Search results for: computer virus classification

Authors: Y. Ruchi, A. Prerna, S. Deepshikha

Abstract:

Amyotrophic lateral sclerosis (ALS), also known as “Lou Gehrig’s disease”. It is a neurodegenerative disease associated with degeneration of motor neurons in the cerebral cortex, brain stem, and spinal cord characterized by distal muscle weakness, atrophy, normal sensation, pyramidal signs and progressive muscular paralysis reflecting. ALS2 is a juvenile autosomal recessive disorder, slowly progressive, that maps to chromosome 2q33 and is associated with mutations in the alsin gene, a putative GTPase regulator. In this paper we have done homology modeling of alsin2 protein using multiple templates (3KCI_A, 4LIM_A, 402W_A, 4D9S_A, and 4DNV_A) designed using the Prime program in Schrödinger software. Further modeled structure is used to identify effective binding sites on the basis of structural and physical properties using sitemap program in Schrödinger software, structural and function analysis is done by using Prosite and ExPASy server that gives insight into conserved domains and motifs that can be used for protein classification. This paper summarizes the structural, functional and binding site property of alsin2 protein. These binding sites can be potential drug target sites and can be used for docking studies.

Keywords: ALS, binding site, homology modeling, neuronal degeneration

Procedia PDF Downloads 387

Authors: Trupti Chavan, Ramachandra Guda, Kameshwar Rao

Abstract:

The lack of data is a pain problem in medical image analysis using a convolutional neural network (CNN). This work uses various spatial transformation techniques to address the medical image augmentation issue for knee detection and localization using an enhanced single shot detector (SSD) network. The spatial transforms like a negative, histogram equalization, power law, sharpening, averaging, gaussian blurring, etc. help to generate more samples, serve as pre-processing methods, and highlight the features of interest. The experimentation is done on the OpenKnee dataset which is a collection of knee images from the openly available online sources. The CNN called enhanced single shot detector (SSD) is utilized for the detection and localization of the knee joint from a given X-ray image. It is an enhanced version of the famous SSD network and is modified in such a way that it will reduce the number of prediction boxes at the output side. It consists of a classification network (VGGNET) and an auxiliary detection network. The performance is measured in mean average precision (mAP), and 99.96% mAP is achieved using the proposed enhanced SSD with spatial transformations. It is also seen that the localization boundary is comparatively more refined and closer to the ground truth in spatial augmentation and gives better detection and localization of knee joints.

Keywords: data augmentation, enhanced SSD, knee detection and localization, medical image analysis, openKnee, Spatial transformations

Procedia PDF Downloads 152

Authors: Mahmut Yildirim

Abstract:

This paper proposed the bidirectional long short-term memory (Bi-LSTM) network-aided deep learning (DL)-based signal detection for Orthogonal frequency division multiplexing with all index modulation (OFDM-AIM), namely Bi-DeepAIM. OFDM-AIM is developed to increase the spectral efficiency of OFDM with index modulation (OFDM-IM), a promising multi-carrier technique for communication systems beyond 5G. In this paper, due to its strong classification ability, Bi-LSTM is considered an alternative to the maximum likelihood (ML) algorithm, which is used for signal detection in the classical OFDM-AIM scheme. The performance of the Bi-DeepAIM is compared with LSTM network-aided DL-based OFDM-AIM (DeepAIM) and classic OFDM-AIM that uses (ML)-based signal detection via BER performance and computational time criteria. Simulation results show that Bi-DeepAIM obtains better bit error rate (BER) performance than DeepAIM and lower computation time in signal detection than ML-AIM.

Keywords: bidirectional long short-term memory, deep learning, maximum likelihood, OFDM with all index modulation, signal detection

Procedia PDF Downloads 70

Authors: Orsolya Kegyes-Brassai, Zsolt Szilvágyi, Ákos Wolf, Richard P. Ray

Abstract:

Local ground conditions have a substantial influence on the seismic response of structures. Their inclusion in seismic hazard assessment and structural design can be realized at different levels of sophistication. However, response results based on more advanced calculation methods e.g. nonlinear or equivalent linear site analysis tend to show significant discrepancies when compared to simpler approaches. This project's main objective was to compare results from several 1-D response programs to Eurocode 8 design spectra. Data from in-situ site investigations were used for assessing local ground conditions at several locations in Hungary. After discussion of the in-situ measurements and calculation methods used, a comprehensive evaluation of all major contributing factors for site response is given. While the Eurocode spectra should account for local ground conditions based on soil classification, there is a wide variation in peak ground acceleration determined from 1-D analyses versus Eurocode. Results show that current Eurocode 8 design spectra may not be conservative enough to account for local ground conditions typical for Hungary.

Keywords: 1-D site response analysis, multichannel analysis of surface waves (MASW), seismic CPT, seismic hazard assessment

Procedia PDF Downloads 243

Authors: Romeo A. Botes, Imelda Smit

Abstract:

The demands of an ever changing and complex higher education environment, along with the profile of modern learners challenge current approaches to assessment and feedback. More learners enter the education system every year. The younger generation expects immediate feedback. At the same time, feedback should be meaningful. The assessment of practical activities in programming poses a particular problem, since both lecturers and learners in the information and computer science discipline acknowledge that paper-based assessment for programming subjects lacks meaningful real-life testing. At the same time, feedback lacks promptness, consistency, comprehensiveness and individualisation. Most of these aspects may be addressed by modern, technology-assisted assessment. The focus of this paper is the continuous development of an artefact that is used to assist the lecturer in the assessment and feedback of practical programming activities in a senior database programming class. The artefact was developed using three Design Science Research cycles. The first implementation allowed one programming activity submission per assessment intervention. This pilot provided valuable insight into the obstacles regarding the implementation of this type of assessment tool. A second implementation improved the initial version to allow multiple programming activity submissions per assessment. The focus of this version is on providing scaffold feedback to the learner – allowing improvement with each subsequent submission. It also has a built-in capability to provide the lecturer with information regarding the key problem areas of each assessment intervention.

