Search results for: machine translation
374 Feature Engineering Based Detection of Buffer Overflow Vulnerability in Source Code Using Deep Neural Networks
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 81373 Radar Fault Diagnosis Strategy Based on Deep Learning
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 81372 Big Data Analytics and Public Policy: A Study in Rural India
Authors: Vasantha Gouri Prathapagiri
Abstract:
Innovations in ICT sector facilitate qualitative life style for citizens across the globe. Countries that facilitate usage of new techniques in ICT, i.e., big data analytics find it easier to fulfil the needs of their citizens. Big data is characterised by its volume, variety, and speed. Analytics involves its processing in a cost effective way in order to draw conclusion for their useful application. Big data also involves into the field of machine learning, artificial intelligence all leading to accuracy in data presentation useful for public policy making. Hence using data analytics in public policy making is a proper way to march towards all round development of any country. The data driven insights can help the government to take important strategic decisions with regard to socio-economic development of her country. Developed nations like UK and USA are already far ahead on the path of digitization with the support of Big Data analytics. India is a huge country and is currently on the path of massive digitization being realised through Digital India Mission. Internet connection per household is on the rise every year. This transforms into a massive data set that has the potential to improvise the public services delivery system into an effective service mechanism for Indian citizens. In fact, when compared to developed nations, this capacity is being underutilized in India. This is particularly true for administrative system in rural areas. The present paper focuses on the need for big data analytics adaptation in Indian rural administration and its contribution towards development of the country on a faster pace. Results of the research focussed on the need for increasing awareness and serious capacity building of the government personnel working for rural development with regard to big data analytics and its utility for development of the country. Multiple public policies are framed and implemented for rural development yet the results are not as effective as they should be. Big data has a major role to play in this context as can assist in improving both policy making and implementation aiming at all round development of the country.Keywords: Digital India Mission, public service delivery system, public policy, Indian administration
Procedia PDF Downloads 154371 Microstructure Dependent Fatigue Crack Growth in Aluminum Alloy
Authors: M. S. Nandana, K. Udaya Bhat, C. M. Manjunatha
Abstract:
In this study aluminum alloy 7010 was subjected to three different ageing treatments i.e., peak ageing (T6), over-ageing (T7451) and retrogression and re ageing (RRA) to study the influence of precipitate microstructure on the fatigue crack growth rate behavior. The microstructural modification was studied by using transmission electron microscope (TEM) to examine the change in the size and morphology of precipitates in the matrix and on the grain boundaries. The standard compact tension (CT) specimens were fabricated and tested under constant amplitude fatigue crack growth tests to evaluate the influence of heat treatment on the fatigue crack growth rate properties. The tests were performed in a computer-controlled servo-hydraulic test machine applying a load ratio, R = 0.1 at a loading frequency of 10 Hz as per ASTM E647. The fatigue crack growth was measured by adopting compliance technique using a CMOD gauge attached to the CT specimen. The average size of the matrix precipitates were found to be of 16-20 nm in T7451, 5-6 nm in RRA and 2-3 nm in T6 conditions respectively. The grain boundary precipitate which was continuous in T6, was disintegrated in RRA and T7451 condition. The PFZ width was lower in RRA compared to T7451 condition. The crack growth rate was higher in T7451 and lowest in RRA treated alloy. The RRA treated alloy also exhibits an increase in threshold stress intensity factor range (∆Kₜₕ). The ∆Kₜₕ measured was 11.1, 10.3 and 5.7 MPam¹/² in RRA, T6 and T7451 alloys respectively. The fatigue crack growth rate in RRA treated alloy was nearly 2-3 times lower than that in T6 and was one order lower than that observed in T7451 condition. The surface roughness of RRA treated alloy was more pronounced when compared to the other conditions. The reduction in fatigue crack growth rate in RRA alloy was majorly due to the increase in roughness and partially due to increase in spacing between the matrix precipitates. The reduction in crack growth rate and increase in threshold stress intensity range is expected to benefit the damage tolerant capability of aircraft structural components under service loads.Keywords: damage tolerance, fatigue, heat treatment, PFZ, RRA
Procedia PDF Downloads 149370 The Role of Home Composting in Waste Management Cost Reduction
Authors: Nahid Hassanshahi, Ayoub Karimi-Jashni, Nasser Talebbeydokhti
Abstract:
Due to the economic and environmental benefits of producing less waste, the US Environmental Protection Agency (EPA) introduces source reduction as one of the most important means to deal with the problems caused by increased landfills and pollution. Waste reduction involves all waste management methods, including source reduction, recycling, and composting, which reduce waste flow to landfills or other disposal facilities. Source reduction of waste can be studied from two perspectives: avoiding waste production, or reducing per capita waste production, and waste deviation that indicates the reduction of waste transfer to landfills. The present paper has investigated home composting as a managerial solution for reduction of waste transfer to landfills. Home composting has many benefits. The use of household waste for the production of compost will result in a much smaller amount of waste being sent to landfills, which in turn will reduce the costs of waste collection, transportation and burial. Reducing the volume of waste for disposal and using them for the production of compost and plant fertilizer might help to recycle the material in a shorter time and to use them effectively in order to preserve the environment and reduce contamination. Producing compost in a home-based manner requires very small piece of land for preparation and recycling compared with other methods. The final product of home-made compost is valuable and helps to grow crops and garden plants. It is also used for modifying the soil structure and maintaining its moisture. The food that is transferred to landfills will spoil and produce leachate after a while. It will also release methane and greenhouse gases. But, composting these materials at home is the best way to manage degradable materials, use them efficiently and reduce environmental pollution. Studies have shown that the benefits of the sale of produced compost and the reduced costs of collecting, transporting, and burying waste can well be responsive to the costs of purchasing home compost machine and the cost of related trainings. Moreover, the process of producing home compost may be profitable within 4 to 5 years and as a result, it will have a major role in reducing waste management.Keywords: compost, home compost, reducing waste, waste management
Procedia PDF Downloads 417369 Development of a Microfluidic Device for Low-Volume Sample Lysis
Authors: Abbas Ali Husseini, Ali Mohammad Yazdani, Fatemeh Ghadiri, Alper Şişman
Abstract:
We developed a microchip device that uses surface acoustic waves for rapid lysis of low level of cell samples. The device incorporates sharp-edge glass microparticles for improved performance. We optimized the lysis conditions for high efficiency and evaluated the device's feasibility for point-of-care applications. The microchip contains a 13-finger pair interdigital transducer with a 30-degree focused angle. It generates high-intensity acoustic beams that converge 6 mm away. The microchip operates at a frequency of 16 MHz, exciting Rayleigh waves with a 250 µm wavelength on the LiNbO3 substrate. Cell lysis occurs when Candida albicans cells and glass particles are placed within the focal area. The high-intensity surface acoustic waves induce centrifugal forces on the cells and glass particles, resulting in cell lysis through lateral forces from the sharp-edge glass particles. We conducted 42 pilot cell lysis experiments to optimize the surface acoustic wave-induced streaming. We varied electrical power, droplet volume, glass particle size, concentration, and lysis time. A regression machine-learning model determined the impact of each parameter on lysis efficiency. Based on these findings, we predicted optimal conditions: electrical signal of 2.5 W, sample volume of 20 µl, glass particle size below 10 µm, concentration of 0.2 µg, and a 5-minute lysis period. Downstream analysis successfully amplified a DNA target fragment directly from the lysate. The study presents an efficient microchip-based cell lysis method employing acoustic streaming and microparticle collisions within microdroplets. Integration of a surface acoustic wave-based lysis chip with an isothermal amplification method enables swift point-of-care applications.Keywords: cell lysis, surface acoustic wave, micro-glass particle, droplet
Procedia PDF Downloads 71368 Gene Expressions in Left Ventricle Heart Tissue of Rat after 150 Mev Proton Irradiation
Abstract:
Introduction: In mediastinal radiotherapy and to a lesser extend also in total-body irradiation (TBI) radiation exposure may lead to development of cardiac diseases. Radiation-induced heart disease is dose-dependent and it is characterized by a loss of cardiac function, associated with progressive heart cells degeneration. We aimed to determine the in-vivo radiation effects on fibronectin, ColaA1, ColaA2, galectin and TGFb1 gene expression levels in left ventricle heart tissues of rats after irradiation. Material and method: Four non-treatment adult Wistar rats as control group (group A) were selected. In group B, 4 adult Wistar rats irradiated to 20 Gy single dose of 150 Mev proton beam locally in heart only. In heart plus lung irradiate group (group C) 4 adult rats was irradiated by 50% of lung laterally plus heart radiation that mentioned in before group. At 8 weeks after radiation animals sacrificed and left ventricle heart dropped in liquid nitrogen for RNA extraction by Absolutely RNA® Miniprep Kit (Stratagen, Cat no. 400800). cDNA was synthesized using M-MLV reverse transcriptase (Life Technologies, Cat no. 28025-013). We used Bio-Rad machine (Bio Rad iQ5 Real Time PCR) for QPCR testing by relative standard curve method. Results: We found that gene expression of fibronectin in group C significantly increased compared to control group, but it was not showed significant change in group B compared to group A. The levels of gene expressions of Cola1 and Cola2 in mRNA did not show any significant changes between normal and radiation groups. Changes of expression of galectin target significantly increased only in group C compared to group A. TGFb1 expressions in group C more than group B showed significant enhancement compared to group A. Conclusion: In summary we can say that 20 Gy of proton exposure of heart tissue may lead to detectable damages in heart cells and may distribute function of them as a component of heart tissue structure in molecular level.Keywords: gene expression, heart damage, proton irradiation, radiotherapy
Procedia PDF Downloads 484367 Exploring the Applications of Neural Networks in the Adaptive Learning Environment
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 169366 Energy Efficiency Approach to Reduce Costs of Ownership of Air Jet Weaving
Authors: Corrado Grassi, Achim Schröter, Yves Gloy, Thomas Gries
Abstract:
Air jet weaving is the most productive, but also the most energy consuming weaving method. Increasing energy costs and environmental impact are constantly a challenge for the manufacturers of weaving machines. Current technological developments concern with low energy costs, low environmental impact, high productivity, and constant product quality. The high degree of energy consumption of the method can be ascribed to the high need of compressed air. An energy efficiency method is applied to the air jet weaving technology. Such method identifies and classifies the main relevant energy consumers and processes from the exergy point of view and it leads to the identification of energy efficiency potentials during the weft insertion process. Starting from the design phase, energy efficiency is considered as the central requirement to be satisfied. The initial phase of the method consists of an analysis of the state of the art of the main weft insertion components in order to point out a prioritization of the high demanding energy components and processes. The identified major components are investigated to reduce the high demand of energy of the weft insertion process. During the interaction of the flow field coming from the relay nozzles within the profiled reed, only a minor part of the stream is really accelerating the weft yarn, hence resulting in large energy inefficiency. Different tools such as FEM analysis, CFD simulation models and experimental analysis are used in order to design a more energy efficient design of the involved components in the filling insertion. A different concept for the metal strip of the profiled reed is developed. The developed metal strip allows a reduction of the machine energy consumption. Based on a parametric and aerodynamic study, the designed reed transmits higher values of the flow power to the filling yarn. The innovative reed fulfills both the requirement of raising energy efficiency and the compliance with the weaving constraints.Keywords: air jet weaving, aerodynamic simulation, energy efficiency, experimental validation, weft insertion
Procedia PDF Downloads 190365 Simulation, Design, and 3D Print of Novel Highly Integrated TEG Device with Improved Thermal Energy Harvest Efficiency
Abstract:
Despite the remarkable advancement of solar cell technology, the challenge of optimizing total solar energy harvest efficiency persists, primarily due to significant heat loss. This excess heat not only diminishes solar panel output efficiency but also curtails its operational lifespan. A promising approach to address this issue is the conversion of surplus heat into electricity. In recent years, there is growing interest in the use of thermoelectric generators (TEG) as a potential solution. The integration of efficient TEG devices holds the promise of augmenting overall energy harvest efficiency while prolonging the longevity of solar panels. While certain research groups have proposed the integration of solar cells and TEG devices, a substantial gap between conceptualization and practical implementation remains, largely attributed to low thermal energy conversion efficiency of TEG devices. To bridge this gap and meet the requisites of practical application, a feasible strategy involves the incorporation of a substantial number of p-n junctions within a confined unit volume. However, the manufacturing of high-density TEG p-n junctions presents a formidable challenge. The prevalent solution often leads to large device sizes to accommodate enough