Search results for: graph convolutional network

Authors: Gulfam Haider, sana danish

Abstract:

Convolutional Neural Networks (CNNs) have emerged as powerful tools for image classification, and the choice of optimizers profoundly affects their performance. The study of optimizers and their adaptations remains a topic of significant importance in machine learning research. While numerous studies have explored and advocated for various optimizers, the efficacy of these optimization techniques is still subject to scrutiny. This work aims to address the challenges surrounding the effectiveness of optimizers by conducting a comprehensive analysis and evaluation. The primary focus of this investigation lies in examining the performance of different optimizers when employed in conjunction with the popular activation function, Rectified Linear Unit (ReLU). By incorporating ReLU, known for its favorable properties in prior research, the aim is to bolster the effectiveness of the optimizers under scrutiny. Specifically, we evaluate the adjustment of these optimizers with both the original Softmax activation function and the modified ReLU activation function, carefully assessing their impact on overall performance. To achieve this, a series of experiments are conducted using a well-established benchmark dataset for image classification tasks, namely the Canadian Institute for Advanced Research dataset (CIFAR-10). The selected optimizers for investigation encompass a range of prominent algorithms, including Adam, Root Mean Squared Propagation (RMSprop), Adaptive Learning Rate Method (Adadelta), Adaptive Gradient Algorithm (Adagrad), and Stochastic Gradient Descent (SGD). The performance analysis encompasses a comprehensive evaluation of the classification accuracy, convergence speed, and robustness of the CNN models trained with each optimizer. Through rigorous experimentation and meticulous assessment, we discern the strengths and weaknesses of the different optimization techniques, providing valuable insights into their suitability for image classification tasks. By conducting this in-depth study, we contribute to the existing body of knowledge surrounding optimizers in CNNs, shedding light on their performance characteristics for image classification. The findings gleaned from this research serve to guide researchers and practitioners in making informed decisions when selecting optimizers and activation functions, thus advancing the state-of-the-art in the field of image classification with convolutional neural networks.

Keywords: deep neural network, optimizers, RMsprop, ReLU, stochastic gradient descent

Procedia PDF Downloads 125

Authors: Jun Seung Woo, Shin Dong Ho

Abstract:

The Image Captioning models with Attention technology have developed significantly compared to previous models, but it is still unsatisfactory in recognizing images. We perform an extensive search over seven interesting Convolutional Neural Networks(CNN) architectures to analyze the behavior of different models for image captioning. We compared seven different CNN Architectures, according to batch size, using on public benchmarks: MS-COCO datasets. In our experimental results, DenseNet and InceptionV3 got about 14% loss and about 160sec training time per epoch. It was the most satisfactory result among the seven CNN architectures after training 50 epochs on GPU.

Keywords: deep learning, image captioning, CNN architectures, densenet, inceptionV3

Procedia PDF Downloads 133

Authors: Prat Yotyawilai, Taratip Suwannasart

Abstract:

Business Process Model and Notation (BPMN) is more important in the business process and creating functional models, and is a standard for OMG, which becomes popular in various organizations and in education. Researches related to software testing based on models are prominent. Although most researches use the UML model in software testing, not many researches use the BPMN Model in creating test cases. Therefore, this research proposes a design of a tool for generating test cases from the BPMN. The model is analyzed and the details of the various components are extracted before creating a flow graph. Both details of components and the flow graph are used in generating test cases.

Keywords: software testing, test case, BPMN, flow graph

Procedia PDF Downloads 555

Authors: Fengjuan Wang, Azilawati Jamaludin

Abstract:

