Search results for: fine-grained dataset

Authors: Briony Gray, Glenn Simpson, Hajira Dambha-Miller, Andrew Farmer

Abstract:

Multimorbidity may be strongly affected by a variety of conditions, factors, and variables requiring higher demands on health and social care services, infrastructure, and expenses. Holding one or more conditions increases one’s risk for development of future conditions; with patients over 65 years old at highest risk. Psychosocial factors such as anxiety and depression are rising exponentially globally, which has been amplified by the COVID19 pandemic. These are highly correlated and predict poorer outcomes when held in coexistence and increase the likelihood of comorbid physical health conditions. While possible future reform of social and healthcare systems may help to alleviate some of these mounting pressures, there remains an urgent need to better understand the potential role health behaviours and psychosocial conditions - such as anxiety and depression – may have on aging populations. Using the UK healthcare scene as a lens for analysis, this study uses big data collected in the UK Longitudinal Study of Aging (ELSA) to examine the role of anxiety and depression in ageing populations (65yrs+). Using logistic regression modelling, results identify the 10 most significant variables correlated with both anxiety and depression from data categorised into the areas of health behaviour, psychosocial, socioeconomic, and life satisfaction (each demonstrated through literature review to be of significance). These are compared with wider global research findings with the aim of better understanding the areas in which social and healthcare reform can support multimorbidity interventions, making suggestions for improved patient-centred care. Scope of future research is outlined, which includes analysis of 59 total multimorbidity variables from the ELSA dataset, going beyond anxiety and depression.

Keywords: multimorbidity, health behaviours, patient centred care, psychosocial factors

Procedia PDF Downloads 75

Authors: Joseph Lillington, Tom Gout, Mike Harrison, Ian Farnan

Abstract:

The vitrification of high-level nuclear waste within borosilicate glass and its incorporation within a multi-barrier repository deep underground is widely accepted as the preferred disposal method. However, for this to happen, any safety case will require validation that the initially localized radionuclides will not be considerably released into the near/far-field. Therefore, accurate mechanistic models are necessary to predict glass dissolution, and these should be robust to a variety of incorporated waste species and leaching test conditions, particularly given substantial variations across international waste-streams. Here, machine learning is used to predict glass material properties (viscosity, density) and glass leaching model parameters from large-scale industrial data. A variety of different machine learning algorithms have been compared to assess performance. Density was predicted solely from composition, whereas viscosity additionally considered temperature. To predict suitable glass leaching model parameters, a large simulated dataset was created by coupling MATLAB and the chemical reactive-transport code HYTEC, considering the state-of-the-art GRAAL model (glass reactivity in allowance of the alteration layer). The trained models were then subsequently applied to the large-scale industrial, experimental data to identify potentially appropriate model parameters. Results indicate that ensemble methods can accurately predict viscosity as a function of temperature and composition across all three industrial datasets. Glass density prediction shows reliable learning performance with predictions primarily being within the experimental uncertainty of the test data. Furthermore, machine learning can predict glass dissolution model parameters behavior, demonstrating potential value in GRAAL model development and in assessing suitable model parameters for large-scale industrial glass dissolution data.

Keywords: machine learning, predictive modelling, pattern recognition, radioactive waste glass

Procedia PDF Downloads 99

Authors: El Harraj Abdeslam, Raissouni Naoufal

Abstract:

The detection of moving objects from a video image sequences is very important for object tracking, activity recognition, and behavior understanding in video surveillance. The most used approach for moving objects detection / tracking is background subtraction algorithms. Many approaches have been suggested for background subtraction. But, these are illumination change sensitive and the solutions proposed to bypass this problem are time consuming. In this paper, we propose a robust yet computationally efficient background subtraction approach and, mainly, focus on the ability to detect moving objects on dynamic scenes, for possible applications in complex and restricted access areas monitoring, where moving and motionless persons must be reliably detected. It consists of three main phases, establishing illumination changes in variance, background/foreground modeling and morphological analysis for noise removing. We handle illumination changes using Contrast Limited Histogram Equalization (CLAHE), which limits the intensity of each pixel to user determined maximum. Thus, it mitigates the degradation due to scene illumination changes and improves the visibility of the video signal. Initially, the background and foreground images are extracted from the video sequence. Then, the background and foreground images are separately enhanced by applying CLAHE. In order to form multi-modal backgrounds we model each channel of a pixel as a mixture of K Gaussians (K=5) using Gaussian Mixture Model (GMM). Finally, we post process the resulting binary foreground mask using morphological erosion and dilation transformations to remove possible noise. For experimental test, we used a standard dataset to challenge the efficiency and accuracy of the proposed method on a diverse set of dynamic scenes.

Keywords: video surveillance, background subtraction, contrast limited histogram equalization, illumination invariance, object tracking, object detection, behavior understanding, dynamic scenes

Procedia PDF Downloads 239

Authors: Dona Ghosh, Zakir Husain

Abstract:

In India, the increase in share of aged has generated many social and economic issues, of which health concerns is a major challenge that society must confront in coming years. Self-reported health (SRH) is a popular health measure in this regard but has been questioned in recent years due to its heavy dependence on the socioeconomic status. So, the validity of SRH, as a measure of health status during old age, is needed to be verified. This paper emphasizes on the self-reported health and related inconsistent responses among elderly in India. The objective of the study is bifurcated into two parts: firstly, to identify the socioeconomic determinants of subjective health status and its change over time; and secondly, to analyse the role of the socioeconomic components in providing inconsistent responses regarding the health status of elderly. Inconsistency in response can rise in two ways: positive response bias (if an individual has a health problem but reports his/her health as good) and negative response bias (if bad health is reported even if there is no health problem). However, in the present study, we focus only on the negative response bias of elderly individuals. To measure the inconsistencies in responses, self-reported health is compared with two types of physical health conditions – existence of chronicle ailment and physical immobility. Using NSS dataset of 60th and 71st rounds, the study found that subjective health has worsened over time in both rural and urban areas. Findings suggest that inconsistency in responses, related to chronic ailment, vary across social classes, living environments, geographical regions, age groups and education levels. On the contrary, variation in inconsistent responses regarding physical mobility is quite rare and difficult to explain by socioeconomic characteristics because most of the indicators are found to be insignificant in this regard. The findings indicate that in case of chronicle ailment, inconsistency between objective and subjective health status largely depends on socioeconomic conditions but the importance of such factors disappears for physical immobility.

