Search results for: feature pyramid network

Authors: Rahul Agarwal, Ashutosh Singh

Abstract:

Cervical cancer is one of the women related cancers which starts from the pre-cancerous cells and a fraction of women with pre-cancers of the cervix will develop cervical cancer. Cervical pre-cancers if treated in pre-invasive stage can prevent almost all true cervical squamous cell carcinoma. The present study investigates the genes and pathways that are involved in the progression of cervical cancer and are responsible in transition from pre-invasive stage to other advanced invasive stages. The study used GDS3292 microarray data to identify the stage specific genes in cervical cancer and further to generate the network of the significant genes. The microarray data GDS3292 consists of the expression profiling of 10 normal cervices, 7 HSILs and 21 SCCs samples. The study identifies 70 upregulated and 37 downregulated genes in HSIL stage while 95 upregulated and 60 downregulated genes in SCC stages. Biological process including cell communication, signal transduction are highly enriched in both HSIL and SCC stages of cervical cancer. Further, the ppi interaction of genes involved in HSIL and SCC stages helps in identifying the interacting partners. This work may lead to the identification of potential diagnostic biomarker which can be utilized for early stage detection.

Keywords: cervical cancer, HSIL, microarray, SCC

Procedia PDF Downloads 234

Authors: Gaelle Candel, David Naccache

Abstract:

t-SNE is an embedding method that the data science community has widely used. It helps two main tasks: to display results by coloring items according to the item class or feature value; and for forensic, giving a first overview of the dataset distribution. Two interesting characteristics of t-SNE are the structure preservation property and the answer to the crowding problem, where all neighbors in high dimensional space cannot be represented correctly in low dimensional space. t-SNE preserves the local neighborhood, and similar items are nicely spaced by adjusting to the local density. These two characteristics produce a meaningful representation, where the cluster area is proportional to its size in number, and relationships between clusters are materialized by closeness on the embedding. This algorithm is non-parametric. The transformation from a high to low dimensional space is described but not learned. Two initializations of the algorithm would lead to two different embeddings. In a forensic approach, analysts would like to compare two or more datasets using their embedding. A naive approach would be to embed all datasets together. However, this process is costly as the complexity of t-SNE is quadratic and would be infeasible for too many datasets. Another approach would be to learn a parametric model over an embedding built with a subset of data. While this approach is highly scalable, points could be mapped at the same exact position, making them indistinguishable. This type of model would be unable to adapt to new outliers nor concept drift. This paper presents a methodology to reuse an embedding to create a new one, where cluster positions are preserved. The optimization process minimizes two costs, one relative to the embedding shape and the second relative to the support embedding’ match. The embedding with the support process can be repeated more than once, with the newly obtained embedding. The successive embedding can be used to study the impact of one variable over the dataset distribution or monitor changes over time. This method has the same complexity as t-SNE per embedding, and memory requirements are only doubled. For a dataset of n elements sorted and split into k subsets, the total embedding complexity would be reduced from O(n²) to O(n²=k), and the memory requirement from n² to 2(n=k)², which enables computation on recent laptops. The method showed promising results on a real-world dataset, allowing to observe the birth, evolution, and death of clusters. The proposed approach facilitates identifying significant trends and changes, which empowers the monitoring high dimensional datasets’ dynamics.

Keywords: concept drift, data visualization, dimension reduction, embedding, monitoring, reusability, t-SNE, unsupervised learning

Procedia PDF Downloads 144

Authors: Maria Areias, Diogo Simões, Ana Figueiredo, Anishur Rahman, Filipa Figueiredo, João Nunes

Abstract:

It is expected that close to 7 billion people will live in urban areas by 2050. In order to improve the sustainability of the territories and its transition towards circular economy, it’s necessary to understand its metabolism and promote and guide the entrepreneurship answer. The study of a macroeconomic metabolism involves the quantification of the inputs, outputs and storage of energy, water, materials and wastes for an urban region. This quantification and analysis representing one opportunity for the promotion of green entrepreneurship. There are several methods to assess the environmental impacts of an urban territory, such as human and environmental risk assessment (HERA), life cycle assessment (LCA), ecological footprint assessment (EF), material flow analysis (MFA), physical input-output table (PIOT), ecological network analysis (ENA), multicriteria decision analysis (MCDA) among others. However, no consensus exists about which of those assessment methods are best to analyze the sustainability of these complex systems. Taking into account the weaknesses and needs identified, the CiiM - Circular Innovation Inter-Municipality project aims to define an uniform and globally accepted methodology through the integration of various methodologies and dynamic approaches to increase the efficiency of macroeconomic metabolisms and promoting entrepreneurship in a circular economy. The pilot territory considered in CiiM project has a total area of 969,428 ha, comprising a total of 897,256 inhabitants (about 41% of the population of the Center Region). The main economic activities in the pilot territory, which contribute to a gross domestic product of 14.4 billion euros, are: social support activities for the elderly; construction of buildings; road transport of goods, retailing in supermarkets and hypermarkets; mass production of other garments; inpatient health facilities; and the manufacture of other components and accessories for motor vehicles. The region's business network is mostly constituted of micro and small companies (similar to the Central Region of Portugal), with a total of 53,708 companies identified in the CIM Region of Coimbra (39 large companies), 28,146 in the CIM Viseu Dão Lafões (22 large companies) and 24,953 in CIM Beiras and Serra da Estrela (13 large companies). For the construction of the database was taking into account data available at the National Institute of Statistics (INE), General Directorate of Energy and Geology (DGEG), Eurostat, Pordata, Strategy and Planning Office (GEP), Portuguese Environment Agency (APA), Commission for Coordination and Regional Development (CCDR) and Inter-municipal Community (CIM), as well as dedicated databases. In addition to the collection of statistical data, it was necessary to identify and characterize the different stakeholder groups in the pilot territory that are relevant to the different metabolism components under analysis. The CIIM project also adds the potential of a Geographic Information System (GIS) so that it is be possible to obtain geospatial results of the territorial metabolisms (rural and urban) of the pilot region. This platform will be a powerful visualization tool of flows of products/services that occur within the region and will support the stakeholders, improving their circular performance and identifying new business ideas and symbiotic partnerships.