Keywords: programming, computer-aided assessment, technology-assisted assessment, programming assessment software, design science research, mixed-method

Procedia PDF Downloads 295

Authors: Xiaoxing Ye

Abstract:

To investigate complex rumen microbial communities, 16S ribosomal RNA (rRNA) sequencing is widely used. Here, we evaluated the impact of bioinformatics pipelines on the observation of OTUs and taxonomic classification of 750 cattle rumen microbial samples by comparing three commonly used pipelines (LotuS, UPARSE, and QIIME) with Usearch. In LotuS-based analyses, 189 archaeal and 3894 bacterial OTUs were observed. The observed OTUs for the Usearch analysis were significantly larger than the LotuS results. We discovered 1495 OTUs for archaea and 92665 OTUs for bacteria using Usearch analysis. In addition, taxonomic assignments were made for the rumen microbial samples. All pipelines had consistent taxonomic annotations from the phylum to the genus level. A difference in relative abundance was calculated for all microbial levels, including Bacteroidetes (QIIME: 72.2%, Usearch: 74.09%), Firmicutes (QIIME: 18.3%, Usearch: 20.20%) for the bacterial phylum, Methanobacteriales (QIIME: 64.2%, Usearch: 45.7%) for the archaeal class, Methanobacteriaceae (QIIME: 35%, Usearch: 45.7%) and Methanomassiliicoccaceae (QIIME: 35%, Usearch: 31.13%) for archaeal family. However, the most prevalent archaeal class varied between these two annotation pipelines. The Thermoplasmata was the top class according to the QIIME annotation, whereas Methanobacteria was the top class according to Usearch.

Keywords: cattle rumen, rumen microbial, 16S rRNA gene sequencing, bioinformatics pipeline

Procedia PDF Downloads 86

Authors: Marian Sagat, Mariana Remesikova

Abstract:

In civil engineering, there is a problem how to industrially produce tensile membrane structures that are non-developable surfaces. Nondevelopable surfaces can only be developed with a certain error and we want to minimize this error. To that goal, the non-developable surfaces are cut into plates along to the geodesic curves. We propose a numerical algorithm for finding approximations of open geodesics on meshes and surfaces based on geodesic curvature flow. For practical reasons, it is important to automatize the choice of the time step. We propose a method for automatic setting of the time step based on the diagonal dominance criterion for the matrix of the linear system obtained by discretization of our partial differential equation model. Practical experiments show reliability of this method. Because approximation of the model is made by numerical method based on classic derivatives, it is necessary to solve obstacles which occur for meshes with sharp corners. We solve this problem for big family of meshes with sharp corners via special rotations which can be seen as partial unfolding of the mesh. In practical applications, it is required that the approximation of geodesic has its vertices only on the edges of the mesh. This problem is solved by a specially designed pointing tracking algorithm. We also partially solve the problem of finding geodesics on meshes with holes. We implemented the whole algorithm in Rhinoceros (commercial 3D computer graphics and computer-aided design software ). It is done by using C# language as C# assembly library for Grasshopper, which is plugin in Rhinoceros.

Keywords: geodesic, geodesic curvature flow, mesh, Rhinoceros software

Procedia PDF Downloads 146

Authors: A. Samer Ezeldin, Jwanda M. El Sarag

Abstract:

Variation orders are of great importance in any construction project. Variation orders are defined as any change in the scope of works of a project that can be an addition omission, or even modification. This paper investigates the variation orders that occur during construction projects in Egypt. The literature review represents a comparison of causes of variation orders among Egypt, Tanzania, Nigeria, Malaysia and the United Kingdom. A classification of occurrence of variation orders due to owner related factors, consultant related factors and other factors are signified in the literature review. These classified events that lead to variation orders were introduced in a survey with 19 events to observe their frequency of occurrence, and their time and cost impacts. The survey data was obtained from 87 participants that included clients, consultants, and contractors and a database of 42 scenarios was created. A model is then developed to help assist project managers in predicting the frequency of variations and account for a budget for any additional costs and minimize any delays that can take place. Two experts with more than 25 years of experience were given the model to verify that the model was working effectively. The model was then validated on a residential compound that was completed in July 2016 to prove that the model actually produces acceptable results.

