Search results for: adaptive deep learning

Authors: Yong Chen

Abstract:

To understand the detailed molecular mechanisms of memory formation in engram cells is one of the most fundamental questions in neuroscience. Recent single-cell RNA-seq (scRNA-seq) and single-nucleus RNA-seq (snRNA-seq) techniques have allowed us to explore the sparsely activated engram ensembles, enabling access to the molecular mechanisms that underlie experience-dependent memory formation and consolidation. However, the absence of specific and powerful computational methods to detect memory-related genes (modules) and their regulatory relationships in the sc/snRNA-seq datasets has strictly limited the analysis of underlying mechanisms and memory coding principles in mammalian brains. Here, we present a deep-learning method named SCENTBOX, to detect memory-related gene modules and causal regulatory relationships among themfromsc/snRNA-seq datasets. SCENTBOX first constructs codifferential expression gene network (CEGN) from case versus control sc/snRNA-seq datasets. It then detects the highly correlated modules of differential expression genes (DEGs) in CEGN. The deep network embedding and attention-based convolutional neural network strategies are employed to precisely detect regulatory relationships among DEG genes in a module. We applied them on scRNA-seq datasets of TRAP; Ai14 mouse neurons with fear memory and detected not only known memory-related genes, but also the modules and potential causal regulations. Our results provided novel regulations within an interesting module, including Arc, Bdnf, Creb, Dusp1, Rgs4, and Btg2. Overall, our methods provide a general computational tool for processing sc/snRNA-seq data from case versus control studie and a systematic investigation of fear-memory-related gene modules.

Keywords: sc/snRNA-seq, memory formation, deep learning, gene module, causal inference

Procedia PDF Downloads 94

Authors: Mingming Jiang

Abstract:

This literature review aims to offer an overview of existing research on learning analytics and learning design, the alignment between the two, and how learning analytics has been leveraged to inform learning design in higher education. Current research suggests a need to create more alignment and integration between learning analytics and learning design in order to not only ground learning analytics on learning sciences but also enable data-driven decisions in learning design to improve learning outcomes. In addition, multiple conceptual frameworks have been proposed to enhance the synergy and alignment between learning analytics and learning design. Future research should explore this synergy further in the unique context of higher education, identifying learning analytics metrics in higher education that can offer insight into learning processes, evaluating the effect of learning analytics outcomes on learning design decision-making in higher education, and designing learning environments in higher education that make the capturing and deployment of learning analytics outcomes more efficient.

Keywords: learning analytics, learning design, big data in higher education, online learning environments

Procedia PDF Downloads 134

Authors: T. Yasmin, M. A. Farrelly, B. C. Rogers

Abstract:

Urban water governance in developing countries faces numerous challenges arising from uncontrolled urban population expansion, water pollution, greater economic push and more recently, climate change impact while undergoing transitioning towards a sustainable system. Sustainability transition requires developing adaptive capacities of the socio-ecological and socio-technical system to be able to deal with complexity. Adaptive capacities deliver strategies to connect individuals, organizations, agencies and institutions at multiple levels for dealing with such complexity. Understanding the level of adaptive capacities for sustainability transformation thus has gained significant research attention within developed countries, much less so in developing countries. Filling this gap, this article develops a conceptual framework for analysing the level of adaptive capacities (if any) within a developing context. This framework then applied to the chronological development of urban water governance strategies in Bangladesh for almost two centuries. The chronological analysis of governance interventions has revealed that crisis (public health, food and natural hazards) became the opportunities and thus opened the windows for experimentation and learning to occur as a deviation from traditional practices. Self-organization and networks thus created the platform for development or disruptions to occur for creating change. Leadership (internal or external) is important for nurturing and upscaling theses development or disruptions towards guiding policy vision and targets as well as championing ground implementation. In the case of Bangladesh, the leadership from the international and national aid organizations and targets have always lead the development whereas more often social capital tools (trust, power relations, cultural norms) act as disruptions. Historically, this has been evident in the development pathways of urban water governance in Bangladesh. Overall this research has shown some level of adaptive capacities is growing for sustainable urban growth in big cities, nevertheless unclear regarding the growth in medium and small cities context.

