Search results for: physical-chemical features

Authors: Skyler Kim

Abstract:

An early diagnosis of leukemia has always been a challenge to doctors and hematologists. On a worldwide basis, it was reported that there were approximately 350,000 new cases in 2012, and diagnosing leukemia was time-consuming and inefficient because of an endemic shortage of flow cytometry equipment in current clinical practice. As the number of medical diagnosis tools increased and a large volume of high-quality data was produced, there was an urgent need for more advanced data analysis methods. One of these methods was the AI approach. This approach has become a major trend in recent years, and several research groups have been working on developing these diagnostic models. However, designing and implementing a leukemia diagnostic system in real clinical environments based on a deep learning approach with larger sets remains complex. Leukemia is a major hematological malignancy that results in mortality and morbidity throughout different ages. We decided to select acute lymphocytic leukemia to develop our diagnostic system since acute lymphocytic leukemia is the most common type of leukemia, accounting for 74% of all children diagnosed with leukemia. The results from this development work can be applied to all other types of leukemia. To develop our model, the Kaggle dataset was used, which consists of 15135 total images, 8491 of these are images of abnormal cells, and 5398 images are normal. In this paper, we design and implement a leukemia diagnostic system in a real clinical environment based on deep learning approaches with larger sets. The proposed diagnostic system has the function of detecting and classifying leukemia. Different from other AI approaches, we explore hybrid architectures to improve the current performance. First, we developed two independent convolutional neural network models: VGG19 and ResNet50. Then, using both VGG19 and ResNet50, we developed a hybrid deep learning architecture employing transfer learning techniques to extract features from each input image. In our approach, fusing the features from specific abstraction layers can be deemed as auxiliary features and lead to further improvement of the classification accuracy. In this approach, features extracted from the lower levels are combined into higher dimension feature maps to help improve the discriminative capability of intermediate features and also overcome the problem of network gradient vanishing or exploding. By comparing VGG19 and ResNet50 and the proposed hybrid model, we concluded that the hybrid model had a significant advantage in accuracy. The detailed results of each model’s performance and their pros and cons will be presented in the conference.

Keywords: acute lymphoblastic leukemia, hybrid model, leukemia diagnostic system, machine learning

Procedia PDF Downloads 187

Authors: Lakshmidevi Sreeramareddy, Komalpreet Kaur, Nane Pothier

Abstract:

Graphical-based passwords have existed for decades. Their major advantage is that they are easier to remember than an alphanumeric password. However, their disadvantage (especially recognition-based passwords) is the smaller password space, making them more vulnerable to brute force attacks. Graphical passwords are also highly susceptible to the shoulder-surfing effect. The gesture-based password method that we developed is a grid-free, template-free method. In this study, we evaluated the gesture-based passwords for usability and vulnerability. The results of the study are significant. We developed a gesture-based password application for data collection. Two modes of data collection were used: Creation mode and Replication mode. In creation mode (Session 1), users were asked to create six different passwords and reenter each password five times. In replication mode, users saw a password image created by some other user for a fixed duration of time. Three different duration timers, such as 5 seconds (Session 2), 10 seconds (Session 3), and 15 seconds (Session 4), were used to mimic the shoulder-surfing attack. After the timer expired, the password image was removed, and users were asked to replicate the password. There were 74, 57, 50, and 44 users participated in Session 1, Session 2, Session 3, and Session 4 respectfully. In this study, the machine learning algorithms have been applied to determine whether the person is a genuine user or an imposter based on the password entered. Five different machine learning algorithms were deployed to compare the performance in user authentication: namely, Decision Trees, Linear Discriminant Analysis, Naive Bayes Classifier, Support Vector Machines (SVMs) with Gaussian Radial Basis Kernel function, and K-Nearest Neighbor. Gesture-based password features vary from one entry to the next. It is difficult to distinguish between a creator and an intruder for authentication. For each password entered by the user, four features were extracted: password score, password length, password speed, and password size. All four features were normalized before being fed to a classifier. Three different classifiers were trained using data from all four sessions. Classifiers A, B, and C were trained and tested using data from the password creation session and the password replication with a timer of 5 seconds, 10 seconds, and 15 seconds, respectively. The classification accuracies for Classifier A using five ML algorithms are 72.5%, 71.3%, 71.9%, 74.4%, and 72.9%, respectively. The classification accuracies for Classifier B using five ML algorithms are 69.7%, 67.9%, 70.2%, 73.8%, and 71.2%, respectively. The classification accuracies for Classifier C using five ML algorithms are 68.1%, 64.9%, 68.4%, 71.5%, and 69.8%, respectively. SVMs with Gaussian Radial Basis Kernel outperform other ML algorithms for gesture-based password authentication. Results confirm that the shorter the duration of the shoulder-surfing attack, the higher the authentication accuracy. In conclusion, behavioral features extracted from the gesture-based passwords lead to less vulnerable user authentication.