Developmental Dyscalculia (DD) is defined as a particular learning difficulty with continuous challenges in learning requisite math skills that cannot be explained by intellectual disability or educational deprivation. Recent studies have increasingly recognized that DD is a heterogeneous, instead of monolithic, learning disorder with not only cognitive and behavioral deficits but so too neural dysfunction. In recent years, neuroimaging studies employed group comparison to explore the neural underpinnings of DD, which contradicted the heterogenous nature of DD and may obfuscate critical individual differences. This research aimed to investigate the neural heterogeneity of DD using case studies with functional near-infrared spectroscopy (fNIRS). A total of 54 aged 6-7 years old of children participated in this study, comprising two comprehensive cognitive assessments, an 8-minute resting state, and an 8-minute one-digit addition task. Nine children met the criteria of DD and scored at or below 85 (i.e., the 16th percentile) on the Mathematics or Math Fluency subtest of the Wechsler Individual Achievement Test, Third Edition (WIAT-III) (both subtest scores were 90 and below). The remaining 45 children formed the typically developing (TD) group. Resting-state data and brain activation in the inferior frontal gyrus (IFG), superior frontal gyrus (SFG), and intraparietal sulcus (IPS) were collected for comparison between each case and the TD group. Graph theory was used to analyze the brain network under the resting state. This theory represents the brain network as a set of nodes--brain regions—and edges—pairwise interactions across areas to reveal the architectural organizations of the nervous network. Next, a single-case methodology developed by Crawford et al. in 2010 was used to compare each case’s brain network indicators and brain activation against 45 TD children’s average data. Results showed that three out of the nine DD children displayed significant deviation from TD children’s brain indicators. Case 1 had inefficient nodal network properties. Case 2 showed inefficient brain network properties and weaker activation in the IFG and IPS areas. Case 3 displayed inefficient brain network properties with no differences in activation patterns. As a rise above, the present study was able to distill differences in architectural organizations and brain activation of DD vis-à-vis TD children using fNIRS and single-case methodology. Although DD is regarded as a heterogeneous learning difficulty, it is noted that all three cases showed lower nodal efficiency in the brain network, which may be one of the neural sources of DD. Importantly, although the current “brain norm” established for the 45 children is tentative, the results from this study provide insights not only for future work in “developmental brain norm” with reliable brain indicators but so too the viability of single-case methodology, which could be used to detect differential brain indicators of DD children for early detection and interventions.

Keywords: brain activation, brain network, case study, developmental dyscalculia, functional near-infrared spectroscopy, graph theory, neural heterogeneity

Procedia PDF Downloads 53

Authors: Michael Moeti, Khuliso Sigama, Thapelo Samuel Matlala

Abstract:

One of the principal challenges that the world is confronted with is insecurity. The crime rate is increasing exponentially, and protecting our physical assets especially in the motorist industry, is becoming impossible when applying our own strength. The need to develop technological solutions that detect and report theft without any human interference is inevitable. This is critical, especially for vehicle owners, to ensure theft detection and speedy identification towards recovery efforts in cases where a vehicle is missing or attempted theft is taking place. The vehicle theft detection system uses Convolutional Neural Network (CNN) to recognize the driver's face captured using an installed mobile phone device. The location identification function uses a Global Positioning System (GPS) to determine the real-time location of the vehicle. Upon identification of the location, Global System for Mobile Communications (GSM) technology is used to report or notify the vehicle owner about the whereabouts of the vehicle. The installed mobile app was implemented by making use of python as it is undoubtedly the best choice in machine learning. It allows easy access to machine learning algorithms through its widely developed library ecosystem. The graphical user interface was developed by making use of JAVA as it is better suited for mobile development. Google's online database (Firebase) was used as a means of storage for the application. The system integration test was performed using a simple percentage analysis. Sixty (60) vehicle owners participated in this study as a sample, and questionnaires were used in order to establish the acceptability of the system developed. The result indicates the efficiency of the proposed system, and consequently, the paper proposes the use of the system can effectively monitor the vehicle at any given place, even if it is driven outside its normal jurisdiction. More so, the system can be used as a database to detect, locate and report missing vehicles to different security agencies.

Keywords: CNN, location identification, tracking, GPS, GSM

Procedia PDF Downloads 167

Authors: Tsega Asresa, Getahun Tigistu, Melaku Bayih

Abstract:

Computer vision is a subfield of artificial intelligence that allows computers and systems to extract meaning from digital images and video. It is used in a wide range of fields of study, including self-driving cars, video surveillance, medical diagnosis, manufacturing, law, agriculture, quality control, health care, facial recognition, and military applications. Aromatic medicinal plants are botanical raw materials used in cosmetics, medicines, health foods, essential oils, decoration, cleaning, and other natural health products for therapeutic and Aromatic culinary purposes. These plants and their products not only serve as a valuable source of income for farmers and entrepreneurs but also going to export for valuable foreign currency exchange. In Ethiopia, there is a lack of technologies for the classification and identification of Aromatic medicinal plant parts and disease type cured by aromatic medicinal plants. Farmers, industry personnel, academicians, and pharmacists find it difficult to identify plant parts and disease types cured by plants before ingredient extraction in the laboratory. Manual plant identification is a time-consuming, labor-intensive, and lengthy process. To alleviate these challenges, few studies have been conducted in the area to address these issues. One way to overcome these problems is to develop a deep learning model for efficient identification of Aromatic medicinal plant parts with their corresponding disease type. The objective of the proposed study is to identify the aromatic medicinal plant parts and their disease type classification using computer vision technology. Therefore, this research initiated a model for the classification of aromatic medicinal plant parts and their disease type by exploring computer vision technology. Morphological characteristics are still the most important tools for the identification of plants. Leaves are the most widely used parts of plants besides roots, flowers, fruits, and latex. For this study, the researcher used RGB leaf images with a size of 128x128 x3. In this study, the researchers trained five cutting-edge models: convolutional neural network, Inception V3, Residual Neural Network, Mobile Network, and Visual Geometry Group. Those models were chosen after a comprehensive review of the best-performing models. The 80/20 percentage split is used to evaluate the model, and classification metrics are used to compare models. The pre-trained Inception V3 model outperforms well, with training and validation accuracy of 99.8% and 98.7%, respectively.

