Abstract:

This paper presents study investigating how to understand better the psychological readiness for mobile learning (m-learning) among Saudi students, while also evaluating m-learning in Saudi Arabia-a topic that has not yet received adequate attention from researchers. Data was acquired through a questionnaire administered to 131 Saudi students at UK universities, in July 2013. The study confirmed that students are confident using mobile devices in their daily lives and that they would welcome more opportunities for mobile learning. The findings indicated that Saudi higher education students are highly familiar with, and are psychologically ready for, m-learning.

Keywords: m-learning, mobile technologies, psychological readiness, higher education

Procedia PDF Downloads 512

Authors: Chryssi Vitsilaki, Apostolos Kostas, Ilias Efthymiou

Abstract:

This paper presents selected best practices on online learning and teaching derived from a novel and innovating Lifelong Learning program through e-Learning, which has during the last five years been set up at the University of the Aegean in Greece. The university, capitalizing on an award-winning, decade-long experience in e-learning and blended learning in undergraduate and postgraduate studies, recently expanded into continuous education and vocational training programs in various cutting-edge fields. So, in this article we present: (a) the academic structure/infrastructure which has been developed for the administrative, organizational and educational support of the e-Learning process, including training the trainers, (b) the mode of design and implementation based on a sound pedagogical framework of open and distance education, and (c) the key results of the assessment of the e-learning process by the participants, as they are used to feedback on continuous organizational and teaching improvement and quality control.

Keywords: distance education, e-learning, life-long programs, synchronous/asynchronous learning

Procedia PDF Downloads 327

Authors: Xabier Olaz Moratinos

Abstract:

Drone navigation is critical, particularly during the initial phases, such as the initial ascension, where pilots may fail due to strong external interferences that could potentially lead to a crash. In this ongoing work, a drone has been successfully trained to perform an ascent of up to 6 meters at speeds with external disturbances pushing it up to 24 mph, with the DQN algorithm managing external forces affecting the system. It has been demonstrated that the system can control its height, position, and stability in all three axes (roll, pitch, and yaw) throughout the process. The learning process is carried out in the Gazebo simulator, which emulates interferences, while ROS is used to communicate with the agent.

Keywords: machine learning, DQN, Gazebo, navigation

Procedia PDF Downloads 71