Keywords: authentication, gesture-based passwords, machine learning algorithms, shoulder-surfing attacks, usability

Procedia PDF Downloads 106

Authors: Kasra Majbouri Yazdi, Adel Majbouri Yazdi, Saeid Khodayi, Jingyu Hou, Wanlei Zhou, Saeed Saedy

Abstract:

Fake news and false information are big challenges of all types of media, especially social media. There is a lot of false information, fake likes, views and duplicated accounts as big social networks such as Facebook and Twitter admitted. Most information appearing on social media is doubtful and in some cases misleading. They need to be detected as soon as possible to avoid a negative impact on society. The dimensions of the fake news datasets are growing rapidly, so to obtain a better result of detecting false information with less computation time and complexity, the dimensions need to be reduced. One of the best techniques of reducing data size is using feature selection method. The aim of this technique is to choose a feature subset from the original set to improve the classification performance. In this paper, a feature selection method is proposed with the integration of K-means clustering and Support Vector Machine (SVM) approaches which work in four steps. First, the similarities between all features are calculated. Then, features are divided into several clusters. Next, the final feature set is selected from all clusters, and finally, fake news is classified based on the final feature subset using the SVM method. The proposed method was evaluated by comparing its performance with other state-of-the-art methods on several specific benchmark datasets and the outcome showed a better classification of false information for our work. The detection performance was improved in two aspects. On the one hand, the detection runtime process decreased, and on the other hand, the classification accuracy increased because of the elimination of redundant features and the reduction of datasets dimensions.

Keywords: clustering, fake news detection, feature selection, machine learning, social media, support vector machine

Procedia PDF Downloads 176

Authors: Farnoosh Alami, Nazgol Azimi

Abstract:

Universities have recently paid much attention to the subject of social entrepreneurship. As a result, many of the highly ranked universities have established centers in this regard. The present research aims to investigate vision and mission of social entrepreneurship centers of the best universities ranked under 50 by Shanghai List 2013. It tries to find the common goals and features of their mission, vision, and activities which lead to their present success. This investigation is based on the web content of the first top 10 universities; among which six had social entrepreneurship centers. This is a qualitative research, and the findings are based on content analysis of documents. The findings confirm that education, research, talent development, innovative solutions, and supporting social innovation, are shared in the vision of these centers. In regard to their missions, social participation, networking, and leader education are the most shared features. Their common activities are focused on five categories of education, research, support, promotion, and networking.

Keywords: comparative study, qualitative research, social entrepreneurship centers, universities in the world

Procedia PDF Downloads 297

Authors: Mohamed Khallaf, Hossein M Rizeei

Abstract:

Simulating wind pattern behavior over proposed urban features is critical in the early stage of the conceptual design of both architectural and urban disciplines. However, it is typically not possible for designers to explore the impact of wind flow profiles across new urban developments due to a lack of real data and inaccurate estimation of building parameters. Modeling the details of existing and proposed urban features and testing them against wind flows is the missing part of the conceptual design puzzle where architectural and urban discipline can focus. This research aims to develop a spatial decision-support design method utilizing LiDAR, GIS, and performance-based wind simulation technology to mitigate wind-related hazards on a design by simulating alternative design scenarios at the pedestrian level prior to its implementation in Sydney, Australia. The result of the experiment demonstrates the capability of the proposed framework to improve pedestrian comfort in relation to wind profile.

Keywords: spatial decision-support design, performance-based wind simulation, LiDAR, GIS

Procedia PDF Downloads 124

Authors: Majed Sanan, Mohammad Rammal, Wassim Rammal

Abstract:

Today, security is a major concern. Intrusion Detection, Prevention Systems and Honeypot can be used to moderate attacks. Many researchers have proposed to use many IDSs ((Intrusion Detection System) time to time. Some of these IDS’s combine their features of two or more IDSs which are called Hybrid Intrusion Detection Systems. Most of the researchers combine the features of Signature based detection methodology and Anomaly based detection methodology. For a signature based IDS, if an attacker attacks slowly and in organized way, the attack may go undetected through the IDS, as signatures include factors based on duration of the events but the actions of attacker do not match. Sometimes, for an unknown attack there is no signature updated or an attacker attack in the mean time when the database is updating. Thus, signature-based IDS fail to detect unknown attacks. Anomaly based IDS suffer from many false-positive readings. So there is a need to hybridize those IDS which can overcome the shortcomings of each other. In this paper we propose a new approach to IDS (Intrusion Detection System) which is more efficient than the traditional IDS (Intrusion Detection System). The IDS is based on Honeypot Technology and Anomaly based Detection Methodology. We have designed Architecture for the IDS in a packet tracer and then implemented it in real time. We have discussed experimental results performed: both the Honeypot and Anomaly based IDS have some shortcomings but if we hybridized these two technologies, the newly proposed Hybrid Intrusion Detection System (HIDS) is capable enough to overcome these shortcomings with much enhanced performance. In this paper, we present a modified Hybrid Intrusion Detection System (HIDS) that combines the positive features of two different detection methodologies - Honeypot methodology and anomaly based intrusion detection methodology. In the experiment, we ran both the Intrusion Detection System individually first and then together and recorded the data from time to time. From the data we can conclude that the resulting IDS are much better in detecting intrusions from the existing IDSs.

