Search results for: deep graphical model

Authors: Muneeb Ullah, Daishihan, Xiadong Young

Abstract:

Deep learning plays a crucial role in identifying COVID-19 and preventing its spread. To improve the accuracy of COVID-19 diagnoses, it is important to have access to a sufficient number of training images of CXRs (chest X-rays) depicting the disease. However, there is currently a shortage of such images. To address this issue, this paper introduces COVID-19 GAN, a model that uses generative adversarial networks (GANs) to generate realistic CXR images of COVID-19, which can be used to train identification models. Initially, a generator model is created that uses digressive channels to generate images of CXR scans for COVID-19. To differentiate between real and fake disease images, an efficient discriminator is developed by combining the dense connectivity strategy and instance normalization. This approach makes use of their feature extraction capabilities on CXR hazy areas. Lastly, the deep regret gradient penalty technique is utilized to ensure stable training of the model. With the use of 4,062 grape leaf disease images, the Leaf GAN model successfully produces 8,124 COVID-19 CXR images. The COVID-19 GAN model produces COVID-19 CXR images that outperform DCGAN and WGAN in terms of the Fréchet inception distance. Experimental findings suggest that the COVID-19 GAN-generated CXR images possess noticeable haziness, offering a promising approach to address the limited training data available for COVID-19 model training. When the dataset was expanded, CNN-based classification models outperformed other models, yielding higher accuracy rates than those of the initial dataset and other augmentation techniques. Among these models, ImagNet exhibited the best recognition accuracy of 99.70% on the testing set. These findings suggest that the proposed augmentation method is a solution to address overfitting issues in disease identification and can enhance identification accuracy effectively.

Keywords: classification, deep learning, medical images, CXR, GAN.

Procedia PDF Downloads 55

Authors: A. Shoiynbek, K. Kozhakhmet, P. Menezes, D. Kuanyshbay, D. Bayazitov

Abstract:

Speech emotion recognition has received increasing research interest all through current years. There was used emotional speech that was collected under controlled conditions in most research work. Actors imitating and artificially producing emotions in front of a microphone noted those records. There are four issues related to that approach, namely, (1) emotions are not natural, and it means that machines are learning to recognize fake emotions. (2) Emotions are very limited by quantity and poor in their variety of speaking. (3) There is language dependency on SER. (4) Consequently, each time when researchers want to start work with SER, they need to find a good emotional database on their language. In this paper, we propose the approach to create an automatic tool for speech emotion extraction based on facial emotion recognition and describe the sequence of actions of the proposed approach. One of the first objectives of the sequence of actions is a speech detection issue. The paper gives a detailed description of the speech detection model based on a fully connected deep neural network for Kazakh and Russian languages. Despite the high results in speech detection for Kazakh and Russian, the described process is suitable for any language. To illustrate the working capacity of the developed model, we have performed an analysis of speech detection and extraction from real tasks.

Keywords: deep neural networks, speech detection, speech emotion recognition, Mel-frequency cepstrum coefficients, collecting speech emotion corpus, collecting speech emotion dataset, Kazakh speech dataset

Procedia PDF Downloads 74

Authors: Shri Krishna Mishra

Abstract:

In an educational context, technology can prompt interactive plight only when it is used in conjunction with interactive plight methods. This study, therefore, examines the relationships among components that contribute to higher levels of interactive plight and execution, such as interactive Plight methods, technology, intrinsic motivation and deep learning. 526 students participated in this study. With structural equation modelling, the authors test the conceptual model and identify satisfactory model fit. The results indicate that interactive Plight methods, technology and intrinsic motivation have significant relationship with interactive Plight; deep learning mediates the relationships of the other variables with Execution.

Keywords: searching the relationship among components, contribute to interactive plight, educational execution, intrinsic motivation

Procedia PDF Downloads 431

Authors: Jinil Patel, Sarthak Patel, Sarthak Thakkar, Deepti Saraswat

Abstract:

Recently, the COVID-19 outbreak has spread across the world, leading the World Health Organization to classify it as a global pandemic. To save the patient’s life, the COVID-19 symptoms have to be identified. But using an AI (Artificial Intelligence) model to identify COVID-19 symptoms within the allotted time was challenging. The RT-PCR test was found to be inadequate in determining the COVID status of a patient. To determine if the patient has COVID-19 or not, a Computed Tomography Scan (CT scan) of patient is a better alternative. It will be challenging to compile and store all the data from various hospitals on the server, though. Federated learning, therefore, aids in resolving this problem. Certain deep learning models help to classify Covid-19. This paper will have detailed work of certain deep learning models like VGG19, ResNet50, MobileNEtv2, and Deep Learning Aggregation (DLA) along with maintaining privacy with encryption.

