Search results for: data stewardship

Authors: Nicholas Milburn, Yu Liang, Dalei Wu

Abstract:

Athlete performance scoring within the climbing do-main presents interesting challenges as the sport does not have an objective way to assign skill. Assessing skill levels within any sport is valuable as it can be used to mark progress while training, and it can help an athlete choose appropriate climbs to attempt. Machine learning-based methods are popular for complex problems like this. The dataset available was composed of dynamic force data recorded during climbing; however, this dataset came with challenges such as data scarcity, imbalance, and it was temporally heterogeneous. Investigated solutions to these challenges include data augmentation, temporal normalization, conversion of time series to the spectral domain, and cross validation strategies. The investigated solutions to the classification problem included light weight machine classifiers KNN and SVM as well as the deep learning with CNN. The best performing model had an 80% accuracy. In conclusion, there seems to be enough information within climbing force data to accurately categorize climbers by skill.

Keywords: classification, climbing, data imbalance, data scarcity, machine learning, time sequence

Procedia PDF Downloads 122

Authors: M. A. I Perera, C. R. Wijesinghe, A. R. Weerasinghe

Abstract:

his study was conducted to review and identify the unsupervised techniques that can be employed to analyze gene expression data in order to identify better subtypes of tumors. Identifying subtypes of cancer help in improving the efficacy and reducing the toxicity of the treatments by identifying clues to find target therapeutics. Process of gene expression data analysis described under three steps as preprocessing, clustering, and cluster validation. Feature selection is important since the genomic data are high dimensional with a large number of features compared to samples. Hierarchical clustering and K Means are often used in the analysis of gene expression data. There are several cluster validation techniques used in validating the clusters. Heatmaps are an effective external validation method that allows comparing the identified classes with clinical variables and visual analysis of the classes.

Keywords: cancer subtypes, gene expression data analysis, clustering, cluster validation

Procedia PDF Downloads 121

Authors: Matthew Montebello

Abstract:

Student engagement within a virtual learning environment generates masses of data points that can significantly contribute to the learning analytics that lead to decision support. Ideally, similar data is collected during student interaction with a physical learning space, and as a consequence, data is present at a large scale, even in relatively small classes. In this paper, we report of such an occurrence during classes held in a HiFlex modality as we investigate the advantages of adopting such a methodology. We plan to take full advantage of the learner-generated data in an attempt to further enhance the effectiveness of the adopted learning environment. This could shed crucial light on operating modalities that higher education institutions around the world will switch to in a post-COVID era.

Keywords: HiFlex, big data in higher education, learning analytics, virtual learning environment

Procedia PDF Downloads 174

Authors: Shahzad Hassan, Kamran Saeed

Abstract:

People are always in search of Wi-Fi hotspots because Internet is a major demand nowadays. But like all other technologies, there is still room for improvement in the Wi-Fi technology with regards to the speed and quality of connectivity. In order to address these aspects, Harald Haas, a professor at the University of Edinburgh, proposed what we know as the Li-Fi (Light Fidelity). Li-Fi is a new technology in the field of wireless communication to provide connectivity within a network environment. It is a two-way mode of wireless communication using light. Basically, the data is transmitted through Light Emitting Diodes which can vary the intensity of light very fast, even faster than the blink of an eye. From the research and experiments conducted so far, it can be said that Li-Fi can increase the speed and reliability of the transfer of data. This paper pays particular attention on the assessment of the performance of this technology. In other words, it is a 5G technology which uses LED as the medium of data transfer. For coverage within the buildings, Wi-Fi is good but Li-Fi can be considered favorable in situations where large amounts of data are to be transferred in areas with electromagnetic interferences. It brings a lot of data related qualities such as efficiency, security as well as large throughputs to the table of wireless communication. All in all, it can be said that Li-Fi is going to be a future phenomenon where the presence of light will mean access to the Internet as well as speedy data transfer.