Keywords: security, intrusion detection, intrusion prevention, honeypot, anomaly-based detection, signature-based detection, cloud computing, kfsensor

Procedia PDF Downloads 382

Authors: Dim En Nyaung, Thin Lai Lai Thein

Abstract:

Due to the rapid increase of Internet, web opinion sources dynamically emerge which is useful for both potential customers and product manufacturers for prediction and decision purposes. These are the user generated contents written in natural languages and are unstructured-free-texts scheme. Therefore, opinion mining techniques become popular to automatically process customer reviews for extracting product features and user opinions expressed over them. Since customer reviews may contain both opinionated and factual sentences, a supervised machine learning technique applies for subjectivity classification to improve the mining performance. In this paper, we dedicate our work is the task of opinion summarization. Therefore, product feature and opinion extraction is critical to opinion summarization, because its effectiveness significantly affects the identification of semantic relationships. The polarity and numeric score of all the features are determined by Senti-WordNet Lexicon. The problem of opinion summarization refers how to relate the opinion words with respect to a certain feature. Probabilistic based model of supervised learning will improve the result that is more flexible and effective.

Keywords: opinion mining, opinion summarization, sentiment analysis, text mining

Procedia PDF Downloads 332

Authors: Kurt Maxwell Kusterer, Sylvain Mignot, Annick Vignes

Abstract:

To our best knowledge YouTube is the largest video hosting platform in the world. A high number of creators, viewers, subscribers and commentators act in this specific eco-system which generates huge sums of money. Views, subscribers, and comments help to increase the popularity of content creators. The most popular creators are sponsored by brands and participate in marketing campaigns. For a few of them, this becomes a financially rewarding profession. This is made possible through the YouTube Partner Program, which shares revenue among creators based on their popularity. We believe that the role of comments in increasing the popularity is to be emphasized. In what follows, YouTube is considered as a bilateral network between the videos and the commentators. Analyzing a detailed data set focused on French YouTubers, we consider each comment as a link between a commentator and a video. Our research question asks what are the predominant features of a video which give it the highest probability to be commented on. Following on from this question, how can we use these features to predict the action of the agent in commenting one video instead of another, considering the characteristics of the commentators, videos, topics, channels, and recommendations. We expect to see that the videos of more popular channels generate higher viewer engagement and thus are more frequently commented. The interest lies in discovering features which have not classically been considered as markers for popularity on the platform. A quick view of our data set shows that 96% of the commentators comment only once on a certain video. Thus, we study a non-weighted bipartite network between commentators and videos built on the sub-sample of 96% of unique comments. A link exists between two nodes when a commentator makes a comment on a video. We run an Exponential Random Graph Model (ERGM) approach to evaluate which characteristics influence the probability of commenting a video. The creation of a link will be explained in terms of common video features, such as duration, quality, number of likes, number of views, etc. Our data is relevant for the period of 2020-2021 and focuses on the French YouTube environment. From this set of 391 588 videos, we extract the channels which can be monetized according to YouTube regulations (channels with at least 1000 subscribers and more than 4000 hours of viewing time during the last twelve months).In the end, we have a data set of 128 462 videos which consist of 4093 channels. Based on these videos, we have a data set of 1 032 771 unique commentators, with a mean of 2 comments per a commentator, a minimum of 1 comment each, and a maximum of 584 comments.

Keywords: YouTube, social networks, economics, consumer behaviour

Procedia PDF Downloads 68

Authors: Omojokun G. Aju, Adedayo O. Sule

Abstract:

ZigBee wireless sensor and control network is one of the most popularly deployed wireless technologies in recent years. This is because ZigBee is an open standard lightweight, low-cost, low-speed, low-power protocol that allows true operability between systems. It is built on existing IEEE 802.15.4 protocol and therefore combines the IEEE 802.15.4 features and newly added features to meet required functionalities thereby finding applications in wide variety of wireless networked systems. ZigBee‘s current focus is on embedded applications of general-purpose, inexpensive, self-organising networks which requires low to medium data rates, high number of nodes and very low power consumption such as home/industrial automation, embedded sensing, medical data collection, smart lighting, safety and security sensor networks, and monitoring systems. Although the ZigBee design specification includes security features to protect data communication confidentiality and integrity, however, when simplicity and low-cost are the goals, security is normally traded-off. A lot of researches have been carried out on ZigBee technology in which emphasis has mainly been placed on ZigBee network performance characteristics such as energy efficiency, throughput, robustness, packet delay and delivery ratio in different scenarios and applications. This paper investigate and analyse the data accuracy, network implementation difficulties and security challenges of ZigBee network applications in star-based and mesh-based topologies with emphases on its home monitoring application using the ZigBee ProBee ZE-10 development boards for the network setup. The paper also expose some factors that need to be considered when designing ZigBee network applications and suggest ways in which ZigBee network can be designed to provide more resilient to network attacks.

Keywords: home monitoring, IEEE 802.14.5, topology, wireless security, wireless sensor network (WSN), ZigBee

Procedia PDF Downloads 382

Authors: Mehdi Attia, Sana Ben Fadhel, Lamjed Bettaieb

Abstract:

Mobile technology has become a part of our daily lives and assist learners (despite their level and age) in their leaning process using various apparatus and mobile devices (laptop, tablets, etc.). This paper presents a new learning framework based on tablets. This solution has been developed and tested in ESPRIT “Ecole Supérieure Privée d’Igénieurie et de Technologies”, a Tunisian school of engineering. This application is named ESSF: Esprit Smart School Framework. In this work, the main features of the proposed solution are listed, particularly its impact on the learners’ evaluation process. Learner’s assessment has always been a critical component of the learning process as it measures students’ knowledge. However, traditional evaluation methods in which the learner is evaluated once or twice each year cannot reflect his real level. This is why a continuous assessment (CA) process becomes necessary. In this context we have proved that ESSF offers many important features that enhance and facilitate the implementation of the CA process.