Keywords: aromatic medicinal plant, computer vision, convolutional neural network, deep learning, plant classification, residual neural network

Procedia PDF Downloads 187

Authors: Meng-Jou Chien, Jheng-Cheng Chen, Chang-Hsiung Tsai

Abstract:

Let maxζG(m) denote the maximum number of edges in a subgraph of graph G induced by m nodes. The n-dimensional augmented cube, denoted as AQn, a variation of the hypercube, possesses some properties superior to those of the hypercube. We study the cases when G is the augmented cube AQn.

Keywords: interconnection network, augmented cube, induced subgraph, bisection width

Procedia PDF Downloads 406

Authors: Hae-Yeoun Lee

Abstract:

Quantification of cardiac function is performed by calculating blood volume and ejection fraction in routine clinical practice. However, these works have been performed by manual contouring,which requires computational costs and varies on the observer. In this paper, an automatic left ventricle segmentation algorithm on cardiac magnetic resonance images (MRI) is presented. Using knowledge on cardiac MRI, a K-mean clustering technique is applied to segment blood region on a coil-sensitivity corrected image. Then, a graph searching technique is used to correct segmentation errors from coil distortion and noises. Finally, blood volume and ejection fraction are calculated. Using cardiac MRI from 15 subjects, the presented algorithm is tested and compared with manual contouring by experts to show outstanding performance.

Keywords: cardiac MRI, graph searching, left ventricle segmentation, K-means clustering

Procedia PDF Downloads 399

Authors: Subhash Chandra Sharma, Mohammad Soleimani

Abstract:

Selection of the correct lubricant for the top of rail application is a complex process. In this paper, the selection of the proper lubricant for a Top-Of-Rail (TOR) lubrication system based on graph theory and matrix approach has been developed. Attributes influencing the selection process and their influence on each other has been represented through a digraph and an equivalent matrix. A matrix function which is called the Permanent Function is derived. By substituting the level of inherent contribution of the influencing parameters and their influence on each other qualitatively, a criterion called Suitability Index is derived. Based on these indices, lubricants can be ranked for their suitability. The proposed model can be useful for maintenance engineers in selecting the best lubricant for a TOR application. The proposed methodology is illustrated step–by-step through an example.

Keywords: lubricant selection, top of rail lubrication, graph-theory, Ranking of lubricants

Procedia PDF Downloads 295

Authors: Thomas Adi Purnomo Sidhi

Abstract:

Nowadays, network is an essential need in almost every part of human daily activities. People now can seamlessly connect to others through the Internet. With advanced technology, our personal data now can be more easily accessed. One of many components we are concerned for delivering the best network is a security issue. This paper is proposing a method that provides more options for security. This research aims to improve network security by focusing on the physical layer which is the first layer of the OSI model. The layer consists of the basic networking hardware transmission technologies of a network. With the use of observation method, the research produces a schematic design for enhancing the network security through the gray code converter.

Keywords: network, network security, grey code, physical layer

Procedia PDF Downloads 504

Authors: E. R. Swart, S. J. Gismondi, N. R. Swart, C. E. Bell

Abstract:

We present an heuristic algorithm that decides graph non-Hamiltonicity. All graphs are directed, each undirected edge regarded as a pair of counter directed arcs. Each of the n! Hamilton cycles in a complete graph on n+1 vertices is mapped to an n-permutation matrix P where p(u,i)=1 if and only if the ith arc in a cycle enters vertex u, starting and ending at vertex n+1. We first create exclusion set E by noting all arcs (u, v) not in G, sufficient to code precisely all cycles excluded from G i.e. cycles not in G use at least one arc not in G. Members are pairs of components of P, {p(u,i),p(v,i+1)}, i=1, n-1. A doubly stochastic-like relaxed LP formulation of the Hamilton cycle decision problem is constructed. Each {p(u,i),p(v,i+1)} in E is coded as variable q(u,i,v,i+1)=0 i.e. shrinks the feasible region. We then implement the Weak Closure Algorithm (WCA) that tests necessary conditions of a matching, together with Boolean closure to decide 0/1 variable assignments. Each {p(u,i),p(v,j)} not in E is tested for membership in E, and if possible, added to E (q(u,i,v,j)=0) to iteratively maximize |E|. If the WCA constructs E to be maximal, the set of all {p(u,i),p(v,j)}, then G is decided non-Hamiltonian. Only non-Hamiltonian G share this maximal property. Ten non-Hamiltonian graphs (10 through 104 vertices) and 2000 randomized 31 vertex non-Hamiltonian graphs are tested and correctly decided non-Hamiltonian. For Hamiltonian G, the complement of E covers a matching, perhaps useful in searching for cycles. We also present an example where the WCA fails.