Keywords: India, aging, self-reported health, inconsistent responses

Procedia PDF Downloads 271

Authors: Mohamed Saad, Huda Ismail

Abstract:

The purpose of this research is to examine the impact of cryptocurrency classification on money laundering crimes and to analyze how the preferences of criminals differ according to the type of digital currency used. Specifically, we aim to explore the roles of stablecoins, utility coins, and privacy tokens in facilitating or hindering money laundering activities and to identify the key factors that influence the choices of criminals in using these cryptocurrencies. To achieve our research objectives, we used a dataset for the most highly traded cryptocurrencies (32 currencies) that were published on the coin market cap for 2022. In addition to conducting a comprehensive review of the existing literature on cryptocurrency and money laundering, with a focus on stablecoins, utility coins, and privacy tokens, Furthermore, we conducted several Multivariate analyses. Our study reveals that the classification of cryptocurrency plays a significant role in money laundering activities, as criminals tend to prefer certain types of digital currencies over others, depending on their specific needs and goals. Specifically, we found that stablecoins are more commonly used in money laundering due to their relatively stable value and low volatility, which makes them less risky to hold and transfer. Utility coins, on the other hand, are less frequently used in money laundering due to their lack of anonymity and limited liquidity. Finally, privacy tokens, such as Monero and Zcash, are increasingly becoming a preferred choice among criminals due to their high degree of privacy and untraceability. In summary, our study highlights the importance of understanding the nuances of cryptocurrency classification in the context of money laundering and provides insights into the preferences of criminals in using digital currencies for illegal activities. Based on our findings, our recommendation to the policymakers is to address the potential misuse of cryptocurrencies for money laundering. By implementing measures to regulate stable coins, strengthening cross-border cooperation, fostering public-private partnerships, and increasing cooperation, policymakers can help prevent and detect money laundering activities involving digital currencies.

Keywords: crime, cryptocurrency, money laundering, tokens.

Procedia PDF Downloads 70

Authors: Rishabh Nagpal, Riya Uchagaonkar, Venkata Naga Narasimha Ashish Mernedi, Ahmed Hambaba

Abstract:

The rapid evolution of technology in the communication sector continually seeks to bridge the gap between different communities, notably between the deaf community and the hearing world. This project develops a comprehensive American Sign Language (ASL) recognition system, leveraging the advanced capabilities of convolutional neural networks (CNNs) and vision transformers (ViTs) to interpret and translate ASL in real-time. The primary objective of this system is to provide an effective communication tool that enables seamless interaction through accurate sign language interpretation. The architecture of the proposed system integrates dual networks -VGG16 for precise spatial feature extraction and vision transformers for contextual understanding of the sign language gestures. The system processes live input, extracting critical features through these sophisticated neural network models, and combines them to enhance gesture recognition accuracy. This integration facilitates a robust understanding of ASL by capturing detailed nuances and broader gesture dynamics. The system is evaluated through a series of tests that measure its efficiency and accuracy in real-world scenarios. Results indicate a high level of precision in recognizing diverse ASL signs, substantiating the potential of this technology in practical applications. Challenges such as enhancing the system’s ability to operate in varied environmental conditions and further expanding the dataset for training were identified and discussed. Future work will refine the model’s adaptability and incorporate haptic feedback to enhance the interactivity and richness of the user experience. This project demonstrates the feasibility of an advanced ASL recognition system and lays the groundwork for future innovations in assistive communication technologies.

Keywords: sign language, computer vision, vision transformer, VGG16, CNN

Procedia PDF Downloads 18

Authors: Sergey V. Vinogradov

Abstract:

Coastal communities are presently facing increased vulnerabilities due to rising sea levels and shifts in global climate patterns, a trend expected to escalate in the long run. To address the needs of government entities, the public sector, and private enterprises, there is an urgent need to thoroughly investigate, assess, and manage the present and projected risks associated with coastal flooding, including storm surges, sea level rise, and nuisance flooding. In response to these challenges, a practical approach to evaluating storm surge inundation risks has been developed. This methodology offers an integrated assessment of potential flood risk in targeted coastal areas. The physical modeling framework involves simulating synthetic storms and utilizing hydrodynamic models that align with projected future climate and ocean conditions. Both publicly available and site-specific data form the basis for a risk assessment methodology designed to translate inundation model outputs into statistically significant projections of expected financial and operational consequences. This integrated approach produces measurable indicators of impacts stemming from floods, encompassing economic and other dimensions. By establishing connections between the frequency of modeled flood events and their consequences across a spectrum of potential future climate conditions, our methodology generates probabilistic risk assessments. These assessments not only account for future uncertainty but also yield comparable metrics, such as expected annual losses for each inundation event. These metrics furnish stakeholders with a dependable dataset to guide strategic planning and inform investments in mitigation. Importantly, the model's adaptability ensures its relevance across diverse coastal environments, even in instances where site-specific data for analysis may be limited.

Keywords: climate, coastal, surge, risk

Procedia PDF Downloads 37

Authors: Ankit Kamboj, Shikha Talwar, Nilesh Powar

Abstract:

To successfully deliver our products in the market, the employees need to be in a safe environment, especially in an industrial and manufacturing environment. The consequences of delinquency in wearing safety glasses while working in industrial plants could be high risk to employees, hence the need to develop a real-time automatic detection system which detects the persons (violators) not wearing safety glasses. In this study a convolutional neural network (CNN) algorithm called faster region based CNN (Faster RCNN) with rotated bounding box has been used for detecting safety glasses on persons; the algorithm has an advantage of detecting safety glasses with different orientation angles on the persons. The proposed method of rotational bounding boxes with a convolutional neural network first detects a person from the images, and then the method detects whether the person is wearing safety glasses or not. The video data is captured at the entrance of restricted zones of the industrial environment (manufacturing plant), which is further converted into images at 2 frames per second. In the first step, the CNN with pre-trained weights on COCO dataset is used for person detection where the detections are cropped as images. Then the safety goggles are labelled on the cropped images using the image labelling tool called roLabelImg, which is used to annotate the ground truth values of rotated objects more accurately, and the annotations obtained are further modified to depict four coordinates of the rectangular bounding box. Next, the faster RCNN with rotated bounding box is used to detect safety goggles, which is then compared with traditional bounding box faster RCNN in terms of detection accuracy (average precision), which shows the effectiveness of the proposed method for detection of rotatory objects. The deep learning benchmarking is done on a Dell workstation with a 16GB Nvidia GPU.