Keywords: construction, cost impact, Egypt, time impact, variation orders

Procedia PDF Downloads 179

Authors: Tayaa Karima, Bouaziz Amina

Abstract:

Today's world is knowledge-based; everything we do is somehow connected with technology which it has a remarkable influence on socio-cultural and economic developments, including educational settings. This type of technology is supported in many teaching/learning setting where the medium of instruction is through computer technology, and particularly involving digital technologies. There has been much debate over the use of computers and the internet in foreign language teaching for more than two decades. Various studies highlights that the integration of Information Communications Technology (ICT) in foreign language teaching will have positive effects on both the teachers and students to help them be aware of the modernized world and meet the current demands of the globalised world. Information and communication technology has been gradually integrated in foreign learning environment as a platform for providing learners with learning opportunities. Thus, the impact of ICT on language teaching and learning has been acknowledged globally, this is because of the fundamental role that it plays in the enhancement of teaching and learning quality, modify the pedagogical practice, and motivate learners. Due to ICT related developments, many Maghreb countries regard ICT as a tool for changes and innovations in education. Therefore, the ministry of education attempted to set up computer laboratories and provide internet connection in the schools. Investment in ICT for educational innovations and improvement purposes has been continuing the need of teacher who will employ it in the classroom as vital role of the curriculum. ICT does not have an educational value in itself, but it becomes precious when teachers use it in learning and teaching process. This paper examines the impacts of ICT on teacher development rather than on teaching quality and highlights some challenges facing using ICT in the language learning/teaching.

Keywords: information communications technology (ICT), integration, foreign language teaching, teacher development, learning opportunity

Procedia PDF Downloads 386

Authors: Fayyaz Rasool, Shakeela Parveen

Abstract:

The studies on genetic diversity of Labeo rohita by using molecular markers were carried out to investigate the genetic structure by RAPAD marker and the levels of polymorphism and similarity amongst the different groups of five populations of wild and farmed types. The samples were collected from different five locations as representatives of wild and hatchery raised populations. RAPAD data for Jaccard’s coefficient by following the un-weighted Pair Group Method with Arithmetic Mean (UPGMA) for Hierarchical Clustering of the similar groups on the basis of similarity amongst the genotypes and the dendrogram generated divided the randomly selected individuals of the five populations into three classes/clusters. The variance decomposition for the optimal classification values remained as 52.11% for within class variation, while 47.89% for the between class differences. The Principal Component Analysis (PCA) for grouping of the different genotypes from the different environmental conditions was done by Spearman Varimax rotation method for bi-plot generation of the co-occurrence of the same genotypes with similar genetic properties and specificity of different primers indicated clearly that the increase in the number of factors or components was correlated with the decrease in eigenvalues. The Kaiser Criterion based upon the eigenvalues greater than one, first two main factors accounted for 58.177% of cumulative variability.

Keywords: variation, clustering, PCA, wild, hatchery, RAPAD, Labeo rohita

Procedia PDF Downloads 447

Authors: Belkacem Selma, Boumediene Selma, Tourkia Guerzou, Abbes Labdelli

Abstract:

Natural disasters have occurred and will continue to cause human and material damage. Therefore, the idea of "preventing" natural disasters will never be possible. However, their prediction is possible with the advancement of technology. Even if natural disasters are effectively inevitable, their consequences may be partly controlled. The rapid growth and progress of artificial intelligence (AI) had a major impact on the prediction of natural disasters and risk assessment which are necessary for effective disaster reduction. The Earthquakes prediction to prevent the loss of human lives and even property damage is an important factor; that is why it is crucial to develop techniques for predicting this natural disaster. This present study aims to analyze the ability of artificial neural networks (ANNs) to predict earthquakes that occur in a given area. The used data describe the problem of high energy (higher than 10^4J) seismic bumps forecasting in a coal mine using two long walls as an example. For this purpose, seismic bumps data obtained from mines has been analyzed. The results obtained show that the ANN with high accuracy was able to predict earthquake parameters; the classification accuracy through neural networks is more than 94%, and that the models developed are efficient and robust and depend only weakly on the initial database.

Keywords: earthquake prediction, ANN, seismic bumps

Procedia PDF Downloads 126

Authors: Ebenezer O. Olaniyi, Oyebade K. Oyedotun, Khashman Adnan

Abstract:

In the food industry high quality production is required within a limited time to meet up with the demand in the society. In this research work, we have developed a model which can be used to replace the human operator due to their low output in production and slow in making decisions as a result of an individual differences in deciding the defective and healthy banana. This model can perform the vision attributes of human operators in deciding if the banana is defective or healthy for food production based. This research work is divided into two phase, the first phase is the image processing where several image processing techniques such as colour conversion, edge detection, thresholding and morphological operation were employed to extract features for training and testing the network in the second phase. These features extracted in the first phase were used in the second phase; the classification system phase where the multilayer perceptron using backpropagation neural network was employed to train the network. After the network has learned and converges, the network was tested with feedforward neural network to determine the performance of the network. From this experiment, a recognition rate of 97% was obtained and the time taken for this experiment was limited which makes the system accurate for use in the food industry.

Keywords: banana, food processing, identification system, neural network

Procedia PDF Downloads 467

Authors: Qasim Idrees, Gao Liangtian, Liu Bo, Miao Yiran

Abstract:

In order to improve the efficiency and accuracy of the pressure hull structure, optimization of underwater vehicle based on response surface methodology, a method for optimizing the design of pressure hull structure was studied. To determine the pressure shell of five dimensions as a design variable, the application of thin shell theory and the Chinese Classification Society (CCS) specification was carried on the preliminary design. In order to optimize variables of the feasible region, different methods were studied and implemented such as Opt LHD method (to determine the design test sample points in the feasible domain space), parametric ABAQUS solution for each sample point response, and the two-order polynomial response for the surface model of the limit load of structures. Based on the ultimate load of the structure and the quality of the shell, the two-generation genetic algorithm was used to solve the response surface, and the Pareto optimal solution set was obtained. The final optimization result was 41.68% higher than that of the initial design, and the shell quality was reduced by about 27.26%. The parametric method can ensure the accuracy of the test and improve the efficiency of optimization.