Keywords: adaptive capacity, Bangladesh, sustainability transformation, water governance

Procedia PDF Downloads 367

Authors: Xiaocong Liu, Huazhen Wang, Ting He, Xiaozheng Li, Weihan Zhang, Jian Chen

Abstract:

The utilization of electronic medical record (EMR) data to establish the disease diagnosis model has become an important research content of biomedical informatics. Deep learning can automatically extract features from the massive data, which brings about breakthroughs in the study of EMR data. The challenge is that deep learning lacks semantic knowledge, which leads to impracticability in medical science. This research proposes a method of incorporating lexical-semantic knowledge from abundant entities into a convolutional neural network (CNN) framework for pediatric disease diagnosis. Firstly, medical terms are vectorized into Lexical Semantic Vectors (LSV), which are concatenated with the embedded word vectors of word2vec to enrich the feature representation. Secondly, the semantic distribution of medical terms serves as Semantic Decision Guide (SDG) for the optimization of deep learning models. The study evaluate the performance of LSV-SDG-CNN model on four kinds of Chinese EMR datasets. Additionally, CNN, LSV-CNN, and SDG-CNN are designed as baseline models for comparison. The experimental results show that LSV-SDG-CNN model outperforms baseline models on four kinds of Chinese EMR datasets. The best configuration of the model yielded an F1 score of 86.20%. The results clearly demonstrate that CNN has been effectively guided and optimized by lexical-semantic knowledge, and LSV-SDG-CNN model improves the disease classification accuracy with a clear margin.

Keywords: convolutional neural network, electronic medical record, feature representation, lexical semantics, semantic decision

Procedia PDF Downloads 111

Authors: M. Abdallah

Abstract:

Urban development requires deep excavations near buildings and other structures. Deep excavation has become more a necessity for better utilization of space as the population of the world has dramatically increased. In Lebanon, some urban areas are very crowded and lack spaces for new buildings and underground projects, which makes the usage of underground space indispensable. In this paper, a numerical modeling is performed using the finite element method to study the deep excavation-diaphragm wall soil-structure interaction in the case of nonlinear soil behavior. The study is focused on a comparison of the results obtained using different support systems. Furthermore, a parametric study is performed according to the remoteness of the structure.

Keywords: deep excavation, ground anchors, interaction soil-structure, struts

Procedia PDF Downloads 384

Authors: Revoti Prasad Bora, Saurabh Yadav, Nikita Katyal

Abstract:

In the context of educational data mining, the use case of extracting information from images containing both text and diagrams is of high importance. Hence, document analysis requires the extraction of diagrams from such images and processes the text and diagrams separately. To the author’s best knowledge, none among plenty of approaches for extracting tables, figures, etc., suffice the need for real-time processing with high accuracy as needed in multiple applications. In the education domain, diagrams can be of varied characteristics viz. line-based i.e. geometric diagrams, chemical bonds, mathematical formulas, etc. There are two broad categories of approaches that try to solve similar problems viz. traditional computer vision based approaches and deep learning approaches. The traditional computer vision based approaches mainly leverage connected components and distance transform based processing and hence perform well in very limited scenarios. The existing deep learning approaches either leverage YOLO or faster-RCNN architectures. These approaches suffer from a performance-accuracy tradeoff. This paper proposes a U-Net based architecture that formulates the diagram extraction as a segmentation problem. The proposed method provides similar accuracy with a much faster extraction time as compared to the mentioned state-of-the-art approaches. Further, the segmentation mask in this approach allows the extraction of diagrams of irregular shapes.

Keywords: computer vision, deep-learning, educational data mining, faster-RCNN, figure extraction, image segmentation, real-time document analysis, text extraction, U-Net, YOLO

Procedia PDF Downloads 107

Authors: Chad Brown

Abstract:

This paper establishes general conditions for the convergence rates of nonparametric sieve estimators with dependent data. We present two key results: one for nonstationary data and another for stationary mixing data. Previous theoretical results often lack practical applicability to deep neural networks (DNNs). Using these conditions, we derive convergence rates for DNN sieve estimators in nonparametric regression settings with both nonstationary and stationary mixing data. The DNN architectures considered adhere to current industry standards, featuring fully connected feedforward networks with rectified linear unit activation functions, unbounded weights, and a width and depth that grows with sample size.

Keywords: sieve extremum estimates, nonparametric estimation, deep learning, neural networks, rectified linear unit, nonstationary processes

Procedia PDF Downloads 8

Authors: Leidy M. Aldana, Jorge E. Camargo

Abstract:

Adversarial Machine Learning has gained importance in recent years as Cybersecurity has gained too, especially malware, it has affected different entities and people in recent years. This paper shows a literature review about defense methods created to prevent adversarial machine learning attacks, firstable it shows an introduction about the context and the description of some terms, in the results section some of the attacks are described, focusing on detecting adversarial examples before coming to the machine learning algorithm and showing other categories that exist in defense. A method with five steps is proposed in the method section in order to define a way to make the literature review; in addition, this paper summarizes the contributions in this research field in the last seven years to identify research directions in this area. About the findings, the category with least quantity of challenges in defense is the Detection of adversarial examples being this one a viable research route with the adaptive approach in attack and defense.