Keywords: CNN, deep learning, faster RCNN, roLabelImg rotated bounding box, safety goggle detection

Procedia PDF Downloads 117

Authors: Pinna Beatrice., Laini Alex, Negro Giovanni, Burgazzi Gemma, Viaroli Pierluigi, Vezza Paolo

Abstract:

Habitat models rarely consider macroinvertebrates as ecological targets in rivers. Available approaches mainly focus on single macroinvertebrate species, not addressing the ecological needs and functionality of the entire community. This research aimed to provide an approach to model the habitat of the macroinvertebrate community. The approach is based on the recently developed Flow-T index, together with a Random Forest (RF) regression, which is employed to apply the Flow-T index at the meso-habitat scale. Using different datasets gathered from both field data collection and 2D hydrodynamic simulations, the model has been calibrated in the Trebbia river (2019 campaign), and then validated in the Trebbia, Taro, and Enza rivers (2020 campaign). The three rivers are characterized by a braiding morphology, gravel riverbeds, and summer low flows. The RF model selected 12 mesohabitat descriptors as important for the macroinvertebrate community. These descriptors belong to different frequency classes of water depth, flow velocity, substrate grain size, and connectivity to the main river channel. The cross-validation R² coefficient (R²𝒸ᵥ) of the training dataset is 0.71 for the Trebbia River (2019), whereas the R² coefficient for the validation datasets (Trebbia, Taro, and Enza Rivers 2020) is 0.63. The agreement between the simulated results and the experimental data shows sufficient accuracy and reliability. The outcomes of the study reveal that the model can identify the ecological response of the macroinvertebrate community to possible flow regime alterations and to possible river morphological modifications. Lastly, the proposed approach allows extending the MesoHABSIM methodology, widely used for the fish habitat assessment, to a different ecological target community. Further applications of the approach can be related to flow design in both perennial and non-perennial rivers, including river reaches in which fish fauna is absent.

Keywords: ecological flows, macroinvertebrate community, mesohabitat, river habitat modeling

Procedia PDF Downloads 77

Authors: Parvaneh Mahmoudi, Ahmad Moeni, Seyed Mojtaba Khayam Nekoei, Mohsen Mardi, Mehrshad Zeinolabedini, Ghasem Hosseini Salekdeh

Abstract:

Saffron (Crocus sativus L.) is one of the most important spice and medicinal plants. The three-branch style of C. sativus flowers are the most important economic part of the plant and known as saffron, which has several medicinal properties. Despite the economic and biological significance of this plant, knowledge about its molecular characteristics is very limited. In the present study, we, for the first time, constructed a comprehensive dataset for C. sativus stigma through de novo transcriptome sequencing. We performed de novo transcriptome sequencing of C. sativus stigma using the Illumina paired-end sequencing technology. A total of 52075128 reads were generated and assembled into 118075 unigenes, with an average length of 629 bp and an N50 of 951 bp. A total of 66171unigenes were identified, among them, 66171 (56%) were annotated in the non-redundant National Center for Biotechnology Information (NCBI) database, 30938 (26%) were annotated in the Swiss-Prot database, 10273 (8.7%) unigenes were mapped to 141 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database, while 52560 (44%) and 40756 (34%) unigenes were assigned to Gen Ontology (GO) categories and Eukaryotic Orthologous Groups of proteins (KOG), respectively. In addition, 65 candidate genes involved in three stages of crocin biosynthesis were identified. Finally, transcriptome sequencing of saffron stigma was used to identify 6779 potential microsatellites (SSRs) molecular markers. High-throughput de novo transcriptome sequencing provided a valuable resource of transcript sequences of C. sativus in public databases. In addition, most of candidate genes potentially involved in crocin biosynthesis were identified which could be further utilized in functional genomics studies. Furthermore, numerous obtained SSRs might contribute to address open questions about the origin of this amphiploid spices with probable little genetic diversity.

Keywords: saffron, transcriptome, NGS, bioinformatic

Procedia PDF Downloads 75

Authors: Massimo Miccoli, Luca Marangoni, Alberto Aniello Scaringi, Alessandro Marceddu, Alessandro Amicone

Abstract:

The susceptibility of deep neural networks (DNNs) to adversarial manipulations is a recognized challenge within the computer vision domain. Adversarial examples, crafted by adding subtle yet malicious alterations to benign images, exploit this vulnerability. Various defense strategies have been proposed to safeguard DNNs against such attacks, stemming from diverse research hypotheses. Building upon prior work, our approach involves the utilization of autoencoder models. Autoencoders, a type of neural network, are trained to learn representations of training data and reconstruct inputs from these representations, typically minimizing reconstruction errors like mean squared error (MSE). Our autoencoder was trained on a dataset of benign examples; learning features specific to them. Consequently, when presented with significantly perturbed adversarial examples, the autoencoder exhibited high reconstruction errors. The architecture of the autoencoder was tailored to the dimensions of the images under evaluation. We considered various image sizes, constructing models differently for 256x256 and 512x512 images. Moreover, the choice of the computer vision model is crucial, as most adversarial attacks are designed with specific AI structures in mind. To mitigate this, we proposed a method to replace image-specific dimensions with a structure independent of both dimensions and neural network models, thereby enhancing robustness. Our multi-modal autoencoder reconstructs the spectral representation of images across the red-green-blue (RGB) color channels. To validate our approach, we conducted experiments using diverse datasets and subjected them to adversarial attacks using models such as ResNet50 and ViT_L_16 from the torch vision library. The autoencoder extracted features used in a classification model, resulting in an MSE (RGB) of 0.014, a classification accuracy of 97.33%, and a precision of 99%.