Keywords: federated learning, COVID-19, CT-scan, homomorphic encryption, ResNet50, VGG-19, MobileNetv2, DLA

Procedia PDF Downloads 48

Authors: Victor Breux, Jérôme Boutet, Alain Goret, Viviane Cattin

Abstract:

Early detection of anomalies in data centers is important to reduce downtimes and the costs of periodic maintenance. However, there is little research on this topic and even fewer on the fusion of sensor data for the detection of abnormal events. The goal of this paper is to propose a method for anomaly detection in data centers by combining sensor data (temperature, humidity, power) and deep learning models. The model described in the paper uses one autoencoder per sensor to reconstruct the inputs. The auto-encoders contain Long-Short Term Memory (LSTM) layers and are trained using the normal samples of the relevant sensors selected by correlation analysis. The difference signal between the input and its reconstruction is then used to classify the samples using feature extraction and a random forest classifier. The data measured by the sensors of a data center between January 2019 and May 2020 are used to train the model, while the data between June 2020 and May 2021 are used to assess it. Performances of the model are assessed a posteriori through F1-score by comparing detected anomalies with the data center’s history. The proposed model outperforms the state-of-the-art reconstruction method, which uses only one autoencoder taking multivariate sequences and detects an anomaly with a threshold on the reconstruction error, with an F1-score of 83.60% compared to 24.16%.

Keywords: anomaly detection, autoencoder, data centers, deep learning

Procedia PDF Downloads 163

Authors: Li Xiaoge, Wang Chaodong

Abstract:

We introduce the first Chinese Coreference Resolution Model based on BERT (CCRM-BERT) and show that it significantly outperforms all previous work. The key idea is to consider the features of the mention, such as part of speech, width of spans, distance between spans, etc. And the influence of each features on the model is analyzed. The model computes mention embeddings that combine BERT with features. Compared to the existing state-of-the-art span-ranking approach, our model significantly improves accuracy on the Chinese OntoNotes benchmark.

Keywords: BERT, coreference resolution, deep learning, nature language processing

Procedia PDF Downloads 174

Authors: Min Wang, Sergey Utev

Abstract:

The lessons from the 2007-09 global financial crisis have driven scientific research, which considers the design of new methodologies and financial models in the global market. The quantum mechanics approach was introduced in the unpredictable stock market modeling. One famous quantum tool is Feynman path integral method, which was used to model insurance risk by Tamturk and Utev and adapted to formalize the path-dependent option pricing by Hao and Utev. The research is based on the path-dependent calculation method, which is motivated by the Feynman path integral method. The path calculation can be studied in two ways, one way is to label, and the other is computational. Labeling is a part of the representation of objects, and generating functions can provide many different ways of representing share price paths. In this paper, the recent works on graphical theoretical construction of individual share price path via matroid is presented. Firstly, a study is done on the knowledge of matroid, relationship between lattice path matroid and Tutte polynomials and ways to connect points in the lattice path matroid and Tutte polynomials is suggested. Secondly, It is found that a general binary tree can be validly constructed from a connected lattice path matroid rather than general lattice path matroid. Lastly, it is suggested that there is a way to represent share price paths via a general binary tree, and an algorithm is developed to construct share price paths from general binary trees. A relationship is also provided between lattice integer points and Tutte polynomials of a transversal matroid. Use this way of connection together with the algorithm, a share price path can be constructed from a given connected lattice path matroid.

Keywords: combinatorial construction, graphical representation, matroid, path calculation, share price, Tutte polynomial

Procedia PDF Downloads 109

Authors: B. Kenny

Abstract:

The complications associated with transfusions are well documented, with the serious hazards of transfusion (SHOT) reporting system continuing to report deaths and serious morbidity due to the transfusion of allogenic blood. Many different sources including the TRICC trial, NHMRC and Cochrane recommending similar transfusion triggers/guidelines. Recent studies found the rate of infection (deep infection, wound infection, chest infection, urinary tract infection, and others) were purely a dose response relationship, increasing the Relative Risk by 3.44. It was also noted that each transfused patient stayed in hospital for one additional day. We hypothesise that providing an approved/standardised, guideline with a graphical summary of decision pathways for anaemic patients will reduce unnecessary transfusions. We retrospectively assessed patients undergoing primary knee or hip arthroplasties over a 4 year period, 1459 patients. Of these, 339 (23.24%) patients received allogenic blood transfusions and 858 units of blood were transfused, 9.14% of patients transfused had haemoglobin levels above 100 g/L, 7.67% of patients were transfused without knowing the haemoglobin level, 24 hours prior to transfusion initiation and 4.5% had possible transfusion reactions. Overall, 17% of allogenic transfusions topatients admitted to the Orthopaedic department within a 4 year period were outside NHMRC and Cochrane guidelines/recommendations. If our transfusion frequency is compared with that of other authors/hospitals, transfusion rates are consistently being high. We subsequently implemented a simple guideline for transfusion initiation. This guideline was then assessed. We found the transfusion rate post transfusion implementation to be significantly lower, without increase in patient morbidity or mortalitiy, p <0.001). Transfusion rates and patient outcome can be optimized by a simple graphical aid for decision making.