Keywords: Hamilton cycle decision problem, computational complexity theory, graph theory, theoretical computer science

Procedia PDF Downloads 373

Authors: Angela U. Makolo, Temitayo A. Olagunju

Abstract:

The knowledge of signaling pathways is central to understanding the biological mechanisms of organisms since it has been identified that in eukaryotic organisms, the number of signaling pathways determines the number of ways the organism will react to external stimuli. Signaling pathways are studied using protein interaction networks constructed from protein-protein interaction data obtained using high throughput experimental procedures. However, these high throughput methods are known to produce very high rates of false positive and negative interactions. In order to construct a useful protein interaction network from this noisy data, computational methods are applied to validate the protein-protein interactions. In this study, a computational technique to identify signaling pathways from a protein interaction network constructed using validated protein-protein interaction data was designed. A weighted interaction graph of the Saccharomyces cerevisiae (Baker’s Yeast) organism using the proteins as the nodes and interactions between them as edges was constructed. The weights were obtained using Bayesian probabilistic network to estimate the posterior probability of interaction between two proteins given the gene expression measurement as biological evidence. Only interactions above a threshold were accepted for the network model. A pathway was formalized as a simple path in the interaction network from a starting protein and an ending protein of interest. We were able to identify some pathway segments, one of which is a segment of the pathway that signals the start of the process of meiosis in S. cerevisiae.

Keywords: Bayesian networks, protein interaction networks, Saccharomyces cerevisiae, signalling pathways

Procedia PDF Downloads 545

Authors: Anika Chebrolu

Abstract:

Mono-targeted drugs can be of limited efficacy against complex diseases. Recently, multi-target drug design has been approached as a promising tool to fight against these challenging diseases. However, the scope of current computational approaches for multi-target drug design is limited. DeepLig presents a de-novo drug discovery platform that uses reinforcement learning to generate and optimize novel, potent, and multitargeted drug candidates against protein targets. DeepLig’s model consists of two networks in interplay: a generative network and a predictive network. The generative network, a Stack- Augmented Recurrent Neural Network, utilizes a stack memory unit to remember and recognize molecular patterns when generating novel ligands from scratch. The generative network passes each newly created ligand to the predictive network, which then uses multiple Graph Attention Networks simultaneously to forecast the average binding affinity of the generated ligand towards multiple target proteins. With each iteration, given feedback from the predictive network, the generative network learns to optimize itself to create molecules with a higher average binding affinity towards multiple proteins. DeepLig was evaluated based on its ability to generate multi-target ligands against two distinct proteins, multi-target ligands against three distinct proteins, and multi-target ligands against two distinct binding pockets on the same protein. With each test case, DeepLig was able to create a library of valid, synthetically accessible, and novel molecules with optimal and equipotent binding energies. We propose that DeepLig provides an effective approach to design multi-targeted drug therapies that can potentially show higher success rates during in-vitro trials.

Keywords: drug design, multitargeticity, de-novo, reinforcement learning

Procedia PDF Downloads 97

Authors: Ayşe Dilek Maden

Abstract:

For a given a simple connected graph, we present some new bounds via a new approach for a special topological index given by the sum of the real number power of the non-zero normalized Laplacian eigenvalues. To use this approach presents an advantage not only to derive old and new bounds on this topic but also gives an idea how some previous results in similar area can be developed.

Keywords: degree Kirchhoff index, normalized Laplacian eigenvalue, spanning tree, simple connected graph

Procedia PDF Downloads 366

Authors: Kenichiro Hida, Shin-Ichi Kuribayashi

Abstract:

NFV-based network implements a variety of network functions with software on general-purpose servers, and this allows the network operator to select any capabilities and locations of network functions without any physical constraints. In this paper, we evaluate the influence of the maximum tolerable network delay on the virtual routing function deployment guidelines which the authors proposed previously. Our evaluation results have revealed the following: (1) the more the maximum tolerable network delay condition becomes severe, the more the number of areas where the route selection function is installed increases and the total network cost increases, (2) the higher the routing function cost relative to the circuit bandwidth cost, the increase ratio of total network cost becomes larger according to the maximum tolerable network delay condition.