Keywords: communication, LED, Li-Fi, Wi-Fi

Procedia PDF Downloads 319

Authors: Renata Kurpiewska-Korbut

Abstract:

Making an assumption that the use and sharing of data generated in humanitarian action constitute a core function of humanitarian organizations, the paper analyzes the position of the largest Polish humanitarian non-governmental organizations in the humanitarian data ecosystem in Ukraine and their approach to non-personal and personal data management since February of 2022. Both expert interviews and document analysis of non-profit organizations providing a direct response in the Ukrainian crisis context, i.e., the Polish Humanitarian Action, Caritas, Polish Medical Mission, Polish Red Cross, and the Polish Center for International Aid and the applicability of theoretical perspective of contingency theory – with its central point that the context or specific set of conditions determining the way of behavior and the choice of methods of action – help to examine the significance of data complexity and adaptive approach to data management by relief organizations in the humanitarian supply chain network. The purpose of this study is to determine how the existence of well-established and accurate internal procedures and good practices of using and sharing data (including safeguards for sensitive data) by the surveyed organizations with comparable human and technological capabilities are implemented and adjusted to Ukrainian humanitarian settings and data infrastructure. The study also poses a fundamental question of whether this crisis experience will have a determining effect on their future performance. The obtained finding indicate that Polish humanitarian organizations in Ukraine, which have their own unique code of conduct and effective managerial data practices determined by contingencies, have limited influence on improving the situational awareness of other assistance providers in the data ecosystem despite their attempts to undertake interagency work in the area of data sharing.

Keywords: humanitarian data ecosystem, humanitarian data management, polish NGOs, Ukraine

Procedia PDF Downloads 69

Authors: Amarasinghe Shashikala, Pramudika Hansini, Fernando Tajan, Rathnayake Piyumi

Abstract:

This study investigates the prominence of the social entrepreneurship (SE) model over the sustainable entrepreneurship model among Sri Lankan exporters in the small and medium enterprise (SME) sector. The primary objective of this study is to explore how the unique socio-economic contextual nuances of the country influence this behavior. The study employs a multiple-case study approach, collecting data from thirteen SEs in the SME sector. The findings reveal a significant alignment between SE and the lifestyle of the people in Sri Lanka, attributed largely to its deep-rooted religious setting and cultural norms. A crucial factor driving the prominence of SE is the predominantly labor-intensive nature of production processes within the exporters of the SME sector. These processes inherently lend themselves to SE, providing employment opportunities and fostering community engagement. Further, SE initiatives substantially resonate with community-centric practices, making them more appealing and accessible to the local populace. In contrast, the findings highlight a dilemma between cost-effectiveness and sustainable entrepreneurship. Transitioning to sustainable export products and production processes is demanded by foreign buyers and acknowledged as essential for environmental stewardship, which often requires capital-intensive makeovers. This investment inevitably raises the overall cost of the export product, making it less competitive in the global market. Interestingly, the study notes a disparity between international demand for sustainable products and the willingness of buyers to pay a premium for them. Despite the growing global preference for eco-friendly options, the findings suggest that the additional costs associated with sustainable entrepreneurship are not adequately reflected in the purchasing behavior of international buyers. The abundance of natural resources coupled with a minimal occurrence of natural catastrophes renders exporters less environmentally sensitive. The absence of robust policy support for environmental preservation exacerbates this inclination. Consequently, exporters exhibit a diminished motivation to incorporate environmental sustainability into their business decisions. Instead, attention is redirected towards factors such as the local population's minimum standards of living, prevalent social issues, governmental corruption and inefficiency, and rural poverty. These elements impel exporters to prioritize social well-being when making business decisions. Notably, the emphasis on social impact, rather than environmental impact, appears to be a generational trend, perpetuating a focus on societal aspects in the realm of business. In conclusion, the manifestation of entrepreneurial behavior within developing nations is notably contingent upon contextual nuances. This investigation contributes to a deeper understanding of the dynamics shaping the prevalence of SE over sustainable entrepreneurship among Sri Lankan exporters in the SME sector. The insights generated have implications for policymakers, industry stakeholders, and academics seeking to navigate the delicate balance between socio-cultural values, economic feasibility, and environmental sustainability in the pursuit of responsible business practices within the export sector.

Keywords: small and medium enterprises, social entrepreneurship, Sri Lanka, sustainable entrepreneurship

Procedia PDF Downloads 38

Authors: Daniel C. Bonzo

Abstract:

Practical considerations lead to the use of unit of analysis within subjects, e.g., bleeding episodes or treatment-related adverse events, in rare disease settings. This is coupled with data augmentation techniques such as extrapolation to enlarge the subject base. In general, one can think about extrapolation of data as extending information and conclusions from one estimand to another estimand. This approach induces hierarchichal clustered data with varying cluster sizes. Extrapolation of clinical trial data is being accepted increasingly by regulatory agencies as a means of generating data in diverse situations during drug development process. Under certain circumstances, data can be extrapolated to a different population, a different but related indication, and different but similar product. We consider here the problem of estimation (point and interval) using a mixed-models approach under an extrapolation. It is proposed that estimators (point and interval) be constructed using weighting schemes for the clusters, e.g., equally weighted and with weights proportional to cluster size. Simulated data generated under varying scenarios are then used to evaluate the performance of this approach. In conclusion, the evaluation result showed that the approach is a useful means for improving statistical inference in rare disease settings and thus aids not only signal detection but risk-benefit evaluation as well.