Keywords: Malware, adversarial, machine learning, defense, attack

Procedia PDF Downloads 31

Authors: Saba Ebrahimi, Saeed Ahmadian, Hedie Ashrafi

Abstract:

World-wide spread of coronavirus grows the concern about planning for the excess demand of hospital services in response to COVID-19 pandemic. The surge in the hospital services demand beyond the current capacity leads to shortage of ICU beds and ventilators in some parts of US. In this study, we forecast the required number of hospital beds and possible shortage of beds in US during COVID-19 pandemic to be used in the planning and hospitalization of new cases. In this paper, we used a data on COVID-19 deaths and patients’ hospitalization besides the data on hospital capacities and utilization in US from publicly available sources and national government websites. we used a novel ensemble modelling of deep learning networks, based on stacking different linear and non-linear layers to predict the shortage in hospital beds. The results showed that our proposed approach can predict the excess hospital beds demand very well and this can be helpful in developing strategies and plans to mitigate this gap.

Keywords: COVID-19, deep learning, ensembled models, hospital capacity planning

Procedia PDF Downloads 133

Authors: Hussein Shaman

Abstract:

This study introduces an innovative design for an RF phase shifter that can maintain a consistent phase shift across a broad spectrum of frequencies. The proposed design integrates an adaptive control system into a reflective-type phase shifter, typically showing frequency-related variations. Adjusting the DC voltage according to the frequency ensures a more reliable phase shift across the frequency span of operation. In contrast, conventional frequency-dependent reflective-type phase shifters may exhibit significant fluctuations in phase shifts exceeding 60 degrees in the same bandwidth. The proposed phase shifter is configured to deliver a 90-degree operation with an expected deviation of around 15 degrees. The fabrication of the phase shifter and adaptive control circuit has been verified through experimentation, with the measured outcomes aligning with the simulation results.

Keywords: phase shifter, adaptive control, varactors, electronic circuits.

Procedia PDF Downloads 36

Authors: Duygu Dere, Mert Ergeneci, Kaan Gokcesu

Abstract:

Google Trends data has gained increasing popularity in the applications of behavioral finance, decision science and risk management. Because of Google’s wide range of use, the Trends statistics provide significant information about the investor sentiment and intention, which can be used as decisive factors for corporate and risk management fields. However, an anomaly, a significant increase or decrease, in a certain query cannot be detected by the state of the art applications of computation due to the random baseline noise of the Trends data, which is modelled as an Additive white Gaussian noise (AWGN). Since through time, the baseline noise power shows a gradual change an adaptive thresholding method is required to track and learn the baseline noise for a correct classification. To this end, we introduce an online method to classify meaningful deviations in Google Trends data. Through extensive experiments, we demonstrate that our method can successfully classify various anomalies for plenty of different data.

Keywords: adaptive data processing, behavioral finance , convex optimization, online learning, soft minimum thresholding

Procedia PDF Downloads 140

Authors: Bouchemha Amel, Chachoui Takieddine, H. Maalem

Abstract:

Automatic detection of targets in a modern communication system RADAR is based primarily on the concept of adaptive CFAR detector. To have an effective detection, we must minimize the influence of disturbances due to the clutter. The detection algorithm adapts the CFAR detection threshold which is proportional to the average power of the clutter, maintaining a constant probability of false alarm. In this article, we analyze the performance of two variants of adaptive algorithms CA-CFAR and OS-CFAR and we compare the thresholds of these detectors in the marine environment (no-Gaussian) with a Weibull distribution.

Keywords: CFAR, threshold, clutter, distribution, Weibull, detection

Procedia PDF Downloads 555

Authors: Ksenia Volkova, Artur Petrosyan, Ignatii Dubyshkin, Alexei Ossadtchi

Abstract:

Brain-computer interfaces are a growing research field producing many implementations that find use in different fields and are used for research and practical purposes. Despite the popularity of the implementations using non-invasive neuroimaging methods, radical improvement of the state channel bandwidth and, thus, decoding accuracy is only possible by using invasive techniques. Electrocorticography (ECoG) is a minimally invasive neuroimaging method that provides highly informative brain activity signals, effective analysis of which requires the use of machine learning methods that are able to learn representations of complex patterns. Deep learning is a family of machine learning algorithms that allow learning representations of data with multiple levels of abstraction. This study explores the potential of deep learning approaches for ECoG processing, decoding movement intentions and the perception of proprioceptive information. To obtain synchronous recording of kinematic movement characteristics and corresponding electrical brain activity, a series of experiments were carried out, during which subjects performed finger movements at their own pace. Finger movements were recorded with a three-axis accelerometer, while ECoG was synchronously registered from the electrode strips that were implanted over the contralateral sensorimotor cortex. Then, multichannel ECoG signals were used to track finger movement trajectory characterized by accelerometer signal. This process was carried out both causally and non-causally, using different position of the ECoG data segment with respect to the accelerometer data stream. The recorded data was split into training and testing sets, containing continuous non-overlapping fragments of the multichannel ECoG. A deep convolutional neural network was implemented and trained, using 1-second segments of ECoG data from the training dataset as input. To assess the decoding accuracy, correlation coefficient r between the output of the model and the accelerometer readings was computed. After optimization of hyperparameters and training, the deep learning model allowed reasonably accurate causal decoding of finger movement with correlation coefficient r = 0.8. In contrast, the classical Wiener-filter like approach was able to achieve only 0.56 in the causal decoding mode. In the noncausal case, the traditional approach reached the accuracy of r = 0.69, which may be due to the presence of additional proprioceptive information. This result demonstrates that the deep neural network was able to effectively find a representation of the complex top-down information related to the actual movement rather than proprioception. The sensitivity analysis shows physiologically plausible pictures of the extent to which individual features (channel, wavelet subband) are utilized during the decoding procedure. In conclusion, the results of this study have demonstrated that a combination of a minimally invasive neuroimaging technique such as ECoG and advanced machine learning approaches allows decoding motion with high accuracy. Such setup provides means for control of devices with a large number of degrees of freedom as well as exploratory studies of the complex neural processes underlying movement execution.

Keywords: brain-computer interface, deep learning, ECoG, movement decoding, sensorimotor cortex

Procedia PDF Downloads 143

Authors: Babak Bahri, Fatemeh Yassaee Meybodi, Changiz Eslahchi

Abstract:

In the pursuit of effective cancer therapies, the exploration of combinatorial drug regimens is crucial to leverage synergistic interactions between drugs, thereby improving treatment efficacy and overcoming drug resistance. However, identifying synergistic drug pairs poses challenges due to the vast combinatorial space and limitations of experimental approaches. This study introduces ClusterSyn, a machine learning (ML)-powered framework for classifying anti-cancer drug synergy scores. ClusterSyn employs a two-step approach involving drug clustering and synergy score prediction using a fully connected deep neural network. For each cell line in the training dataset, a drug graph is constructed, with nodes representing drugs and edge weights denoting synergy scores between drug pairs. Drugs are clustered using the Markov clustering (MCL) algorithm, and vectors representing the similarity of drug pairs to each cluster are input into the deep neural network for synergy score prediction (synergy or antagonism). Clustering results demonstrate effective grouping of drugs based on synergy scores, aligning similar synergy profiles. Subsequently, neural network predictions and synergy scores of the two drugs on others within their clusters are used to predict the synergy score of the considered drug pair. This approach facilitates comparative analysis with clustering and regression-based methods, revealing the superior performance of ClusterSyn over state-of-the-art methods like DeepSynergy and DeepDDS on diverse datasets such as Oniel and Almanac. The results highlight the remarkable potential of ClusterSyn as a versatile tool for predicting anti-cancer drug synergy scores.

Keywords: drug synergy, clustering, prediction, machine learning., deep learning

Procedia PDF Downloads 48

Authors: Asimina Paparrigopoulou, John Pavlopoulos, Maria Konstantinidou

Abstract:

Dating papyri accurately is crucial not only to editing their texts but also for our understanding of palaeography and the history of writing, ancient scholarship, material culture, networks in antiquity, etc. Most ancient manuscripts offer little evidence regarding the time of their production, forcing papyrologists to date them on palaeographical grounds, a method often criticized for its subjectivity. By experimenting with data obtained from the Collaborative Database of Dateable Greek Bookhands and the PapPal online collections of objectively dated Greek papyri, this study shows that deep learning dating models, pre-trained on generic images, can achieve accurate chronological estimates for a test subset (67,97% accuracy for book hands and 55,25% for documents). To compare the estimates of these models with those of humans, experts were asked to complete a questionnaire with samples of literary and documentary hands that had to be sorted chronologically by century. The same samples were dated by the models in question. The results are presented and analysed.

Keywords: image classification, papyri images, dating

Procedia PDF Downloads 59

Authors: Sarita Agarwal, Deepika Delsa Dean

Abstract:

Like the task of transferring photo images to artistic images, image-to-image translation aims to translate the data to the imitated data which belongs to the target domain. Neural Style Transfer and CycleGAN are two well-known deep learning architectures used for photo image-to-art image transfer. However, studies involving these two models concentrate on one-to-one domain translation, not one-to-multi domains translation. Our study tries to investigate deep learning architectures, which can be controlled to yield multiple artistic style translation only by adding a conditional vector. We have expanded CycleGAN and constructed Conditional CycleGAN for 5 kinds of categories translation. Our study found that the architecture inserting conditional vector into the middle layer of the Generator could output multiple artistic images.

