Adaptive Visual Learning Using Augmented Reality and Machine Learning Techniques

2020 ◽  
Vol 17 (11) ◽  
pp. 4952-4956
Author(s):  
Satyajit Ghana ◽  
Shikhar Singh ◽  
Aryan Jalali ◽  
Vivek Badani ◽  
Sahana P. Shankar
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Logistic Regression ◽  
Augmented Reality ◽  
Student Performance ◽  
Optical Character Recognition ◽  
Conceptual Knowledge ◽  
Learning Experience ◽  
Visual Learning ◽  
Machine Learning Techniques

The current curriculum forces students to understand topics by visualizing axonometric structures in their cognitive minds depending upon the conceptual texts and information. This methodology is inconsistent as the idea of visualization through conceptual knowledge is dependent on the level of reasoning and IQ (Intelligence Quotient) a student possesses. It is usually common for a student to misinterpret an information due to lack of reasoning and imaginative skills. Our educational model aims to diminish this intellectual barrier by incorporating Augmented Reality (AR) and Machine Learning (ML) techniques together and create an Adaptive Visual Learning experience for students. A mobile interface with OCR (Optical Character Recognition) and TTS (Text-To-Speech) feature is given to make this whole process simple and easy to use for any student. In this paper, two ML techniques Logistic Regression and Neural Network are applied in order to enhance and modify the existing educational system by removing the intellectual barrier involved due to neurodiversity. A comparative study is performed between the two ML algorithms, where in Logistic Regression performed better than the Neural Network. This form of adaptive visual learning aims to boost student performance in academia.

scholarly journals Classification Comparative Analysis for Detection of Brain Tumor Using Neural Network, Logistic Regression & KNN Classifier with VGG19 Convolution Neural Network Feature Extraction

2021 ◽  
Author(s):  
Vijaya Kamble ◽  
Rohin Daruwala
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Feature Extraction ◽  
Logistic Regression ◽  
Brain Tumor ◽  
Medical Image Analysis ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Data Set ◽  
Knn Classifier

In recent years due to advancements in digital imaging machine learning techniques are used in medical image analysis for the prognosis and diagnosis of various abnormalities in the human body. Various Machine learning algorithms, convolution and deep neural networks are used for classification, detection and prediction of various brain tumors. The proposed approach is a different comparative classification analysis approach which is based on three different classification namely KNN classifier,Logistic regression & neural network as classifier. It is based on a deep learning feature extraction technique using VGG19. This VGG 19-layer image recognition model trained on Imgenet. Generally, MRI data sequences are analyzed in terms of different modalities and every modality contains rich tissue information. So, feature exaction from MRI sequences is very important task for brain tumor classification. Our approach demonstrated fair classification on BRATS Benchmarks 2018 data set with different modalities and sizes of images,results are without any human annotations. Based on selected classifiers all the classifiers gives accuracy above 90%. It is good compared to other state of art methods.

Abstract 27: Machine Learning Models for Outcome Prediction of Out-Of-Hospital Cardiac Arrest of Presumed Cardiac Cause Using the All-Japan Utstein Registry

Circulation ◽  
2019 ◽  
Vol 140 (Suppl_2) ◽  
Author(s):  
Tomohisa Seki ◽  
Tomoyoshi Tamura ◽  
Kazuhiko Ohe ◽  
Masaru Suzuki
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Logistic Regression ◽  
Cardiac Arrest ◽  
Deep Neural Network ◽  
Outcome Prediction ◽  
Machine Learning Techniques ◽  
Test Set ◽  
Learning Techniques ◽  
Hospital Cardiac Arrest

Background: Outcome prediction for patients with out-of-hospital cardiac arrest (OHCA) using prehospital information has been one of the major challenges in resuscitation medicine. Recently, machine learning techniques have been shown to be highly effective in predicting outcomes using clinical registries. In this study, we aimed to establish a prediction model for outcomes of OHCA of presumed cardiac cause using machine learning techniques. Methods: We analyzed data from the All-Japan Utstein Registry of the Fire and Disaster Management Agency between 2005 and 2016. Of 1,423,338 cases, data of OHCA patients aged ≥18 years with presumed cardiac etiology were retrieved and divided into two groups: training set, n = 584,748 (between 2005 and 2013) and test set, n = 223,314 (between 2014 and 2016). The endpoints were neurologic outcome at 1-month and survival at 1-month. Of 47 variables evaluated during the prehospital course, 19 variables (e.g.,sex, age, ECG waveform, and practice of bystander CPR) were used for outcome prediction. Performances of logistic regression, random forests, and deep neural network were examined in this study. Results: For prediction of neurologic outcomes (cerebral performance category 1 or 2) using the test set, the generated models showed area under the receiver operating characteristic curve (AUROC) values of 0.942 (95% confidence interval [CI] 0.941-0.943), 0.947 (95% CI 0.946-0.948), and 0.948 (95% CI 0.948-0.950) in logistic regression, random forest, and deep neural network, respectively. For survival prediction, the generated models showed AUROC values of 0.901 (95% CI 0.900-0.902), 0.913 (95% CI 0.912-0.914), and 0.912 (95% CI 0.911-0.913) in logistic regression, random forest, and deep neural network, respectively. Conclusions: Machine learning techniques using prehospital variables showed favorable prediction capability for 1-month neurologic outcome and survival in OHCA of presumed cardiac cause.