Keywords: clustered data, estimand, extrapolation, mixed model

Procedia PDF Downloads 111

Authors: Celal Dudak, Aslı Utku, Burak Yağlioğlu

Abstract:

Uninterrupted and continuous satellite communication through the whole orbit time is becoming more indispensable every day. Data relay systems are developed and built for various high/low data rate information exchanges like TDRSS of USA and EDRSS of Europe. In these missions, a couple of task-dedicated communication satellites exist. In this regard, for Turkey a data relay system is attempted to be defined exchanging low data rate information (i.e. TTC) for Earth-observing LEO satellites appointing commercial GEO communication satellites all over the world. First, justification of this attempt is given, demonstrating duration enhancements in the link. Discussion of preference of RF communication is, also, given instead of laser communication. Then, preferred communication GEOs – including TURKSAT4A already belonging to Turkey- are given, together with the coverage enhancements through STK simulations and the corresponding link budget. Also, a block diagram of the communication system is given on the LEO satellite.

Keywords: communication, GEO satellite, data relay system, coverage

Procedia PDF Downloads 415

Authors: Tanawat Hongthai, Dusit Thanapatay

Abstract:

This paper presents the development of encrypted near field communication (NFC) data exchange format transmission in an NFC passive tag for the feasibility of implementing a genuine product authentication. We propose a research encryption and checking the genuine product into four major categories; concept, infrastructure, development and applications. This result shows the passive NFC-forum Type 2 tag can be configured to be compatible with the NFC data exchange format (NDEF), which can be automatically partially data updated when there is NFC field.

Keywords: near field communication, NFC data exchange format, checking the genuine product, encrypted NFC

Procedia PDF Downloads 256

Authors: Yung Gi Wu

Abstract:

With the growing of computer and network, digital data can be spread to anywhere in the world quickly. In addition, digital data can also be copied or tampered easily so that the security issue becomes an important topic in the protection of digital data. Digital watermark is a method to protect the ownership of digital data. Embedding the watermark will influence the quality certainly. In this paper, Vector Quantization (VQ) is used to embed the watermark into the image to fulfill the goal of data hiding. This kind of watermarking is invisible which means that the users will not conscious the existing of embedded watermark even though the embedded image has tiny difference compared to the original image. Meanwhile, VQ needs a lot of computation burden so that we adopt a fast VQ encoding scheme by partial distortion searching (PDS) and mean approximation scheme to speed up the data hiding process. The watermarks we hide to the image could be gray, bi-level and color images. Texts are also can be regarded as watermark to embed. In order to test the robustness of the system, we adopt Photoshop to fulfill sharpen, cropping and altering to check if the extracted watermark is still recognizable. Experimental results demonstrate that the proposed system can resist the above three kinds of tampering in general cases.

Keywords: data hiding, vector quantization, watermark, color image

Procedia PDF Downloads 338

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

Abstract:

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

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

Procedia PDF Downloads 164

Authors: Pavel H. Llamocca, Victoria Lopez

Abstract:

The main objective of our work is the comparative analysis of environmental data from Open Data bases, belonging to different governments. This means that you have to integrate data from various different sources. Nowadays, many governments have the intention of publishing thousands of data sets for people and organizations to use them. In this way, the quantity of applications based on Open Data is increasing. However each government has its own procedures to publish its data, and it causes a variety of formats of data sets because there are no international standards to specify the formats of the data sets from Open Data bases. Due to this variety of formats, we must build a data integration process that is able to put together all kind of formats. There are some software tools developed in order to give support to the integration process, e.g. Data Tamer, Data Wrangler. The problem with these tools is that they need data scientist interaction to take part in the integration process as a final step. In our case we don’t want to depend on a data scientist, because environmental data are usually similar and these processes can be automated by programming. The main idea of our tool is to build Hadoop procedures adapted to data sources per each government in order to achieve an automated integration. Our work focus in environment data like temperature, energy consumption, air quality, solar radiation, speeds of wind, etc. Since 2 years, the government of Madrid is publishing its Open Data bases relative to environment indicators in real time. In the same way, other governments have published Open Data sets relative to the environment (like Andalucia or Bilbao). But all of those data sets have different formats and our solution is able to integrate all of them, furthermore it allows the user to make and visualize some analysis over the real-time data. Once the integration task is done, all the data from any government has the same format and the analysis process can be initiated in a computational better way. So the tool presented in this work has two goals: 1. Integration process; and 2. Graphic and analytic interface. As a first approach, the integration process was developed using Java and Oracle and the graphic and analytic interface with Java (jsp). However, in order to open our software tool, as second approach, we also developed an implementation with R language as mature open source technology. R is a really powerful open source programming language that allows us to process and analyze a huge amount of data with high performance. There are also some R libraries for the building of a graphic interface like shiny. A performance comparison between both implementations was made and no significant differences were found. In addition, our work provides with an Official Real-Time Integrated Data Set about Environment Data in Spain to any developer in order that they can build their own applications.