Keywords: genetic counseling, FMR1 gene, fragile x-associated primary ovarian insufficiency, premutation

Procedia PDF Downloads 99

Authors: Rizwan Rizwan

Abstract:

Differentially private ({DP}) training preserves the data privacy but often leads to slower convergence and lower accuracy, along with notable mis-calibration compared to non-private training. Analyzing {DP} training through a continuous-time approach with the neural tangent kernel ({NTK}). The {NTK} helps characterize per sample {(PS)} gradient clipping and the incorporation of noise during {DP} training across arbitrary network architectures as well as loss functions. Our analysis reveals that noise addition impacts privacy risk exclusively, leaving convergence and calibration unaffected. In contrast, {PS} gradient clipping (flat styles, layerwise styles) influences convergence as well as calibration but not privacy risk. Models with a small clipping norm generally achieve optimal accuracy but exhibit poor calibration, making them less reliable. Conversely, {DP} models that are trained with a large clipping norm maintain the similar accuracy and same privacy guarantee, yet they demonstrate notably improved calibration.

Keywords: deep learning, convergence, differential privacy, calibration

Procedia PDF Downloads 17

Authors: Carly Madge, Melanie Zurba, Ryan Bullock

Abstract:

The Boreal Forest has experienced some of the most drastic climate change-induced temperature rises in Canada, with average winter temperatures increasing by 3°C since 1948. One of the main concerns of the province of Saskatchewan, and particularly wildfire managers, is the increased risk of wildfires due to climate change. With these concerns in mind Sakaw Askiy Management Inc., a forestry corporation located in Prince Albert, Saskatchewan with operations in the Boreal Forest biome, is developing wildfire risk reduction strategies that are supported by the shareholders of the corporation as well as the stakeholders of the Prince Albert Forest Management Area (which includes citizens, hunters, trappers, cottage owners, and outfitters). In the past, wildfire management strategies implemented through harvesting have been received with skepticism by some community members of Prince Albert. Engagement of the stakeholders of the Prince Albert Management Area through the development of the wildfire risk reduction strategies aims to reduce this skepticism and rebuild some of the trust that has been lost between industry and community. This research project works with the framework of social learning, which is defined as the learning that occurs when individuals come together to form a group with the purpose of understanding environmental challenges and determining appropriate responses to them. The project evaluates the social learning processes that occur through the development of the risk reduction strategies and how the learning has allowed Sakaw to work towards implementing the strategies into their forest harvesting plans. The incorporation of wildfire risk reduction strategies works to increase the adaptive capacity of Sakaw, which in this case refers to the ability to adjust to climate change, moderate potential damages, take advantage of opportunities, and cope with consequences. Using semi-structured interviews and wildfire workshop meetings shareholders and stakeholders shared their knowledge of wildfire, their main wildfire concerns, and changes they would like to see made in the Prince Albert Forest Management Area. Interviews and topics discussed in the workshops were inductively coded for themes related to learning, adaptive capacity, areas of concern, and preferred methods of wildfire risk reduction strategies. Analysis determined that some of the learning that has occurred has resulted through social interactions and the development of networks oriented towards wildfire and wildfire risk reduction strategies. Participants have learned new knowledge and skills regarding wildfire risk reduction. The formation of wildfire networks increases access to information on wildfire and the social capital (trust and strengthened relations) of wildfire personnel. Both factors can be attributed to increases in adaptive capacity. Interview results were shared with the General Manager of Sakaw, where the areas of concern and preferred strategies of wildfire risk reduction will be considered and accounted for in the implementation of new harvesting plans. This research also augments the growing conceptual and empirical evidence of the important role of learning and networks in regional wildfire risk management efforts.