Keywords: NFV (Network Functions Virtualization), resource allocation, virtual routing function, minimum total network cost

Procedia PDF Downloads 247

Authors: Jane E. Alam Solangi

Abstract:

The contribution of the study is to explore the important part of humorous dialogues in cellular network advertisements. This promotes the message of valuable construction and promotion of network companies in Pakistan that employ different and broad techniques to give promotion to selling products. It merely instigates the consumers to buy it. The results of the study after analysis of its collected data gives a vision that advertisers of network advertisements use humorous dialogues as a significant device to the greater level. The source of entertainment in the advertisement is accompanied by the texts and humorous discourses to influence buying decisions of the consumers. Therefore, it tends to neutralize personal and social based values. The earlier contribution of scholars presented that the technical employment of humorous devices leads to the successful market of the relevant products. In order to analyze the humorous discourse devices, the approach of multi-modality of Fairclough (1989) is used. It is accompanied by the framework of Kress and van Leeuwen’s (1996). It analyzes the visual graph of the grammar. The overall findings in the study verified the role of humorous devices in the captivation of consumers’ decision to buy the product that interests them. Therefore, the role of humor acts as a breaker of the monotonous rhythm of advertisements.

Keywords: advertisements, devices, humorous, multi-modality, networks, Pakistan

Procedia PDF Downloads 103

Authors: Gokhan Kilic, Pelin Celik

Abstract:

Structural health monitoring (SHM) is widely used in evaluating the state and health of membrane structures. In the past, in order to collect data and send it to a data collection unit on membrane structures, wire sensors had to be put as part of the SHM process. However, this study recommends using wireless sensors instead of traditional wire ones to construct an economical, useful, and easy-to-install membrane structure health monitoring system. Every wireless sensor uses a software translation program that is connected to the monitoring server. Operational neural networks (ONNs) have recently been developed to solve the shortcomings of convolutional neural networks (CNNs), such as the network's resemblance to the linear neuron model. The results of using ONNs for monitoring to evaluate the structural health of a membrane are presented in this work.

Keywords: wireless sensor network, non-destructive testing, operational neural networks, membrane structures, dynamic monitoring

Procedia PDF Downloads 92

Authors: Huayuan Lu, Chunfang Yang, Ma Zhu, Baojun Qi, Yaqiong Qiao, Jiangqian Xu

Abstract:

Image geolocalization has great application prospects in fields such as autonomous driving and virtual/augmented reality. In practical application scenarios, the size of the image to be located is not fixed; it is impractical to train different networks for all possible sizes. When its size does not match the size of the input of the descriptor extraction model, existing image geolocalization methods usually directly scale or crop the image in some common ways. This will result in the loss of some information important to the geolocalization task, thus affecting the performance of the image geolocalization method. For example, excessive down-sampling can lead to blurred building contour, and inappropriate cropping can lead to the loss of key semantic elements, resulting in incorrect geolocation results. To address this problem, this paper designs a learnable image resizer and proposes an arbitrary-sized image geolocation method. (1) The designed learnable image resizer employs the self-attention mechanism to enhance the geometric features of the resized image. Firstly, it applies bilinear interpolation to the input image and its feature maps to obtain the initial resized image and the resized feature maps. Then, SKNet (selective kernel net) is used to approximate the best receptive field, thus keeping the geometric shapes as the original image. And SENet (squeeze and extraction net) is used to automatically select the feature maps with strong contour information, enhancing the geometric features. Finally, the enhanced geometric features are fused with the initial resized image, to obtain the final resized images. (2) The proposed image geolocalization method embeds the above image resizer as a fronting layer of the descriptor extraction network. It not only enables the network to be compatible with arbitrary-sized input images but also enhances the geometric features that are crucial to the image geolocalization task. Moreover, the triplet attention mechanism is added after the first convolutional layer of the backbone network to optimize the utilization of geometric elements extracted by the first convolutional layer. Finally, the local features extracted by the backbone network are aggregated to form image descriptors for image geolocalization. The proposed method was evaluated on several mainstream datasets, such as Pittsburgh30K, Tokyo24/7, and Places365. The results show that the proposed method has excellent size compatibility and compares favorably to recently mainstream geolocalization methods.