p-n junctions, consequently complicating integration with solar cells. Recently, the adoption of 3D printing technology has emerged as a promising solution to address this challenge by fabricating high-density p-n arrays. Despite this, further developmental efforts are necessary. Presently, the primary focus is on the 3D printing of vertically layered TEG devices, wherein p-n junction density remains constrained by spatial limitations and the constraints of 3D printing techniques. This study proposes a novel device configuration featuring horizontally arrayed p-n junctions of Bi2Te3. The structural design of the device is subjected to simulation through the Finite Element Method (FEM) within COMSOL Multiphysics software. Various device configurations are simulated to identify optimal device structure. Based on the simulation results, a new TEG device is fabricated utilizing 3D Selective laser melting (SLM) printing technology. Fusion 360 facilitates the translation of the COMSOL device structure into a 3D print file. The horizontal design offers a unique advantage, enabling the fabrication of densely packed, three-dimensional p-n junction arrays. The fabrication process entails printing a singular row of horizontal p-n junctions using the 3D SLM printing technique in a single layer. Subsequently, successive rows of p-n junction arrays are printed within the same layer, interconnected by thermally conductive copper. This sequence is replicated across multiple layers, separated by thermal insulating glass. This integration created in a highly compact three-dimensional TEG device with high density p-n junctions. The fabricated TEG device is then attached to the bottom of the solar cell using thermal glue. The whole device is characterized, with output data closely matching with COMSOL simulation results. Future research endeavors will encompass the refinement of thermoelectric materials. This includes the advancement of high-resolution 3D printing techniques tailored to diverse thermoelectric materials, along with the optimization of material microstructures such as porosity and doping. The objective is to achieve an optimal and highly integrated PV-TEG device that can substantially increase the solar energy harvest efficiency.Keywords: thermoelectric, finite element method, 3d print, energy conversion
Procedia PDF Downloads 56364 Milling Simulations with a 3-DOF Flexible Planar Robot
Authors: Hoai Nam Huynh, Edouard Rivière-Lorphèvre, Olivier Verlinden
Abstract:
Manufacturing technologies are becoming continuously more diversified over the years. The increasing use of robots for various applications such as assembling, painting, welding has also affected the field of machining. Machining robots can deal with larger workspaces than conventional machine-tools at a lower cost and thus represent a very promising alternative for machining applications. Furthermore, their inherent structure ensures them a great flexibility of motion to reach any location on the workpiece with the desired orientation. Nevertheless, machining robots suffer from a lack of stiffness at their joints restricting their use to applications involving low cutting forces especially finishing operations. Vibratory instabilities may also happen while machining and deteriorate the precision leading to scrap parts. Some researchers are therefore concerned with the identification of optimal parameters in robotic machining. This paper continues the development of a virtual robotic machining simulator in order to find optimized cutting parameters in terms of depth of cut or feed per tooth for example. The simulation environment combines an in-house milling routine (DyStaMill) achieving the computation of cutting forces and material removal with an in-house multibody library (EasyDyn) which is used to build a dynamic model of a 3-DOF planar robot with flexible links. The position of the robot end-effector submitted to milling forces is controlled through an inverse kinematics scheme while controlling the position of its joints separately. Each joint is actuated through a servomotor for which the transfer function has been computed in order to tune the corresponding controller. The output results feature the evolution of the cutting forces when the robot structure is deformable or not and the tracking errors of the end-effector. Illustrations of the resulting machined surfaces are also presented. The consideration of the links flexibility has highlighted an increase of the cutting forces magnitude. This proof of concept will aim to enrich the database of results in robotic machining for potential improvements in production.Keywords: control, milling, multibody, robotic, simulation
Procedia PDF Downloads 240363 Healthcare-SignNet: Advanced Video Classification for Medical Sign Language Recognition Using CNN and RNN Models
Authors: Chithra A. V., Somoshree Datta, Sandeep Nithyanandan
Abstract:
Sign Language Recognition (SLR) is the process of interpreting and translating sign language into spoken or written language using technological systems. It involves recognizing hand gestures, facial expressions, and body movements that makeup sign language communication. The primary goal of SLR is to facilitate communication between hearing- and speech-impaired communities and those who do not understand sign language. Due to the increased awareness and greater recognition of the rights and needs of the hearing- and speech-impaired community, sign language recognition has gained significant importance over the past 10 years. Technological advancements in the fields of Artificial Intelligence and Machine Learning have made it more practical and feasible to create accurate SLR systems. This paper presents a distinct approach to SLR by framing it as a video classification problem using Deep Learning (DL), whereby a combination of Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) has been used. This research targets the integration of sign language recognition into healthcare settings, aiming to improve communication between medical professionals and patients with hearing impairments. The spatial features from each video frame are extracted using a CNN, which captures essential elements such as hand shapes, movements, and facial expressions. These features are then fed into an RNN network that learns the temporal dependencies and patterns inherent in sign language sequences. The INCLUDE dataset has been enhanced with more videos from the healthcare domain and the model is evaluated on the same. Our model achieves 91% accuracy, representing state-of-the-art performance in this domain. The results highlight the effectiveness of treating SLR as a video classification task with the CNN-RNN architecture. This approach not only improves recognition accuracy but also offers a scalable solution for real-time SLR applications, significantly advancing the field of accessible communication technologies.Keywords: sign language recognition, deep learning, convolution neural network, recurrent neural network
Procedia PDF Downloads 14362 Trip Reduction in Turbo Machinery
Authors: Pranay Mathur, Carlo Michelassi, Simi Karatha, Gilda Pedoto
Abstract:
Industrial plant uptime is top most importance for reliable, profitable & sustainable operation. Trip and failed start has major impact on plant reliability and all plant operators focussed on efforts required to minimise the trips & failed starts. The performance of these CTQs are measured with 2 metrics, MTBT(Mean time between trips) and SR (Starting reliability). These metrics helps to identify top failure modes and identify units need more effort to improve plant reliability. Baker Hughes Trip reduction program structured to reduce these unwanted trip 1. Real time machine operational parameters remotely available and capturing the signature of malfunction including related boundary condition. 2. Real time alerting system based on analytics available remotely. 3. Remote access to trip logs and alarms from control system to identify the cause of events. 4. Continuous support to field engineers by remotely connecting with subject matter expert. 5. Live tracking of key CTQs 6. Benchmark against fleet 7. Break down to the cause of failure to component level 8. Investigate top contributor, identify design and operational root cause 9. Implement corrective and preventive action 10. Assessing effectiveness of implemented solution using reliability growth models. 11. Develop analytics for predictive maintenance With this approach , Baker Hughes team is able to support customer in achieving their Reliability Key performance Indicators for monitored units, huge cost savings for plant operators. This Presentation explains these approach while providing successful case studies, in particular where 12nos. of LNG and Pipeline operators with about 140 gas compressing line-ups has adopted these techniques and significantly reduce the number of trips and improved MTBTKeywords: reliability, availability, sustainability, digital infrastructure, weibull, effectiveness, automation, trips, fail start
Procedia PDF Downloads 68361 Understanding the Classification of Rain Microstructure and Estimation of Z-R Relationship using a Micro Rain Radar in Tropical Region
Authors: Tomiwa, Akinyemi Clement
Abstract:
Tropical regions experience diverse and complex precipitation patterns, posing significant challenges for accurate rainfall estimation and forecasting. This study addresses the problem of effectively classifying tropical rain types and refining the Z-R (Reflectivity-Rain Rate) relationship to enhance rainfall estimation accuracy. Through a combination of remote sensing, meteorological analysis, and machine learning, the research aims to develop an advanced classification framework capable of distinguishing between different types of tropical rain based on their unique characteristics. This involves utilizing high-resolution satellite imagery, radar data, and atmospheric parameters to categorize precipitation events into distinct classes, providing a comprehensive understanding of tropical rain systems. Additionally, the study seeks to improve the Z-R relationship, a crucial aspect of rainfall estimation. One year of rainfall data was analyzed using a Micro Rain Radar (MRR) located at The Federal University of Technology Akure, Nigeria, measuring rainfall parameters from ground level to a height of 4.8 km with a vertical resolution of 0.16 km. Rain rates were classified into low (stratiform) and high (convective) based on various microstructural attributes such as rain rates, liquid water content, Drop Size Distribution (DSD), average fall speed of the drops, and radar reflectivity. By integrating diverse datasets and employing advanced statistical techniques, the study aims to enhance the precision of Z-R models, offering a more reliable means of estimating rainfall rates from radar reflectivity data. This refined Z-R relationship holds significant potential for improving our understanding of tropical rain systems and enhancing forecasting accuracy in regions prone to heavy precipitation.Keywords: remote sensing, precipitation, drop size distribution, micro rain radar
Procedia PDF Downloads 22360 Corporate Digital Responsibility in Construction Engineering-Construction 4.0: Ethical Guidelines for Digitization and Artificial Intelligence
Authors: Weber-Lewerenz Bianca
Abstract:
Digitization is developing fast and has become a powerful tool for digital planning, construction, and operations. Its transformation bears high potentials for companies, is critical for success, and thus, requires responsible handling. This study provides an assessment of calls made in the sustainable development goals by the United Nations (SDGs), White Papers on AI by international institutions, EU-Commission and German Government requesting for the consideration and protection of values and fundamental rights, the careful demarcation between machine (artificial) and human intelligence and the careful use of such technologies. The study discusses digitization and the impacts of artificial intelligence (AI) in construction engineering from an ethical perspective by generating data via conducting case studies and interviewing experts as part of the qualitative method. This research evaluates critically opportunities and risks revolving around corporate digital responsibility (CDR) in the construction industry. To the author's knowledge, no study has set out to investigate how CDR in construction could be conceptualized, especially in relation to the digitization and AI, to mitigate digital transformation both in large, medium-sized, and small companies. No study addressed the key research question: Where can CDR be allocated, how shall its adequate ethical framework be designed to support digital innovations in order to make full use of the potentials of digitization and AI? Now is the right timing for constructive approaches and apply ethics-by-design in order to develop and implement a safe and efficient AI. This represents the first study in construction engineering applying a holistic, interdisciplinary, inclusive approach to provide guidelines for orientation, examine benefits of AI and define ethical principles as the key driver for success, resources-cost-time efficiency, and sustainability using digital technologies and AI in construction engineering to enhance digital transformation. Innovative corporate organizations starting new business models are more likely to succeed than those dominated by conservative, traditional attitudes.Keywords: construction engineering, digitization, digital transformation, artificial intelligence, ethics, corporate digital responsibility, digital innovation
Procedia PDF Downloads 238359 Enhancing Robustness in Federated Learning through Decentralized Oracle Consensus and Adaptive Evaluation
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 55358 Improving Cell Type Identification of Single Cell Data by Iterative Graph-Based Noise Filtering
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 106357 News Reading Practices: Traditional Media versus New Media
Authors: Nuran Öze
Abstract:
People always want to be aware of what is happening around them. The nature of man constantly triggers the need for gathering information because of curiosity. The media has emerged to save people the need for information. It is known that the media has changed with the technological developments over time, diversified and, people's information needs are provided in different ways. Today, the Internet has become an integral part of everyday life. The invasion of the Internet into everyday life practices at this level affects every aspect of life. These effects cause people to change their life practices. Technological developments have always influenced of people, the way they reach information. Looking at the history of the media, the breaking point about the dissemination of information is seen as the invention of the machine of the printing press. This adventure that started with written media has now become a multi-dimensional structure. Written, audio, visual media has now changed shape with new technologies. Especially emerging of the internet to everyday life, of course, has effects on media field. 