Keywords: parameterization, response surface, structure optimization, pressure hull

Procedia PDF Downloads 232

Authors: Fatma El-Zahraa El-Taher, Ayman Taha, Jane Courtney, Susan Mckeever

Abstract:

Understanding road networks plays an important role in navigation applications such as self-driving vehicles and route planning for individual journeys. Intersections of roads are essential components of road networks. Understanding the features of an intersection, from a simple T-junction to larger multi-road junctions, is critical to decisions such as crossing roads or selecting the safest routes. The identification and profiling of intersections from satellite images is a challenging task. While deep learning approaches offer the state-of-the-art in image classification and detection, the availability of training datasets is a bottleneck in this approach. In this paper, a labelled satellite image dataset for the intersection recognition problem is presented. It consists of 14,692 satellite images of Washington DC, USA. To support other users of the dataset, an automated download and labelling script is provided for dataset replication. The challenges of construction and fine-grained feature labelling of a satellite image dataset is examined, including the issue of how to address features that are spread across multiple images. Finally, the accuracy of the detection of intersections in satellite images is evaluated.

Keywords: satellite images, remote sensing images, data acquisition, autonomous vehicles

Procedia PDF Downloads 143

Authors: Joao Garretto, R. J. Yarwood, Vamsi Borra, Frank Li

Abstract:

Bluetooth Low Energy (BLE) has emerged as one of the main wireless communication technologies used in low-power electronics, such as wearables, beacons, and Internet of Things (IoT) devices. BLE’s energy efficiency characteristic, smart mobiles interoperability, and Over the Air (OTA) capabilities are essential features for ultralow-power devices, which are usually designed with size and cost constraints. Most current research regarding the power analysis of BLE devices focuses on the theoretical aspects of the advertising and scanning cycles, with most results being presented in the form of mathematical models and computer software simulations. Such computer modeling and simulations are important for the comprehension of the technology, but hardware measurement is essential for the understanding of how BLE devices behave in real operation. In addition, recent literature focuses mostly on the BLE technology, leaving possible applications and its analysis out of scope. In this paper, a coin cell battery-powered BLE Data Acquisition Device, with a 4-in-1 sensor and one accelerometer, is proposed and evaluated with respect to its Power Consumption. First, evaluations of the device in advertising mode with the sensors turned off completely, followed by the power analysis when each of the sensors is individually turned on and data is being transmitted, and concluding with the power consumption evaluation when both sensors are on and respectively broadcasting the data to a mobile phone. The results presented in this paper are real-time measurements of the electrical current consumption of the BLE device, where the energy levels that are demonstrated are matched to the BLE behavior and sensor activity.

Keywords: bluetooth low energy, power analysis, BLE advertising cycle, wireless sensor node

Procedia PDF Downloads 90

Authors: Iftikhar A. Tayubi, Aabdulrahman Alsubhi, Abdullah Althrwi

Abstract:

The deoxyribonucleic acid text provides a single source of high-quality Cryptography about Deoxyribonucleic acid sequence for structural biologists. We will provide an intuitive, well-organized and user-friendly web interface that allows users to encrypt and decrypt Deoxy Ribonucleic Acid sequence text. It includes complex, securing by using Algorithm to encrypt and decrypt Deoxy Ribonucleic Acid sequence. The utility of this Deoxy Ribonucleic Acid Sequence Text is that, it can provide a user-friendly interface for users to Encrypt and Decrypt store the information about Deoxy Ribonucleic Acid sequence. These interfaces created in this project will satisfy the demands of the scientific community by providing fully encrypt of Deoxy Ribonucleic Acid sequence during this website. We have adopted a methodology by using C# and Active Server Page.NET for programming which is smart and secure. Deoxy Ribonucleic Acid sequence text is a wonderful piece of equipment for encrypting large quantities of data, efficiently. The users can thus navigate from one encoding and store orange text, depending on the field for user’s interest. Algorithm classification allows a user to Protect the deoxy ribonucleic acid sequence from change, whether an alteration or error occurred during the Deoxy Ribonucleic Acid sequence data transfer. It will check the integrity of the Deoxy Ribonucleic Acid sequence data during the access.

Keywords: algorithm, ASP.NET, DNA, encrypt, decrypt

Procedia PDF Downloads 230

Authors: Daria Reczynska, Justyna Jarczak, Michal Czopowicz, Danuta Sloniewska, Karina Horbanczuk, Wieslaw Jarmuz, Jaroslaw Kaba, Emilia Bagnicka

Abstract:

Since people, animals and plants are constantly exposed to pathogens they have developed very complex systems of defense. Among ca. 1000 antimicrobial peptides from different families so far identified, approximately 30 belonging to cathelicidin family can be found in mammals. Cathelicidins probably constitute the first line of defense because they can act at a physiological salt concentration which is present in healthy tissues. Moreover, the low salt concentration which is present in infected tissues inhibits their activity. In goat bactenecin 7.5 (BAC7.5), bactenecin 5 (BAC5), myeloid antimicrobial peptide 28 (MAP28), myeloid antimicrobial peptide 34 (MAP34 A and B), goat bactenecin3.4 (ChBac3.4) were identified. Caprine arthritis-encephalitis (CAE) caused by small ruminant lentivirus (SRLV) is economic problem. The main CAE symptoms are weight loss, arthritis, pneumonia and mastitis (significant elevation of the somatic cell count and deterioration of some technological parameters). The study was conducted on 24 dairy goats. The animals were divided into two groups: experimental (SRLV-infected) and control (non-infected). The blood samples were collected five times: on the 1st, 7th, 30th, 90th and 150thday of lactation. The levels of transcripts of BAC7.5, BAC5, MAP28 and MAP34 genes in blood leucocytes were measured using qPCR method. There were no differences in mRNA levels of studied genes between stages of lactation. The differences were observed in expressions of BAC5, MAP28 and MAP34 genes with lower levels in the experimental group. There was no difference in BAC7.5 expression between groups. The decreased levels of transcripts of cathelicidin genes in blood leucocytes of SRLV-infected goats may indicate the disturbances of homeostasis in organisms. It can be concluded that SRLV infection seems to inhibit expression of cathelicidin genes. The study was financed by a grant from the National Scientific Center No. UMO-2013/09/B/NZ/03514.

Keywords: goat, CAEV, cathelicidins, blood leukocytes, gene expression

Procedia PDF Downloads 282

Authors: Zerihun Ketema Tadesse, Dabbu Suman Reddy

Abstract:

Brain-Computer Interfaces (BCIs) give the possibility for disabled people to communicate and control devices. This work aims at developing steady-state visual evoked potential (SSVEP)-based BCI for patients with limb disabilities. In hospitals, devices like nurse emergency call devices, lights, and TV sets are what patients use most frequently, but these devices are operated manually or using the remote control. Thus, disabled patients are not able to operate these devices by themselves. Hence, SSVEP-based BCI system that can allow disabled patients to control nurse calling device and other devices is proposed in this work. Portable LED visual stimulator that flickers at specific frequencies of 7Hz, 8Hz, 9Hz and 10Hz were developed as part of this project. Disabled patients can stare at specific flickering LED of visual stimulator and Emotiv EPOC used to acquire EEG signal in a non-invasive way. The acquired EEG signal can be processed to generate various control signals depending upon the amplitude and duration of signal components. MATLAB software is used for signal processing and analysis and also for command generation. Arduino is used as a hardware interface device to receive and transmit command signals to the experimental setup. Therefore, this study is focused on the design and development of Steady-state visually evoked potential (SSVEP)-based BCI for limb disabled patients, which helps them to operate and control devices in the hospital room/wards.

Keywords: SSVEP-BCI, Limb Disabled Patients, LED Visual Stimulator, EEG signal, control devices, hospital room/wards

Procedia PDF Downloads 220

Authors: Luis Fonseca, Filipe Cabral Pinto, Susana Sargento

Abstract:

The increase of the world population, especially in large urban centers, has resulted in new challenges particularly with the control and optimization of public safety. Thus, in the present work, a solution is proposed for the prediction of criminal occurrences in a city based on historical data of incidents and demographic information. The entire research and implementation will be presented start with the data collection from its original source, the treatment and transformations applied to them, choice and the evaluation and implementation of the Machine Learning model up to the application layer. Classification models will be implemented to predict criminal risk for a given time interval and location. Machine Learning algorithms such as Random Forest, Neural Networks, K-Nearest Neighbors and Logistic Regression will be used to predict occurrences, and their performance will be compared according to the data processing and transformation used. The results show that the use of Machine Learning techniques helps to anticipate criminal occurrences, which contributed to the reinforcement of public security. Finally, the models were implemented on a platform that will provide an API to enable other entities to make requests for predictions in real-time. An application will also be presented where it is possible to show criminal predictions visually.

Keywords: crime prediction, machine learning, public safety, smart city

Procedia PDF Downloads 110

Authors: Khaled Nagaty, Heba Nagaty, Gerard McKee

Abstract:

A signature is a behavioral biometric that is used for authenticating users in most financial and legal transactions. Signatures can be easily forged by skilled forgers. Therefore, it is essential to verify whether a signature is genuine or forged. The aim of any signature verification algorithm is to accommodate the differences between signatures of the same person and increase the ability to discriminate between signatures of different persons. This work presented in this paper proposes an automatic signature verification system to indicate whether a signature is genuine or not. The system comprises four phases: (1) The pre-processing phase in which image scaling, binarization, image rotation, dilation, thinning, and connecting ridge breaks are applied. (2) The feature extraction phase in which global and local features are extracted. The local features are minutiae points, curvature orientation, and curve plateau. The global features are signature area, signature aspect ratio, and Hu moments. (3) The post-processing phase, in which false minutiae are removed. (4) The classification phase in which features are enhanced before feeding it into the classifier. k-nearest neighbors and support vector machines are used. The classifier was trained on a benchmark dataset to compare the performance of the proposed offline signature verification system against the state-of-the-art. The accuracy of the proposed system is 92.3%.