Keywords: circular economy tools, life cycle assessment macroeconomic metabolism, multicriteria decision analysis, decision support tools, circular entrepreneurship, industrial and regional symbiosis

Procedia PDF Downloads 101

Authors: Meet Shah, Ankita Aditya, Dhruv Bindra, V. S. Omkar, Aashruti Seervi

Abstract:

In today's world, we need documents everywhere for a smooth workflow in the identification process or any other security aspects. The current system and techniques which are used for identification need one thing, that is ‘proof of existence’, which involves valid documents, for example, educational, financial, etc. The main issue with the current identity access management system and digital identification process is that the system is centralized in their network, which makes it inefficient. The paper presents the system which resolves all these cited issues. It is based on ‘blockchain’ technology, which is a 'decentralized system'. It allows transactions in a decentralized and immutable manner. The primary notion of the model is to ‘have everything with nothing’. It involves inter-linking required documents of a person with a single identity card so that a person can go anywhere without having the required documents with him/her. The person just needs to be physically present at a place wherein documents are necessary, and using a fingerprint impression and an iris scan print, the rest of the verification will progress. Furthermore, some technical overheads and advancements are listed. This paper also aims to layout its far-vision scenario of blockchain and its impact on future trends.

Keywords: blockchain, decentralized system, fingerprint impression, identity management, iris scan

Procedia PDF Downloads 129

Authors: Shriya Shukla, Lachin Fernando

Abstract:

Pneumonia, a leading cause of mortality in young children globally, presents significant diagnostic challenges, particularly in resource-limited settings. This study presents an approach to diagnosing pediatric pneumonia using Chest X-Ray (CXR) images, employing a lightweight MobileNetV2 model enhanced with synthetic data augmentation. Addressing the challenge of dataset scarcity and imbalance, the study used a Variational Autoencoder-Generative Adversarial Network (VAE-GAN) to generate synthetic CXR images, improving the representation of normal cases in the pediatric dataset. This approach not only addresses the issues of data imbalance and scarcity prevalent in medical imaging but also provides a more accessible and reliable diagnostic tool for early pneumonia detection. The augmented data improved the model’s accuracy and generalization, achieving an overall accuracy of 95% in pneumonia detection. These findings highlight the efficacy of the MobileNetV2 model, offering a computationally efficient yet robust solution well-suited for resource-constrained environments such as mobile health applications. This study demonstrates the potential of synthetic data augmentation in enhancing medical image analysis for critical conditions like pediatric pneumonia.

Keywords: pneumonia, MobileNetV2, image classification, GAN, VAE, deep learning

Procedia PDF Downloads 126

Authors: Sufia Khowaja, Nishat Fatima

Abstract:

Social networking sites are significant for visually impaired to overcome the unique challenges they face and access the resources they need to succeed in their education and beyond which might be difficult to obtain through traditional means. It provides them an opportunity to build relationships, stay connected with their support network as well as to develop social skills which give them emotional support to fell less isolated. In this connection the study is conducted with the aim to determine the use of social networking sites, purpose of using and activities performed by visually impaired at Delhi University, Delhi, Jawaharlal Nehru University, Delhi and Jamia Milia Islamia, Delhi. The study followed survey technique in which structured interview is followed to collect data from 137 visually impaired students and analysed using ‘SPSS ver23’. The findings of the study revealed that mostly used social networking sites are whatsapp by 89.23% students of DU, 95.12% of JNU, 87.09% of JMI, followed by e-mail by 78.46% of DU, 78.04% of JNU, 64.51%; youtube by 73.84% DU, 90.24% JNU, 80.64% JMI. Purpose for using these sites is for academics mentioned by 96.92% DU, 100% JNU, 93.54% JMI. Activities performed on sites are sending and receiving messaging 96.92% DU, 92.68% JNU, 93.55% JMI, communicating with friends and family as well as getting academic information. Findings of the study will be helpful for libraries to disseminate their services and resources as well as latest updates to their visually impaired users with the help of most used tools.

Keywords: social networking sites, visually impaired, Delhi University, Jawaharlal Nehru University, Jamia Milia Islamia

Procedia PDF Downloads 90

Authors: Jay L. Fu

Abstract:

Patients with Alzheimer's disease progressively lose their memory and thinking skills and, eventually, the ability to carry out simple daily tasks. The disease is irreversible, but early detection and treatment can slow down the disease progression. In this research, publicly available MRI data and demographic data from 373 MRI imaging sessions were utilized to build models to predict dementia. Various machine learning models, including logistic regression, k-nearest neighbor, support vector machine, random forest, and neural network, were developed. Data were divided into training and testing sets, where training sets were used to build the predictive model, and testing sets were used to assess the accuracy of prediction. Key risk factors were identified, and various models were compared to come forward with the best prediction model. Among these models, the random forest model appeared to be the best model with an accuracy of 90.34%. MMSE, nWBV, and gender were the three most important contributing factors to the detection of Alzheimer’s. Among all the models used, the percent in which at least 4 of the 5 models shared the same diagnosis for a testing input was 90.42%. These machine learning models allow early detection of Alzheimer’s with good accuracy, which ultimately leads to early treatment of these patients.