Keywords: continuous assessment, mobile learning, tablet based learning, smart school, ESSF

Procedia PDF Downloads 333

Authors: Rashini Maduka, C. R. Wijesinghe, A. R. Weerasinghe

Abstract:

Drug-drug interactions (DDIs) can happen when two or more drugs are taken together. Today DDIs have become a serious health issue due to adverse drug effects. In vivo and in vitro methods for identifying DDIs are time-consuming and costly. Therefore, in-silico-based approaches are preferred in DDI identification. Most machine learning models for DDI prediction are used chemical and biological drug properties as features. However, some drug features are not available and costly to extract. Therefore, it is better to make automatic feature engineering. Furthermore, people who have diabetes already suffer from other diseases and take more than one medicine together. Then adverse drug effects may happen to diabetic patients and cause unpleasant reactions in the body. In this study, we present a model with a graph convolutional autoencoder and a graph decoder using a dataset from DrugBank version 5.1.3. The main objective of the model is to identify unknown interactions between antidiabetic drugs and the drugs taken by diabetic patients for other diseases. We considered automatic feature engineering and used Known DDIs only as the input for the model. Our model has achieved 0.86 in AUC and 0.86 in AP.

Keywords: drug-drug interaction prediction, graph embedding, graph convolutional networks, adverse drug effects

Procedia PDF Downloads 100

Authors: Rezvan Hosseinian

Abstract:

Aim: Systemic sclerosis (SSc) is a chronic connective tissue disease with the clinical hallmark of skin thickening and tethering. The correlation of musculoskeletal features with other parameters should be considered in SSc patients. Methods: We reviewed the records of all patients who had more than one visit and standard anteroposterior radiography of hand. We used univariate analysis, and factors with p<0.05 were included in logistic regression to find out dependent factors. Results: Overall, 180 SSc patients were enrolled in our study, 161 (89.4%) of whom were women. The median age (IQR) was 47.0 years (16), and 52% had a diffuse subtype of the disease. In multivariate analysis, tendon friction rubs (TFRs) were associated with the presence of calcinosis, muscle tenderness, and flexion contracture (FC) on physical examination (p<0.05). Arthritis showed no differences in the two subtypes of the disease (p=0.98), and in multivariate analysis, there were no correlations between radiographic arthritis and serological and clinical features. The radiographic results indicated that disease duration correlated with joint erosion, acro-osteolysis, resorption of the distal ulna, calcinosis and radiologic FC (p< 0.05). Acro-osteolysis was more frequent in the dcSSc subtype, TFRs, and anti-TOPO I antibody. Radiologic FC showed an association with skin score, calcinosis and haematocrit <30% (p<0.05). Joint flexion on radiography was associated with disease duration, modified Rodnan skin score, calcinosis, and low hematocrit (P<0.01). Conclusion: Disease duration was a main dependent factor for developing joint erosion, acro-osteolysis, bone resorption, calcinosis, and flexion contracture on hand radiography. Acro-osteolysis presented in the severe form of the disease. Acro-osteolysis was the only dependent variable associated with bone demineralization.

Keywords: disease subsets, hand radiography, joint erosion, sclerosis

Procedia PDF Downloads 90

Authors: Nasrin Azarbani

Abstract:

Aim: Systemic sclerosis (SSc) is a chronic connective tissue disease with the clinical hallmark of skin thickening and tethering. Correlation of musculoskeletal features with other parameters should be considered in SSc patients. Methods: We reviewed the records of all patients who had more than one visit and standard anteroposterior radiography of hand. We used univariate analysis, and factors with p<0.05 were included in logistic regression to find out dependent factors. Results: Overall, 180 SSc patients were enrolled in our study, 161 (89.4%) of whom were women. Median age (IQR) was 47.0 years (16), and 52% had diffuse subtype of the disease. In multivariate analysis, tendon friction rubs (TFRs) was associated with the presence of calcinosis, muscle tenderness, and flexion contracture (FC) on physical examination (p<0.05). Arthritis showed no differences in the two subtypes of the disease (p=0.98), and in multivariate analysis, there were no correlations between radiographic arthritis and serological and clinical features. The radiographic results indicated that disease duration correlated with joint erosion, acro-osteolysis, resorption of distal ulna, calcinosis and radiologic FC (p< 0.05). Acro-osteolysis was more frequent in the dcSSc subtype, TFRs, and anti-TOPO I antibody. Radiologic FC showed an association with skin score, calcinosis and haematocrit <30% (p<0.05). Joint flexion on radiography was associated with disease duration, modified Rodnan skin score, calcinosis, and low haematocrit (P<0.01). Conclusion: Disease duration was a main dependent factor for developing joint erosion, acro-osteolysis, bone resorption, calcinosis, and flexion contracture on hand radiography. Acro-osteolysis presented in the severe form of the disease. Acro-osteolysis was the only dependent variable associated with bone demineralization.