Keywords: adversarial attacks, malicious images detector, binary classifier, multimodal transformer autoencoder

Procedia PDF Downloads 88

Authors: Ebipatei Victoria Tunyan, T. A. Cao, Cheol Young Ock

Abstract:

Detecting subjectively biased statements is a vital task. This is because this kind of bias, when present in the text or other forms of information dissemination media such as news, social media, scientific texts, and encyclopedias, can weaken trust in the information and stir conflicts amongst consumers. Subjective bias detection is also critical for many Natural Language Processing (NLP) tasks like sentiment analysis, opinion identification, and bias neutralization. Having a system that can adequately detect subjectivity in text will boost research in the above-mentioned areas significantly. It can also come in handy for platforms like Wikipedia, where the use of neutral language is of importance. The goal of this work is to identify the subjectively biased language in text on a sentence level. With machine learning, we can solve complex AI problems, making it a good fit for the problem of subjective bias detection. A key step in this approach is to train a classifier based on BERT (Bidirectional Encoder Representations from Transformers) as upstream model. BERT by itself can be used as a classifier; however, in this study, we use BERT as data preprocessor as well as an embedding generator for a Bi-LSTM (Bidirectional Long Short-Term Memory) network incorporated with attention mechanism. This approach produces a deeper and better classifier. We evaluate the effectiveness of our model using the Wiki Neutrality Corpus (WNC), which was compiled from Wikipedia edits that removed various biased instances from sentences as a benchmark dataset, with which we also compare our model to existing approaches. Experimental analysis indicates an improved performance, as our model achieved state-of-the-art accuracy in detecting subjective bias. This study focuses on the English language, but the model can be fine-tuned to accommodate other languages.

Keywords: subjective bias detection, machine learning, BERT–BiLSTM–Attention, text classification, natural language processing

Procedia PDF Downloads 112

Authors: Ghufran Ahmed, Muhammad Osama Siddiqui, Shahbaz Siddiqui, Rauf Ahmad Shams Malick, Faisal Khan, Mubashir Khan

Abstract:

In this paper, a machine learning-based approach for early prediction of diseases in cows is proposed. Different ML algos are applied to extract useful patterns from the available dataset. Technology has changed today’s world in every aspect of life. Similarly, advanced technologies have been developed in livestock and dairy farming to monitor dairy cows in various aspects. Dairy cattle monitoring is crucial as it plays a significant role in milk production around the globe. Moreover, it has become necessary for farmers to adopt the latest early prediction technologies as the food demand is increasing with population growth. This highlight the importance of state-ofthe-art technologies in analyzing how important technology is in analyzing dairy cows’ activities. It is not easy to predict the activities of a large number of cows on the farm, so, the system has made it very convenient for the farmers., as it provides all the solutions under one roof. The cattle industry’s productivity is boosted as the early diagnosis of any disease on a cattle farm is detected and hence it is treated early. It is done on behalf of the machine learning output received. The learning models are already set which interpret the data collected in a centralized system. Basically, we will run different algorithms on behalf of the data set received to analyze milk quality, and track cows’ health, location, and safety. This deep learning algorithm draws patterns from the data, which makes it easier for farmers to study any animal’s behavioral changes. With the emergence of machine learning algorithms and the Internet of Things, accurate tracking of animals is possible as the rate of error is minimized. As a result, milk productivity is increased. IoT with ML capability has given a new phase to the cattle farming industry by increasing the yield in the most cost-effective and time-saving manner.

Keywords: IoT, machine learning, health care, dairy cows

Procedia PDF Downloads 43

Authors: Bliss Singhal

Abstract:

Epilepsy is a prevalent neurological disorder affecting approximately 50 million individuals worldwide and 1.2 million Americans. There exist millions of pediatric patients with intractable epilepsy, a condition in which seizures fail to come under control. The occurrence of seizures can result in physical injury, disorientation, unconsciousness, and additional symptoms that could impede children's ability to participate in everyday tasks. Predicting seizures can help parents and healthcare providers take precautions, prevent risky situations, and mentally prepare children to minimize anxiety and nervousness associated with the uncertainty of a seizure. This research proposes a comprehensive framework to predict seizures in pediatric patients by evaluating machine learning algorithms on unimodal neuroimaging data consisting of electroencephalogram signals. The bandpass filtering and independent component analysis proved to be effective in reducing the noise and artifacts from the dataset. Various machine learning algorithms’ performance is evaluated on important metrics such as accuracy, precision, specificity, sensitivity, F1 score and MCC. The results show that the deep learning algorithms are more successful in predicting seizures than logistic Regression, and k nearest neighbors. The recurrent neural network (RNN) gave the highest precision and F1 Score, long short-term memory (LSTM) outperformed RNN in accuracy and convolutional neural network (CNN) resulted in the highest Specificity. This research has significant implications for healthcare providers in proactively managing seizure occurrence in pediatric patients, potentially transforming clinical practices, and improving pediatric care.

Keywords: intractable epilepsy, seizure, deep learning, prediction, electroencephalogram channels

Procedia PDF Downloads 66

Authors: Gayathri Nagarajan, L. D. Dhinesh Babu

Abstract:

Health care is one of the prominent industries that generate voluminous data thereby finding the need of machine learning techniques with big data solutions for efficient processing and prediction. Missing data, incomplete data, real time streaming data, sensitive data, privacy, heterogeneity are few of the common challenges to be addressed for efficient processing and mining of health care data. In comparison with other applications, accuracy and fast processing are of higher importance for health care applications as they are related to the human life directly. Though there are many machine learning techniques and big data solutions used for efficient processing and prediction in health care data, different techniques and different frameworks are proved to be effective for different applications largely depending on the characteristics of the datasets. In this paper, we present a framework that uses ensemble machine learning technique gradient boosted trees for data classification in health care big data. The framework is built on Spark platform which is fast in comparison with other traditional frameworks. Unlike other works that focus on a single technique, our work presents a comparison of six different machine learning techniques along with gradient boosted trees on datasets of different characteristics. Five benchmark health care datasets are considered for experimentation, and the results of different machine learning techniques are discussed in comparison with gradient boosted trees. The metric chosen for comparison is misclassification error rate and the run time of the algorithms. The goal of this paper is to i) Compare the performance of gradient boosted trees with other machine learning techniques in Spark platform specifically for health care big data and ii) Discuss the results from the experiments conducted on datasets of different characteristics thereby drawing inference and conclusion. The experimental results show that the accuracy is largely dependent on the characteristics of the datasets for other machine learning techniques whereas gradient boosting trees yields reasonably stable results in terms of accuracy without largely depending on the dataset characteristics.