Keywords: open data, R language, data integration, environmental data

Procedia PDF Downloads 288

Authors: Zahid Ullah, Atlas Khan

Abstract:

The rapid growth of data in various domains has created a pressing need for effective methods to transform this data into meaningful knowledge. In this era of big data, mathematical and statistical innovations play a crucial role in unlocking insights and facilitating informed decision-making in data analytics. This abstract aims to explore the transformative potential of these innovations and their impact on converting raw data into actionable knowledge. Drawing upon a comprehensive review of existing literature, this research investigates the cutting-edge mathematical and statistical techniques that enable the conversion of data into knowledge. By evaluating their underlying principles, strengths, and limitations, we aim to identify the most promising innovations in data analytics. To demonstrate the practical applications of these innovations, real-world datasets will be utilized through case studies or simulations. This empirical approach will showcase how mathematical and statistical innovations can extract patterns, trends, and insights from complex data, enabling evidence-based decision-making across diverse domains. Furthermore, a comparative analysis will be conducted to assess the performance, scalability, interpretability, and adaptability of different innovations. By benchmarking against established techniques, we aim to validate the effectiveness and superiority of the proposed mathematical and statistical innovations in data analytics. Ethical considerations surrounding data analytics, such as privacy, security, bias, and fairness, will be addressed throughout the research. Guidelines and best practices will be developed to ensure the responsible and ethical use of mathematical and statistical innovations in data analytics. The expected contributions of this research include advancements in mathematical and statistical sciences, improved data analysis techniques, enhanced decision-making processes, and practical implications for industries and policymakers. The outcomes will guide the adoption and implementation of mathematical and statistical innovations, empowering stakeholders to transform data into actionable knowledge and drive meaningful outcomes.

Keywords: data analytics, mathematical innovations, knowledge extraction, decision-making

Procedia PDF Downloads 49

Authors: Lu Si, Jie Yu, Shasha Li, Jun Ma, Lei Luo, Qingbo Wu, Yongqi Ma, Zhengji Liu

Abstract:

Instance selection (IS) technique is used to reduce the data size to improve the performance of data mining methods. Recently, to process very large data set, several proposed methods divide the training set into some disjoint subsets and apply IS algorithms independently to each subset. In this paper, we analyze the limitation of these methods and give our viewpoint about how to divide and conquer in IS procedure. Then, based on fast condensed nearest neighbor (FCNN) rule, we propose a large data sets instance selection method with MapReduce framework. Besides ensuring the prediction accuracy and reduction rate, it has two desirable properties: First, it reduces the work load in the aggregation node; Second and most important, it produces the same result with the sequential version, which other parallel methods cannot achieve. We evaluate the performance of FCNN-MR on one small data set and two large data sets. The experimental results show that it is effective and practical.

Keywords: instance selection, data reduction, MapReduce, kNN

Procedia PDF Downloads 233

Authors: Trevor Toy, Josef Langerman

Abstract:

Around a quarter of the world’s data is generated by financial with an estimated 708.5 billion global non-cash transactions reached between 2018 and. And with Open Banking still a rapidly developing concept within the financial industry, there is an opportunity to create a secure mechanism for connecting its stakeholders to openly, legitimately and consensually share the data required to enable it. Integration and data sharing of anonymised transactional data are still operated in silos and centralised between the large corporate entities in the ecosystem that have the resources to do so. Smaller fintechs generating data and businesses looking to consume data are largely excluded from the process. Therefore there is a growing demand for accessible transactional data for analytical purposes and also to support the rapid global adoption of Open Banking. The following research has provided a solution framework that aims to provide a secure decentralised marketplace for 1.) data providers to list their transactional data, 2.) data consumers to find and access that data, and 3.) data subjects (the individuals making the transactions that generate the data) to manage and sell the data that relates to themselves. The platform also provides an integrated system for downstream transactional-related data from merchants, enriching the data product available to build a comprehensive view of a data subject’s spending habits. A robust and sustainable data market can be developed by providing a more accessible mechanism for data producers to monetise their data investments and encouraging data subjects to share their data through the same financial incentives. At the centre of the platform is the market mechanism that connects the data providers and their data subjects to the data consumers. This core component of the platform is developed on a decentralised blockchain contract with a market layer that manages transaction, user, pricing, payment, tagging, contract, control, and lineage features that pertain to the user interactions on the platform. One of the platform’s key features is enabling the participation and management of personal data by the individuals from whom the data is being generated. This framework developed a proof-of-concept on the Etheruem blockchain base where an individual can securely manage access to their own personal data and that individual’s identifiable relationship to the card-based transaction data provided by financial institutions. This gives data consumers access to a complete view of transactional spending behaviour in correlation to key demographic information. This platform solution can ultimately support the growth, prosperity, and development of economies, businesses, communities, and individuals by providing accessible and relevant transactional data for big data analytics and open banking.

Keywords: big data markets, open banking, blockchain, personal data management

Procedia PDF Downloads 49

Authors: T. Mcgraw, F. N. Morin, N. Desai

Abstract:

Background: Antimicrobial resistance is a critical issue in global health care and a significant contributor to increased patient morbidity and mortality. Suspected urinary tract infection (UTI) is a key area of inappropriate antibiotic prescription in pediatrics. Management patterns of infectious diseases have been shown to vary by provider type within a single setting. The aim of this study was to assess compliance with national UTI management guidelines by nurse practitioners in a pediatric emergency department (ED). Methods: This was a post-hoc analysis of a retrospective cohort study to review and evaluate visits to a tertiary care freestanding pediatric emergency department. Patients were included if they were 60 days to 36 months old and discharged with a diagnosis of UTI or ‘rule-out UTI’ between July 2015 and July 2020. Primary outcome measure was proportion of visits seen by Nurse Practitioners (NP) which were associated with national guideline compliance in the diagnosis and treatment of suspected UTI. We performed descriptive statistics and comparative analyses to determine differences in practice patterns between NPs, and physicians. Results: A total of 636 charts were reviewed, of which 402 patients met inclusion criteria. 17 patients were treated by NPs, 385 were treated by either Pediatric Emergency Medicine physicians (PEM) or non-PEM physicians. Overall, the proportion of infants receiving guideline-compliant care was 25.9% (21.8-30.4%). Of those who were prescribed antibiotics, 79.6% (74.7-83.8%) received first line guideline recommended therapy and 58.9% (53.8-63.8%) received fully compliant therapy with respect to age, dose, duration, and frequency. In patients treated by NPs, 16/17 (94%(95% CI:73.0-99.0)) required antibiotics, 15/16 (93%(95% CI: 71.7-98.9)) were treated with first line agent (cephalexin), 8/16 (50%(95% CI:28-72)) were guideline compliant of dose and duration. 5/8 (63%(95% CI:30.6-86.3)) were noncompliant for dose being too high. There was no difference in receiving guideline compliant empiric antibiotic therapy between physicians and nurse practitioners (OR: 0.837 CI: 0.302-2.69). Conclusion: In this post-hoc analysis, guideline noncompliance by nurse practitioners is common in children tested and treated for UTIs in a pediatric emergency department. Care by a Nurse Practitioner was not associated with greater rate of noncompliance than care by a Pediatric Emergency Medicine physician. Future appropriately powered studies may focus on confirming these results.

Keywords: antibiotic stewardship, infectious disease, nurse practitioner, urinary tract infection

Procedia PDF Downloads 81

Authors: Dmitriy Kuptsov

Abstract:

Tree data structures, such as binary or in general k-ary trees, are essential in computer science. The applications of these data structures can range from data search and retrieval to sorting and ranking algorithms. Naive implementations of these data structures can consume prohibitively large volumes of random access memory limiting their applicability in certain solutions. Thus, in these cases, more advanced representation of these data structures is essential. In this paper we present the design of the compact version of ternary tree data structure and demonstrate the results for the experimental evaluation using static dictionary problem. We compare these results with the results for binary and regular ternary trees. The conducted evaluation study shows that our design, in the best case, consumes up to 12 times less memory (for the dictionary used in our experimental evaluation) than a regular ternary tree and in certain configuration shows performance comparable to regular ternary trees. We have evaluated the performance of the algorithms using both 32 and 64 bit operating systems.