Keywords: signature, ridge breaks, minutiae, orientation

Procedia PDF Downloads 144

Authors: Mo-Se Lee, Cheol-Hwi Ahn, Kee-Young Kwahk, Hyunchul Ahn

Abstract:

Over the past decade, deep learning has been in spotlight among various machine learning algorithms. In particular, CNN (Convolutional Neural Network), which is known as effective solution for recognizing and classifying images, has been popularly applied to classification and prediction problems in various fields. In this study, we try to apply CNN to stock market prediction, one of the most challenging tasks in the machine learning research. In specific, we propose to apply CNN as the binary classifier that predicts stock market direction (up or down) by using a graph as its input. That is, our proposal is to build a machine learning algorithm that mimics a person who looks at the graph and predicts whether the trend will go up or down. Our proposed model consists of four steps. In the first step, it divides the dataset into 5 days, 10 days, 15 days, and 20 days. And then, it creates graphs for each interval in step 2. In the next step, CNN classifiers are trained using the graphs generated in the previous step. In step 4, it optimizes the hyper parameters of the trained model by using the validation dataset. To validate our model, we will apply it to the prediction of KOSPI200 for 1,986 days in eight years (from 2009 to 2016). The experimental dataset will include 14 technical indicators such as CCI, Momentum, ROC and daily closing price of KOSPI200 of Korean stock market.

Keywords: convolutional neural network, deep learning, Korean stock market, stock market prediction

Procedia PDF Downloads 424

Authors: Cheng-Jhe Lee, Chiun-Chieh Hsu

Abstract:

Due to the rapid development of the Internet and the increased availability of digital documents, the excessive information on the Internet has led to information overflow problem. In order to solve these problems for effective information retrieval, document clustering in text mining becomes a popular research topic. Clustering is the unsupervised classification of data items into groups without the need of training data. Many conventional document clustering methods perform inefficiently for large document collections because they were originally designed for relational database. Therefore they are impractical in real-world document clustering and require special handling for high dimensionality and high volume. We propose the FIHC (Frequent Itemset-based Hierarchical Clustering) method, which is a hierarchical clustering method developed for document clustering, where the intuition of FIHC is that there exist some common words for each cluster. FIHC uses such words to cluster documents and builds hierarchical topic tree. In this paper, we combine FIHC algorithm with ontology to solve the semantic problem and mine the meaning behind the words in documents. Furthermore, we use the closed frequent itemsets instead of only use frequent itemsets, which increases efficiency and scalability. The experimental results show that our method is more accurate than those of well-known document clustering algorithms.

Keywords: FIHC, documents clustering, ontology, closed frequent itemset

Procedia PDF Downloads 397

Authors: Mina Adel Shokry Fahim, Jūratė Sužiedelytė Visockienė

Abstract:

With the growing concern over air pollution (AP), it is clear that this has gained more prominence than ever before. The level of consciousness has increased and a sense of knowledge now has to be forwarded as a duty by those enlightened enough to disseminate it to others. This realisation often comes after an understanding of how poor air quality indices (AQI) damage human health. The study focuses on assessing air pollution prediction models specifically for Lithuania, addressing a substantial need for empirical research within the region. Concentrating on Vilnius, it specifically examines particulate matter concentrations 10 micrometers or less in diameter (PM10). Utilizing Gaussian Process Regression (GPR) and Regression Tree Ensemble, and Regression Tree methodologies, predictive forecasting models are validated and tested using hourly data from January 2020 to December 2022. The study explores the classification of AP data into anthropogenic and natural sources, the impact of AP on human health, and its connection to cardiovascular diseases. The study revealed varying levels of accuracy among the models, with GPR achieving the highest accuracy, indicated by an RMSE of 4.14 in validation and 3.89 in testing.

Keywords: air pollution, anthropogenic and natural sources, machine learning, Gaussian process regression, tree ensemble, forecasting models, particulate matter

Procedia PDF Downloads 51

Authors: Muhammad Bilal Shaikh

Abstract:

Mineral processing plants play a pivotal role in extracting valuable minerals from raw ores, contributing significantly to various industries. However, the optimization of plant throughput remains a complex challenge, necessitating innovative approaches for increased efficiency and productivity. This research paper investigates the application of Multimodal Artificial Intelligence (MAI) techniques to address this challenge, aiming to improve overall plant throughput in mineral processing operations. The integration of multimodal AI leverages a combination of diverse data sources, including sensor data, images, and textual information, to provide a holistic understanding of the complex processes involved in mineral extraction. The paper explores the synergies between various AI modalities, such as machine learning, computer vision, and natural language processing, to create a comprehensive and adaptive system for optimizing mineral processing plants. The primary focus of the research is on developing advanced predictive models that can accurately forecast various parameters affecting plant throughput. Utilizing historical process data, machine learning algorithms are trained to identify patterns, correlations, and dependencies within the intricate network of mineral processing operations. This enables real-time decision-making and process optimization, ultimately leading to enhanced plant throughput. Incorporating computer vision into the multimodal AI framework allows for the analysis of visual data from sensors and cameras positioned throughout the plant. This visual input aids in monitoring equipment conditions, identifying anomalies, and optimizing the flow of raw materials. The combination of machine learning and computer vision enables the creation of predictive maintenance strategies, reducing downtime and improving the overall reliability of mineral processing plants. Furthermore, the integration of natural language processing facilitates the extraction of valuable insights from unstructured textual data, such as maintenance logs, research papers, and operator reports. By understanding and analyzing this textual information, the multimodal AI system can identify trends, potential bottlenecks, and areas for improvement in plant operations. This comprehensive approach enables a more nuanced understanding of the factors influencing throughput and allows for targeted interventions. The research also explores the challenges associated with implementing multimodal AI in mineral processing plants, including data integration, model interpretability, and scalability. Addressing these challenges is crucial for the successful deployment of AI solutions in real-world industrial settings. To validate the effectiveness of the proposed multimodal AI framework, the research conducts case studies in collaboration with mineral processing plants. The results demonstrate tangible improvements in plant throughput, efficiency, and cost-effectiveness. The paper concludes with insights into the broader implications of implementing multimodal AI in mineral processing and its potential to revolutionize the industry by providing a robust, adaptive, and data-driven approach to optimizing plant operations. In summary, this research contributes to the evolving field of mineral processing by showcasing the transformative potential of multimodal artificial intelligence in enhancing plant throughput. The proposed framework offers a holistic solution that integrates machine learning, computer vision, and natural language processing to address the intricacies of mineral extraction processes, paving the way for a more efficient and sustainable future in the mineral processing industry.