Keywords: big data analytics, ensemble machine learning, gradient boosted trees, Spark platform

Procedia PDF Downloads 227

Authors: Abhisek Dash

Abstract:

This study had three neural networks, one for number segmentation, one for number detection and one for number recognition all of which are coupled to one another. All networks were trained on the MNIST dataset and were convolutional. It was assumed that the images had lighter background and darker foreground. The segmentation network took 28x28 images as input and had sixteen outputs. Segmentation training starts when a dark pixel is encountered. Taking a window(7x7) over that pixel as focus, the eight neighborhood of the focus was checked for further dark pixels. The segmentation network was then trained to move in those directions which had dark pixels. To this end the segmentation network had 16 outputs. They were arranged as “go east”, ”don’t go east ”, “go south east”, “don’t go south east”, “go south”, “don’t go south” and so on w.r.t focus window. The focus window was resized into a 28x28 image and the network was trained to consider those neighborhoods which had dark pixels. The neighborhoods which had dark pixels were pushed into a queue in a particular order. The neighborhoods were then popped one at a time stitched to the existing partial image of the number one at a time and trained on which neighborhoods to consider when the new partial image was presented. The above process was repeated until the image was fully covered by the 7x7 neighborhoods and there were no more uncovered black pixels. During testing the network scans and looks for the first dark pixel. From here on the network predicts which neighborhoods to consider and segments the image. After this step the group of neighborhoods are passed into the detection network. The detection network took 28x28 images as input and had two outputs denoting whether a number was detected or not. Since the ground truth of the bounds of a number was known during training the detection network outputted in favor of number not found until the bounds were not met and vice versa. The recognition network was a standard CNN that also took 28x28 images and had 10 outputs for recognition of numbers from 0 to 9. This network was activated only when the detection network votes in favor of number detected. The above methodology could segment connected and overlapping numbers. Additionally the recognition unit was only invoked when a number was detected which minimized false positives. It also eliminated the need for rules of thumb as segmentation is learned. The strategy can also be extended to other characters as well.

Keywords: convolutional neural networks, OCR, text detection, text segmentation

Procedia PDF Downloads 140

Authors: Ibrahim Khan, Waqas Khalid

Abstract:

The Healthcare sector is known to consume a higher proportion of total energy consumption in the HVAC market owing to an excessive cooling and heating requirement in maintaining human thermal comfort in indoor conditions, catering to patients undergoing treatment in hospital wards, rooms, and intensive care units. The indoor thermal comfort conditions in selected hospitals of Islamabad, Pakistan, were measured on a real-time basis with the collection of first-hand experimental data using calibrated sensors measuring Ambient Temperature, Wet Bulb Globe Temperature, Relative Humidity, Air Velocity, Light Intensity and CO2 levels. The Experimental data recorded was analyzed in conjunction with the Thermal Comfort Questionnaire Surveys, where the participants, including patients, doctors, nurses, and hospital staff, were assessed based on their thermal sensation, acceptability, preference, and comfort responses. The Recorded Dataset, including experimental and survey-based responses, was further analyzed in the development of a correlation between operative temperature, operative relative humidity, and other measured operative parameters with the predicted mean vote and adaptive predicted mean vote, with the adaptive temperature and adaptive relative humidity estimated using the seasonal data set gathered for both summer – hot and dry, and hot and humid as well as winter – cold and dry, and cold and humid climate conditions. The Machine Learning Logistic Regression Algorithm was incorporated to train the operative experimental data parameters and develop a correlation between patient sensations and the thermal environmental parameters for which a new ML-based adaptive thermal comfort model was proposed and developed in our study. Finally, the accuracy of our model was determined using the K-fold cross-validation.

Keywords: predicted mean vote, thermal comfort, energy management, logistic regression, machine learning

Procedia PDF Downloads 43

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 70

Authors: Rasha Ahmadi

Abstract:

In December 2019, a coronavirus, currently identified as SARS-CoV-2, produced a series of acute atypical respiratory illnesses in Wuhan, Hubei Province, China. The sickness induced by this virus was named COVID-19. The virus is transmittable between humans and has caused pandemics worldwide. The number of death tolls continues to climb and a huge number of countries have been obliged to perform social isolation and lockdown. Lack of focused therapy continues to be a problem. Epidemiological research showed that senior patients were more susceptible to severe diseases, whereas children tend to have milder symptoms. In this study, we focus on other possible side effects of COVID-19 and more detailed treatment strategies. Using bioinformatics analysis, we first isolated the gene expression profile of patients with severe COVID-19 from the GEO database. Patients' blood samples were used in the GSE183071 dataset. We then categorized the genes with high and low expression. In the next step, we uploaded the genes separately to the Enrichr database and evaluated our data for signs and symptoms as well as related medication regimens. The results showed that 138 genes with high expression and 108 genes with low expression were observed differentially in the severe COVID-19 VS control group. Symptoms and diseases such as embolism and thrombosis of the abdominal aorta, ankylosing spondylitis, suicidal ideation or attempt, regional enteritis were observed in genes with high expression and in genes with low expression of acute and subacute forms of ischemic heart, CNS infection and poliomyelitis, synovitis and tenosynovitis. Following the detection of diseases and possible signs and symptoms, Carmustine, Bithionol, Leflunomide were evaluated more significantly for high-expression genes and Chlorambucil, Ifosfamide, Hydroxyurea, Bisphenol for low-expression genes. In general, examining the different and invisible aspects of COVID-19 and identifying possible treatments can help us significantly in the emergency and hospitalization of patients.