Keywords: image geolocalization, self-attention mechanism, image resizer, geometric feature

Procedia PDF Downloads 214

Authors: Boumesbah Asma, Chergui Mohamed El-amine

Abstract:

Since it has been shown that the multi-objective minimum spanning tree problem (MOST) is NP-hard even with two criteria, we propose in this study a hybrid NSGA-II algorithm with an exact mutation operator, which is only used with low probability, to find an approximation to the Pareto front of the problem. In a connected graph G, a spanning tree T of G being a connected and cycle-free graph, if k edges of G\T are added to T, we obtain a partial graph H of G inducing a reduced size multi-objective spanning tree problem compared to the initial one. With a weak probability for the mutation operator, an exact method for solving the reduced MOST problem considering the graph H is then used to give birth to several mutated solutions from a spanning tree T. Then, the selection operator of NSGA-II is activated to obtain the Pareto front approximation. Finally, an adaptation of the VNS metaheuristic is called for further improvements on this front. It allows finding good individuals to counterbalance the diversification and the intensification during the optimization search process. Experimental comparison studies with an exact method show promising results and indicate that the proposed algorithm is efficient.

Keywords: minimum spanning tree, multiple objective linear optimization, combinatorial optimization, non-sorting genetic algorithm, variable neighborhood search

Procedia PDF Downloads 91

Authors: Bubai Maji, Monorama Swain

Abstract:

Speech is considered an essential and most natural medium for the interaction between machines and humans. However, extracting effective features for speech emotion recognition (SER) is remains challenging. The present studies show that the temporal information captured but high-level temporal-feature learning is yet to be investigated. In this paper, we present an efficient novel method using the Self-attention (SA) mechanism in a combination of Convolutional Neural Network (CNN) and Bi-directional Gated Recurrent Unit (Bi-GRU) network to learn high-level temporal-feature. In order to further enhance the representation of the high-level temporal-feature, we integrate a Bi-GRU output with learnable weights features by SA, and improve the performance. We evaluate our proposed method on our created SITB-OSED and IEMOCAP databases. We report that the experimental results of our proposed method achieve state-of-the-art performance on both databases.

Keywords: Bi-GRU, 1D-CNNs, self-attention, speech emotion recognition

Procedia PDF Downloads 113

Authors: Goutam Kumar Ghorai, Sandip Sadhukhan, Arpita Sarkar, Debprasad Sinha, G. Sarkar, Ashis K. Dhara

Abstract:

Diabetes mellitus is one of the most common chronic diseases in all countries and continues to increase in numbers significantly. Diabetic retinopathy (DR) is damage to the retina that occurs with long-term diabetes. DR is a major cause of blindness in the Indian population. Therefore, its early diagnosis is of utmost importance towards preventing progression towards imminent irreversible loss of vision, particularly in the huge population across rural India. The barriers to eye examination of all diabetic patients are socioeconomic factors, lack of referrals, poor access to the healthcare system, lack of knowledge, insufficient number of ophthalmologists, and lack of networking between physicians, diabetologists and ophthalmologists. A few diabetic patients often visit a healthcare facility for their general checkup, but their eye condition remains largely undetected until the patient is symptomatic. This work aims to focus on the design and development of a fully automated intelligent decision system for screening retinal fundus images towards detection of the pathophysiology caused by microaneurysm in the early stage of the diseases. Automated detection of microaneurysm is a challenging problem due to the variation in color and the variation introduced by the field of view, inhomogeneous illumination, and pathological abnormalities. We have developed aconvolutional neural network for efficient detection of microaneurysm. A loss function is also developed to handle severe class imbalance due to very small size of microaneurysms compared to background. The network is able to locate the salient region containing microaneurysms in case of noisy images captured by non-mydriatic cameras. The ground truth of microaneurysms is created by expert ophthalmologists for MESSIDOR database as well as private database, collected from Indian patients. The network is trained from scratch using the fundus images of MESSIDOR database. The proposed method is evaluated on DIARETDB1 and the private database. The method is successful in detection of microaneurysms for dilated and non-dilated types of fundus images acquired from different medical centres. The proposed algorithm could be used for development of AI based affordable and accessible system, to provide service at grass root-level primary healthcare units spread across the country to cater to the need of the rural people unaware of the severe impact of DR.

Keywords: retinal fundus image, deep convolutional neural network, early detection of microaneurysms, screening of diabetic retinopathy

Procedia PDF Downloads 142

Authors: Itumeleng Phage

Abstract:

Students’ comprehension of graphs may be affected by the characteristics of the discipline in which the graph is used, the type of the task as well as the background of the students who are the readers or interpreters of the graph. This research study investigated these aspects of the graph comprehension of 152 first-year undergraduate physics students by comparing their responses to corresponding tasks in the mathematics and physics disciplines. The discipline characteristics were analysed for four task-related constructs namely coordinates, representations, area and slope. Students’ responses to corresponding visual decoding and judgement tasks set in mathematics and kinematics contexts were statistically compared. The effects of the participants’ gender, year of school completion and study course were determined as reader characteristics. The results of the empirical study indicated that participants generally transferred their mathematics knowledge on coordinates and representation of straight line graphs to the physics contexts, but not in the cases of parabolic and hyperbolic functions or area under graphs. Insufficient understanding of the slope concept contributed to weak performances on this construct in both mathematics and physics contexts. Discipline characteristics seem to play a vital role in students’ understanding, while reader characteristics had insignificant to medium effects on their responses.