Keywords: multimodal AI, computer vision, NLP, mineral processing, mining

Procedia PDF Downloads 65

Authors: Nikola Tomasevic, Marko Batic, Sanja Vranes

Abstract:

Emerging Semantic Web technologies can be seen as the next step in evolution of the intelligent facility management systems. Particularly, this considers increased usage of open source and/or standardized concepts for data classification and semantic interpretation. To deliver such facility management systems, providing the comprehensive integration and interoperability platform in from of the facility data model is a prerequisite. In this paper, one of the possible modelling approaches to provide such integrative facility data model which was based on the ontology modelling concept was presented. Complete ontology development process, starting from the input data acquisition, ontology concepts definition and finally ontology concepts population, was described. At the beginning, the core facility ontology was developed representing the generic facility infrastructure comprised of the common facility concepts relevant from the facility management perspective. To develop the data model of a specific facility infrastructure, first extension and then population of the core facility ontology was performed. For the development of the full-blown facility data models, Malpensa and Fiumicino airports in Italy, two major European air-traffic hubs, were chosen as a test-bed platform. Furthermore, the way how these ontology models supported the integration and interoperability of the overall airport energy management system was analyzed as well.

Keywords: airport ontology, energy management, facility data model, ontology modeling

Procedia PDF Downloads 446

Authors: Tung-Kuang Wu, Hsing-Pei Hsieh, Ying-Ru Meng

Abstract:

Special Education in Taiwan has been facing difficulties including shortage of teachers and lack in resources. Some students need to receive special education are thus not identified or admitted. Fortunately, information technologies can be applied to relieve some of the difficulties. For example, on-line multimedia courseware can be used to assist the learning of special education students and take pretty much workload from special education teachers. However, there may exist cognitive variations between students in special or regular educations, which suggests the design of online courseware requires different considerations. This study aims to investigate the difference in focus of attention (FOA) between special and regular education students of primary school in viewing the computer screen. The study is essential as it helps courseware developers in determining where to put learning elements that matter the most on the right position of screen. It may also assist special education specialists to better understand the subtle differences among various subtypes of learning disabilities. This study involves 76 special education students (among them, 39 are students with mental retardation, MR, and 37 are students with learning disabilities, LDs) and 42 regular education students. The participants were asked to view a computer screen showing a picture partitioned into 3 × 3 areas with each area filled with text or icon. The subjects were then instructed to mark on the prior given paper sheets, which are also partitioned into 3 × 3 grids, the areas corresponding to the pictures on the computer screen that they first set their eyes on. The data are then collected and analyzed. Major findings are listed: 1. In both text and icon scenario, significant differences exist in the first preferred FOA between special and regular education students. The first FOA for the former is mainly on area 1 (upper left area, 53.8% / 51.3% for MR / LDs students in text scenario; and 53.8% / 56.8% for MR / LDs students in icons scenario), while the latter on area 5 (middle area, 50.0% and 57.1% in text and icons scenarios). 2. The second most preferred area in text scenario for students with MR and LDs are area 2 (upper-middle, 20.5%) and 5 (middle area, 24.3%). In icons scenario, the results are similar, but lesser in percentage. 3. Students with LDs that show similar preference (either in text or icons scenarios) in FOA to regular education students tend to be of some specific sub-type of learning disabilities. For instance, students with LDs that chose area 5 (middle area, either in text or icon scenario) as their FOA are mostly ones that have reading or writing disability. Also, three (out of 13) subjects in this category, after going through the rediagnosis process, were excluded from being learning disabilities. In summary, the findings suggest when designing multimedia courseware for students with MR and LDs, the essential learning elements should be placed on area 1, 2 and 5. In addition, FOV preference may also potentially be used as an indicator for diagnosing students with LDs.

Keywords: focus of attention, learning disabilities, mental retardation, on-line multimedia courseware, special education

Procedia PDF Downloads 163

Authors: Win Thandar Soe, Innocent Mpawenimana, Mathieu Di Fazio, Cécile Belleudy, Aung Ze Ya

Abstract:

Intelligent electronic equipment and automation network is the brain of high-tech energy management systems in critical role of smart homes dominance. Smart home is a technology integration for greater comfort, autonomy, reduced cost, and energy saving as well. These services can be provided to home owners for managing their home appliances locally or remotely and consequently allow them to automate intelligently and responsibly their consumption by individual or collective control systems. In this study, three smart plugs are described and one of them tested on typical household appliances. This article proposes to collect the data from the wireless technology and to extract some smart data for energy management system. This smart data is to quantify for three kinds of load: intermittent load, phantom load and continuous load. Phantom load is a waste power that is one of unnoticed power of each appliance while connected or disconnected to the main. Intermittent load and continuous load take in to consideration the power and using time of home appliances. By analysing the classification of loads, this smart data will be provided to reduce the communication of wireless sensor network for energy management system.