Keywords: adaptive capacity, community-engagement, social learning, wildfire risk reduction

Procedia PDF Downloads 119

Authors: Liliana Villavicencio, Yohn Garcia, Pallavi Singh, Luis Fernando Cruz, Wilfrido Moreno

Abstract:

Psychophysiological adaptive automation is a concept that combines human physiological data and computer algorithms to create personalized interfaces and experiences for users. This approach aims to enhance human learning by adapting to individual needs and preferences and optimizing the interaction between humans and machines. According to neurosciences, the working memory demand during the student learning process is modified when the student is learning a new subject or topic, managing and/or fulfilling a specific task goal. A sudden increase in working memory demand modifies the level of students’ attention, engagement, and cognitive load. The proposed psychophysiological adaptive automation system will adapt the task requirements to optimize cognitive load, the process output variable, by monitoring the student's brain activity. Cognitive load changes according to the student’s previous knowledge, the type of task, the difficulty level of the task, and the overall psychophysiological state of the student. Scaling the measured cognitive load as low, medium, or high; the system will assign a task difficulty level to the next task according to the ratio between the previous-task difficulty level and student stress. For instance, if a student becomes stressed or overwhelmed during a particular task, the system detects this through signal measurements such as brain waves, heart rate variability, or any other psychophysiological variables analyzed to adjust the task difficulty level. The control of engagement and stress are considered internal variables for the hypermedia system which selects between three different types of instructional material. This work assesses the feasibility of a fuzzy controller to track a student's physiological responses and adjust the learning content and pace accordingly. Using an industrial automation approach, the proposed fuzzy logic controller is based on linguistic rules that complement the instrumentation of the system to monitor and control the delivery of instructional material to the students. From the test results, it can be proved that the implemented fuzzy controller can satisfactorily regulate the delivery of academic content based on the working memory demand without compromising students’ health. This work has a potential application in the instructional design of virtual reality environments for training and education.

Keywords: fuzzy logic controller, hypermedia control system, personalized education, psychophysiological adaptive automation

Procedia PDF Downloads 55

Authors: Alireza Alesaadi

Abstract:

Reliability analysis is a powerful method for determining the weak points of the electrical networks. In designing of electrical network, it is tried to design the most reliable network with minimal system shutting down, but it is usually associated with increasing the cost. In this paper, using adaptive genetic algorithm, a method was presented that provides the most reliable system with a certain economical cost. Finally, the proposed method is applied to a sample network and results will be analyzed.

Keywords: reliability, adaptive genetic algorithm, electrical network, communication engineering

Procedia PDF Downloads 469

Authors: Hossein Kheiri, Bashir Naderi, Mohamad Reza Niknam

Abstract:

In this paper we study the optimal synchronization of chaotic T-system with complete uncertain parameter. Optimal control laws and parameter estimation rules are obtained by using Hamilton-Jacobi-Bellman (HJB) technique and Lyapunov stability theorem. The derived control laws are optimal adaptive control and make the states of drive and response systems asymptotically synchronized. Numerical simulation shows the effectiveness and feasibility of the proposed method.

Keywords: Lyapunov stability, synchronization, chaos, optimal control, adaptive control

Procedia PDF Downloads 451

Authors: Deawan Rakin Ahamed Remal, Sinthia Chowdhury, Sharun Akter Khushbu, Sheak Rashed Haider Noori

Abstract:

Native Speech to text synthesis has its own leverage for the purpose of mankind. The extensive nature of art to speaking different accents is common but the purpose of communication between two different accent types of people is quite difficult. This problem will be motivated by the extraction of the wrong perception of language meaning. Thus, many existing automatic speech recognition has been placed to detect text. Overall study of this paper mentions a review of NSTTR (Native Speech Text to Text Recognition) synthesis compared with Text to Text recognition. Review has exposed many text to text recognition systems that are at a very early stage to comply with the system by native speech recognition. Many discussions started about the progression of chatbots, linguistic theory another is rule based approach. In the Recent years Deep learning is an overwhelming chapter for text to text learning to detect language nature. To the best of our knowledge, In the sub continent a huge number of people speak in Bangla language but they have different accents in different regions therefore study has been elaborate contradictory discussion achievement of existing works and findings of future needs in Bangla language acoustic accent.

Keywords: TTR, NSTTR, text to text recognition, deep learning, natural language processing

Procedia PDF Downloads 105

Authors: Jacqueline Rose T. Alipo-on, Francesca Isabelle F. Escobar, Myles Joshua T. Tan, Hezerul Abdul Karim, Nouar Al Dahoul

Abstract:

Electrocardiogram (ECG) signal analysis and processing are crucial in the diagnosis of cardiovascular diseases, which are considered one of the leading causes of mortality worldwide. However, the traditional rule-based analysis of large volumes of ECG data is time-consuming, labor-intensive, and prone to human errors. With the advancement of the programming paradigm, algorithms such as machine learning have been increasingly used to perform an analysis of ECG signals. In this paper, various deep learning algorithms were adapted to classify five classes of heartbeat types. The dataset used in this work is the synthetic MIT-BIH Arrhythmia dataset produced from generative adversarial networks (GANs). Various deep learning models such as ResNet-50 convolutional neural network (CNN), 1-D CNN, and long short-term memory (LSTM) were evaluated and compared. ResNet-50 was found to outperform other models in terms of recall and F1 score using a five-fold average score of 98.88% and 98.87%, respectively. 1-D CNN, on the other hand, was found to have the highest average precision of 98.93%.