Keywords: kinematics graph, discipline characteristics, constructs, coordinates, representations, area and slope

Procedia PDF Downloads 260

Authors: Zhengrong Wu, Haibo Yang

Abstract:

In the context of escalating global climate change and environmental degradation, the complexity and frequency of natural disasters are continually increasing. Confronted with an abundance of information regarding natural disasters, traditional knowledge graph construction methods, which heavily rely on grammatical rules and prior knowledge, demonstrate suboptimal performance in processing complex, multi-source disaster information. This study, drawing upon past natural disaster reports, disaster-related literature in both English and Chinese, and data from various disaster monitoring stations, constructs question-answer templates based on large language models. Utilizing the P-Tune method, the ChatGLM2-6B model is fine-tuned, leading to the development of a disaster knowledge graph based on large language models. This serves as a knowledge database support for disaster emergency response.

Keywords: large language model, knowledge graph, disaster, deep learning

Procedia PDF Downloads 56

Authors: Somyung Oh, Jeonghyeon Ha, Kyungwon Lee, Sejong Oh

Abstract:

Microarray is a general scheme to find differentially expressed genes for target concept. The output is expressed by heat map, and biologists analyze related terms of gene ontology to find some characteristics of differentially expressed genes. In this paper, we propose integrated visualization tool for heat map and gene ontology graph. Previous two methods are used by static manner and separated way. Proposed visualization tool integrates them and users can interactively manage it. Users may easily find and confirm related terms of gene ontology for given differentially expressed genes. Proposed tool also visualize connections between genes on heat map and gene ontology graph. We expect biologists to find new meaningful topics by proposed tool.

Keywords: heat map, gene ontology, microarray, differentially expressed gene

Procedia PDF Downloads 316

Authors: Parisa Golbayani, Dan Wang, Ionut¸ Florescu

Abstract:

In this work we implement machine learning techniques to financial statement reports in order to asses company’s credit rating. Specifically, the work analyzes the performance of four neural network architectures (MLP, CNN, CNN2D, LSTM) in predicting corporate credit rating as issued by Standard and Poor’s. The paper focuses on companies from the energy, financial, and healthcare sectors in the US. The goal of this analysis is to improve application of machine learning algorithms to credit assessment. To accomplish this, the study investigates three questions. First, we investigate if the algorithms perform better when using a selected subset of important features or whether better performance is obtained by allowing the algorithms to select features themselves. Second, we address the temporal aspect inherent in financial data and study whether it is important for the results obtained by a machine learning algorithm. Third, we aim to answer if one of the four particular neural network architectures considered consistently outperforms the others, and if so under which conditions. This work frames the problem as several case studies to answer these questions and analyze the results using ANOVA and multiple comparison testing procedures.

Keywords: convolutional neural network, long short term memory, multilayer perceptron, credit rating

Procedia PDF Downloads 235

Authors: Yang Zhou, Heli Sun, Jianbin Huang, Jizhong Zhao, Shaojie Qiao

Abstract:

Traffic prediction is a fundamental problem in urban environment, facilitating the smart management of various businesses, such as taxi dispatching, bike relocation, and stampede alert. Most earlier methods rely on identifying the intrinsic spatial-temporal correlation to forecast. However, the complex nature of this problem entails a more sophisticated solution that can simultaneously capture the mutual influence of both adjacent and far-flung areas, with the information of time-dimension also incorporated seamlessly. To tackle this difficulty, we propose a new multi-phase architecture, DSTDS (Dual Spatial-Temporal Data Scheme for traffic prediction), that aims to reveal the underlying relationship that determines future traffic trend. First, a graph-based neural network with an attention mechanism is devised to obtain the static features of the road network. Then, a multi-granularity recurrent neural network is built in conjunction with the knowledge from a grid-based model. Subsequently, the preceding output is fed into a spatial-temporal super-resolution module. With this 3-phase structure, we carry out extensive experiments on several real-world datasets to demonstrate the effectiveness of our approach, which surpasses several state-of-the-art methods.