Keywords: Alzheimer's disease, clinical diagnosis, magnetic resonance imaging, machine learning prediction

Procedia PDF Downloads 143

Authors: Hassan Hussin Zwida

Abstract:

The liquified LPG (Liquefied Petroleum Gas) drying system at the Complex is designed to remove water and mercaptans from the LPG stream. Upon saturation of the desiccant beds, a regeneration cycle becomes necessary. The original design routed the regeneration gas, produced during the LPG dryer heating cycle, to the sulfur recovery unit to the incineration. However, concerns regarding high temperatures and potential unit disruptions led to a modification where the gas is currently vented to the acid flare for the initial hour before being diverted to the LP network fuel gas system. While this addresses the temperature concerns, it generates significant smoke due to the presence of liquid hydrocarbons. This paper proposes an approach to recover the regeneration gas and redirect it back to the gas plant's (off-gas compressors) instead of sending it to the AC (Acid Flare), by utilizing the existing pipe 6” and connected to off gas compressor KO (Knock-Out ) Drums . This option is simple to operate, flexible, environment-friendly solution as long-term solution, lower in capital expenditure and increase the company's profitability. The feasibility of this proposal is supported by dynamic simulations. The simulations suggest the possibility of operating two out of the three off-gas compressors and LPG (Liquefied petroleum gas) as a liquid phase, is foreseen to be carried over and gathered at the bottom level of the KO (Knock-Out) Drum.

Keywords: thermal incinerator, off-gas compressors, environment, knock-out drums, acid flare

Procedia PDF Downloads 51

Authors: Felix Boehnisch, Alexander Lutz

Abstract:

Based on the number of users and the amount of content posted daily, Facebook is considered the largest social network in the world. This content includes images or text posts from companies but also private persons, which are also commented on by other users. However, it can sometimes be difficult for companies to keep track of all the posts and the reactions to them, especially when there are several posts a day that contain hundreds to thousands of comments. To facilitate this, the following paper deals with the possible applications of sentiment analysis to social media comments in order to be able to support the work in social media marketing. In a first step, post comments were divided into positive and negative by a subjective rating, then the same comments were checked for their polarity value by the two german python libraries TextBlobDE and SentiWS and also grouped into positive, negative, or even neutral. As a control, the subjective classifications were compared with the machine-generated ones by a confusion matrix, and relevant quality criteria were determined. The accuracy of both libraries was not really meaningful, with 60% to 66%. However, many words or sentences were not evaluated at all, so there seems to be room for optimization to possibly get more accurate results. In future studies, the use of these specific German libraries can be optimized to gain better insights by either applying them to stricter cleaned data or by adding a sentiment value to emojis, which have been removed from the comments in advance, as they are not contained in the libraries.

Keywords: Facebook, German libraries, polarity, sentiment analysis, social media comments

Procedia PDF Downloads 182

Authors: Soofiyan Atar, Adil Shaikh, Sahil Rajpurkar, Pragnesh Bhalala, Aniket Desai, Irfan Siddavatam

Abstract:

Deep reinforcement learning (deep RL) algorithms leverage the symbolic power of complex controllers by automating it by mapping sensory inputs to low-level actions. Deep RL eliminates the complex robot dynamics with minimal engineering. Deep RL provides high-risk involvement by directly implementing it in real-world scenarios and also high sensitivity towards hyperparameters. Tuning of hyperparameters on a pneumatic quadruped robot becomes very expensive through trial-and-error learning. This paper presents an automated learning control for a pneumatic quadruped robot using sample efficient deep Q learning, enabling minimal tuning and very few trials to learn the neural network. Long training hours may degrade the pneumatic cylinder due to jerk actions originated through stochastic weights. We applied this method to the pneumatic quadruped robot, which resulted in a hopping gait. In our process, we eliminated the use of a simulator and acquired a stable gait. This approach evolves so that the resultant gait matures more sturdy towards any stochastic changes in the environment. We further show that our algorithm performed very well as compared to programmed gait using robot dynamics.

Keywords: model-based reinforcement learning, gait stability, supervised learning, pneumatic quadruped

Procedia PDF Downloads 316

Authors: Samuel O. Oduyela

Abstract:

Over the last ten years, many studies have examined the media coverage of terrorism across the world. Nevertheless, most of these studies have been inclined to the western narrative, more so in relation to the international media. This study departs from that partiality to explore the Nigerian press and its coverage of the Boko Haram. The study intends to illustrate how the Nigerian press has reported its homegrown terrorism within its borders. On 14 April 2014, the Shekau-led Boko Haram kidnapped over 200 female students from Chibok in the Borno State. This study analyses a structured sample of news stories, feature articles, editorial comments, and opinions from the Nation newspaper. The study examined the representation of the Chibok girls kidnaps by concentrating on four main viewpoints. The news framing of the Chibok girls’ kidnap under Presidents Goodluck Jonathan (2014) and Mohammadu Buhari (2016-2018), the sourcing model present in the news reporting of the kidnap and the challenges Nation reporters face in reporting Boko Haram. The study adopted the use of qualitative news framing analysis to provide further insights into significant developments established from the examination of news contents. The study found that the news reportage mainly focused on the government response to Chibok girls kidnap, international press and Boko Haram. Boko Haram was also framed, as a political conspiracy, as prevailing, and as instilling fear. Political, and economic influence appeared to be a significant determinant of the reportage. The study found that the Nation newspaper's portrayal of the crisis under President Jonathan differed significantly from under President Buhari. While the newspaper framed the action of President Jonathan as lacklustre, dismissive, and confusing, it was less critical of President Buhari's government's handling of the crisis. The Nation newspaper failed to promote or explore non-violent approaches. News reports of the kidnap, thus, were presented mainly from a political and ethnoreligious perspective. The study also raised questions of what roles should journalists play in covering conflicts? Should they merely report comments on and interpret it, or should they be actors in the resolution or, more importantly, the prevention of conflicts? The study underlined the need for the independence of the media, more training for journalists to advance a more nuanced and conflict-sensitive news coverage in the Nigerian context.

Keywords: boko haram, chibok girls kidnap, conflict in nigeria, media framing

Procedia PDF Downloads 150

Authors: Jialin Ma, Jia Meng

Abstract:

Genomic features, the genome-based coordinates, are commonly used for the representation of biological features such as genes, RNA transcripts and transcription factor binding sites. For the analysis of RNA-related genomic features, such as RNA modification sites, a common task is to correlate these features with transcript components (5'UTR, CDS, 3'UTR) to explore their distribution characteristics in terms of transcriptomic coordinates, e.g., to examine whether a specific type of biological feature is enriched near transcription start sites. Existing approaches for performing these tasks involve the manipulation of a gene database, conversion from genome-based coordinate to transcript-based coordinate, and visualization methods that are capable of showing RNA transcript components and distribution of the features. These steps are complicated and time consuming, and this is especially true for researchers who are not familiar with relevant tools. To overcome this obstacle, we develop a dedicated web app TARF, which represents web toolkit for annotating RNA-related genomic features. TARF web tool intends to provide a web-based way to easily annotate and visualize RNA-related genomic features. Once a user has uploaded the features with BED format and specified a built-in transcript database or uploaded a customized gene database with GTF format, the tool could fulfill its three main functions. First, it adds annotation on gene and RNA transcript components. For every features provided by the user, the overlapping with RNA transcript components are identified, and the information is combined in one table which is available for copy and download. Summary statistics about ambiguous belongings are also carried out. Second, the tool provides a convenient visualization method of the features on single gene/transcript level. For the selected gene, the tool shows the features with gene model on genome-based view, and also maps the features to transcript-based coordinate and show the distribution against one single spliced RNA transcript. Third, a global transcriptomic view of the genomic features is generated utilizing the Guitar R/Bioconductor package. The distribution of features on RNA transcripts are normalized with respect to RNA transcript landmarks and the enrichment of the features on different RNA transcript components is demonstrated. We tested the newly developed TARF toolkit with 3 different types of genomics features related to chromatin H3K4me3, RNA N6-methyladenosine (m6A) and RNA 5-methylcytosine (m5C), which are obtained from ChIP-Seq, MeRIP-Seq and RNA BS-Seq data, respectively. TARF successfully revealed their respective distribution characteristics, i.e. H3K4me3, m6A and m5C are enriched near transcription starting sites, stop codons and 5’UTRs, respectively. Overall, TARF is a useful web toolkit for annotation and visualization of RNA-related genomic features, and should help simplify the analysis of various RNA-related genomic features, especially those related RNA modifications.

Keywords: RNA-related genomic features, annotation, visualization, web server

Procedia PDF Downloads 208

Authors: Bandar Alahmadi, Lethia Jackson

Abstract:

Machine Learning model is used today in many real-life applications. The safety and security of such model is important, so the results of the model are as accurate as possible. One challenge of machine learning model security is the adversarial examples attack. Adversarial examples are designed by the attacker to cause the machine learning model to misclassify the input. We propose a method to generate adversarial examples to attack image classifiers. We are modifying the successfully classified images, so a classifier misclassifies them after the modification. In our method, we do not update the whole image, but instead we detect the important region, modify it, place it back to the original image, and then run it through a classifier. The algorithm modifies the detected region using two methods. First, it will add abstract image matrix on back of the detected image matrix. Then, it will perform a rotation attack to rotate the detected region around its axes, and embed the trace of image in image background. Finally, the attacked region is placed in its original position, from where it was removed, and a smoothing filter is applied to smooth the background with foreground. We test our method in cascade classifier, and the algorithm is efficient, the classifier confident has dropped to almost zero. We also try it in CNN (Convolutional neural network) with higher setting and the algorithm was successfully worked.

Keywords: adversarial examples, attack, computer vision, image processing

Procedia PDF Downloads 339

Authors: Hamidreza Ghazvinian, Khosro Ghazvinian, Touba Khodaiean

Abstract:

The importance of evaporation estimation in water resources and agricultural studies is undeniable. Pan evaporation are used as an indicator to determine the evaporation of lakes and reservoirs around the world due to the ease of interpreting its data. In this research, intelligent models were investigated in estimating pan evaporation on a daily basis. Shahroud and Mayamey were considered as the studied cities. These two cities are located in Semnan province in Iran. The mentioned cities have dry weather conditions that are susceptible to high evaporation potential. Meteorological data of 11 years of synoptic stations of Shahrood and Mayamey cities were used. The intelligent models used in this study are Artificial Neural Network (ANN), Least Squares Support Vector Machine (LSSVM), and M5 tree models. Meteorological parameters of minimum and maximum air temperature (Tmax, Tmin), wind speed (WS), sunshine hours (SH), air pressure (PA), relative humidity (RH) as selected input data and evaporation data from pan (EP) to The output data was considered. 70% of data is used at the education level, and 30 % of the data is used at the test level. Models used with explanation coefficient evaluation (R2) Root of Mean Squares Error (RMSE) and Mean Absolute Error (MAE). The results for the two Shahroud and Mayamey stations showed that the above three models' operations are rather appropriate.