Keywords: algorithms, data structures, succinct ternary tree, per- formance evaluation

Procedia PDF Downloads 140

Authors: Ahmed I. Alutabi, Naghmeh Dezhabad, Sudhakar Ganti

Abstract:

Data centers have been growing in size and dema nd continuously in the last two decades. Planning for the deployment of resources has been shallow and always resorted to over-provisioning. Data center operators try to maximize the availability of their services by allocating multiple of the needed resources. One resource that has been wasted, with little thought, has been energy. In recent years, programmable resource allocation has paved the way to allow for more efficient and robust data centers. In this work, we examine the predictability of resource usage in a data center environment. We use a number of models that cover a wide spectrum of machine learning categories. Then we establish a framework to guarantee the client service level agreement (SLA). Our results show that using prediction can cut energy loss by up to 55%.

Keywords: machine learning, artificial intelligence, prediction, data center, resource allocation, green computing

Procedia PDF Downloads 80

Authors: Prabhakar C. Dhavale, Meenakshi M. Pawar

Abstract:

In this paper, everyday tons of data is embedded on digital media or distributed over the internet. The data is so distributed that it can easily be replicated without error, putting the rights of their owners at risk. Even when encrypted for distribution, data can easily be decrypted and copied. One way to discourage illegal duplication is to insert information known as watermark, into potentially valuable data in such a way that it is impossible to separate the watermark from the data. These challenges motivated researchers to carry out intense research in the field of watermarking. A watermark is a form, image or text that is impressed onto paper, which provides evidence of its authenticity. Digital watermarking is an extension of the same concept. There are two types of watermarks visible watermark and invisible watermark. In this project, we have concentrated on implementing watermark in image. The main consideration for any watermarking scheme is its robustness to various attacks

Keywords: watermarking, digital, DCT-DWT, security

Procedia PDF Downloads 393

Authors: Neerav Kumar, Naumaan Nayyar, Sharath Kashyap

Abstract:

Most companies see an increase in the adoption of machine learning (ML) applications across internal and external-facing use cases. ML applications vend output either in batch or real-time patterns. A complete batch ML pipeline architecture comprises data sourcing, feature engineering, model training, model deployment, model output vending into a data store for downstream application. Due to unclear role expectations, we have observed that scientists specializing in building and optimizing models are investing significant efforts into building the other components of the architecture, which we do not believe is the best use of scientists’ bandwidth. We propose a system architecture created using AWS services that bring industry best practices to managing the workflow and simplifies the process of model deployment and end-to-end data integration for an ML application. This narrows down the scope of scientists’ work to model building and refinement while specialized data engineers take over the deployment, pipeline orchestration, data quality, data permission system, etc. The pipeline infrastructure is built and deployed as code (using terraform, cdk, cloudformation, etc.) which makes it easy to replicate and/or extend the architecture to other models that are used in an organization.

Keywords: data pipeline, machine learning, AWS, architecture, batch machine learning

Procedia PDF Downloads 35

Authors: André Smith, Amr Abdel-Dayem

Abstract:

The stereo matching problem, while having been present for several decades, continues to be an active area of research. The goal of this research is to find correspondences between elements found in a set of stereoscopic images. With these pairings, it is possible to infer the distance of objects within a scene, relative to the observer. Advancements in this field have led to experimentations with various techniques, from graph-cut energy minimization to artificial neural networks. At the basis of these techniques is a cost function, which is used to evaluate the likelihood of a particular match between points in each image. While at its core, the cost is based on comparing the image pixel data; there is a general lack of consistency as to what image data representation to use. This paper presents an experimental analysis to compare the effectiveness of more common image data representations. The goal is to determine the effectiveness of these data representations to reduce the cost for the correct correspondence relative to other possible matches.

Keywords: colour data, local stereo matching, stereo correspondence, disparity map

Procedia PDF Downloads 349

Authors: Waad Al Saleemi, Badriya Al Adawi, Zaaima Al Jabri, Sahim Al Ghafri, Jalila Al Hadhramia

Abstract:

Objectives: Using standard lab methods, a positive blood culture requires a minimum of two days (two occasions of overnight incubation) to obtain a final identification (ID) and antimicrobial susceptibility results (AST) report. In this study, we aimed to evaluate the accuracy and precision of identification and antimicrobial susceptibility testing of an alternative method (direct method) that will reduce the turnaround time by 24 hours. This method involves the direct inoculation of positive blood culture broths into the Phoenix system using serum separation tubes (SST). Method: This prospective study included monomicrobial-positive blood cultures obtained from January 2022 to May 2023 in SQUH. Blood cultures containing a mixture of organisms, fungi, or anaerobic organisms were excluded from this study. The result of the new “direct method” under study was compared with the current “standard method” used in the lab. The accuracy and precision were evaluated for the ID and AST using Clinical and Laboratory Standards Institute (CLSI) recommendations. The categorical agreement, essential agreement, and the rates of very major errors (VME), major errors (ME), and minor errors (MIE) for both gram-negative and gram-positive bacteria were calculated. Passing criteria were set according to CLSI. Result: The results of ID and AST were available for a total of 158 isolates. Of 77 isolates of gram-negative bacteria, 71 (92%) were correctly identified at the species level. Of 70 isolates of gram-positive bacteria, 47(67%) isolates were correctly identified. For gram-negative bacteria, the essential agreement of the direct method was ≥92% when compared to the standard method, while the categorical agreement was ≥91% for all tested antibiotics. The precision of ID and AST were noted to be 100% for all tested isolates. For gram-positive bacteria, the essential agreement was >93%, while the categorical agreement was >92% for all tested antibiotics except moxifloxacin. Many antibiotics were noted to have an unacceptable higher rate of very major errors including penicillin, cotrimoxazole, clindamycin, ciprofloxacin, and moxifloxacin. However, no error was observed in the results of vancomycin, linezolid, and daptomycin. Conclusion: The direct method of ID and AST for positive blood cultures using SST is reliable for gram negative bacteria. It will significantly decrease the turnaround time and will facilitate antimicrobial stewardship.

Keywords: bloodstream infection, oman, direct ast, blood culture, rapid identification, antimicrobial susceptibility, phoenix, direct inoculation

Procedia PDF Downloads 27

Authors: Karima Qayumi, Alex Norta

Abstract:

The rapid generation of high volume and a broad variety of data from the application of new technologies pose challenges for the generation of business-intelligence. Most organizations and business owners need to extract data from multiple sources and apply analytical methods for the purposes of developing their business. Therefore, the recently decentralized data management environment is relying on a distributed computing paradigm. While data are stored in highly distributed systems, the implementation of distributed data-mining techniques is a challenge. The aim of this technique is to gather knowledge from every domain and all the datasets stemming from distributed resources. As agent technologies offer significant contributions for managing the complexity of distributed systems, we consider this for next-generation data-mining processes. To demonstrate agent-based business intelligence operations, we use agent-oriented modeling techniques to develop a new artifact for mining massive datasets.

Keywords: agent-oriented modeling (AOM), business intelligence model (BIM), distributed data mining (DDM), multi-agent system (MAS)

Procedia PDF Downloads 400

Authors: Qing K. Zhu

Abstract:

Very Large Scale Integration (VLSI) design becomes very complex due to the continuous integration of millions of gates in one chip based on Moore’s law. Designers have encountered numerous report files during design iterations using timing and noise analysis tools. This paper presented our work using data mining techniques combined with HTML tables to extract and represent critical timing/noise data. When we apply this data-mining tool in real applications, the running speed is important. The software employs table look-up techniques in the programming for the reasonable running speed based on performance testing results. We added several advanced features for the application in one industry chip design.

Keywords: VLSI design, data mining, big data, HTML forms, web, VLSI, EDA, timing, noise

Procedia PDF Downloads 229

Authors: Anuruddha Jagoda, Samiddhi Samarakoon, Anil Jasinghe

Abstract:

In its simplest form, data quality can be defined as 'fitness for use' and it is a concept with multi-dimensions. Emergency Departments(ED) require information to treat patients and on the other hand it is the primary source of information regarding accidents, injuries, emergencies etc. Also, it is the starting point of various patient registries, databases and surveillance systems. This interventional study was carried out to improve data quality at the ED of the National Hospital of Sri Lanka (NHSL) by introducing an e health solution to improve data quality. The NHSL is the premier trauma care centre in Sri Lanka. The study consisted of three components. A research study was conducted to assess the quality of data in relation to selected five dimensions of data quality namely accuracy, completeness, timeliness, legibility and reliability. The intervention was to develop and deploy an electronic emergency department information system (eEDIS). Post assessment of the intervention confirmed that all five dimensions of data quality had improved. The most significant improvements are noticed in accuracy and timeliness dimensions.