Keywords: transfusion, morbidity, mortality, neck of femur, fracture, arthroplasty, rehabilitation

Procedia PDF Downloads 219

Authors: Adrià Arbués-Sangüesa, Coloma Ballester, Gloria Haro

Abstract:

Tracking sports players is a widely challenging scenario, specially in single-feed videos recorded in tight courts, where cluttering and occlusions cannot be avoided. This paper presents an analysis of several geometric and semantic visual features to detect and track basketball players. An ablation study is carried out and then used to remark that a robust tracker can be built with Deep Learning features, without the need of extracting contextual ones, such as proximity or color similarity, nor applying camera stabilization techniques. The presented tracker consists of: (1) a detection step, which uses a pretrained deep learning model to estimate the players pose, followed by (2) a tracking step, which leverages pose and semantic information from the output of a convolutional layer in a VGG network. Its performance is analyzed in terms of MOTA over a basketball dataset with more than 10k instances.

Keywords: basketball, deep learning, feature extraction, single-camera, tracking

Procedia PDF Downloads 113

Authors: Aibek Kukpayev, Dhawal Shah

Abstract:

Combustion of sour fuels containing high amount of sulfur leads to the formation of sulfur oxides, which adversely harm the environment and has a negative impact on human health. Considering this, several legislations have been imposed to bring down the sulfur content in fuel to less than 10 ppm. In recent years, novel deep eutectic solvents (DESs) have been developed to achieve deep desulfurization, particularly to extract thiophenic compounds from liquid fuels. These novel DESs, considered as analogous to ionic liquids are green, eco-friendly, inexpensive, and sustainable. We herein, using molecular dynamic simulation, analyze the interactions of metal-based DESs with model oil consisting of thiophenic compounds. The DES used consists of polyethylene glycol (PEG-200) as a hydrogen bond donor, choline chloride (ChCl) or tetrabutyl ammonium chloride (TBAC) as a hydrogen bond acceptor, and cobalt chloride (CoCl₂) as metal salt. In particular, the combination of ChCl: PEG-200:CoCl₂ at a ratio 1:2:1 and the combination of TBAC:PEG-200:CoCl₂ at a ratio 1:2:0.25 were simulated, separately, with model oil consisting of octane and thiophenes at 25ᵒC and 1 bar. The results of molecular dynamics simulations were analyzed in terms of interaction energies between different components. The simulations revealed a stronger interaction between DESs/thiophenes as compared with octane/thiophenes, suggestive of an efficient desulfurization process. In addition, our analysis suggests that the choice of hydrogen bond acceptor strongly influences the efficiency of the desulfurization process. Taken together, the results also show the importance of the metal ion, although present in small amount, in the process, and the role of the polymer in desulfurization of the model fuel.

Keywords: deep eutectic solvents, desulfurization, molecular dynamics simulations, thiophenes

Procedia PDF Downloads 118

Authors: Ons Slimene, Aroua Taamallah, Maha Khemaja

Abstract:

In recent years, human activity recognition (HAR) has been a key area of research due to its diverse applications. It has garnered increasing attention in the field of computer vision. HAR plays an important role in people’s daily lives as it has the ability to learn advanced knowledge about human activities from data. In HAR, activities are usually represented by exploiting different types of sensors, such as embedded sensors or visual sensors. However, these sensors have limitations, such as local obstacles, image-related obstacles, sensor unreliability, and consumer concerns. Recently, several deep learning-based approaches have been proposed for HAR and these approaches are classified into two categories based on the type of data used: vision-based approaches and sensor-based approaches. This research paper highlights the importance of multimodal data fusion from skeleton data obtained from videos and data generated by embedded sensors using deep neural networks for achieving HAR. We propose a deep multimodal fusion network based on a twostream architecture. These two streams use the Convolutional Neural Network combined with the Bidirectional LSTM (CNN BILSTM) to process skeleton data and data generated by embedded sensors and the fusion at the feature level is considered. The proposed model was evaluated on a public OPPORTUNITY++ dataset and produced a accuracy of 96.77%.

Keywords: human activity recognition, action recognition, sensors, vision, human-centric sensing, deep learning, context-awareness

Procedia PDF Downloads 71

Authors: Phanida Phukoetphim, Asaad Y. Shamseldin

Abstract:

In this study, the novel Stochastic Gradient Boosting (SGB) combination method is addressed for producing daily river flows from four different rain-runoff models of Ohinemuri catchment, New Zealand. The selected rainfall-runoff models are two empirical black-box models: linear perturbation model and linear varying gain factor model, two conceptual models: soil moisture accounting and routing model and Nedbør-Afrstrømnings model. In this study, the simple average combination method and the weighted average combination method were used as a benchmark for comparing the results of the novel SGB combination method. The models and combination results are evaluated using statistical and graphical criteria. Overall results of this study show that the use of combination technique can certainly improve the simulated river flows of four selected models for Ohinemuri catchment, New Zealand. The results also indicate that the novel SGB combination method is capable of accurate prediction when used in a combination method of the simulated river flows in New Zealand.