Keywords: pan evaporation, intelligent methods, shahroud, mayamey

Procedia PDF Downloads 74

Authors: Emelia Noronha, Milind Malshe

Abstract:

The phatic function of communication is a vital element of any conversation. This research paper looks into this function with respect to personal blogs maintained by Indian bloggers. This paper is a study into the phenomenon of phatic communication maintained by bloggers through their blogs. Based on a linguistic analysis of the posts of twenty eight Indian bloggers, writing in English, studied over a period of three years, the study indicates that though the blogging phenomenon is not conversational in the same manner as face-to-face communication, it does make ample provision for feedback that is conversational in nature. Ordinary day to day offline conversations use conventionalized phatic utterances; those on the social media are in a perpetual mode of innovation and experimentation in order to sustain contact with its readers. These innovative methods and means are the focus of this study. Though the personal blogger aims to chronicle his/her personal life through the blog, the socializing function is crucial to these bloggers. In comparison to the western personal blogs which focus on the presentation of the ‘bounded individual self’, we find Indian personal bloggers engage in the presentation of their ‘social selves’. These bloggers yearn to reach out to the readers on the internet and the phatic function serves to initiate, sustain and renew social ties on the blogosphere thereby consolidating the social network of readers and bloggers.

Keywords: personal blogs, phatic, social-selves, blog readers

Procedia PDF Downloads 362

Authors: O. A. Ajigini, E. N. Mwim

Abstract:

Cybersecurity is the protection of computers, programs, networks, and data from attack, damage, unauthorised, unintended access, change, or destruction. Organisations collect, process and store their confidential and sensitive information on computers and transmit this data across networks to other computers. Moreover, the advent of internet technologies has led to various cyberattacks resulting in dangerous consequences for organisations. Therefore, with the increase in the volume and sophistication of cyberattacks, there is a need to develop models and make recommendations for the management of cybersecurity threats in organisations. This paper reports on various threats that cause malicious damage to organisations in cyberspace and provides measures on how these threats can be eliminated or reduced. The paper explores various aspects of protection measures against cybersecurity threats such as handling of sensitive data, network security, protection of information assets and cybersecurity awareness. The paper posits a model and recommendations on how to manage cybersecurity threats in organisations effectively. The model and the recommendations can then be utilised by organisations to manage the threats affecting their cyberspace. The paper provides valuable information to assist organisations in managing their cybersecurity threats and hence protect their computers, programs, networks and data in cyberspace. The paper aims to assist organisations to protect their information assets and data from cyberthreats as part of the contributions toward community engagement.

Keywords: confidential information, cyberattacks, cybersecurity, cyberspace, sensitive information

Procedia PDF Downloads 259

Authors: Olivia Borgue, Massimo Panarotto, Ola Isaksson

Abstract:

This article presents a neural networks-based strategy for reducing the need for support structures when designing for additive manufacturing (AM). Additive manufacturing is a relatively new and immature industrial technology, and the information to make confident decisions when designing for AM is limited. This lack of information impacts especially the early stages of engineering design, for instance, it is difficult to actively consider the support structures needed for manufacturing a part. This difficulty is related to the challenge of designing a product geometry accounting for customer requirements, manufacturing constraints and minimization of support structure. The approach presented in this article proposes an automatized geometry modification technique for reducing the use of the support structures while designing for AM. This strategy starts with a neural network-based strategy for shape recognition to achieve product classification, using an STL file of the product as input. Based on the classification, an automatic part geometry modification based on MATLAB© is implemented. At the end of the process, the strategy presents different geometry modification alternatives depending on the type of product to be designed. The geometry alternatives are then evaluated adopting a QFD-like decision support tool.

Keywords: additive manufacturing, engineering design, geometry modification optimization, neural networks

Procedia PDF Downloads 253

Authors: Arindam Chaudhuri

Abstract:

Classification of data has been actively used for most effective and efficient means of conveying knowledge and information to users. The prima face has always been upon techniques for extracting useful knowledge from data such that returns are maximized. With emergence of huge datasets the existing classification techniques often fail to produce desirable results. The challenge lies in analyzing and understanding characteristics of massive data sets by retrieving useful geometric and statistical patterns. We propose a supervised parallel fuzzy rough support vector machine (PFRSVM) for data classification in cloud environment. The classification is performed by PFRSVM using hyperbolic tangent kernel. The fuzzy rough set model takes care of sensitiveness of noisy samples and handles impreciseness in training samples bringing robustness to results. The membership function is function of center and radius of each class in feature space and is represented with kernel. It plays an important role towards sampling the decision surface. The success of PFRSVM is governed by choosing appropriate parameter values. The training samples are either linear or nonlinear separable. The different input points make unique contributions to decision surface. The algorithm is parallelized with a view to reduce training times. The system is built on support vector machine library using Hadoop implementation of MapReduce. The algorithm is tested on large data sets to check its feasibility and convergence. The performance of classifier is also assessed in terms of number of support vectors. The challenges encountered towards implementing big data classification in machine learning frameworks are also discussed. The experiments are done on the cloud environment available at University of Technology and Management, India. The results are illustrated for Gaussian RBF and Bayesian kernels. The effect of variability in prediction and generalization of PFRSVM is examined with respect to values of parameter C. It effectively resolves outliers’ effects, imbalance and overlapping class problems, normalizes to unseen data and relaxes dependency between features and labels. The average classification accuracy for PFRSVM is better than other classifiers for both Gaussian RBF and Bayesian kernels. The experimental results on both synthetic and real data sets clearly demonstrate the superiority of the proposed technique.