Keywords: electronic health records, electronic emergency department information system, emergency department, data quality

Procedia PDF Downloads 249

Authors: Basma Khelfa, Antoine Tordeux, Ibrahima Ba

Abstract:

We present a descriptive analysis data of lane-changing events in multi-lane roads. The data are provided from The Highway Drone Dataset (HighD), which are microscopic trajectories in highway. This paper describes and analyses the role of the different parameters and their significance. Thanks to HighD data, we aim to find the most frequent reasons that motivate drivers to change lanes. We used the programming language R for the processing of these data. We analyze the involvement and relationship of different variables of each parameter of the ego vehicle and the four vehicles surrounding it, i.e., distance, speed difference, time gap, and acceleration. This was studied according to the class of the vehicle (car or truck), and according to the maneuver it undertook (overtaking or falling back).

Keywords: autonomous driving, physical traffic model, prediction model, statistical learning process

Procedia PDF Downloads 232

Authors: Shoukat Ali, Abdul Qadir Jandga, Amjad Hussain

Abstract:

Objective: The main purpose of this study is to compare the Percent Depth dose (PDD) and In-plane and cross-plane profiles of Varian Golden beam data to the measured data of 6 and 18 MV photons for the commissioning of Eclipse treatment planning system. Introduction: Commissioning of treatment planning system requires an extensive acquisition of beam data for the clinical use of linear accelerators. Accurate dose delivery require to enter the PDDs, Profiles and dose rate tables for open and wedges fields into treatment planning system, enabling to calculate the MUs and dose distribution. Varian offers a generic set of beam data as a reference data, however not recommend for clinical use. In this study, we compared the generic beam data with the measured beam data to evaluate the reliability of generic beam data to be used for the clinical purpose. Methods and Material: PDDs and Profiles of Open and Wedge fields for different field sizes and at different depths measured as per Varian’s algorithm commissioning guideline. The measurement performed with PTW 3D-scanning water phantom with semi-flex ion chamber and MEPHYSTO software. The online available Varian Golden Beam Data compared with the measured data to evaluate the accuracy of the golden beam data to be used for the commissioning of Eclipse treatment planning system. Results: The deviation between measured vs. golden beam data was in the range of 2% max. In PDDs, the deviation increases more in the deeper depths than the shallower depths. Similarly, profiles have the same trend of increasing deviation at large field sizes and increasing depths. Conclusion: Study shows that the percentage deviation between measured and golden beam data is within the acceptable tolerance and therefore can be used for the commissioning process; however, verification of small subset of acquired data with the golden beam data should be mandatory before clinical use.

Keywords: percent depth dose, flatness, symmetry, golden beam data

Procedia PDF Downloads 459

Authors: Selvakumar Ulaganathan, Ivo Couckuyt, Francesco Ferranti, Tom Dhaene, Eric Laermans

Abstract:

Variable-fidelity surrogate modelling offers an efficient way to approximate function data available in multiple degrees of accuracy each with varying computational cost. In this paper, a Kriging-based variable-fidelity surrogate modelling approach is introduced to approximate such deterministic data. Initially, individual Kriging surrogate models, which are enhanced with gradient data of different degrees of accuracy, are constructed. Then these Gradient enhanced Kriging surrogate models are strategically coupled using a recursive CoKriging formulation to provide an accurate surrogate model for the highest fidelity data. While, intuitively, gradient data is useful to enhance the accuracy of surrogate models, the primary motivation behind this work is to investigate if it is also worthwhile incorporating gradient data of varying degrees of accuracy.

Keywords: Kriging, CoKriging, Surrogate modelling, Variable- fidelity modelling, Gradients

Procedia PDF Downloads 532

Authors: Xiaoyan Dai, Hsieh Yisan

Abstract:

Barcode processing of captured images is a huge challenge, as different shooting conditions can result in different barcode appearances. This paper proposes a deep learning-based barcode detection using synthetic-to-real data augmentation. We first augment barcodes themselves; we then augment images containing the barcodes to generate a large variety of data that is close to the actual shooting environments. Comparisons with previous works and evaluations with our original data show that this approach achieves state-of-the-art performance in various real images. In addition, the system uses hybrid resolution for barcode “scan” and is applicable to real-time applications.

Keywords: barcode detection, data augmentation, deep learning, image-based processing

Procedia PDF Downloads 134

Authors: Rahul Malhotra, Anurag Sharma

Abstract:

Fiber based access networks can deliver performance that can support the increasing demands for high speed connections. One of the new technologies that have emerged in recent years is Passive Optical Networks. This paper is targeted to show the simultaneous delivery of triple play service (data, voice, and video). The comparative investigation and suitability of various data rates is presented. It is demonstrated that as we increase the data rate, number of users to be accommodated decreases due to increase in bit error rate.

Keywords: FTTH, quad play, play service, access networks, data rate

Procedia PDF Downloads 385