Keywords: sclerosis, disease subsets, joint erosion, musculoskeletal

Procedia PDF Downloads 66

Authors: Shu-Min Tsao, Chung-Ming Lo, Shao-Chun Chen

Abstract:

Most diabetes patients tend to suffer from its complication of retina diseases. Therefore, early detection and early treatment are important. In clinical examinations, using color fundus image was the most convenient and available examination method. According to the exudates appeared in the retinal image, the status of retina can be confirmed. However, the routine screening of diabetic retinopathy by color fundus images would bring time-consuming tasks to physicians. This study thus proposed a computer-aided exudate diagnosis for the screening of diabetic retinopathy. After removing vessels and optic disc in the retinal image, six quantitative features including region number, region area, and gray-scale values etc… were extracted from the remaining regions for classification. As results, all six features were evaluated to be statistically significant (p-value < 0.001). The accuracy of classifying the retinal images into normal and diabetic retinopathy achieved 82%. Based on this system, the clinical workload could be reduced. The examination procedure may also be improved to be more efficient.

Keywords: computer-aided diagnosis, diabetic retinopathy, exudate, image processing

Procedia PDF Downloads 267

Authors: Javier A. Dominguez Caballero, Graeme A. Manson, Matthew B. Marshall

Abstract:

Cutting tools with ceramic inserts are often used in the process of machining many types of superalloy, mainly due to their high strength and thermal resistance. Nevertheless, during the cutting process, the plastic flow wear generated in these inserts enhances and propagates cracks due to high temperature and high mechanical stress. This leads to a very variable failure of the cutting tool. This article explores the relationship between the continuous wear that ceramic SiAlON (solid solutions based on the Si3N4 structure) inserts experience during a high-speed machining process and the evolution of sparks created during the same process. These sparks were analysed through pictures of the cutting process recorded using an SLR camera. Features relating to the intensity and area of the cutting sparks were extracted from the individual pictures using image processing techniques. These features were then related to the ceramic insert’s crater wear area.

Keywords: ceramic cutting tools, high speed machining, image processing, tool condition monitoring, tool wear

Procedia PDF Downloads 298

Authors: Rebecca J. Hafner, Daniel Read, David Elmes

Abstract:

The current research applies decision making theory in order to address the problem of increasing uptake of energy-efficient technologies in the market place, where uptake is currently slower than one might predict following rational choice models. Specifically, in two studies we apply the alignable/non-alignable features effect and explore the impact of varying information structure on the consumers’ preference for standard versus energy efficient technologies. As researchers in the Interdisciplinary centre for Storage, Transformation and Upgrading of Thermal Energy (i-STUTE) are currently developing energy efficient heating systems for homes and businesses, we focus on the context of home heating choice, and compare preference for a standard condensing boiler versus an energy efficient heat pump, according to experimental manipulations in the structure of prior information. In Study 1, we find that people prefer stronger alignable features when options are similar; an effect which is mediated by an increased tendency to infer missing information is the same. Yet, in contrast to previous research, we find no effects of alignability on option preference when options differ. The advanced methodological approach used here, which is the first study of its kind to randomly allocate features as either alignable or non-alignable, highlights potential design effects in previous work. Study 2 is designed to explore the interaction between alignability and construal level as an explanation for the shift in attentional focus when options differ. Theoretical and applied implications for promoting energy efficient technologies are discussed.

Keywords: energy-efficient technologies, decision-making, alignability effects, construal level theory, CO2 reduction

Procedia PDF Downloads 329

Authors: Yagya Raj Pandeya, Joonwhoan Lee

Abstract:

The dynamic behavior of music and video makes it difficult to evaluate musical instrument playing in a video by computer system. Any television or film video clip with music information are rich sources for analyzing musical instruments using modern machine learning technologies. In this research, we integrate the audio and video information sources using convolutional neural network (CNN) and pass network learned features through recurrent neural network (RNN) to preserve the dynamic behaviors of audio and video. We use different pre-trained CNN for music and video feature extraction and then fine tune each model. The music network use 2D convolutional network and video network use 3D convolution (C3D). Finally, we concatenate each music and video feature by preserving the time varying features. The long short term memory (LSTM) network is used for long-term dynamic feature characterization and then use late fusion with generalized mean. The proposed network performs better performance to recognize the musical instrument using audio-video multimodal neural network.

Keywords: multimodal, 3D convolution, music-video feature extraction, generalized mean

Procedia PDF Downloads 215

Authors: T. Bensana, S. Mekhilef

Abstract:

The methodology of vibration based condition monitoring technology has been developing at a rapid stage in the recent years suiting to the maintenance of sophisticated and complicated machines. The ability of wavelet analysis to efficiently detect non-stationary, non-periodic, transient features of the vibration signal makes it a demanding tool for condition monitoring. This paper presents a methodology for fault diagnosis of rolling element bearings based on wavelet envelope power spectrum technique is analysed in both the time and frequency domains. In the time domain the auto-correlation of the wavelet de-noised signal is applied to evaluate the period of the fault pulses. However, in the frequency domain the wavelet envelope power spectrum has been used to identify the fault frequencies with the single sided complex Laplace wavelet as the mother wavelet function. Results show the superiority of the proposed method and its effectiveness in extracting fault features from the raw vibration signal.