Keywords: FRSVM, Hadoop, MapReduce, PFRSVM

Procedia PDF Downloads 490

Authors: Melodi Mbuyi Mata, Blessing Olamide Taiwo, Afolabi Ayodele David

Abstract:

The purpose of this research is to create empirical models for assessing the profitability of granite aggregate production in Akure, Ondo state aggregate quarries. In addition, an artificial neural network (ANN) model and multivariate predicting models for granite profitability were developed in the study. A formal survey questionnaire was used to collect data for the study. The data extracted from the case study mine for this study includes granite marketing operations, royalty, production costs, and mine production information. The following methods were used to achieve the goal of this study: descriptive statistics, MATLAB 2017, and SPSS16.0 software in analyzing and modeling the data collected from granite traders in the study areas. The ANN and Multi Variant Regression models' prediction accuracy was compared using a coefficient of determination (R²), Root mean square error (RMSE), and mean square error (MSE). Due to the high prediction error, the model evaluation indices revealed that the ANN model was suitable for predicting generated profit in a typical quarry. More quarries in Nigeria's southwest region and other geopolitical zones should be considered to improve ANN prediction accuracy.

Keywords: national development, granite, profitability assessment, ANN models

Procedia PDF Downloads 101

Authors: L. Edirisinghe, P. Mukherjee, H. Edirisinghe

Abstract:

Virtual Container Yard (VCY) is a modern concept that helps to reduce the empty container repositioning cost of carriers. The concept of VCY is based on container interchange between shipping lines. Although this mechanism has been theoretically accepted by the shipping community as a feasible solution, it has not yet achieved the necessary momentum among container shipping lines (CSL). This paper investigates whether there is any legal influence on this industry myopia about the VCY. It is believed that this is the first publication that focuses on the legal aspects of container exchange between carriers. Not much literature on this subject is available. This study establishes with statistical evidence that there is a phobia prevailing in the shipping industry that exchanging containers with other carriers may lead to various legal implications. The complexity of exchange is two faceted. CSLs assume that offering a container to another carrier (obviously, a competitor in terms of commercial context) or using a container offered by another carrier may lead to undue legal implications. This research reveals that this fear is reflected through four types of perceived components, namely: shipping associate; warehouse associate; network associate; and trading associate. These components carry eighteen subcomponents that comprehensively cover the entire process of a container shipment. The statistical explanation has been supported through regression analysis; INCO terms were used to illustrate the shipping process.

Keywords: virtual container yard, legal, maritime law, inventory

Procedia PDF Downloads 165

Authors: Prashansa Singh, Manjot Kaur, Rohit Bajaj

Abstract:

The proliferation of wireless sensor networks and Internet of Things (IoT) devices in the quickly changing digital landscape has highlighted the urgent need for strong security solutions that can handle these systems’ limited resources. A key solution to this problem is the emergence of ultralightweight security protocols, which provide strong security features while respecting the strict computational, energy, and memory constraints imposed on these kinds of devices. This in-depth analysis explores the field of ultralightweight security protocols, offering a thorough examination of their evolution, salient features, and the particular security issues they resolve. We carefully examine and contrast different protocols, pointing out their advantages and disadvantages as well as the compromises between resource limitations and security resilience. We also study these protocols’ application domains, including the Internet of Things, RFID systems, and wireless sensor networks, to name a few. In addition, the review highlights recent developments and advancements in the field, pointing out new trends and possible avenues for future research. This paper aims to be a useful resource for researchers, practitioners, and developers, guiding the design and implementation of safe, effective, and scalable systems in the Internet of Things era by providing a comprehensive overview of ultralightweight security protocols.

Keywords: wireless sensor network, machine-to-machine, MQTT broker, server, ultralightweight, TCP/IP

Procedia PDF Downloads 82

Authors: Dermot O'Brien, Vasileios Christaras, Georgios Fontaras, Igor Nai Fovino, Ioannis Kounelis

Abstract:

With the rise of the Internet of Things (IoT), 5G, and blockchain (BC) technologies, vehicles are becoming ever increasingly connected and are already transmitting substantial amounts of data to the original equipment manufacturers (OEMs) servers. This data could be used to help detect mileage fraud and enable more accurate vehicle emissions monitoring. This would not only help regulators but could enable applications such as permitting efficient drivers to pay less tax, geofencing for air quality improvement, as well as pollution tolling and trading platforms for transport-related businesses and EU citizens. Other applications could include traffic management and shared mobility systems. BC enables the transmission of data with additional security and removes single points of failure while maintaining data provenance, identity ownership, and the possibility to retain varying levels of privacy depending on the requirements of the applied use case. This research performs simulations of vehicles interacting with European member state authorities and European Commission BC nodes that are running hyperleger fabric and explores whether the technology is currently feasible for transport applications such as the emission monitoring use-case.