Keywords: phenotypes, drug regimens, gene expression profiles, bioinformatics analysis, severe COVID-19

Procedia PDF Downloads 121

Authors: C. A. Bradley, D. Patsios, D. Berridge

Abstract:

This paper presents evidence to better understand the risk and protective factors related to sibling conflict and the patterns of association between sibling conflict and negative adjustment outcomes by incorporating additional familial and societal factors within statistical models of risk and adjustment. It was conducted through the secondary analysis of a large representative cross-sectional dataset of children in the UK. The original study includes proxy interviews for young children and self-report interviews for adolescents. The study applies an ecological systems framework for the analyses. Hierarchical regression models assess risk and protective factors and adjustment outcomes associated with sibling conflict. Interactions reveal differential effect between contextual risk factors and the social context of influence. The general pattern of findings suggested that, although factors affecting likelihood of experiencing sibling conflict were often determined by child age, some remained consistent across childhood. These factors were often conditional on each other, reinforcing the importance of an ecological framework. Across both age-groups, sibling conflict was associated with siblings closer in age; male sibling groups; most advantaged socio-economic group; and exposure to community violence, such as witnessing violent assault or robbery. The study develops the evidence base on the influence of ethnicity and socio-economic group on sibling conflict by exploring interactions between social context. It also identifies key new areas of influence – such as family structure, disability, and community violence in exacerbating or reducing risk of conflict. The study found negative associations between sibling conflict and young children’s mental well-being and adolescents' mental well-being and anti-social behaviour, but also more context specific associations – such as sibling conflict moderating the negative impact of adversity and high risk experiences for young children such as parental violence toward the child.

Keywords: adjustment, conflict, ecological systems, family systems, risk and protective factors, sibling

Procedia PDF Downloads 92

Authors: Julie V. Logan, Preston T. Webster, Kevin B. Woller, Christian P. Morath, Michael P. Short

Abstract:

Semiconductors, as the base of critical electronic systems, are exposed to damaging radiation while operating in space, nuclear reactors, and particle accelerator environments. What innate property allows some semiconductors to sustain little damage while others accumulate defects rapidly with dose is, at present, poorly understood. This limits the extent to which radiation tolerance can be implemented as a design criterion. To address this problem of determining the driver of semiconductor radiation tolerance, the first step is to generate a dataset of the relative radiation tolerance of a large range of semiconductors (exposed to the same radiation damage and characterized in the same way). To accomplish this, Rutherford backscatter channeling experiments are used to compare the displaced lattice atom buildup in InAs, InP, GaP, GaN, ZnO, MgO, and Si as a function of step-wise alpha particle dose. With this experimental information on radiation-induced incorporation of interstitial defects in hand, hybrid density functional theory electron densities (and their derived quantities) are calculated, and their gradient and Laplacian are evaluated to obtain key fundamental information about the interactions in each material. It is shown that simple, undifferentiated values (which are typically used to describe bond strength) are insufficient to predict radiation tolerance. Instead, the curvature of the electron density at bond critical points provides a measure of radiation tolerance consistent with the experimental results obtained. This curvature and associated forces surrounding bond critical points disfavors localization of displaced lattice atoms at these points, favoring their diffusion toward perfect lattice positions. With this criterion to predict radiation tolerance, simple density functional theory simulations can be conducted on potential new materials to gain insight into how they may operate in demanding high radiation environments.

Keywords: density functional theory, GaN, GaP, InAs, InP, MgO, radiation tolerance, rutherford backscatter channeling

Procedia PDF Downloads 154

Authors: Bin Feng, Zhulin Zong

Abstract:

Radar systems are critical in the modern military, aviation, and maritime operations, and their proper functioning is essential for the success of these operations. However, due to the complexity and sensitivity of radar systems, they are susceptible to various faults that can significantly affect their performance. Traditional radar fault diagnosis strategies rely on expert knowledge and rule-based approaches, which are often limited in effectiveness and require a lot of time and resources. Deep learning has recently emerged as a promising approach for fault diagnosis due to its ability to learn features and patterns from large amounts of data automatically. In this paper, we propose a radar fault diagnosis strategy based on deep learning that can accurately identify and classify faults in radar systems. Our approach uses convolutional neural networks (CNN) to extract features from radar signals and fault classify the features. The proposed strategy is trained and validated on a dataset of measured radar signals with various types of faults. The results show that it achieves high accuracy in fault diagnosis. To further evaluate the effectiveness of the proposed strategy, we compare it with traditional rule-based approaches and other machine learning-based methods, including decision trees, support vector machines (SVMs), and random forests. The results demonstrate that our deep learning-based approach outperforms the traditional approaches in terms of accuracy and efficiency. Finally, we discuss the potential applications and limitations of the proposed strategy, as well as future research directions. Our study highlights the importance and potential of deep learning for radar fault diagnosis. It suggests that it can be a valuable tool for improving the performance and reliability of radar systems. In summary, this paper presents a radar fault diagnosis strategy based on deep learning that achieves high accuracy and efficiency in identifying and classifying faults in radar systems. The proposed strategy has significant potential for practical applications and can pave the way for further research.

Keywords: radar system, fault diagnosis, deep learning, radar fault

Procedia PDF Downloads 65

Authors: Oscar E. Cariceo, Claudia V. Casal

Abstract:

Machine learning offers a set of techniques to promote social work interventions and can lead to support decisions of practitioners in order to predict new behaviors based on data produced by the organizations, services agencies, users, clients or individuals. Machine learning techniques include a set of generalizable algorithms that are data-driven, which means that rules and solutions are derived by examining data, based on the patterns that are present within any data set. In other words, the goal of machine learning is teaching computers through 'examples', by training data to test specifics hypothesis and predict what would be a certain outcome, based on a current scenario and improve that experience. Machine learning can be classified into two general categories depending on the nature of the problem that this technique needs to tackle. First, supervised learning involves a dataset that is already known in terms of their output. Supervising learning problems are categorized, into regression problems, which involve a prediction from quantitative variables, using a continuous function; and classification problems, which seek predict results from discrete qualitative variables. For social work research, machine learning generates predictions as a key element to improving social interventions on complex social issues by providing better inference from data and establishing more precise estimated effects, for example in services that seek to improve their outcomes. This paper exposes the results of a classification algorithm to predict cyberbullying among adolescents. Data were retrieved from the National Polyvictimization Survey conducted by the government of Chile in 2017. A logistic regression model was created to predict if an adolescent would experience cyberbullying based on the interaction and behavior of gender, age, grade, type of school, and self-esteem sentiments. The model can predict with an accuracy of 59.8% if an adolescent will suffer cyberbullying. These results can help to promote programs to avoid cyberbullying at schools and improve evidence based practice.