Keywords: preventive maintenance, fault diagnostics, rolling element bearings, wavelet de-noising

Procedia PDF Downloads 378

Authors: Umair Rashid

Abstract:

Normally softmax loss is used as the supervision signal in face recognition (FR) system, and it boosts the separability of features. In the last two years, a number of techniques have been proposed by reformulating the original softmax loss to enhance the discriminating power of Deep Convolutional Neural Networks (DCNNs) for FR system. To learn angularly discriminative features Cosine-Margin based softmax has been adjusted as monotonically decreasing angular function, that is the main challenge for angular based softmax. On that issue, we propose monotonically decreasing element for Cosine-Margin based softmax and also, we discussed the effect of different monotonically decreasing parameters on angular Margin softmax for FR system. We train the model on publicly available dataset CASIA- WebFace via our proposed monotonically decreasing parameters for cosine function and the tests on YouTube Faces (YTF, Labeled Face in the Wild (LFW), VGGFace1 and VGGFace2 attain the state-of-the-art performance.

Keywords: deep convolutional neural networks, cosine margin face recognition, softmax loss, monotonically decreasing parameter

Procedia PDF Downloads 101

Authors: Sajjad Ahmed, Knut Hinkelmann, Flavio Corradini

Abstract:

Fake news detection research is still in the early stage as this is a relatively new phenomenon in the interest raised by society. Machine learning helps to solve complex problems and to build AI systems nowadays and especially in those cases where we have tacit knowledge or the knowledge that is not known. We used machine learning algorithms and for identification of fake news; we applied three classifiers; Passive Aggressive, Naïve Bayes, and Support Vector Machine. Simple classification is not completely correct in fake news detection because classification methods are not specialized for fake news. With the integration of machine learning and text-based processing, we can detect fake news and build classifiers that can classify the news data. Text classification mainly focuses on extracting various features of text and after that incorporating those features into classification. The big challenge in this area is the lack of an efficient way to differentiate between fake and non-fake due to the unavailability of corpora. We applied three different machine learning classifiers on two publicly available datasets. Experimental analysis based on the existing dataset indicates a very encouraging and improved performance.

Keywords: fake news detection, natural language processing, machine learning, classification techniques.

Procedia PDF Downloads 167

Authors: Eda Jazexhiu-Postoli, Gladiola Hoxha, Ada Simeoni, Sonila Biba

Abstract:

Background Neonatal hypotonia represents a commonly encountered issue in the Neonatal Intensive Care Unit and newborn nursery. The differential diagnosis is broad, encompassing chromosome abnormalities, primary muscular dystrophies, neuropathies and inborn errors of metabolism. Aim of study Our study describes some of the main clinical features of hypotonia in newborns and presents clinical cases of neonatal hypotonia we treated in our Neonatal unit in the last 3 years. Case reports Four neonates born in our hospital presented with hypotonia after birth, one preterm newborn 35-36 weeks of gestational age and three other term newborns (38-39 weeks of gestational age). Prenatal data revealed a decrease in fetal movements in both cases. Intrapartum meconium-stained amniotic fluid was found in 75% of our hypotonic newborns. Clinical features included inability to establish effective respiratory movements and need for resuscitation in the delivery room, respiratory distress syndrome, feeding difficulties and need for oro-gastric tube feeding, dysmorphic features, hoarse voice and moderate to severe muscular hypotonia. The genetic workup revealed the diagnosis of Autosomal Recessive Congenital Myasthenic Syndrome 1-B, Sotos Syndrome, Spinal Muscular Atrophy Type 1 and Transient Hypotonia of the Newborn. Two out of four hypotonic neonates were transferred to the Pediatric Intensive Care Unit and died at the age of three to five months old. Conclusion Hypotonia is a concerning finding in neonatal care and it is suggested by decreased intrauterine fetal movements, failure to establish first breaths, respiratory distress and feeding difficulties in the neonate. Prognosis is determined by its etiology and time of diagnosis and intervention.

Keywords: hypotonic neonate, respiratory distress, feeding difficulties, fetal movements

Procedia PDF Downloads 115

Authors: Trude Tørset, María Díez Gutiérrez

Abstract:

The west coast of Norway features numerous islands and fjords. Ferry services connect the roads when these features make the construction challenging. Currently, scientific effort is designated to assess potential ferry replacement projects along the European road E-39. The inconvenience of ferry dependency is imprecisely represented in the transport models, thus transport analyses of ferry replacement projects appear as guesstimates rather than reliable input to decision-making processes of such costly projects. Trips including ferry connections imply more inconvenient elements than just travel time and cost. The goal of this paper is to understand and explain the extra inconveniences associated to the dependency of the ferry. The first scientific approach is to identify the characteristics of the ferry travelers and their trips’ features, as well as whether the ferry represents an obstacle for some specific trip types. In doing so, a survey was conducted in 2011 in eight E-39 ferries and in 2013 in 18 ferries connecting different road categories. More than 20,000 passengers answered with their trip and socioeconomic characteristics. The travel patterns in the different ferry connections were compared. The analysis showed that the trip features differed based on the location of the ferry connections, yet independently of the road category. Additionally, the patterns were compared to the national travel survey to detect differences in the travel patterns due to the use of the ferry connections. The results showed that the share of commuting trips within the same travel time was lower if the ferry was part of the trip. The second scientific approach is to know how the different travelers perceive potential benefits for a ferry replacement project. In the 2011 survey, some of the questions were about the relevance of nine different benefits this project might bring. Travelers identified the better access to public services and job market as the most valuable benefits, followed by the reduced planning of the trip. In 2016, a follow-up survey in some of the ferry connections was carried out in order to investigate variations in travelers’ perceptions. The growing interest in ferry replacement projects might make travelers more aware of the potential benefits these would bring to their daily lives. This paper describes the travel behaviour of travelers using a ferry connection as part of their trips, as well as the potential inconveniences associated to these trips. The findings might provide valuable input to further development of transport models, concept evaluations and cost benefit analysis methods.