Keywords: heartbeat classification, convolutional neural network, electrocardiogram signals, generative adversarial networks, long short-term memory, ResNet-50

Procedia PDF Downloads 94

Authors: Wann-Yun Shieh, Chin-Man Wang, Ya-Cheng Shieh

Abstract:

This study aims to develop an objective scoring system to evaluate hand-eye coordination and hand dexterity for Parkinson’s disease. This system contains three boards, and each of them is implemented with the sensors to sense a user’s finger operations. The operations include the peg test, the block test, and the blind block test. A user has to use the vision, hearing, and tactile abilities to finish these operations, and the board will record the results automatically. These results can help the physicians to evaluate a user’s reaction, coordination, dexterity function. The results will be collected to a cloud database for further analysis and statistics. A researcher can use this system to obtain systematic, graphic reports for an individual or a group of users. Particularly, a deep learning model is developed to learn the features of the data from different users. This model will help the physicians to assess the Parkinson’s disease symptoms by a more intellective algorithm.

Keywords: deep learning, hand-eye coordination, reaction, hand dexterity

Procedia PDF Downloads 42

Authors: Mst Shapna Akter, Hossain Shahriar

Abstract:

One of the most important challenges in the field of software code audit is the presence of vulnerabilities in software source code. Every year, more and more software flaws are found, either internally in proprietary code or revealed publicly. These flaws are highly likely exploited and lead to system compromise, data leakage, or denial of service. C and C++ open-source code are now available in order to create a largescale, machine-learning system for function-level vulnerability identification. We assembled a sizable dataset of millions of opensource functions that point to potential exploits. We developed an efficient and scalable vulnerability detection method based on deep neural network models that learn features extracted from the source codes. The source code is first converted into a minimal intermediate representation to remove the pointless components and shorten the dependency. Moreover, we keep the semantic and syntactic information using state-of-the-art word embedding algorithms such as glove and fastText. The embedded vectors are subsequently fed into deep learning networks such as LSTM, BilSTM, LSTM-Autoencoder, word2vec, BERT, and GPT-2 to classify the possible vulnerabilities. Furthermore, we proposed a neural network model which can overcome issues associated with traditional neural networks. Evaluation metrics such as f1 score, precision, recall, accuracy, and total execution time have been used to measure the performance. We made a comparative analysis between results derived from features containing a minimal text representation and semantic and syntactic information. We found that all of the deep learning models provide comparatively higher accuracy when we use semantic and syntactic information as the features but require higher execution time as the word embedding the algorithm puts on a bit of complexity to the overall system.

Keywords: cyber security, vulnerability detection, neural networks, feature extraction

Procedia PDF Downloads 57

Authors: Bessaad Taieb, Benbouali Abderrahmen

Abstract:

Fuzzy controllers are a powerful tool for controlling complex processes. However, its robustness capacity remains moderately limited because it loses its property for large ranges of parametric variations. In this paper, the proposed control method is designed, based on a fuzzy adaptive controller used as a remedy for this problem. For increase the robustness of the vector control and to maintain the performance of the five-phase asynchronous machine despite the presence of disturbances (variation of rotor resistance, rotor inertia variations, sudden variations in the load etc.), by applying the method of behaviour model control (BMC). The results of simulation show that the fuzzy adaptive control provides best performance and has a more robustness as the fuzzy (FLC) and as a conventional (PI) controller.

Keywords: fuzzy adaptive control, behaviour model control, vector control, five-phase asynchronous machine

Procedia PDF Downloads 66

Authors: Akshay Paranjape, Nils Plettenberg, Robert Schmitt

Abstract:

Traditional optimization methods like evolutionary algorithms are widely used in production processes to find an optimal or near-optimal solution of control parameters based on the simulated environment space of a process. These algorithms are computationally intensive and therefore do not provide the opportunity for real-time optimization. This paper utilizes the Deep Reinforcement Learning (DRL) framework to find an optimal or near-optimal solution for control parameters. A model based on maximum a posteriori policy optimization (Hybrid-MPO) that can handle both numerical and categorical parameters is used as a benchmark for comparison. A comparative study shows that DRL can find optimal solutions of similar quality as compared to evolutionary algorithms while requiring significantly less time making them preferable for real-time optimization. The results are confirmed in a large-scale validation study on datasets from production and other fields. A trained XGBoost model is used as a surrogate for process simulation. Finally, multiple ways to improve the model are discussed.