Keywords: cyberbullying, evidence based practice, machine learning, social work research

Procedia PDF Downloads 153

Authors: Virginie Hergault, Siham Chebbah, Bertrand Frere

Abstract:

Following in-house objectives, Central laboratory of Paris police Prefecture conducted a general review on models and Computational Fluid Dynamics (CFD) codes used to simulate pollutant dispersion in the atmosphere. Starting from that review and considering main features of Large Eddy Simulation, Central Laboratory Of Paris Police Prefecture (LCPP) postulates that the Fire Dynamics Simulator (FDS) model, from National Institute of Standards and Technology (NIST), should be well suited for air pollutant dispersion modeling. This paper focuses on the implementation and the evaluation of FDS in the frame of the European COST ES1006 Action. This action aimed at quantifying the performance of modeling approaches. In this paper, the CUTE dataset carried out in the city of Hamburg, and its mock-up has been used. We have performed a comparison of FDS results with wind tunnel measurements from CUTE trials on the one hand, and, on the other, with the models results involved in the COST Action. The most time-consuming part of creating input data for simulations is the transfer of obstacle geometry information to the format required by SDS. Thus, we have developed Python codes to convert automatically building and topographic data to the FDS input file. In order to evaluate the predictions of FDS with observations, statistical performance measures have been used. These metrics include the fractional bias (FB), the normalized mean square error (NMSE) and the fraction of predictions within a factor of two of observations (FAC2). As well as the CFD models tested in the COST Action, FDS results demonstrate a good agreement with measured concentrations. Furthermore, the metrics assessment indicate that FB and NMSE meet the tolerance acceptable.

Keywords: numerical simulations, atmospheric dispersion, cost ES1006 action, CFD model, cute experiments, wind tunnel data, numerical results

Procedia PDF Downloads 120

Authors: Baladitya Swaika, Rahul Khatry

Abstract:

Computer Adaptive Tests (CATs) is one of the most efficient ways for testing the cognitive abilities of students. CATs are based on Item Response Theory (IRT) which is based on item selection and ability estimation using statistical methods of maximum information selection/selection from posterior and maximum-likelihood (ML)/maximum a posteriori (MAP) estimators respectively. This study aims at combining both classical and Bayesian approaches to IRT to create a dataset which is then fed to a neural network which automates the process of ability estimation and then comparing it to traditional CAT models designed using IRT. This study uses python as the base coding language, pymc for statistical modelling of the IRT and scikit-learn for neural network implementations. On creation of the model and on comparison, it is found that the Neural Network based model performs 7-10% worse than the IRT model for score estimations. Although performing poorly, compared to the IRT model, the neural network model can be beneficially used in back-ends for reducing time complexity as the IRT model would have to re-calculate the ability every-time it gets a request whereas the prediction from a neural network could be done in a single step for an existing trained Regressor. This study also proposes a new kind of framework whereby the neural network model could be used to incorporate feature sets, other than the normal IRT feature set and use a neural network’s capacity of learning unknown functions to give rise to better CAT models. Categorical features like test type, etc. could be learnt and incorporated in IRT functions with the help of techniques like logistic regression and can be used to learn functions and expressed as models which may not be trivial to be expressed via equations. This kind of a framework, when implemented would be highly advantageous in psychometrics and cognitive assessments. This study gives a brief overview as to how neural networks can be used in adaptive testing, not only by reducing time-complexity but also by being able to incorporate newer and better datasets which would eventually lead to higher quality testing.

Keywords: computer adaptive tests, item response theory, machine learning, neural networks

Procedia PDF Downloads 160

Authors: Luis Alvarado, Victor Poblete, Isaac Gonzalez, Yetzabeth Gonzalez

Abstract:

In MIREX 2016 (http://www.music-ir.org/mirex), the deep neural network (DNN)-Deep Chroma Extractor, proposed by Korzeniowski and Wiedmer, reached the highest score in an audio chord recognition task. In the present paper, this tool is assessed under acoustic reverberant environments and distinct source-microphone distances. The evaluation dataset comprises The Beatles and Queen datasets. These datasets are sequentially re-recorded with a single microphone in a real reverberant chamber at four reverberation times (0 -anechoic-, 1, 2, and 3 s, approximately), as well as four source-microphone distances (32, 64, 128, and 256 cm). It is expected that the performance of the trained DNN will dramatically decrease under these acoustic conditions with signals degraded by room reverberation and distance to the source. Recently, the effect of the bio-inspired Locally-Normalized Cepstral Coefficients (LNCC), has been assessed in a text independent speaker verification task using speech signals degraded by additive noise at different signal-to-noise ratios with variations of recording distance, and it has also been assessed under reverberant conditions with variations of recording distance. LNCC showed a performance so high as the state-of-the-art Mel Frequency Cepstral Coefficient filters. Based on these results, this paper proposes a variation of locally-normalized triangular filters called Locally-Normalized Quarter Tone (LNQT) filters. By using the LNQT spectrogram, robustness improvements of the trained Deep Chroma Extractor are expected, compared with classical triangular filters, and thus compensating the music signal degradation improving the accuracy of the chord recognition system.

Keywords: chord recognition, deep neural networks, feature extraction, music information retrieval

Procedia PDF Downloads 216

Authors: Ismaila Olabisi Ogundiji, Hakeem Mayowa Olujide, Qasim Usamot

Abstract:

The main reasons for site-specific management for agricultural inputs are to increase the profitability of crop production, to protect the environment and to improve products’ quality. Information about the variability of different soil attributes within a field is highly essential for the decision-making process. Lack of fast and accurate acquisition of soil characteristics remains one of the biggest limitations of precision agriculture due to being expensive and time-consuming. Adaptive sampling has been proven as an accurate and affordable sampling technique for planning within a field for site-specific management of agricultural inputs. This study employed spatial uncertainty of soil apparent electrical conductivity (ECa) estimates to identify adaptive re-survey areas in the field. The original dataset was grouped into validation and calibration groups where the calibration group was sub-grouped into three sets of different measurements pass intervals. A conditional simulation was performed on the field ECa to evaluate the ECa spatial uncertainty estimates by the use of the geostatistical technique. The grouping of high-uncertainty areas for each set was done using image segmentation in MATLAB, then, high and low area value-separate was identified. Finally, an adaptive re-survey was carried out on those areas of high-uncertainty. Adding adaptive re-surveying significantly minimized the time required for resampling whole field and resulted in ECa with minimal error. For the most spacious transect, the root mean square error (RMSE) yielded from an initial crude sampling survey was minimized after an adaptive re-survey, which was close to that value of the ECa yielded with an all-field re-survey. The estimated sampling time for the adaptive re-survey was found to be 45% lesser than that of all-field re-survey. The results indicate that designing adaptive sampling through spatial uncertainty models significantly mitigates sampling cost, and there was still conformity in the accuracy of the observations.

