Search results for: imbalance data

Authors: Inkyu Kim, SangMan Moon

Abstract:

This IEEE 802.11b protocol provides up to 11Mbps data rate, whereas aerospace industry wants to seek higher data rate COTS data link system in the UAV. The Total Maximum Throughput (TMT) and delay time are studied on many researchers in the past years This paper provides theoretical data throughput performance of UAV formation flight data link using the existing 802.11b performance theory. We operate the UAV formation flight with more than 30 quad copters with 802.11b protocol. We may be predicting that UAV formation flight numbers have to bound data link protocol performance limitations.

Keywords: UAV datalink, UAV formation flight datalink, UAV WLAN datalink application, UAV IEEE 802.11b datalink application

Procedia PDF Downloads 362

Authors: Radoslav Židek, Veronika Šidlová, Radovan Kasarda, Birgit Fuerst-Waltl

Abstract:

Machine learning represents a set of topics dealing with the creation and evaluation of algorithms that facilitate pattern recognition, classification, and prediction, based on models derived from existing data. The data can present identification patterns which are used to classify into groups. The result of the analysis is the pattern which can be used for identification of data set without the need to obtain input data used for creation of this pattern. An important requirement in this process is careful data preparation validation of model used and its suitable interpretation. For breeders, it is important to know the origin of animals from the point of the genetic diversity. In case of missing pedigree information, other methods can be used for traceability of animal´s origin. Genetic diversity written in genetic data is holding relatively useful information to identify animals originated from individual countries. We can conclude that the application of data mining for molecular genetic data using supervised learning is an appropriate tool for hypothesis testing and identifying an individual.

Keywords: genetic data, Pinzgau cattle, supervised learning, machine learning

Procedia PDF Downloads 524

Authors: Nidhi Gopal

Abstract:

Routing is the process of moving a packet of data from source to destination and enables messages to pass from one computer to another and eventually reach the target machine. A router is a networking device that forwards data packets between computer networks. It is connected to two or more data lines from different networks (as opposed to a network switch, which connects data lines from one single network). This paper mainly emphasizes upon the study of router device, its top level architecture, and how various sub-modules of router i.e. Register, FIFO, FSM and Synchronizer are synthesized, and simulated and finally connected to its top module.

Keywords: data packets, networking, router, routing

Procedia PDF Downloads 772

Authors: Julius Onyancha, Valentina Plekhanova

Abstract:

One of the significant issues facing web users is the amount of noise in web data which hinders the process of finding useful information in relation to their dynamic interests. Current research works consider noise as any data that does not form part of the main web page and propose noise web data reduction tools which mainly focus on eliminating noise in relation to the content and layout of web data. This paper argues that not all data that form part of the main web page is of a user interest and not all noise data is actually noise to a given user. Therefore, learning of noise web data allocated to the user requests ensures not only reduction of noisiness level in a web user profile, but also a decrease in the loss of useful information hence improves the quality of a web user profile. Noise Web Data Learning (NWDL) tool/algorithm capable of learning noise web data in web user profile is proposed. The proposed work considers elimination of noise data in relation to dynamic user interest. In order to validate the performance of the proposed work, an experimental design setup is presented. The results obtained are compared with the current algorithms applied in noise web data reduction process. The experimental results show that the proposed work considers the dynamic change of user interest prior to elimination of noise data. The proposed work contributes towards improving the quality of a web user profile by reducing the amount of useful information eliminated as noise.

Keywords: web log data, web user profile, user interest, noise web data learning, machine learning

Procedia PDF Downloads 240

Authors: Yolanda Indah Permatasari, Sudarsono Hardjosoekarto

Abstract:

Public private partnership (PPP) scheme was considered as an alternative source of funding for infrastructure provision. However, the performance of PPP scheme and interest of private sector to participate in the provision of infrastructure was still practically low. This phenomenon motivates the research to reconstruct the form of collaborative governance at the policy level from the perspective of transaction cost of the PPP scheme. Soft-system methodology (SSM)-based action research was used as this research methodology. The result of this study concludes that the emergence of transaction cost sources at the policy level is caused by the absence of a law that governs infrastructure investment, especially the implementation of PPP scheme. This absence is causing the imbalance in risk allocation and risk mitigation between the public and private sector. Thus, this research recommended the formulation of infrastructure investment law that aims to minimize asymmetry information, to anticipate the principal-principal problems, and to provide legal basis that ensures risk certainty and guarantee fair risk allocation between public and private sector.