Keywords: multi-model combination, rainfall-runoff modeling, stochastic gradient boosting, bioinformatics

Procedia PDF Downloads 309

Authors: Mária Csernoch, Piroska Biró

Abstract:

The legends about “user-friendly” and “easy-to-use” birotical tools (computer-related office tools) have been spreading and misleading end-users. This approach has led us to the extremely high number of incorrect documents, causing serious financial losses in the creating, modifying, and retrieving processes. Our research proved that there are at least two sources of this underachievement: (1) The lack of the definition of the correctly edited, formatted documents. Consequently, end-users do not know whether their methods and results are correct or not. They are not aware of their ignorance. They are so ignorant that their ignorance does not allow them to realize their lack of knowledge. (2) The end-users’ problem-solving methods. We have found that in non-traditional programming environments end-users apply, almost exclusively, surface approach metacognitive methods to carry out their computer related activities, which are proved less effective than deep approach methods. Based on these findings we have developed deep approach methods which are based on and adapted from traditional programming languages. In this study, we focus on the most popular type of birotical documents, the text-based documents. We have provided the definition of the correctly edited text, and based on this definition, adapted the debugging method known in programming. According to the method, before the realization of text editing, a thorough debugging of already existing texts and the categorization of errors are carried out. With this method in advance to real text editing users learn the requirements of text-based documents and also of the correctly formatted text. The method has been proved much more effective than the previously applied surface approach methods. The advantages of the method are that the real text handling requires much less human and computer sources than clicking aimlessly in the GUI (Graphical User Interface), and the data retrieval is much more effective than from error-prone documents.

Keywords: deep approach metacognitive methods, error-prone birotical documents, financial losses, human and computer resources

Procedia PDF Downloads 363

Authors: Barham J. Nareeman, Ilham I. Mohammed

Abstract:

Excavations in urban developed area are generally supported by deep excavation walls such as; diaphragm wall, bored piles, soldier piles and sheet piles. In some cases, these walls may be braced by internal braces or tie back anchors. Tie back anchors are by far the predominant method for wall support, the large working space inside the excavation provided by a tieback anchor system has a significant construction advantage. This paper aims to analyze a deep excavation bracing system of contiguous pile wall braced by pre-stressed tie back anchors, which is a part of a huge residential building project, located in Turkey/Gaziantep province. The contiguous pile wall will be constructed with a length of 270 m that consists of 285 piles, each having a diameter of 80 cm, and a center to center spacing of 95 cm. The deformation analysis was carried out by a finite element analysis tool using PLAXIS. In the analysis, beam element method together with an elastic perfect plastic soil model and Soil Hardening Model was used to design the contiguous pile wall, the tieback anchor system, and the soil. The two soil clusters which are limestone and a filled soil were modelled with both Hardening soil and Mohr Coulomb models. According to the basic design, both soil clusters are modelled as drained condition. The simulation results show that the maximum horizontal movement of the walls and the maximum settlement of the ground are convenient with 300 individual case histories which are ranging between 1.2mm and 2.3mm for walls, and 15mm and 6.5mm for the settlements. It was concluded that tied-back contiguous pile wall can be satisfactorily modelled using Hardening soil model.

Keywords: deep excavation, finite element, pre-stressed tie back anchors, contiguous pile wall, PLAXIS, horizontal deflection, ground settlement

Procedia PDF Downloads 226

Authors: Gong Zhilin, Jing Yang, Jian Yin

Abstract:

The purpose of this paper is to develop a novel system for credit card fraud detection based on sequential modeling of data using hybrid deep learning models. The projected model encapsulates five major phases are pre-processing, imbalance-data handling, feature extraction, optimal feature selection, and fraud detection with an ensemble classifier. The collected raw data (input) is pre-processed to enhance the quality of the data through alleviation of the missing data, noisy data as well as null values. The pre-processed data are class imbalanced in nature, and therefore they are handled effectively with the K-means clustering-based SMOTE model. From the balanced class data, the most relevant features like improved Principal Component Analysis (PCA), statistical features (mean, median, standard deviation) and higher-order statistical features (skewness and kurtosis). Among the extracted features, the most optimal features are selected with the Self-improved Arithmetic Optimization Algorithm (SI-AOA). This SI-AOA model is the conceptual improvement of the standard Arithmetic Optimization Algorithm. The deep learning models like Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and optimized Quantum Deep Neural Network (QDNN). The LSTM and CNN are trained with the extracted optimal features. The outcomes from LSTM and CNN will enter as input to optimized QDNN that provides the final detection outcome. Since the QDNN is the ultimate detector, its weight function is fine-tuned with the Self-improved Arithmetic Optimization Algorithm (SI-AOA).