Keywords: energy management, load profile, smart plug, wireless sensor network

Procedia PDF Downloads 271

Authors: Ebru Turgal, Beyza Doganay Erdogan

Abstract:

Machine learning aims to model the relationship between the response and features. Medical decision-making researchers would like to make decisions about patients’ course and treatment, by examining the repeated measurements over time. Boosting approach is now being used in machine learning area for these aims as an influential tool. The aim of this study is to show the usage of multivariate tree boosting in this field. The main reason for utilizing this approach in the field of decision-making is the ease solutions of complex relationships. To show how multivariate tree boosting method can be used to identify important features and feature-time interaction, we used the data, which was collected retrospectively from Ankara University Chest Diseases Department records. Dataset includes repeated PF ratio measurements. The follow-up time is planned for 120 hours. A set of different models is tested. In conclusion, main idea of classification with weighed combination of classifiers is a reliable method which was shown with simulations several times. Furthermore, time varying variables will be taken into consideration within this concept and it could be possible to make accurate decisions about regression and survival problems.

Keywords: boosted multivariate trees, longitudinal data, multivariate regression tree, panel data

Procedia PDF Downloads 201

Authors: L. Meers, L. Contalbrigo, V. Stevens, O. Ulitina, S. Laufer, W. E. Samuels, S. Normando

Abstract:

Since the COVID-19 pandemic started, there has been concern that domestic animals may help spread SARS-Cov-2. This concern also greatly affected human-animal interaction projects such as animal-assisted interventions (AAI). As a result, institutions and AAI practitioners developed new safety protocols and procedures to control the spread of the SARS-Cov-2 virus during AAI sessions and to guarantee safety for their clients and animals. However, little is known yet about the impact on animals' needs and the possible welfare issues due to these lifestyle adaptions. Fifteen therapists in Flanders, Belgium, who were currently conducting canine-assisted interventions, conducted unstructured observations on how their dogs' (11 mixed breeds, 3 Labradors, 1 terrier aged 2 – 12 years) behaviors changed due to institutional COVID-19 safety protocols. Most (80%) of the respondents reported that their dogs showed sniffing or sneezing after smelling disinfected areas. Two (13%) dogs responded with vomiting and gagging, and three (20%) dogs urinated over disinfected areas. All protocols advise social distancing between participants and animals. When held back, eight (53%) dogs showed self-calming behaviors. Respondents reported that most (73%) dogs responded with flight reactions when seeing humans wearing facial masks. When practitioners threw their used masks in open dustbins, five (33%) dogs tried to take them out with their mouths and play with them; two (13%) Labradors tried to eat them. Taking the dogs' temperatures was the most frequently (53%) used method to supervise their health. However, all dogs showed behaviors as ducking the tail, trying to escape, or biting the animal handler during this procedure. We interpret these results to suggest that dogs tended to react with stress and confusion to the changes in AAI practices they're part of. The health and safety protocols that institutions used were largely borne from recommendations made to protect humans. The participating practitioners appeared to use their knowledge of dog behavior and safety to modify them as best they could—but with more significant concern directed towards the other humans. Given their inter-relatedness and mutual importance for welfare, we advocate for integrated human and animal health and welfare assessments and protocols to provide a framework for "One health" approaches in animal-assisted interventions.

Keywords: animal-assisted therapy, COVID-19 protocol, one health, welfare

Procedia PDF Downloads 200

Authors: Dang Thuy Binh, Truong Thi Oanh, Le Phan Khanh Hung, Luong thi Tuong Vy

Abstract:

One of the greatest environmental challenges facing Asia is the management and conservation of the marine biodiversity threaten by fisheries overexploitation, pollution, habitat destruction, and climate change. To date, a few molecular taxonomical studies has been conducted on marine fauna in Vietnam. The purpose of this study was to clarify the phylogeny of economic and ecological reef fish species in Vietnam Reef fish species covering Labridae, Scaridae, Nemipteridae, Serranidae, Acanthuridae, Lutjanidae, Lethrinidae, Mullidae, Balistidae, Pseudochromidae, Pinguipedidae, Fistulariidae, Holocentridae, Synodontidae, and Pomacentridae representing 28 genera were collected from South and Center, Vietnam. Combine with Genbank sequences, a phylogenetic tree was constructed based on 16S gene of mitochondrial DNA using maximum parsimony, maximum likelihood, and Bayesian inference approaches. The phylogram showed the well-resolved clades at genus and family level. Perciformes is the major order of reef fish species in Vietnam. The monophyly of Perciformes is not strongly supported as it was clustered in the same clade with Tetraodontiformes syngnathiformes and Beryciformes. Continue sampling of commercial fish species and classification based on morphology and genetics to build DNA barcoding of fish species in Vietnam is really necessary.

Keywords: reef fish, 16s rDNA, Vietnam, phylogeny

Procedia PDF Downloads 437