scholarly journals Profitable Algorithmic Trading Strategy

2021 ◽  
Vol 9 (12) ◽  
pp. 2424-2433
Author(s):  
Prince Nathan S
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Logistic Regression ◽  
Data Analysis ◽  
Data Science ◽  
Principal Component ◽  
Machine Learning Techniques ◽  
Learning Models ◽  
Data Set ◽  
Machine Learning Models

Abstract: Cryptocurrency has drastically increased its growth in recent years and Bitcoin (BTC) is a very popular type of currency among all the other types of cryptocurrencies which is been used in most of the sectors nowadays for trading, transactions, bookings, etc. In this paper, we aim to predict the change in bitcoin prices by using machine learning techniques on data from Investing.com. We interpret the output and accuracy rate using various machine learning models. To see whether to buy or sell the bitcoin we created exploratory data analysis from a year of data set and predict the next 5 days change using machine learning models like logistic Regression, Logistic Regression with PCA (Principal Component Analysis), and Neural network. Keywords: Data Science, Machine Learning, Regression, PCA, Neural Network, Data Analysis

scholarly journals Detection and Severity Evaluation of Combined Rail Defects Using Deep Learning

Vibration ◽  
2021 ◽  
Vol 4 (2) ◽  
pp. 341-356
Author(s):  
Jessada Sresakoolchai ◽  
Sakdirat Kaewunruen
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Deep Learning ◽  
Mean Absolute Error ◽  
Absolute Error ◽  
Machine Learning Techniques ◽  
Rolling Stock ◽  
Raw Data ◽  
Learning Techniques ◽  
Combined Defects

Various techniques have been developed to detect railway defects. One of the popular techniques is machine learning. This unprecedented study applies deep learning, which is a branch of machine learning techniques, to detect and evaluate the severity of rail combined defects. The combined defects in the study are settlement and dipped joint. Features used to detect and evaluate the severity of combined defects are axle box accelerations simulated using a verified rolling stock dynamic behavior simulation called D-Track. A total of 1650 simulations are run to generate numerical data. Deep learning techniques used in the study are deep neural network (DNN), convolutional neural network (CNN), and recurrent neural network (RNN). Simulated data are used in two ways: simplified data and raw data. Simplified data are used to develop the DNN model, while raw data are used to develop the CNN and RNN model. For simplified data, features are extracted from raw data, which are the weight of rolling stock, the speed of rolling stock, and three peak and bottom accelerations from two wheels of rolling stock. In total, there are 14 features used as simplified data for developing the DNN model. For raw data, time-domain accelerations are used directly to develop the CNN and RNN models without processing and data extraction. Hyperparameter tuning is performed to ensure that the performance of each model is optimized. Grid search is used for performing hyperparameter tuning. To detect the combined defects, the study proposes two approaches. The first approach uses one model to detect settlement and dipped joint, and the second approach uses two models to detect settlement and dipped joint separately. The results show that the CNN models of both approaches provide the same accuracy of 99%, so one model is good enough to detect settlement and dipped joint. To evaluate the severity of the combined defects, the study applies classification and regression concepts. Classification is used to evaluate the severity by categorizing defects into light, medium, and severe classes, and regression is used to estimate the size of defects. From the study, the CNN model is suitable for evaluating dipped joint severity with an accuracy of 84% and mean absolute error (MAE) of 1.25 mm, and the RNN model is suitable for evaluating settlement severity with an accuracy of 99% and mean absolute error (MAE) of 1.58 mm.

scholarly journals Burst Pressure Prediction of API 5L X-Grade Dented Pipelines Using Deep Neural Network

2020 ◽  
Vol 8 (10) ◽  
pp. 766
Author(s):  
Dohan Oh ◽  
Julia Race ◽  
Selda Oterkus ◽  
Bonguk Koo
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Artificial Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Deep Neural Network ◽  
Machine Learning Techniques ◽  
Burst Pressure ◽  
Comparison Results ◽  
Artificial Neural

Mechanical damage is recognized as a problem that reduces the performance of oil and gas pipelines and has been the subject of continuous research. The artificial neural network in the spotlight recently is expected to be another solution to solve the problems relating to the pipelines. The deep neural network, which is on the basis of artificial neural network algorithm and is a method amongst various machine learning methods, is applied in this study. The applicability of machine learning techniques such as deep neural network for the prediction of burst pressure has been investigated for dented API 5L X-grade pipelines. To this end, supervised learning is employed, and the deep neural network model has four layers with three hidden layers, and the neural network uses the fully connected layer. The burst pressure computed by deep neural network model has been compared with the results of finite element analysis based parametric study, and the burst pressure calculated by the experimental results. According to the comparison results, it showed good agreement. Therefore, it is concluded that deep neural networks can be another solution for predicting the burst pressure of API 5L X-grade dented pipelines.

