2692-4048ISSN (Online)
2770-0070ISSN (Print)
Internet of Things (IoT) based healthcare applications have grown exponentially over the past decade. With the increasing number of fatalities due to cardiovascular diseases (CVD), it is the need of the hour to detect any signs of cardiac abnormalities as early as possible. This calls for automation in the detection and classification of said cardiac abnormalities by physicians. The problem here is that there is not enough data to train Deep Learning models to classify ECG signals accurately because of the sensitive nature of data and the rarity of certain cases involved in CVDs. In this paper, we propose a framework that involves Generative Adversarial Networks (GAN) to create synthetic training data for the classes with fewer data points to improve the performance of Deep Learning models trained with the dataset. With data being input from sensors via the cloud and this model to classify the ECG signals, we expect the framework to be functional, accurate, and efficient.
Read MoreDoi: https://doi.org/10.54216/FPA.020201
Vol. 2 Issue. 2 PP. 42-49, (2020)
Whenever we visit a showroom to buy a watch, we are unable to try all color variants of watches due to unavailability, and also, if a watch is tried by too many customers, it loses its shine and finishing. We tried to overcome this problem with our Augmented reality app called “Arzara ."This Augmented reality app lets customers see how a particular watch model looks on their hand without physically trying the Watch. It will not only prevent the degradation of the polish of watches due to repeated wear by different customers at offline stores but also this new Augmented Reality watch app will take off online watch shopping. It will make it easier for customers to make decisions and purchase watches.
Read MoreDoi: https://doi.org/10.54216/FPA.020202
Vol. 2 Issue. 2 PP. 50-56, (2020)
Human facial emotion recognition has attracted interest in the field of Artificial Intelligence. The emotions on a human face depict what's going on inside the mind. Facial expression recognition is the part of Facial recognition which is gaining more importance, and the need for it has increased tremendously. Though there are methods to identify expressions using machine learning and Artificial Intelligence techniques, this work attempts to use convolution neural networks to recognize expressions and classify the expressions into 6 emotional categories. Various datasets are investigated and explored for training expression recognition models are explained in this paper, and the models which are used in this paper are VGG 19 and RESNET 18. We included facial emotional recognition with gender identification also. In this project, we have used the fer2013 and ck+ datasets and ultimately achieved 73% and 94% around accuracies, respectively.
Read MoreDoi: https://doi.org/10.54216/FPA.020203
Vol. 2 Issue. 2 PP. 57-63, (2020)
Performance capture of human beings has been used to animate 3D characters for movies and games for several decades now. Traditional performance capture methods require a costly dedicated setup which usually consists of more than one sensor placed at a distance from the subject, hence requiring a large amount of budget and space to accommodate. This lowers its feasibility and portability by a huge amount. Egocentric (first-person/wearable) cameras, however, are attached to the body and hence are mobile. With the rise of acceptance of wearable technology by the general public, wearable cameras have gotten cheaper too. We can make use of their excessive portability in the performance capture domain. However, working with egocentric images is a mammoth task as the views are severely distorted due to the first-person perspective, and the body parts farther from the camera are highly prone to be occluded. In this paper, we review the existing state-of-the-art methods of performance capture using egocentric-based views.
Read MoreDoi: https://doi.org/10.54216/FPA.020204
Vol. 2 Issue. 2 PP. 64-73, (2020)
The concept Sentiment means the feeling, behavior, belief, or attitude towards something that almost being embedded. sentiment analysis is the process of analyzing, extracting, studying, and classifying the various reviews, opinions are given by people, and human’s emictions into positive, negative, neutral. It is considered one of the most significant scientific branches that aim to determine the behavior of the speaker, the attitude of the writer according to some topic, or the overall emotional reaction to website, document, event, interaction, products, or services. many users can share every day various opinions on different topics that may be detected or embedded by using micro-blogging which considered a rich resource for sentiment analysis and belief mining such as Facebook, Twitter, forums, and Blogs. recently a huge number of posted comments, tweets, and reviews of different social media websites include rich information in addition to most of the on-line shopping sites provide the opportunity to customers to write reviews about products in order to enhance the sales of those products and to improve both of product quality and customer satisfaction. manual analysis of these large reviews is practically impossible thus it is needed to discover an automated approach to solving such a hard process. In the Middle East and particularly in the Arab world, social media websites continue to be the top-visited websites especially with the current social and political changes in this part of the world. the main objective of that research is to differentiate between various algorithms and techniques of sentiment analysis and classification dependent on the Arabic language as a little number of researchers discusses that point relevant to the Arabic language. Different algorithms and techniques of data mining such as Support Vector Machine (SVM), Naïve Bayes (NB), Bayesian Network (BN), Decision tree (DT), k-nearest neighbor (KNN), Maximum Entropy (ME), and Neural Network (NN) in addition to many other alternative techniques which are used for analyzing and classifying textual data. For the reasons of difficulties in analyzing and mining a large number of linguistic words for their Those techniques are estimated based on the Arabic language due to its richness and diversity. The comparison between data mining techniques showed that the most accurate technique is the support vector machine (SVM) algorithm. every successful sentiment depends on two essential analysis tools are language and culture.
Read MoreDoi: https://doi.org/10.54216/FPA.020205
Vol. 2 Issue. 2 PP. 74-87, (2020)