Keywords: credit card, data mining, fraud detection, money transactions

Procedia PDF Downloads 104

Authors: Simon Espigolé, Igor Shalyminov, Helen Hastie

Abstract:

Deep machine learning techniques including deep neural networks (DNN) have been used to model language and dialogue for conversational agents to perform tasks, such as giving technical support and also for general chit-chat. They have been shown to be capable of generating long, diverse and coherent sentences in end-to-end dialogue systems and natural language generation. However, these systems tend to imitate the training data and will only generate the concepts and language within the scope of what they have been trained on. This work explores how deep neural networks can be used in a task that would normally require human creativity, whereby the human would read the movie description and/or watch the movie and come up with a compelling, interesting movie title. This task differs from simple summarization in that the movie title may not necessarily be derivable from the content or semantics of the movie description. Here, we train a type of DNN called a sequence-to-sequence model (seq2seq) that takes as input a short textual movie description and some information on e.g. genre of the movie. It then learns to output a movie title. The idea is that the DNN will learn certain techniques and approaches that the human movie titler may deploy that may not be immediately obvious to the human-eye. To give an example of a generated movie title, for the movie synopsis: ‘A hitman concludes his legacy with one more job, only to discover he may be the one getting hit.’; the original, true title is ‘The Driver’ and the one generated by the model is ‘The Masquerade’. A human evaluation was conducted where the DNN output was compared to the true human-generated title, as well as a number of baselines, on three 5-point Likert scales: ‘creativity’, ‘naturalness’ and ‘suitability’. Subjects were also asked which of the two systems they preferred. The scores of the DNN model were comparable to the scores of the human-generated movie title, with means m=3.11, m=3.12, respectively. There is room for improvement in these models as they were rated significantly less ‘natural’ and ‘suitable’ when compared to the human title. In addition, the human-generated title was preferred overall 58% of the time when pitted against the DNN model. These results, however, are encouraging given the comparison with a highly-considered, well-crafted human-generated movie title. Movie titles go through a rigorous process of assessment by experts and focus groups, who have watched the movie. This process is in place due to the large amount of money at stake and the importance of creating an effective title that captures the audiences’ attention. Our work shows progress towards automating this process, which in turn may lead to a better understanding of creativity itself.

Keywords: creativity, deep machine learning, natural language generation, movies

Procedia PDF Downloads 304

Authors: Chiau Yuan Lim, Hayyiratul Fatimah Mohd Zaid, Fai Kait Chong

Abstract:

Deep Eutectic Solvent (DES) is used in various applications due to its simplicity in synthesis procedure, biodegradable, inexpensive and easily available chemical ingredients. Graphene Oxide is a popular catalyst that being used in various processes due to its stacking carbon sheets in layer which theoretically rapid up the catalytic processes. In this study, choline chloride based DESs were synthesized and ChCl-PEG(1:4) was found to be the most effective DES in performing desulfurization, which it is able to remove up to 47.4% of the sulfur content in the model oil in just 10 minutes, and up to 95% of sulfur content after repeat the process for six times. ChCl-PEG(1:4) able to perform up to 32.7% desulfurization on real diesel after 6 multiple stages. Thus, future research works should focus on removing the impurities on real diesel before utilising DESs in petroleum field.

Keywords: choline chloride, deep eutectic solvent, fuel desulfurization, graphene oxide

Procedia PDF Downloads 125

Authors: Ethan James

Abstract:

Over one billion people worldwide suffer from some level of vision loss or blindness as a result of progressive retinal diseases. Many patients, particularly in developing areas, are incorrectly diagnosed or undiagnosed whatsoever due to unconventional diagnostic tools and screening methods. Artificial intelligence (AI) based on deep learning (DL) convolutional neural networks (CNN) have recently gained a high interest in ophthalmology for its computer-imaging diagnosis, disease prognosis, and risk assessment. Optical coherence tomography (OCT) is a popular imaging technique used to capture high-resolution cross-sections of retinas. In ophthalmology, DL has been applied to fundus photographs, optical coherence tomography, and visual fields, achieving robust classification performance in the detection of various retinal diseases including macular degeneration, diabetic retinopathy, and retinitis pigmentosa. However, there is no complete diagnostic model to analyze these retinal images that provide a diagnostic accuracy above 90%. Thus, the purpose of this project was to develop an AI model that utilizes machine learning techniques to automatically diagnose specific retinal diseases from OCT scans. The algorithm consists of neural network architecture that was trained from a dataset of over 20,000 real-world OCT images to train the robust model to utilize residual neural networks with cyclic pooling. This DL model can ultimately aid ophthalmologists in diagnosing patients with these retinal diseases more quickly and more accurately, therefore facilitating earlier treatment, which results in improved post-treatment outcomes.