scholarly journals A Mining Frame Work of CO-PO Attainment using Deep Learning Techniques

2020 ◽  
Vol 9 (4S2) ◽  
pp. 60-63
Keyword(s):  
Machine Learning ◽  
Student Performance ◽  
Answer Sheet ◽  
Input Image ◽  
Machine Learning Techniques ◽  
Optimal Model ◽  
Program Outcome ◽  
Learning Techniques ◽  
Frame Work ◽  
Machine Learning Models

student performance measured in CO-PO (Course Outcome and Program Outcome) attainment for OMR based answer sheet automation playing very curtail role in pupil concert analysis in this approach. In the proposed work, marks evaluation sheet is consider as input image, then apply frame cropping technique to extract the marks filled table by subdividing into cells as individual images by frame cropping technique. In order to recognition of hand written digit in each frame, various machine learning models are adopted, trained. Experimental results from proposed work show that convolutional neural network excels higher in identification digits from frames. The outputs are then converted to CSV version, which is used to evaluate CO-PO attainment for each learner. The experiments have been conducted and tested in proposed work on various machine learning techniques and compared the results to pick the optimal model

scholarly journals Malicious URL Detection using Logistic Regression

2021 ◽  
Author(s):  
Rohit Rayala ◽  
Sashank Pasumarthi ◽  
Rohith Kuppa ◽  
S R KARTHIK
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Machine Learning Techniques ◽  
Learning Techniques

Paper is based on a model that is built to detect malicious URLs using machine learning techniques.

scholarly journals Predicting Student Failure in University Examination using Machine Learning Algorithms

2020 ◽  
Vol 9 (5) ◽  
pp. 956-959
Keyword(s):  
Machine Learning ◽  
Data Mining ◽  
Performance Management ◽  
Student Performance ◽  
Learning Algorithms ◽  
Educational Data Mining ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Social Characteristics ◽  
Student Failure

Student Performance Management is one of the key pillars of the higher education institutions since it directly impacts the student’s career prospects and college rankings. This paper follows the path of learning analytics and educational data mining by applying machine learning techniques in student data for identifying students who are at the more likely to fail in the university examinations and thus providing needed interventions for improved student performance. The Paper uses data mining approach with 10 fold cross validation to classify students based on predictors which are demographic and social characteristics of the students. This paper compares five popular machine learning algorithms Rep Tree, Jrip, Random Forest, Random Tree, Naive Bayes algorithms based on overall classifier accuracy as well as other class specific indicators i.e. precision, recall, f-measure. Results proved that Rep tree algorithm outperformed other machine learning algorithms in classifying students who are at more likely to fail in the examinations.

scholarly journals First Year Student Perception and Experience of Online Topographical Anatomy Laboratory Classes using Zoom Technology during the COVID-19 Pandemic

2021 ◽  
Vol 29 (3) ◽  
Author(s):  
Rudi Klein ◽  
Chiara Tomassoni ◽  
Gayathri Rajaaman ◽  
Maxwell Winchester ◽  
Norman Eizenberg ◽  
...  
Keyword(s):  
Student Performance ◽  
Learning Experience ◽  
Student Perception ◽  
Visual Learning ◽  
Functional Anatomy ◽  
Gross Anatomy ◽  
First Year ◽  
Topographical Anatomy ◽  
Anatomy Laboratory ◽  
Student Grades

During semester one of 2020, the units ‘Functional Anatomy of the Trunk’ and ‘Functional Anatomy of the Limbs’ which focus on human topographical anatomy were re-designed into an online delivery format and taught remotely in response to the COVID-19 lockdown. It was expected that the move to remote teaching would negatively impact student perception and learning experience, in particular that of the cadaver-based laboratory work. The aim of this study was to investigate whether the replacement of traditional face-to-face cadaver-based anatomy laboratories with an online version using digital anatomy resources and Zoom technology as the communication platform would achieve comparable student learning experience and outcomes. First Year Students (n=69) enrolled in these units were invited to participate in this study and were asked at the conclusion of each unit to complete an anonymous opinion-based survey via Qualtrics. The Qualtrics data, student grades and Learning Management System (LMS) statistics were analysed. Results indicate that student perception of the online gross anatomy laboratory learning was positive and that it had complemented their learning. Most students agreed that as a visual learning resource, it provided an improved understanding of anatomy and helped with the application of anatomical knowledge. Interestingly, student performance showed a similar range of marks compared with previous years. However, students strongly agreed that the online 2D learning experience had significant limitations when compared to live use of cadavers in laboratories.

Sign in / Sign up

Export Citation Format

Share Document