Keywords: soil electrical conductivity, adaptive sampling, conditional simulation, spatial uncertainty, site-specific management

Procedia PDF Downloads 114

Authors: Wenyan Xu, Rundong Wang, Chen Li, Yonghong Hu, Zhonghua Lu

Abstract:

In the field of quantitative trading, it is common to find patterns in short-term volatile trends of the market. These patterns are known as High-Frequency (HF) risk factors, serving as effective indicators of future stock price volatility. However, in the past, these risk factors were usually generated by traditional financial models, and the validity of these risk factors is heavily based on domain-specific knowledge manually added instead of extensive market data. Inspired by symbolic regression (SR), the task of inferring mathematical laws from existing data, we take the extraction of formulaic risk factors from high-frequency trading (HFT) market data as an SR task. In this paper, we challenge the procedure of manually constructing risk factors and propose an end-to-end methodology, Intraday Risk Factor Transformer (IRFT) to directly predict the full formulaic factors, constants included. Specifically, we utilize a hybrid symbolic-numeric vocabulary where symbolic tokens denote operators/stock features and numeric tokens denote constants. Then, we train a Transformer model on the HFT dataset to directly generate complete formulaic HF risk factors without relying on the skeleton, which is a parametric function using a pre-defined list of operators – typically, the math operations (+, ×, /) and functions(√x, log x, cos x). It determines the general shape of the stock volatility law up to a choice of constants, e.g., f(x) = tan(ax+b) (x is the stock price). We further refine predicted constants(a,b) using the Broyden–Fletcher–Goldfarb–Shanno algorithm (BFGS) as informed guesses to mitigate non-linear issues. Compared to the 10 approaches in SRBench, which is a living benchmark for SR, IRFT gains a 30% excess investment return on the HS300 and S&P500 datasets, with inference times orders of magnitude faster than theirs in HF risk factor mining tasks.

Keywords: transformer, factor-mining language model, highfrequency risk factor collections

Procedia PDF Downloads 11

Authors: Peiming Li

Abstract:

This paper presents an innovative blockchain-based approach to enhance the reliability and efficiency of federated learning systems. By integrating a decentralized oracle consensus mechanism into the federated learning framework, we address key challenges of data and model integrity. Our approach utilizes a network of redundant oracles, functioning as independent validators within an epoch-based training system in the federated learning model. In federated learning, data is decentralized, residing on various participants' devices. This scenario often leads to concerns about data integrity and model quality. Our solution employs blockchain technology to establish a transparent and tamper-proof environment, ensuring secure data sharing and aggregation. The decentralized oracles, a concept borrowed from blockchain systems, act as unbiased validators. They assess the contributions of each participant using a Hidden Markov Model (HMM), which is crucial for evaluating the consistency of participant inputs and safeguarding against model poisoning and malicious activities. Our methodology's distinct feature is its epoch-based training. An epoch here refers to a specific training phase where data is updated and assessed for quality and relevance. The redundant oracles work in concert to validate data updates during these epochs, enhancing the system's resilience to security threats and data corruption. The effectiveness of this system was tested using the Mnist dataset, a standard in machine learning for benchmarking. Results demonstrate that our blockchain-oriented federated learning approach significantly boosts system resilience, addressing the common challenges of federated environments. This paper aims to make these advanced concepts accessible, even to those with a limited background in blockchain or federated learning. We provide a foundational understanding of how blockchain technology can revolutionize data integrity in decentralized systems and explain the role of oracles in maintaining model accuracy and reliability.

Keywords: federated learning system, block chain, decentralized oracles, hidden markov model

Procedia PDF Downloads 43

Authors: Annika Stechemesser, Rachel Pounds, Emma Lucas, Chris Dawson, Julia Lipecki, Pavle Vrljicak, Jan Brosens, Sean Kehoe, Jason Yap, Lawrence Young, Sascha Ott

Abstract:

Advances in technology make it now possible to retrieve the genetic information of thousands of single cancerous cells. One of the key challenges in single cell analysis of cancerous tissue is to determine the number of different cell types and their characteristic genes within the sample to better understand the tumors and their reaction to different treatments. For this analysis to be possible, it is crucial to filter out background noise as it can severely blur the downstream analysis and give misleading results. In-depth analysis of the state-of-the-art filtering methods for single cell data showed that they do, in some cases, not separate noisy and normal cells sufficiently. We introduced an algorithm that filters and clusters single cell data simultaneously without relying on certain genes or thresholds chosen by eye. It detects communities in a Shared Nearest Neighbor similarity network, which captures the similarities and dissimilarities of the cells by optimizing the modularity and then identifies and removes vertices with a weak clustering belonging. This strategy is based on the fact that noisy data instances are very likely to be similar to true cell types but do not match any of these wells. Once the clustering is complete, we apply a set of evaluation metrics on the cluster level and accept or reject clusters based on the outcome. The performance of our algorithm was tested on three datasets and led to convincing results. We were able to replicate the results on a Peripheral Blood Mononuclear Cells dataset. Furthermore, we applied the algorithm to two samples of ovarian cancer from the same patient before and after chemotherapy. Comparing the standard approach to our algorithm, we found a hidden cell type in the ovarian postchemotherapy data with interesting marker genes that are potentially relevant for medical research.

Keywords: cancer research, graph theory, machine learning, single cell analysis

Procedia PDF Downloads 87