Keywords: artificial intelligence, deep learning, imaging, medical devices, ophthalmic devices, ophthalmology, retina

Procedia PDF Downloads 152

Authors: Paula Verdugo-Hernandez, Patricio Cumsille

Abstract:

We analyze a first-class on the convergence of real number sequences, named hereafter sequences, to foster exploration and discovery of concepts through graphical representations before engaging students in proving. The main goal was to differentiate between sequences and continuous functions-of-a-real-variable and better understand concepts at an initial stage. We applied the analytic frame of mathematical working spaces, which we expect to contribute to extending to sequences since, as far as we know, it has only developed for other objects, and which is relevant to analyze how mathematical work is built systematically by connecting the epistemological and cognitive perspectives, and involving the semiotic, instrumental, and discursive dimensions.

Keywords: convergence, graphical representations, mathematical working spaces, paradigms of real analysis, real number sequences

Procedia PDF Downloads 121

Authors: Shimaa Holail, Tamer Saleh, Xiongwu Xiao

Abstract:

Building change detection (BCD) from satellite imagery is an essential topic in urbanization monitoring, agricultural land management, and updating geospatial databases. Recently, methods for detecting changes based on deep learning have made significant progress and impressive results. However, it has the problem of being insensitive to changes in buildings with complex spectral differences, and the features being extracted are not discriminatory enough, resulting in incomplete buildings and irregular boundaries. To overcome these problems, we propose a dual Siamese network based on the Unet model with the addition of a deep supervision strategy (DS) in this paper. This network consists of a backbone (encoder) based on ImageNet pre-training, a fusion block, and feature pyramid networks (FPN) to enhance the step-by-step information of the changing regions and obtain a more accurate BCD map. To train the proposed method, we created a new dataset (EGY-BCD) of high-resolution and multi-temporal aerial images captured over New Cairo in Egypt to detect building changes for this purpose. The experimental results showed that the proposed method is effective and performs well with the EGY-BCD dataset regarding the overall accuracy, F1-score, and mIoU, which were 91.6 %, 80.1 %, and 73.5 %, respectively.

Keywords: building change detection, deep supervision, semantic segmentation, EGY-BCD dataset

Procedia PDF Downloads 76

Authors: T. Godwin

Abstract:

A call-taxi system is a type of taxi service where a taxi could be requested through a phone call or mobile app. A schematic functioning of a call-taxi system is modeled using Petri net, which provides the necessary conditions for a taxi to be assigned by a dispatcher to pick a customer as well as the conditions for the taxi to be released by the customer. A Petri net is a graphical modeling tool used to understand sequences, concurrences, and confluences of activities in the working of discrete event systems. It uses tokens on a directed bipartite multi-graph to simulate the activities of a system. The Petri net model is translated into a simulation model and a call-taxi system is simulated. The simulation model helps in evaluating the operation of a call-taxi system based on the fleet size as well as the operating policies for call-taxi assignment and empty call-taxi repositioning. The developed Petri net based simulation model can be used to decide the fleet size as well as the call-taxi assignment policies for a call-taxi system.

Keywords: call-taxi, discrete event system, petri net, simulation modeling

Procedia PDF Downloads 401

Authors: Deepika Christopher, Garima Anand

Abstract:

To determine the best model for churn prediction in the telecom industry, this paper compares 11 machine learning algorithms, namely Logistic Regression, Support Vector Machine, Random Forest, Decision Tree, XGBoost, LightGBM, Cat Boost, AdaBoost, Extra Trees, Deep Neural Network, and Hybrid Model (MLPClassifier). It also aims to pinpoint the top three factors that lead to customer churn and conducts customer segmentation to identify vulnerable groups. According to the data, the Logistic Regression model performs the best, with an F1 score of 0.6215, 81.76% accuracy, 68.95% precision, and 56.57% recall. The top three attributes that cause churn are found to be tenure, Internet Service Fiber optic, and Internet Service DSL; conversely, the top three models in this article that perform the best are Logistic Regression, Deep Neural Network, and AdaBoost. The K means algorithm is applied to establish and analyze four different customer clusters. This study has effectively identified customers that are at risk of churn and may be utilized to develop and execute strategies that lower customer attrition.

Keywords: attrition, retention, predictive modeling, customer segmentation, telecommunications

Procedia PDF Downloads 28

Authors: Yifan Fan, Xudong Luo, Pingping Lin

Abstract:

Nowadays, dialogue systems increasingly become the way for humans to access many computer systems. So, humans can interact with computers in natural language. A dialogue system consists of three parts: understanding what humans say in natural language, managing dialogue, and generating responses in natural language. In this paper, we survey deep learning based methods for dialogue management, response generation and dialogue evaluation. Specifically, these methods are based on neural network, long short-term memory network, deep reinforcement learning, pre-training and generative adversarial network. We compare these methods and point out the further research directions.