Keywords: public governance, public private partnership, soft system methodology, transaction cost

Procedia PDF Downloads 115

Authors: Zeba Mahmood

Abstract:

The modern business organizations are adopting technological advancement to achieve competitive edge and satisfy their consumer. The development in the field of Information technology systems has changed the way of conducting business today. Business operations today rely more on the data they obtained and this data is continuously increasing in volume. The data stored in different locations is difficult to find and use without the effective implementation of Data mining and Knowledge management techniques. Organizations who smartly identify, obtain and then convert data in useful formats for their decision making and operational improvements create additional value for their customers and enhance their operational capabilities. Marketers and Customer relationship departments of firm use Data mining techniques to make relevant decisions, this paper emphasizes on the identification of different data mining and Knowledge management techniques that are applied to different business industries. The challenges and issues of execution of these techniques are also discussed and critically analyzed in this paper.

Keywords: knowledge, knowledge management, knowledge discovery in databases, business, operational, information, data mining

Procedia PDF Downloads 509

Authors: Adarsh Shroff

Abstract:

Big data is a collection of dataset so large and complex that it becomes difficult to process using data base management tools. To perform operations like search, analysis, visualization on big data by using data mining; which is the process of extraction of patterns or knowledge from large data set. In recent years, the data mining applications become stale and obsolete over time. Incremental processing is a promising approach to refreshing mining results. It utilizes previously saved states to avoid the expense of re-computation from scratch. This project uses i2MapReduce, an incremental processing extension to Map Reduce, the most widely used framework for mining big data. I2MapReduce performs key-value pair level incremental processing rather than task level re-computation, supports not only one-step computation but also more sophisticated iterative computation, which is widely used in data mining applications, and incorporates a set of novel techniques to reduce I/O overhead for accessing preserved fine-grain computation states. To optimize the mining results, evaluate i2MapReduce using a one-step algorithm and three iterative algorithms with diverse computation characteristics for efficient mining.

Keywords: big data, map reduce, incremental processing, iterative computation

Procedia PDF Downloads 322

Authors: Jerry Q. Cheng

Abstract:

Currently, in analyzing large-scale recurrent event data, there are many challenges such as memory limitations, unscalable computing time, etc. In this research, a divide-and-conquer method is proposed using parametric frailty models. Specifically, the data is randomly divided into many subsets, and the maximum likelihood estimator from each individual data set is obtained. Then a weighted method is proposed to combine these individual estimators as the final estimator. It is shown that this divide-and-conquer estimator is asymptotically equivalent to the estimator based on the full data. Simulation studies are conducted to demonstrate the performance of this proposed method. This approach is applied to a large real dataset of repeated heart failure hospitalizations.

Keywords: big data analytics, divide-and-conquer, recurrent event data, statistical computing

Procedia PDF Downloads 138

Authors: Harneet Walia, Morteza Zihayat

Abstract:

Customer acquisition has become one of the critical issues of any business in the 21st century; having a healthy customer base is the essential asset of the bank business. Term deposits act as a major source of cheap funds for the banks to invest and benefit from interest rate arbitrage. To attract customers, the marketing campaigns at most financial institutions consist of multiple outbound telephonic calls with more than one contact to a customer which is a very time-consuming process. Therefore, customized direct marketing has become more critical than ever for attracting new clients. As customer acquisition is becoming more difficult to archive, having an intelligent and redefined list is necessary to sell a product smartly. Our aim of this research is to increase the effectiveness of campaigns by predicting customers who will most likely subscribe to the fixed deposit and suggest the most suitable month to reach out to customers. We design a Time Aware Upsell Prediction Framework (TAUPF) using two different approaches, with an aim to find the best approach and technique to build the prediction model. TAUPF is implemented using Upsell Prediction Approach (UPA) and Clustered Upsell Prediction Approach (CUPA). We also address the data imbalance problem by examining and comparing different methods of sampling (Up-sampling and down-sampling). Our results have shown building such a model is quite feasible and profitable for the financial institutions. The Time Aware Upsell Prediction Framework (TAUPF) can be easily used in any industry such as telecom, automobile, tourism, etc. where the TAUPF (Clustered Upsell Prediction Approach (CUPA) or Upsell Prediction Approach (UPA)) holds valid. In our case, CUPA books more reliable. As proven in our research, one of the most important challenges is to define measures which have enough predictive power as the subscription to a fixed deposit depends on highly ambiguous situations and cannot be easily isolated. While we have shown the practicality of time-aware upsell prediction model where financial institutions can benefit from contacting the customers at the specified month, further research needs to be done to understand the specific time of the day. In addition, a further empirical/pilot study on real live customer needs to be conducted to prove the effectiveness of the model in the real world.