Keywords: traffic prediction, spatial-temporal, recurrent neural network, dual data scheme

Procedia PDF Downloads 117

Authors: Hira Lal Gope, Hidekazu Fukai

Abstract:

The aim of this study is to develop a system which can identify and sort peaberries automatically at low cost for coffee producers in developing countries. In this paper, the focus is on the classification of peaberries and normal coffee beans using image processing and machine learning techniques. The peaberry is not bad and not a normal bean. The peaberry is born in an only single seed, relatively round seed from a coffee cherry instead of the usual flat-sided pair of beans. It has another value and flavor. To make the taste of the coffee better, it is necessary to separate the peaberry and normal bean before green coffee beans roasting. Otherwise, the taste of total beans will be mixed, and it will be bad. In roaster procedure time, all the beans shape, size, and weight must be unique; otherwise, the larger bean will take more time for roasting inside. The peaberry has a different size and different shape even though they have the same weight as normal beans. The peaberry roasts slower than other normal beans. Therefore, neither technique provides a good option to select the peaberries. Defect beans, e.g., sour, broken, black, and fade bean, are easy to check and pick up manually by hand. On the other hand, the peaberry pick up is very difficult even for trained specialists because the shape and color of the peaberry are similar to normal beans. In this study, we use image processing and machine learning techniques to discriminate the normal and peaberry bean as a part of the sorting system. As the first step, we applied Deep Convolutional Neural Networks (CNN) and Support Vector Machine (SVM) as machine learning techniques to discriminate the peaberry and normal bean. As a result, better performance was obtained with CNN than with SVM for the discrimination of the peaberry. The trained artificial neural network with high performance CPU and GPU in this work will be simply installed into the inexpensive and low in calculation Raspberry Pi system. We assume that this system will be used in under developed countries. The study evaluates and compares the feasibility of the methods in terms of accuracy of classification and processing speed.

Keywords: convolutional neural networks, coffee bean, peaberry, sorting, support vector machine

Procedia PDF Downloads 144

Authors: Chien-kuo Chang, Bo-wei Wu, Yi-yun Tang, Min-chiu Wu

Abstract:

This paper proposes a power equipment diagnosis system based on partial discharge (PD), which is characterized by increasing the readability of experimental data and the convenience of operation. This system integrates a variety of analysis programs of different data formats and different programming languages and then establishes a set of interfaces that can follow and expand the structure, which is also helpful for subsequent maintenance and innovation. This study shows a case of using the developed Convolutional Neural Networks (CNN) to integrate with this system, using the designed model architecture to simplify the complex training process. It is expected that the simplified training process can be used to establish an adaptive deep learning experimental structure. By selecting different test data for repeated training, the accuracy of the identification system can be enhanced. On this platform, the measurement status and partial discharge pattern of each equipment can be checked in real time, and the function of real-time identification can be set, and various training models can be used to carry out real-time partial discharge insulation defect identification and insulation state diagnosis. When the electric power equipment entering the dangerous period, replace equipment early to avoid unexpected electrical accidents.

Keywords: partial discharge, convolutional neural network, partial discharge analysis platform, adaptive deep learning

Procedia PDF Downloads 78

Authors: Manel Brichni, Abdelhak-Djamel Seriai

Abstract:

Systems are increasingly complex. To reduce their complexity, an abstract view of the system can simplify its development. To overcome this problem, we propose a method to decompose systems into subsystems while reducing their coupling. These subsystems represent components. Consisting of an existing object-oriented systems, the main idea of our approach is based on modelling as graphs all entities of an oriented object source code. Such modelling is easy to handle, so we can apply restructuring algorithms based on graph metrics. The particularity of our approach consists in integrating in addition to standard metrics, such as coupling and cohesion, some graph metrics giving more precision during the components identication. To treat this problem, we relied on the ROMANTIC approach that proposed a component-based software architecture recovery from an object oriented system.

Keywords: software reengineering, software component and interfaces, metrics, graphs

Procedia PDF Downloads 501

Authors: Nihed Bahria El Asghar, Imen Jouili, Mounir Frikha

Abstract:

Inter-datacenter transport network is very bandwidth and delay demanding. The data transferred over such a network is also highly QoS-exigent mostly because a huge volume of data should be transported transparently with regard to the application user. To avoid the data transfer failure, a backup path should be reserved. No re-routing delay should be observed. A dedicated 1+1 protection is however not applicable in inter-datacenter transport network because of the huge spare capacity. In this context, we propose a survivable virtual network with minimal backup based on network coding (1 ⊕ 1) and solve it using a modified Dijkstra-based heuristic.

Keywords: network coding, dedicated protection, spare capacity, inter-datacenters transport network

Procedia PDF Downloads 447