Keywords: ferry connections, ferry trip, inconvenience costs, travel behaviour

Procedia PDF Downloads 227

Authors: M. Michael Gromiha, A. Mary Thangakani, Sandeep Kumar, D. Velmurugan

Abstract:

The main cause of several neurodegenerative diseases such as Alzhemier, Parkinson, and spongiform encephalopathies is formation of amyloid fibrils and plaques in proteins. We have analyzed different sets of proteins and peptides to understand the influence of sequence-based features on protein aggregation process. The comparison of 373 pairs of homologous mesophilic and thermophilic proteins showed that aggregation-prone regions (APRs) are present in both. But, the thermophilic protein monomers show greater ability to ‘stow away’ the APRs in their hydrophobic cores and protect them from solvent exposure. The comparison of amyloid forming and amorphous b-aggregating hexapeptides suggested distinct preferences for specific residues at the six positions as well as all possible combinations of nine residue pairs. The compositions of residues at different positions and residue pairs have been converted into energy potentials and utilized for distinguishing between amyloid forming and amorphous b-aggregating peptides. Our method could correctly identify the amyloid forming peptides at an accuracy of 95-100% in different datasets of peptides.

Keywords: aggregation, amyloids, thermophilic proteins, amino acid residues, machine learning techniques

Procedia PDF Downloads 614

Authors: Saumya Srivastava, Rina Maiti

Abstract:

In order to achieve a better performance of vehicle detection in a complex environment, we present an efficient approach for a multi-vehicle detection system using an adaptive sliding window technique. For a given frame, image segmentation is carried out to establish the region of interest. Gradient computation followed by thresholding, denoising, and morphological operations is performed to extract the binary search image. Near-region field and far-region field are defined to generate hypotheses using the adaptive sliding window technique on the resultant binary search image. For each vehicle candidate, features are extracted using a histogram of oriented gradients, and a pre-trained support vector machine is applied for hypothesis verification. Later, the Kalman filter is used for tracking the vanishing point. The experimental results show that the method is robust and effective on various roads and driving scenarios. The algorithm was tested on highways and urban roads in India.

Keywords: gradient, vehicle detection, histograms of oriented gradients, support vector machine

Procedia PDF Downloads 124

Authors: Wilfred Jeyatheese Jeyaraj

Abstract:

The Sri Lankan legislative acts dictate several recommendations to improve accessibility of services for the visually challenged. But the main consideration here is the feasibility and extent to which these endorsements have been implemented in actuality within Sri Lankan academic libraries. This paper tends to assess the existent issues that impediment the implementation of accessibility features for the visually challenged in Sri Lankan academic libraries. Visually challenged students continually walk through immense challenges to step forth into their university life. Reaching their undergrad stage of their academic phase, they should be entitled to access information resources with ease and with equality in comparison to the sighted users of a university library. The current university libraries in Sri Lanka, have well improved services that they render to their users. But, what lacks in this scenario is the consideration as to whether these features offered by libraries are user-friendly and easily accessible by the visually challenged users as well. Hence, this paper tends to analyze the inhibitions in delivering services oriented towards the visually challenged and the sighted, and propose feasible alternatives to create a neutral high-end university library environment.

Keywords: accessibility, university library, Sri Lanka, visually-challenged

Procedia PDF Downloads 289

Authors: Elnaz Lashgari, Emel Demircan

Abstract:

Electromyography (EMG) is one of the most important interfaces between humans and robots for rehabilitation. Decoding this signal helps to recognize muscle activation and converts it into smooth motion for the robots. Detecting each muscle’s pattern during walking and running is vital for improving the quality of a patient’s life. In this study, EMG data from 10 muscles in 10 subjects at 4 different speeds were analyzed. EMG signals are nonlinear with high dimensionality. To deal with this challenge, we extracted some features in time-frequency domain and used manifold learning and Laplacian Eigenmaps algorithm to find the intrinsic features that represent data in low-dimensional space. We then used the Bayesian classifier to identify various patterns of EMG signals for different muscles across a range of running speeds. The best result for vastus medialis muscle corresponds to 97.87±0.69 for sensitivity and 88.37±0.79 for specificity with 97.07±0.29 accuracy using Bayesian classifier. The results of this study provide important insight into human movement and its application for robotics research.