'New media' has appeared which contains most of traditional media features in its'. While in the one hand this transformation enables captures a harmony between traditional and new media, on the other hand, new media and traditional media are rivaling each other. The purpose of this study is to examine the problematic relationship between traditional media and new media through the news reading practices of individuals. This study can be evaluated as a kind of media sociology. To reach this aim, two different field researches will be done besides literature review. The research will be conducted in Northern Cyprus. Northern Cyprus Northern Cyprus is located in the Mediterranean Sea. North Cyprus is a country which is not recognized by any country except Turkey. Despite this, takes its share from all technological developments take place in the world. One of the field researches will consist of the questionnaires to be applied on media readers' news reading practices. This survey will be conducted in a social media environment. The second field survey will be conducted in the form of interviews with general editorials or news directors in traditional media. In the second field survey, in-depth interview method will be applied. As a result of these investigations, supporting sides between the new media and the traditional media and directions which contrast with each other will be revealed. In addition to that, it will try to understand the attitudes and perceptions of readers about the traditional media and the new media in this study.Keywords: new media, news, North Cyprus, traditional media
Procedia PDF Downloads 222356 An Efficient Motion Recognition System Based on LMA Technique and a Discrete Hidden Markov Model
Authors: Insaf Ajili, Malik Mallem, Jean-Yves Didier
Abstract:
Human motion recognition has been extensively increased in recent years due to its importance in a wide range of applications, such as human-computer interaction, intelligent surveillance, augmented reality, content-based video compression and retrieval, etc. However, it is still regarded as a challenging task especially in realistic scenarios. It can be seen as a general machine learning problem which requires an effective human motion representation and an efficient learning method. In this work, we introduce a descriptor based on Laban Movement Analysis technique, a formal and universal language for human movement, to capture both quantitative and qualitative aspects of movement. We use Discrete Hidden Markov Model (DHMM) for training and classification motions. We improve the classification algorithm by proposing two DHMMs for each motion class to process the motion sequence in two different directions, forward and backward. Such modification allows avoiding the misclassification that can happen when recognizing similar motions. Two experiments are conducted. In the first one, we evaluate our method on a public dataset, the Microsoft Research Cambridge-12 Kinect gesture data set (MSRC-12) which is a widely used dataset for evaluating action/gesture recognition methods. In the second experiment, we build a dataset composed of 10 gestures(Introduce yourself, waving, Dance, move, turn left, turn right, stop, sit down, increase velocity, decrease velocity) performed by 20 persons. The evaluation of the system includes testing the efficiency of our descriptor vector based on LMA with basic DHMM method and comparing the recognition results of the modified DHMM with the original one. Experiment results demonstrate that our method outperforms most of existing methods that used the MSRC-12 dataset, and a near perfect classification rate in our dataset.Keywords: human motion recognition, motion representation, Laban Movement Analysis, Discrete Hidden Markov Model
Procedia PDF Downloads 201355 Hand Gesture Detection via EmguCV Canny Pruning
Authors: N. N. Mosola, S. J. Molete, L. S. Masoebe, M. Letsae
Abstract:
Hand gesture recognition is a technique used to locate, detect, and recognize a hand gesture. Detection and recognition are concepts of Artificial Intelligence (AI). AI concepts are applicable in Human Computer Interaction (HCI), Expert systems (ES), etc. Hand gesture recognition can be used in sign language interpretation. Sign language is a visual communication tool. This tool is used mostly by deaf societies and those with speech disorder. Communication barriers exist when societies with speech disorder interact with others. This research aims to build a hand recognition system for Lesotho’s Sesotho and English language interpretation. The system will help to bridge the communication problems encountered by the mentioned societies. The system has various processing modules. The modules consist of a hand detection engine, image processing engine, feature extraction, and sign recognition. Detection is a process of identifying an object. The proposed system uses Canny pruning Haar and Haarcascade detection algorithms. Canny pruning implements the Canny edge detection. This is an optimal image processing algorithm. It is used to detect edges of an object. The system employs a skin detection algorithm. The skin detection performs background subtraction, computes the convex hull, and the centroid to assist in the detection process. Recognition is a process of gesture classification. Template matching classifies each hand gesture in real-time. The system was tested using various experiments. The results obtained show that time, distance, and light are factors that affect the rate of detection and ultimately recognition. Detection rate is directly proportional to the distance of the hand from the camera. Different lighting conditions were considered. The more the light intensity, the faster the detection rate. Based on the results obtained from this research, the applied methodologies are efficient and provide a plausible solution towards a light-weight, inexpensive system which can be used for sign language interpretation.Keywords: canny pruning, hand recognition, machine learning, skin tracking
Procedia PDF Downloads 177354 Deep Cryogenic Treatment With Subsequent Aging Applied to Martensitic Stainless Steel: Evaluation of Hardness, Tenacity and Microstructure
Authors: Victor Manuel Alcántara Alza
Abstract:
The way in which the application of the deep cryogenic treatment DCT(-196°C) affects, applied with subsequent aging, was investigated, regarding the mechanical properties of hardness, toughness and microstructure, applied to martensitic stainless steels, with the aim of establishing a different methodology compared to the traditional DCT cryogenic treatment with subsequent tempering. For this experimental study, a muffle furnace was used, first subjecting the specimens to deep cryogenization in a liquid Nitrogen bath/4h, after being previously austenitized at the following temperatures: 1020-1030-1040-1050 (°C) / 1 hour; and then tempered in oil. A first group of cryogenic samples were subjected to subsequent aging at 150°C, with immersion times: 2.5 -5- 10 - 20 - 50 – 100 (h). The next group was subjected to subsequent tempering at temperatures: 480-500-510-520-530-540 (°C)/ 2h. The hardness tests were carried out under standards, using a Universal Durometer, and the readings were made on the HRC scale. The Impact Resistance tests were carried out in a Charpy machine following the ASTM E 23 – 93ª standard. Measurements were taken in joules. Microscopy was performed at the optical level using a 1000X microscope. It was found: For the entire aging interval, the samples austenitized at 1050°C present greater hardness than austenitized at 1040°C, with the maximum peak aged being at 30h. In all cases, the aged samples exceed the hardness of the tempered samples, even in their minimum values. In post-tempered samples, the tempering temperature hardly have effect on the impact strength of material. In the Cryogenic Treatment: DCT + subsequent aging, the maximum hardness value (58.7 HRC) is linked to an impact toughness value (54J) obtained with aging time of 39h, which is considered an optimal condition. The higher hardness of steel after the DCT treatment is attributed to the transformation of retained austenite into martensite. The microstructure is composed mainly of lath martensite; and the original grain size of the austenite can be appreciated. The choice of the combination: Hardness-toughness, is subject to the required service conditions of steel.Keywords: deep cryogenic treatment; aged precipitation; martensitic steels;, mechanical properties; martensitic steels, hardness, carbides precipitaion