Keywords: future transportation systems, technological innovations, policy approaches for transportation future, economic and regulatory trends, blockchain

Procedia PDF Downloads 177

Authors: Dalin Si, Azizan Aziz, Bertrand Lasternas

Abstract:

To encourage building owners to purchase electricity at the wholesale market and reduce building peak demand, this study aims to develop models that predict day-ahead hourly electricity consumption and demand using artificial neural network (ANN) and support vector machine (SVM). All prediction models are built in Python, with tool Scikit-learn and Pybrain. The input data for both consumption and demand prediction are time stamp, outdoor dry bulb temperature, relative humidity, air handling unit (AHU), supply air temperature and solar radiation. Solar radiation, which is unavailable a day-ahead, is predicted at first, and then this estimation is used as an input to predict consumption and demand. Models to predict consumption and demand are trained in both SVM and ANN, and depend on cooling or heating, weekdays or weekends. The results show that ANN is the better option for both consumption and demand prediction. It can achieve 15.50% to 20.03% coefficient of variance of root mean square error (CVRMSE) for consumption prediction and 22.89% to 32.42% CVRMSE for demand prediction, respectively. To conclude, the presented models have potential to help building owners to purchase electricity at the wholesale market, but they are not robust when used in demand response control.

Keywords: building energy prediction, data mining, demand response, electricity market

Procedia PDF Downloads 316

Authors: Nantakrit Yodpijit, Teppakorn Sittiwanchai

Abstract:

The number of older adults is rising rapidly. The world’s population becomes aging. Falls is one of common and major health problems in the elderly. Falls may lead to acute and chronic injuries and deaths. The fall-prone individuals are at greater risk for decreased quality of life, lowered productivity and poverty, social problems, and additional health problems. A number of studies on falls prevention using fall detection system have been conducted. Many available technologies for fall detection system are laboratory-based and can incur substantial costs for falls prevention. The utilization of alternative technologies can potentially reduce costs. This paper presents the new design and development of a wearable-based fall detection system using an Accelerometer and Gyroscope as motion sensors for the detection of body orientation and movement. Algorithms are developed to differentiate between Activities of Daily Living (ADL) and falls by comparing Threshold-based values with Artificial Neural Networks (ANN). Results indicate the possibility of using the new threshold-based method with neural network algorithm to reduce the number of false positive (false alarm) and improve the accuracy of fall detection system.

Keywords: aging, algorithm, artificial neural networks (ANN), fall detection system, motion sensorsthreshold

Procedia PDF Downloads 496

Authors: Bel G Raggad

Abstract:

An e-Learning setting is a distributed computing environment where information resources can be connected to any public network. Public networks are very unsecure which can compromise the reliability of an e-Learning environment. This study is only concerned with the intrusion detection aspect of e-Learning security and how incident responses are planned. The literature reported great advances in intrusion detection system (ids) but neglected to study an important ids weakness: suspected events are detected but an intrusion is not determined because it is not defined in ids databases. We propose an incident response generator (DIRG) that produces incident responses when the working ids system suspects an event that does not correspond to a known intrusion. Data involved in intrusion detection when ample uncertainty is present is often not suitable to formal statistical models including Bayesian. We instead adopt Dempster and Shafer theory to process intrusion data for the unknown event. The DIRG engine transforms data into a belief structure using incident scenarios deduced by the security administrator. Belief values associated with various incident scenarios are then derived and evaluated to choose the most appropriate scenario for which an automatic incident response is generated. This article provides a numerical example demonstrating the working of the DIRG system.

Keywords: decision support system, distributed computing, e-Learning security, incident response, intrusion detection, security risk, statefull inspection

Procedia PDF Downloads 437

Authors: Giselle Maggie Fer Castañeda, Diego Iván González

Abstract:

The objective of this work was to study the autoshaped choice behavior in the Donahoe, Burgos and Palmer (DBP) neural network model and analyze it under the matching law. Autoshaped choice can be viewed as a form of economic behavior defined as the preference between alternatives according to their relative outcomes. The Donahoe, Burgos and Palmer (DBP) model is a connectionist proposal that unifies operant and Pavlovian conditioning. This model has been used for more than three decades as a neurobehavioral explanation of conditioning phenomena, as well as a generator of predictions suitable for experimental testing with non-human animals and humans. The study consisted of different simulations in which, in each one, a ratio of reinforcement was established for two alternatives, and the responses (i.e., activations) in each of them were measured. Choice studies with animals have demonstrated that the data generally conform closely to the generalized matching law equation, which states that the response ratio equals proportionally to the reinforcement ratio; therefore, it was expected to find similar results with the neural networks of the Donahoe, Burgos and Palmer (DBP) model since these networks have simulated and predicted various conditioning phenomena. The results were analyzed by the generalized matching law equation, and it was observed that under some contingencies, the data from the networks adjusted approximately to what was established by the equation. Implications and limitations are discussed.

Keywords: matching law, neural networks, computational models, behavioral sciences

Procedia PDF Downloads 74

Authors: Zhiwen Pan, Jesus Pacheco, Salim Hariri, Yiqiang Chen, Bozhi Liu

Abstract:

The Internet of Things (IoT) will lead to the development of advanced Smart Home services that are pervasive, cost-effective, and can be accessed by home occupants from anywhere and at any time. However, advanced smart home applications will introduce grand security challenges due to the increase in the attack surface. Current approaches do not handle cybersecurity from a holistic point of view; hence, a systematic cybersecurity mechanism needs to be adopted when designing smart home applications. In this paper, we present a generic intrusion detection methodology to detect and mitigate the anomaly behaviors happened in Smart Home Systems (SHS). By utilizing our Smart Home Context Data Structure, the heterogeneous information and services acquired from SHS are mapped in context attributes which can describe the context of smart home operation precisely and accurately. Runtime models for describing usage patterns of home assets are developed based on characterization functions. A threat-aware action management methodology, used to efficiently mitigate anomaly behaviors, is proposed at the end. Our preliminary experimental results show that our methodology can be used to detect and mitigate known and unknown threats, as well as to protect SHS premises and services.