Keywords: electromyography, manifold learning, ISOMAP, Laplacian Eigenmaps, locally linear embedding

Procedia PDF Downloads 361

Authors: Tsz Ching Jasmine Lam

Abstract:

Linguistic landscape (LL) can serve as both top-down and bottom-up approaches to understanding language planning policy in various dimensions. It can reflect the language identities, motives and contestations perceived by stakeholders of different decision-making levels. Prior studies adopted the bottom-up approach to investigate the language practice and ideologies reflected by the design and linguistic features observed in the linguistic landscapes in ethnically and linguistically diverse areas, like Medan in Russia and Seoul in Korea. As Hong Kong is also a trilingual city with an inclusive combination of nationalities, this paper is intended to take it as a case study to explore the de facto language ideologies reflected by LL at the micro-level. We would look into the catering industry from a holistic perspective by reviewing the food menus of 66 restaurants located in diversified districts and serving different types of cuisines. This bottom-up LL research reveals that business owners and the public share the language ideologies of perceiving English as a prestigious language, multilingualism and traditional Chinese as a standard character.

Keywords: bottom-up, language ideologies, language planning policy, language policy, language identities, linguistic landscape

Procedia PDF Downloads 78

Authors: Musaab Salman Sultan

Abstract:

The surface and bulk magnetization behavior of template released isolated ferromagnetic Ni₆₀Fe₄₀ nanowires of relatively thick diameters (~200 nm), deposited from a dilute suspension onto pre-patterned insulating chips have been investigated experimentally, using a highly sensitive Magneto-Optical Ker Effect (MOKE) magnetometry and Magneto-Resistance (MR) measurements, respectively. The MR data were consistent with the theoretical predictions of the anisotropic magneto-resistance (AMR) effect. The MR measurements, in all the angles of investigations, showed large features and a series of nonmonotonic "continuous small features" in the resistance profiles. The extracted switching fields from these features and from MOKE loops were compared with each other and with the switching fields reported in the literature that adopted the same analytical techniques on the similar compositions and dimensions of nanowires. A large difference between MOKE and MR measurments was noticed. The disparate between MOKE and MR results is attributed to the variance in the micro-magnetic structure of the surface and the bulk of such ferromagnetic nanowires. This result was ascertained using micro-magnetic simulations on an individual: cylindrical and rectangular cross sections NiFe nanowires, with the same diameter/thickness of the experimental wires, using the Object Oriented Micro-magnetic Framework (OOMMF) package where the simulated loops showed different switching events, indicating that such wires have different magnetic states in the reversal process and the micro-magnetic spin structures during switching behavior was complicated. These results further supported the difference between surface and bulk magnetization behavior in these nanowires. This work suggests that a combination of MOKE and MR measurements is required to fully understand the magnetization behavior of such relatively thick isolated cylindrical ferromagnetic nanowires.

Keywords: MOKE magnetometry, MR measurements, OOMMF package, micromagnetic simulations, ferromagnetic nanowires, surface magnetic properties

Procedia PDF Downloads 250

Authors: Artur Daniel Ramos Modolo

Abstract:

Popular science magazines are an important way to communicate scientific information to lay audience in science. Since the popularization of social networking sites (SNSs) such as Facebook and Twitter, these magazines are trying to adapt their content to these new media. In this study, one hundred posts of popular science magazines on Facebook are analyzed regarding the use of genres of communication and readers’ reactions. The quantitative analysis of these features considers the variety of genres and how the users of Facebook answer to them (liking, sharing and commenting). The first hypothesis was that these magazines used the genres of communication posted on Facebook both to marketing and informational purposes and that these mixed intentions have an impact in the number of readers’ reactions. In order to analyze these features, twenty timeline posts published by five magazines: Cosmos, Galileu, New Scientist, Scientific American and Superinteressante were gathered during the period of three days (6th November 2015–8th November 2015). This research shows that the hyperlinks posted by these magazines created ways to diversify the communication genres used on their pages and, at the same time, revealed that, overall, readers react quantitatively different to these genres.

Keywords: Facebook, genres of communication, likes, popular science magazines, social networking sites

Procedia PDF Downloads 402

Authors: Yifan Hou, Haibo Liu, Le Jiang, Wandong Su, Binqing Wang

Abstract:

Road defect detection is crucial for modern urban management and infrastructure maintenance. Traditional road defect detection methods mostly rely on manual labor, which is not only inefficient but also difficult to ensure their reliability. However, existing deep learning-based road defect detection models have poor detection performance in complex environments and lack robustness to multi-scale targets. To address this challenge, this paper proposes a distinct detection framework based on the one stage algorithm network structure. This article designs a deep feature extraction network based on RCSDarknet, which applies channel shuffling to enhance information fusion between tensors. Through repeated stacking of RCS modules, the information flow between different channels of adjacent layer features is enhanced to improve the model's ability to capture target spatial features. In addition, a multi-scale feature fusion mechanism with weighted dual flow paths was adopted to fuse spatial features of different scales, thereby further improving the detection performance of the model at different scales. To validate the performance of the proposed algorithm, we tested it using the RDD2022 dataset. The experimental results show that the enhancement algorithm achieved 84.14% mAP, which is 1.06% higher than the currently advanced YOLOv8 algorithm. Through visualization analysis of the results, it can also be seen that our proposed algorithm has good performance in detecting targets of different scales in complex scenes. The above experimental results demonstrate the effectiveness and superiority of the proposed algorithm, providing valuable insights for advancing real-time road defect detection methods.

Keywords: roads, defect detection, visualization, deep learning

Procedia PDF Downloads 6