Procedia PDF Downloads 70353 Performance Evaluation of Production Schedules Based on Process Mining
Authors: Kwan Hee Han
Abstract:
External environment of enterprise is rapidly changing majorly by global competition, cost reduction pressures, and new technology. In these situations, production scheduling function plays a critical role to meet customer requirements and to attain the goal of operational efficiency. It deals with short-term decision making in the production process of the whole supply chain. The major task of production scheduling is to seek a balance between customer orders and limited resources. In manufacturing companies, this task is so difficult because it should efficiently utilize resource capacity under the careful consideration of many interacting constraints. At present, many computerized software solutions have been utilized in many enterprises to generate a realistic production schedule to overcome the complexity of schedule generation. However, most production scheduling systems do not provide sufficient information about the validity of the generated schedule except limited statistics. Process mining only recently emerged as a sub-discipline of both data mining and business process management. Process mining techniques enable the useful analysis of a wide variety of processes such as process discovery, conformance checking, and bottleneck analysis. In this study, the performance of generated production schedule is evaluated by mining event log data of production scheduling software system by using the process mining techniques since every software system generates event logs for the further use such as security investigation, auditing and error bugging. An application of process mining approach is proposed for the validation of the goodness of production schedule generated by scheduling software systems in this study. By using process mining techniques, major evaluation criteria such as utilization of workstation, existence of bottleneck workstations, critical process route patterns, and work load balance of each machine over time are measured, and finally, the goodness of production schedule is evaluated. By using the proposed process mining approach for evaluating the performance of generated production schedule, the quality of production schedule of manufacturing enterprises can be improved.Keywords: data mining, event log, process mining, production scheduling
Procedia PDF Downloads 274352 Finite Element Method (FEM) Simulation, design and 3D Print of Novel Highly Integrated PV-TEG Device with Improved Solar Energy Harvest Efficiency
Abstract:
Despite the remarkable advancement of solar cell technology, the challenge of optimizing total solar energy harvest efficiency persists, primarily due to significant heat loss. This excess heat not only diminishes solar panel output efficiency but also curtails its operational lifespan. A promising approach to address this issue is the conversion of surplus heat into electricity. In recent years, there is growing interest in the use of thermoelectric generators (TEG) as a potential solution. The integration of efficient TEG devices holds the promise of augmenting overall energy harvest efficiency while prolonging the longevity of solar panels. While certain research groups have proposed the integration of solar cells and TEG devices, a substantial gap between conceptualization and practical implementation remains, largely attributed to low thermal energy conversion efficiency of TEG devices. To bridge this gap and meet the requisites of practical application, a feasible strategy involves the incorporation of a substantial number of p-n junctions within a confined unit volume. However, the manufacturing of high-density TEG p-n junctions presents a formidable challenge. The prevalent solution often leads to large device sizes to accommodate enough p-n junctions, consequently complicating integration with solar cells. Recently, the adoption of 3D printing technology has emerged as a promising solution to address this challenge by fabricating high-density p-n arrays. Despite this, further developmental efforts are necessary. Presently, the primary focus is on the 3D printing of vertically layered TEG devices, wherein p-n junction density remains constrained by spatial limitations and the constraints of 3D printing techniques. This study proposes a novel device configuration featuring horizontally arrayed p-n junctions of Bi2Te3. The structural design of the device is subjected to simulation through the Finite Element Method (FEM) within COMSOL Multiphysics software. Various device configurations are simulated to identify optimal device structure. Based on the simulation results, a new TEG device is fabricated utilizing 3D Selective laser melting (SLM) printing technology. Fusion 360 facilitates the translation of the COMSOL device structure into a 3D print file. The horizontal design offers a unique advantage, enabling the fabrication of densely packed, three-dimensional p-n junction arrays. The fabrication process entails printing a singular row of horizontal p-n junctions using the 3D SLM printing technique in a single layer. Subsequently, successive rows of p-n junction arrays are printed within the same layer, interconnected by thermally conductive copper. This sequence is replicated across multiple layers, separated by thermal insulating glass. This integration created in a highly compact three-dimensional TEG device with high density p-n junctions. The fabricated TEG device is then attached to the bottom of the solar cell using thermal glue. The whole device is characterized, with output data closely matching with COMSOL simulation results. Future research endeavors will encompass the refinement of thermoelectric materials. This includes the advancement of high-resolution 3D printing techniques tailored to diverse thermoelectric materials, along with the optimization of material microstructures such as porosity and doping. The objective is to achieve an optimal and highly integrated PV-TEG device that can substantially increase the solar energy harvest efficiency.Keywords: thermoelectric, finite element method, 3d print, energy conversion
Procedia PDF Downloads 63351 Interfacial Adhesion and Properties Improvement of Polyethylene/Thermoplastic Starch Blend Compatibilized by Stearic Acid-Grafted-Starch
Authors: Nattaporn Khanoonkon, Rangrong Yoksan, Amod A. Ogale
Abstract:
Polyethylene (PE) is one of the most petroleum-based thermoplastic materials used in many applications including packaging due to its cheap, light-weight, chemically inert and capable to be converted into various shapes and sizes of products. Although PE is a commercially potential material, its non-biodegradability caused environmental problems. At present, bio-based polymers become more interesting owing to its bio-degradability, non-toxicity, and renewability as well as being eco-friendly. Thermoplastic starch (TPS) is a bio-based and biodegradable plastic produced from the plasticization of starch under applying heat and shear force. In many researches, TPS was blended with petroleum-based polymers including PE in order to reduce the cost and the use of those polymers. However, the phase separation between hydrophobic PE and hydrophilic TPS limited the amount of TPS incorporated. The immiscibility of two different polarity polymers can be diminished by adding compatibilizer. PE-based compatibilizers, e.g. polyethylene-grafted-maleic anhydride, polyethylene-co-vinyl alcohol, etc. have been applied for the PE/TPS blend system in order to improve their miscibility. Until now, there is no report about the utilization of starch-based compatibilizer for PE/TPS blend system. The aims of the present research were therefore to synthesize a new starch-based compatibilizer, i.e. stearic acid-grafted starch (SA-g-starch) and to study the effect of SA-g-starch on chemical interaction, morphological properties, tensile properties and water vapor as well as oxygen barrier properties of the PE/TPS blend films. PE/TPS blends without and with incorporating SA-g-starch with a content of 1, 3 and 5 part(s) per hundred parts of starch (phr) were prepared using a twin screw extruder and then blown into films using a film blowing machine. Incorporating 1 phr and 3 phr of SA-g-starch could improve miscibility of the two polymers as confirmed from the reduction of TPS phase size and the good dispersion of TPS phase in PE matrix. In addition, the blend containing SA-g-starch with contents of 1 phr and 3 phr exhibited higher tensile strength and extensibility, as well as lower water vapor and oxygen permeabilities than the naked blend. The above results suggested that SA-g-starch could be potentially applied as a compatibilizer for the PE/TPS blend system.Keywords: blend, compatibilizer, polyethylene, thermoplastic starch
Procedia PDF Downloads 436350 Austempered Compacted Graphite Irons: Influence of Austempering Temperature on Microstructure and Microscratch Behavior
Authors: Rohollah Ghasemi, Arvin Ghorbani
Abstract:
This study investigates the effect of austempering temperature on microstructure and scratch behavior of the austempered heat-treated compacted graphite irons. The as-cast was used as base material for heat treatment practices. The samples were extracted from as-cast ferritic CGI pieces and were heat treated under austenitising temperature of 900°C for 60 minutes which followed by quenching in salt-bath at different austempering temperatures of 275°C, 325°C and 375°C. For all heat treatments, an austempering holding time of 30 minutes was selected for this study. Light optical microscope (LOM) and scanning electron microscope (SEM) and electron back scattered diffraction (EBSD) analysis confirmed the ausferritic matrix formed in all heat-treated samples. Microscratches were performed under the load of 200, 600 and 1000 mN using a sphero-conical diamond indenter with a tip radius of 50 μm and induced cone angle 90° at a speed of 10 μm/s at room temperature ~25°C. An instrumented nanoindentation machine was used for performing nanoindentation hardness measurement and microscratch testing. Hardness measurements and scratch resistance showed a significant increase in Brinell, Vickers, and nanoindentation hardness values as well as microscratch resistance of the heat-treated samples compared to the as-cast ferritic sample. The increase in hardness and improvement in microscratch resistance are associated with the formation of the ausferrite matrix consisted of carbon-saturated retained austenite and acicular ferrite in austempered matrix. The maximum hardness was observed for samples austempered at 275°C which resulted in the formation of very fine acicular ferrite. In addition, nanohardness values showed a quite significant variation in the matrix due to the presence of acicular ferrite and carbon-saturated retained austenite. It was also observed that the increase of austempering temperature resulted in increase of volume of the carbon-saturated retained austenite and decrease of hardness values.Keywords: austempered CGI, austempering, scratch testing, scratch plastic deformation, scratch hardness
Procedia PDF Downloads 129349 On Cloud Computing: A Review of the Features
Authors: Assem Abdel Hamed Mousa
Abstract:
The Internet of Things probably already influences your life. And if it doesn’t, it soon will, say computer scientists; Ubiquitous computing names the third wave in computing, just now beginning. First were mainframes, each shared by lots of people. Now we are in the personal computing era, person and machine staring uneasily at each other across the desktop. Next comes ubiquitous computing, or the age of calm technology, when technology recedes into the background of our lives. Alan Kay of Apple calls this "Third Paradigm" computing. Ubiquitous computing is essentially the term for human interaction with computers in virtually everything. Ubiquitous computing is roughly the opposite of virtual reality. Where virtual reality puts people inside a computer-generated world, ubiquitous computing forces the computer to live out here in the world with people. Virtual reality is primarily a horse power problem; ubiquitous computing is a very difficult integration of human factors, computer science, engineering, and social sciences. The approach: Activate the world. Provide hundreds of wireless computing devices per person per office, of all scales (from 1" displays to wall sized). This has required new work in operating systems, user interfaces, networks, wireless, displays, and many other areas. We call our work "ubiquitous computing". This is different from PDA's, dynabooks, or information at your fingertips. It is invisible; everywhere computing that does not live on a personal device of any sort, but is in the woodwork everywhere. The initial incarnation of ubiquitous computing was in the form of "tabs", "pads", and "boards" built at Xerox PARC, 1988-1994. Several papers describe this work, and there are web pages for the Tabs and for the Boards (which are a commercial product now): Ubiquitous computing will drastically reduce the cost of digital devices and tasks for the average consumer. With labor intensive components such as processors and hard drives stored in the remote data centers powering the cloud , and with pooled resources giving individual consumers the benefits of economies of scale, monthly fees similar to a cable bill for services that feed into a consumer’s phone.Keywords: internet, cloud computing, ubiquitous computing, big data
Procedia PDF Downloads 377348 Discerning Divergent Nodes in Social Networks
Authors: Mehran Asadi, Afrand Agah
Abstract:
In data mining, partitioning is used as a fundamental tool for classification. With the help of partitioning, we study the structure of data, which allows us to envision decision rules, which can be applied to classification trees. In this research, we used online social network dataset and all of its attributes (e.g., Node features, labels, etc.) to determine what constitutes an above average chance of being a divergent node. We used the R statistical computing language to conduct the analyses in this report. The data were found on the UC Irvine Machine Learning Repository. This research introduces the basic concepts of classification in online social networks. In this work, we utilize overfitting and describe different approaches for evaluation and performance comparison of different classification methods. In classification, the main objective is to categorize different items and assign them into different groups based on their properties and similarities. In data mining, recursive partitioning is being utilized to probe the structure of a data set, which allow us to envision decision rules and apply them to classify data into several groups. Estimating densities is hard, especially in high dimensions, with limited data. Of course, we do not know the densities, but we could estimate them using classical techniques. First, we calculated the correlation matrix of the dataset to see if any predictors are highly correlated with one another. By calculating the correlation coefficients for the predictor variables, we see that density is strongly correlated with transitivity. We initialized a data frame to easily compare the quality of the result classification methods and utilized decision trees (with k-fold cross validation to prune the tree). The method performed on this dataset is decision trees. Decision tree is a non-parametric classification method, which uses a set of rules to predict that each observation belongs to the most commonly occurring class label of the training data. Our method aggregates many decision trees to create an optimized model that is not susceptible to overfitting. When using a decision tree, however, it is important to use cross-validation to prune the tree in order to narrow it down to the most important variables.Keywords: online social networks, data mining, social cloud computing, interaction and collaboration