Keywords: reinforcement learning, evolutionary algorithms, production process optimization, real-time optimization, hybrid-MPO

Procedia PDF Downloads 83

Authors: Khaing Yadana Swe, Lillie Dewan

Abstract:

At present, the cascade PID control is widely used to control the super-heating temperature (main steam temperature). As the main steam temperature has the characteristics of large inertia, large time-delay, and time varying, etc., conventional PID control strategy can not achieve good control performance. In order to overcome the bad performance and deficiencies of main steam temperature control system, Model Free Adaptive Control (MFAC) P cascade control system is proposed in this paper. By substituting MFAC in PID of the main control loop of the main steam temperature control, it can overcome time delays, non-linearity, disturbance and time variation.

Keywords: model-free adaptive control, cascade control, adaptive control, PID

Procedia PDF Downloads 580

Authors: Cesar Vazquez-Gonzalez, Sophie Avila-Foucat, Leonardo Ortiz-Lozano, Patricia Moreno-Casasola, Alejandro Granados-Barba

Abstract:

In the last twenty years, academic and scientific literature has been focused on understanding the processes and factors of coastal social-ecological systems vulnerability and resilience. Some scholars argue that resilience and vulnerability are isolated concepts due to their epistemological origin, while others note the existence of a strong resilience-vulnerability relationship. Here we present an ordinal logistic regression model based on the analytical framework about dynamic resilience-vulnerability interaction along adaptive cycle of complex systems and disasters process phases (during, recovery and learning). In this way, we demonstrate that 1) during the disturbance, absorptive capacity (resilience as a core of attributes) and external response capacity explain the probability of households capitals to diminish the damage, and exposure sets the thresholds about the amount of disturbance that households can absorb, 2) at recovery, absorptive capacity and external response capacity explain the probability of households capitals to recovery faster (resilience as an outcome) from damage, and 3) at learning, adaptive capacity (resilience as a core of attributes) explains the probability of households adaptation measures based on the enhancement of physical capital. As a result, during the disturbance phase, exposure has the greatest weight in the probability of capital’s damage, and households with absorptive and external response capacity elements absorbed the impact of floods in comparison with households without these elements. At the recovery phase, households with absorptive and external response capacity showed a faster recovery on their capital; however, the damage sets the thresholds of recovery time. More importantly, diversity in financial capital increases the probability of recovering other capital, but it becomes a liability so that the probability of recovering the household finances in a longer time increases. At learning-reorganizing phase, adaptation (modifications to the house) increases the probability of having less damage on physical capital; however, it is not very relevant. As conclusion, resilience is an outcome but also core of attributes that interacts with vulnerability along the adaptive cycle and disaster process phases. Absorptive capacity can diminish the damage experienced by floods; however, when exposure overcomes thresholds, both absorptive and external response capacity are not enough. In the same way, absorptive and external response capacity diminish the recovery time of capital, but the damage sets the thresholds in where households are not capable of recovering their capital.

Keywords: absorptive capacity, adaptive capacity, capital, floods, recovery-learning, social-ecological systems

Procedia PDF Downloads 110

Authors: Ogar Ofut Tumenayu, Olga Shabalina

Abstract:

This paper describes the application of two types of pedagogical agents’ technology with different functions in an adaptive educational game with the sole aim of improving learning and enhancing interactivities in Digital Educational Games (DEG). This idea could promote the elimination of some problems of DEG, like isolation in game-based learning, by introducing a tutor and tutee pedagogical agents. We present an analysis of a learning companion interacting in a peer tutoring environment as a step toward improving social interactions in the educational game environment. We show that tutor and tutee agents use different interventions and interactive approaches: the tutor agent is engaged in tracking the learner’s activities and inferring the learning state, while the tutee agent initiates interactions with the learner at the appropriate times and in appropriate manners. In order to provide motivation to prevent mistakes and clarity a game task, the tutor agent uses the help dialog tool to provide assistance, while the tutee agent provides collaboration assistance by using the hind tool. We presented our idea on a prototype game called “Pyramid Programming Game,” a 2D game that was developed using Libgdx. The game's Pyramid component symbolizes a programming task that is presented to the player in the form of a puzzle. During gameplay, the Agents can instruct, direct, inspire, and communicate emotions. They can also rapidly alter the instructional pattern in response to the learner's performance and knowledge. The pyramid must be effectively destroyed in order to win the game. The game also teaches and illustrates the advantages of utilizing educational agents such as TrA and TeA to assist and motivate students. Our findings support the idea that the functionality of a pedagogical agent should be dualized into an instructional and learner’s companion agent in order to enhance interactivity in a game-based environment.

Keywords: tutor agent, tutee agent, learner’s companion interaction, agent collaboration

Procedia PDF Downloads 42