Keywords: Internet of Things, network security, context awareness, intrusion detection

Procedia PDF Downloads 191

Authors: Ahmed Krobba, Meriem Touzene, Mohamed Debeyche

Abstract:

The wireless communications system (5G) will provide more diverse applications and higher quality services for users compared to the long-term evolution 4G (LTE). 5G uses a higher carrier frequency, which suffers from information loss in 5G coverage. Most 5G users often cannot obtain high-quality communications due to transmission channel noise and channel complexity. Physical Uplink Control Channel (PUCCH-NR: Physical Uplink Control Channel New Radio) plays a crucial role in 5G NR telecommunication technology, which is mainly used to transmit link control information uplink (UCI: Uplink Control Information. This study based of evaluating the performance of channel physical uplink control PUCCH-NR under low Signal-to-Noise Ratios with various antenna numbers reception. We propose the artificial intelligence approach based on deep neural networks (Deep Learning) to estimate the PUCCH-NR channel in comparison with this approach with different conventional methods such as least-square (LS) and minimum-mean-square-error (MMSE). To evaluate the channel performance we use the block error rate (BLER) as an evaluation criterion of the communication system. The results show that the deep neural networks method gives best performance compared with MMSE and LS

Keywords: 5G network, uplink (Uplink), PUCCH channel, NR-PUCCH channel, deep learning

Procedia PDF Downloads 84

Authors: Devi Eka Septiyani Arifin, Jrjeng Ruan

Abstract:

As an approach to manipulate the performance of polymer thin film, epitaxy crystallization within polymer blends of poly(vinylidene fluoride) (PVDF) and its copolymer poly(vinylidene fluoride-trifluoroethylene) P(VDF-TrFE) was studied in this research, which involves the competition between phase separation and crystal growth of constitutive semicrystalline polymers. The unique piezoelectric feature of poly(vinylidene fluoride) crystalline phase is derived from the packing of molecular chains in all-trans conformation, which spatially arranges all the substituted fluorene atoms on one side of the molecular chain and hydrogen atoms on the other side. Therefore, the net dipole moment is induced across the lateral packing of molecular chains. Nevertheless, due to the mutual repulsion among fluorene atoms, this all-trans molecular conformation is not stable, and ready to change above curie temperature, where thermal energy is sufficient to cause segmental rotation. This research attempts to explore whether the epitaxial interactions between piezoelectric crystals and crystal lattice of hexamethylbenzene (HMB) crystalline platelet is able to stabilize this metastable all-trans molecular conformation or not. As an aromatic crystalline compound, the melt of HMB was surprisingly found able to dissolve the poly(vinylidene fluoride), resulting in homogeneous eutectic solution. Thus, after quenching this binary eutectic mixture to room temperature, subsequent heating or annealing processes were designed to explore the involve phase separation and crystallization behavior. The phase transition behaviors were observed in-situ by X-ray diffraction and differential scanning calorimetry (DSC). The molecular packing was observed via transmission electron microscope (TEM) and the principles of electron diffraction were brought to study the internal crystal structure epitaxially developed within thin films. Obtained results clearly indicated the occurrence of heteroepitaxy of PVDF/PVDF-TrFE on HMB crystalline platelet. Both the concentration of poly(vinylidene fluoride) and the mixing ratios of these two constitutive polymers have been adopted as the influential factors for studying the competition between the epitaxial crystallization of PVDF and P(VDF-TrFE) on HMB crystalline. Furthermore, the involved epitaxial relationship is to be deciphered and studied as a potential factor capable of guiding the wide spread of piezoelectric crystalline form.

Keywords: epitaxy, crystallization, crystalline platelet, thin film and mixing ratio

Procedia PDF Downloads 223

Authors: Elsayed A. Issawy, Anwar H. Radwan

Abstract:

Recent crustal deformations studies in Egypt are applied on the most active areas with relation to seismic activity. Temporal gravity variations in parallel with the geodetic technique (GPS) were used to monitor recent crustal movements in Egypt since 1997. The non-tidal gravity changes were constrained by the vertical component of surface movements derived from the GPS observations. The gravity changes were used to understand the surface tectonics and geodynamic modelling of the Greater Cairo region after the occurrence of an earthquake of 1992. It was found that there is a certain relation showed by increasing of gravity values before the main seismic activity. As example, relative considerable increase of gravity values was noticed for the network between the epochs of 2000 and 2004. Otherwise, the temporal gravity variations were reported a considerable decrease in gravity values between the two campaigns of 2004 and 2007 for the same stations. This behaviour could explain by compressive deformation and strain build-up stage before the South western Cairo earthquake (July 31, 2005 with magnitude of 4.3) and the stress release stage occurred after the main event. The geodetic measurements showed that, the estimated horizontal velocities for almost of points are 5.5 mm/year in approximately NW direction.

Keywords: temporal gravity variations, geodynamics, greater Cairo, recent crustal movements, earthquakes

Procedia PDF Downloads 368