Procedia PDF Downloads 148347 System Dietadhoc® - A Fusion of Human-Centred Design and Agile Development for the Explainability of AI Techniques Based on Nutritional and Clinical Data
Authors: Michelangelo Sofo, Giuseppe Labianca
Abstract:
In recent years, the scientific community's interest in the exploratory analysis of biomedical data has increased exponentially. Considering the field of research of nutritional biologists, the curative process, based on the analysis of clinical data, is a very delicate operation due to the fact that there are multiple solutions for the management of pathologies in the food sector (for example can recall intolerances and allergies, management of cholesterol metabolism, diabetic pathologies, arterial hypertension, up to obesity and breathing and sleep problems). In this regard, in this research work a system was created capable of evaluating various dietary regimes for specific patient pathologies. The system is founded on a mathematical-numerical model and has been created tailored for the real working needs of an expert in human nutrition using the human-centered design (ISO 9241-210), therefore it is in step with continuous scientific progress in the field and evolves through the experience of managed clinical cases (machine learning process). DietAdhoc® is a decision support system nutrition specialists for patients of both sexes (from 18 years of age) developed with an agile methodology. Its task consists in drawing up the biomedical and clinical profile of the specific patient by applying two algorithmic optimization approaches on nutritional data and a symbolic solution, obtained by transforming the relational database underlying the system into a deductive database. For all three solution approaches, particular emphasis has been given to the explainability of the suggested clinical decisions through flexible and customizable user interfaces. Furthermore, the system has multiple software modules based on time series and visual analytics techniques that allow to evaluate the complete picture of the situation and the evolution of the diet assigned for specific pathologies.Keywords: medical decision support, physiological data extraction, data driven diagnosis, human centered AI, symbiotic AI paradigm
Procedia PDF Downloads 10346 “CheckPrivate”: Artificial Intelligence Powered Mobile Application to Enhance the Well-Being of Sextual Transmitted Diseases Patients in Sri Lanka under Cultural Barriers
Authors: Warnakulasuriya Arachichige Malisha Ann Rosary Fernando, Udalamatta Gamage Omila Chalanka Jinadasa, Bihini Pabasara Amandi Amarasinghe, Manul Thisuraka Mandalawatta, Uthpala Samarakoon, Manori Gamage
Abstract:
The surge in sexually transmitted diseases (STDs) has become a critical public health crisis demanding urgent attention and action. Like many other nations, Sri Lanka is grappling with a significant increase in STDs due to a lack of education and awareness regarding their dangers. Presently, the available applications for tracking and managing STDs cover only a limited number of easily detectable infections, resulting in a significant gap in effectively controlling their spread. To address this gap and combat the rising STD rates, it is essential to leverage technology and data. Employing technology to enhance the tracking and management of STDs is vital to prevent their further propagation and to enable early intervention and treatment. This requires adopting a comprehensive approach that involves raising public awareness about the perils of STDs, improving access to affordable healthcare services for early detection and treatment, and utilizing advanced technology and data analysis. The proposed mobile application aims to cater to a broad range of users, including STD patients, recovered individuals, and those unaware of their STD status. By harnessing cutting-edge technologies like image detection, symptom-based identification, prevention methods, doctor and clinic recommendations, and virtual counselor chat, the application offers a holistic approach to STD management. In conclusion, the escalating STD rates in Sri Lanka and across the globe require immediate action. The integration of technology-driven solutions, along with comprehensive education and healthcare accessibility, is the key to curbing the spread of STDs and promoting better overall public health.Keywords: STD, machine learning, NLP, artificial intelligence
Procedia PDF Downloads 76345 Determination of Community Based Reference Interval of Aspartate Aminotransferase to Platelet Ratio Index (APRI) among Healthy Populations in Mekelle City Tigray, Northern Ethiopia
Authors: Getachew Belay Kassahun
Abstract:
Background: Aspartate aminotransferase to Platelet Ratio Index (APRI) currently becomes a biomarker for screening liver fibrosis since liver biopsy procedure is invasive and variation in pathological interpretation. Clinical Laboratory Standard Institute recommends establishing age, sex and environment specific reference interval for biomarkers in a homogenous population. The current study was aimed to derive community based reference interval of APRI aged between 12 and 60 years old in Mekelle city Tigrai, Northern Ethiopia. Method: Six hundred eighty eight study participants were collected from three districts in Mekelle city. The 3 districts were selected through random sampling technique and sample size to kebelles (small administration) were distributed proportional to household number in each district. Lottery method was used at household level if more than 2 study participants to each age partition were found. A community based cross sectional in a total of 534 study participants, 264 male and 270 females, were included in the final laboratory and data analysis but around 154 study participants were excluded through exclusion criteria. Aspartate aminotransferase was analyzed through Biosystem chemistry analyzer and Sysmix machine was used to analyze platelet. Man Whitney U test non parametric stastical tool was used to appreciate stastical difference among gender after excluding the outliers through Box and Whisker. Result: The study appreciated stastical difference among gender for APRI reference interval. The combined, male and female reference interval in the current study was 0.098-0.390, 0.133-0.428 and 0.090-0.319 respectively. The upper and lower reference interval of males was higher than females in all age partition and there was no stastical difference (p-value (<0.05)) between age partition. Conclusion: The current study showed using sex specific reference interval is significant to APRI biomarker in clinical practice for result interpretation.Keywords: reference interval, aspartate aminotransferase to platelet ratio Index, Ethiopia, tigray
Procedia PDF Downloads 102