Keywords: dialogue management, response generation, deep learning, evaluation

Procedia PDF Downloads 135

Authors: Atef Ahmad Bleibel

Abstract:

Most codes of practice, like ACI 318-14, require the use of strut-and-tie modeling to analyze and design reinforced concrete deep beams. Though, investigations that conducted on deep beams do not include ring deep beams of influential parameters. This work presents an analytical parametric study using strut-and-tie modeling stated by ACI 318-14 to predict load capacity of 20 reinforced concrete ring deep beam specimens with different parameters. The parameters that were under consideration in the current work are ring diameter (Dc), number of supports (NS), width of ring beam (bw), concrete compressive strength (f'c) and width of bearing plate (Bp). It is found that the load capacity decreases by about 14-36% when ring diameter increases by about 25-75%. It is also found that load capacity increases by about 62-189% when number of supports increases by about 33-100%, while the load capacity increases by about 25-75% when the beam ring width increases by about 25-75%. Finally, it is found that load capacity increases by about 24-76% when compressive strength increases by about 24-76%, while the load capacity increases by about 5-16% when Bp increases by about 25-75%.

Keywords: load parameters, reinforced concrete, ring deep beam, strut and tie

Procedia PDF Downloads 81

Authors: Sneha Thakur, Sanjeev Karmakar

Abstract:

This paper proposed hybrid Particle Swarm Optimization (PSO) – Long-Short Term Memory (LSTM) model for groundwater level prediction. The evaluation of the performance is realized using the parameters: root mean square error (RMSE) and mean absolute error (MAE). Ground water level forecasting will be very effective for planning water harvesting. Proper calculation of water level forecasting can overcome the problem of drought and flood to some extent. The objective of this work is to develop a ground water level forecasting model using deep learning technique integrated with optimization technique PSO by applying 29 years data of Chhattisgarh state, In-dia. It is important to find the precise forecasting in case of ground water level so that various water resource planning and water harvesting can be managed effectively.

Keywords: long short-term memory, particle swarm optimization, prediction, deep learning, groundwater level

Procedia PDF Downloads 48

Authors: Heewon Jeong, Seongpyo Kim, Joon Ha Kim

Abstract:

Over the last few tens of years, the coastal regions of Korea have experienced red tide algal blooms, which are harmful and toxic to both humans and marine organisms due to their potential threat. It was accelerated owing to eutrophication by human activities, certain oceanic processes, and climate change. Previous studies have tried to monitoring and predicting the algae concentration of the ocean with the bio-optical algorithms applied to color images of the satellite. However, the accurate estimation of algal blooms remains problems to challenges because of the complexity of coastal waters. Therefore, this study suggests a new method to identify the concentration of red tide algal bloom from images of geostationary ocean color imager (GOCI) which are representing the water environment of the sea in Korea. The method employed GOCI images, which took the water leaving radiances centered at 443nm, 490nm and 660nm respectively, as well as observed weather data (i.e., humidity, temperature and atmospheric pressure) for the database to apply optical characteristics of algae and train deep learning algorithm. Convolution neural network (CNN) was used to extract the significant features from the images. And then artificial neural network (ANN) was used to estimate the concentration of algae from the extracted features. For training of the deep learning model, backpropagation learning strategy is developed. The established methods were tested and compared with the performances of GOCI data processing system (GDPS), which is based on standard image processing algorithms and optical algorithms. The model had better performance to estimate algae concentration than the GDPS which is impossible to estimate greater than 5mg/m³. Thus, deep learning model trained successfully to assess algae concentration in spite of the complexity of water environment. Furthermore, the results of this system and methodology can be used to improve the performances of remote sensing. Acknowledgement: This work was supported by the 'Climate Technology Development and Application' research project (#K07731) through a grant provided by GIST in 2017.

Keywords: deep learning, algae concentration, remote sensing, satellite

Procedia PDF Downloads 157

Authors: Umporn Wutchana

Abstract:

An open ended problem and unstructured interview had been used to explore students’ conceptual and procedural understanding of vector components. The open ended problem had been designed based on research instrument used in previous physics education research. Without physical context, we asked students to find out magnitude and draw graphical form of vector components. The open ended problem was given to 211 first year students of faculty of science during the third (summer) semester in 2014 academic year. The students spent approximately 15 minutes of their second time of the General Physics I course to complete the open ended problem after they had failed. Consequently, their responses were classified based on the similarity of errors performed in the responses. Then, an unstructured interview was conducted. 7 students were randomly selected and asked to reason and explain their answers. The study results showed that 53% of 211 students provided correct numerical magnitude of vector components while 10.9% of them confused and punctuated the magnitude of vectors in x- with y-components. Others 20.4% provided just symbols and the last 15.6% gave no answer. When asking to draw graphical form of vector components, only 10% of 211 students made corrections. A majority of them produced errors and revealed alternative conceptions. 46.5% drew longer and/or shorter magnitude of vector components. 43.1% drew vectors in different forms or wrote down other symbols. Results from the unstructured interview indicated that some students just memorized the method to get numerical magnitude of x- and y-components. About graphical form of component vectors, some students though that the length of component vectors should be shorter than those of the given one. So then, it could be combined to be equal length of the given vectors while others though that component vectors should has the same length as the given vectors. It was likely to be that many students did not develop a strong foundation of understanding in vector components but just learn by memorizing its solution or the way to compute its magnitude and attribute little meaning to such concept.

Keywords: graphical vectors, vectors, vector components, misconceptions, alternative conceptions

Procedia PDF Downloads 169

Authors: Shaibu Baanni Azumah, Ignatius Tindjina, Stella Obanyi, Tara N. Wood

Abstract:

This study examined the effect of Urea Deep Placement (UDP) technology on the output of irrigated rice farmers in the northern region of Ghana. Multi-stage sampling technique was used to select 142 rice farmers from the Golinga and Bontanga irrigation schemes, around Tamale. A treatment effect model was estimated at two stages; firstly, to determine the factors that influenced farmers’ decision to adopt the UDP technology and secondly, to determine the effect of the adoption of the UDP technology on the output of rice farmers. The significant variables that influenced rice farmers’ adoption of the UPD technology were sex of the farmer, land ownership, off-farm activity, extension service, farmer group participation and training. The results also revealed that farm size and the adoption of UDP technology significantly influenced the output of rice farmers in the northern region of Ghana. In addition to the potential of the technology to improve yields, it also presents an employment opportunity for women and youth, who are engaged in the deep placement of Urea Super Granules (USG), as well as in the transplantation of rice. It is recommended that the government of Ghana work closely with the IFDC to embed the UDP technology in the national agricultural programmes and policies. The study also recommends an effective collaboration between the government, through the Ministry of Food and Agriculture (MoFA) and the International Fertilizer Development Center (IFDC) to train agricultural extension agents on UDP technology in the rice producing areas of the country.

Keywords: Northern Ghana, output , irrigation rice farmers, treatment effect model, urea deep placement

Procedia PDF Downloads 398

Authors: Ozgu Hafizoglu

Abstract:

Analogy is an essential tool of human cognition that enables connecting diffuse and diverse systems with attributional, deep structural, casual relations that are essential to learning, to innovation in artificial worlds, and to discovery in science. Cognitive Model of Analogy (CMA) leads and creates information pattern transfer within and between domains and disciplines in science. This paper demonstrates the Cognitive Model of Analogy (CMA) as an evolutionary approach to scientific research. The model puts forward the challenges of deep uncertainty about the future, emphasizing the need for flexibility of the system in order to enable reasoning methodology to adapt to changing conditions. In this paper, the model of analogical reasoning is created based on brain cells, their fractal, and operational forms within the system itself. Visualization techniques are used to show correspondences. Distinct phases of the problem-solving processes are divided thusly: encoding, mapping, inference, and response. The system is revealed relevant to brain activation considering each of these phases with an emphasis on achieving a better visualization of the brain cells: glial cells, axons, axon terminals, and neurons, relative to matching conditions of analogical reasoning and relational information. It’s found that encoding, mapping, inference, and response processes in four-term analogical reasoning are corresponding with the fractal and operational forms of brain cells: glial, axons, and neurons.

Keywords: analogy, analogical reasoning, cognitive model, brain and glials

Procedia PDF Downloads 163

Authors: Migena Mana, Ahmed Khalid Syed, Abdul Malik, Nikhil Cherian

Abstract:

In recent years, advancements in deep learning enabled researchers to tackle the problem of self-driving cars. Car companies use huge datasets to train their deep learning models to make autonomous cars a reality. However, this approach has certain drawbacks in that the state space of possible actions for a car is so huge that there cannot be a dataset for every possible road scenario. To overcome this problem, the concept of reinforcement learning (RL) is being investigated in this research. Since the problem of autonomous driving can be modeled in a simulation, it lends itself naturally to the domain of reinforcement learning. The advantage of this approach is that we can model different and complex road scenarios in a simulation without having to deploy in the real world. The autonomous agent can learn to drive by finding the optimal policy. This learned model can then be easily deployed in a real-world setting. In this project, we focus on three RL algorithms: Q-learning, Deep Deterministic Policy Gradient (DDPG), and Proximal Policy Optimization (PPO). To model the environment, we have used TORCS (The Open Racing Car Simulator), which provides us with a strong foundation to test our model. The inputs to the algorithms are the sensor data provided by the simulator such as velocity, distance from side pavement, etc. The outcome of this research project is a comparative analysis of these algorithms. Based on the comparison, the PPO algorithm gives the best results. When using PPO algorithm, the reward is greater, and the acceleration, steering angle and braking are more stable compared to the other algorithms, which means that the agent learns to drive in a better and more efficient way in this case. Additionally, we have come up with a dataset taken from the training of the agent with DDPG and PPO algorithms. It contains all the steps of the agent during one full training in the form: (all input values, acceleration, steering angle, break, loss, reward). This study can serve as a base for further complex road scenarios. Furthermore, it can be enlarged in the field of computer vision, using the images to find the best policy.

Keywords: autonomous driving, DDPG (deep deterministic policy gradient), PPO (proximal policy optimization), reinforcement learning

Procedia PDF Downloads 120