Keywords: customer acquisition, predictive analysis, targeted marketing, time-aware analysis

Procedia PDF Downloads 100

Authors: Fayomi Oluyemi, Ajayi Lady

Abstract:

This article raises the question of political leadership in the context of womens' roles and responsibilities in Nigeria. The leadership question in Nigeria is disquieting to both academics and policy actors. In a democratic society like Nigeria, the parameters for a well-deserved leadership position is characterised by variables of equity, competence, transparency, accountability, selflessness, and commitment to the tenets of democracy, but the failure of leadership is pervasive in all spheres of socio-political sectors in Nigeria. The paper appraises the activities of Nigerian women in the socio-political arena in Nigeria. It traces their leadership roles from pre-colonial through post-colonial eras with emphasis on 1914 till date. It is argued in the paper that gender imbalance in leadership is a bane to peaceful co-existence and development in Nigeria. It is a truism that gender-blind and gender biased political agendas can distort leadership activities. The extent of their contributions of the few outstanding women’s relative tranquility is highlighted in the theoretical discourse. The methodology adopted for this study is an exploratory study employing the extended case method (ECM). The study was carried out among some selected Nigerian women politicians and academics. Because of ECM's robustness as a qualitative research design, it has helped this study in identifying the challenges of these women thematically and also in constructing valid and reliable measures of the constructs. The study made use of ethnography and triangulation, the latter of which is used by qualitative researchers to check and establish validity in their studies by analyzing a research question from multiple perspectives, specifically Investigator triangulation which involves using several different investigators in the analysis process. Typically, this manifests as the evaluation team consisting of colleagues within a field of study wherein each investigator examines the question of political leadership with the same qualitative method (interview, observation, case study, or focus groups). In addition, data was collated through documentary sources like journals, books, magazines, newspapers, and internet materials. The arguments of this paper center on gender equity of both sexes in socio-political representation and effective participation. The paper concludes with the need to effectively maintain gender balance in leadership in order to enhance lasting peace and unity in Nigeria.

Keywords: gender, politics, leadership, women

Procedia PDF Downloads 422

Authors: Ashiff Khan, A. Seetharaman, Abhijit Dasgupta

Abstract:

The new era of big data (BD) is influencing chemical industries tremendously, providing several opportunities to reshape the way they operate and help them shift towards intelligent manufacturing. Given the availability of free software and the large amount of real-time data generated and stored in process plants, chemical industries are still in the early stages of big data adoption. The industry is just starting to realize the importance of the large amount of data it owns to make the right decisions and support its strategies. This article explores the importance of professional competencies and data science that influence BD in chemical industries to help it move towards intelligent manufacturing fast and reliable. This article utilizes a literature review and identifies potential applications in the chemical industry to move from conventional methods to a data-driven approach. The scope of this document is limited to the adoption of BD in chemical industries and the variables identified in this article. To achieve this objective, government, academia, and industry must work together to overcome all present and future challenges.

Keywords: chemical engineering, big data analytics, industrial revolution, professional competence, data science

Procedia PDF Downloads 58

Authors: M. Sumana, K. S. Hareesha

Abstract:

Secure computations are essential while performing privacy preserving data mining. Distributed privacy preserving data mining involve two to more sites that cannot pool in their data to a third party due to the violation of law regarding the individual. Hence in order to model the private data without compromising privacy and information loss, secure multiparty computations are used. Secure computations of product, mean, variance, dot product, sigmoid function using the additive and multiplicative homomorphic property is discussed. The computations are performed on vertically partitioned data with a single site holding the class value.

Keywords: homomorphic property, secure product, secure mean and variance, secure dot product, vertically partitioned data

Procedia PDF Downloads 393

Authors: Liron Hoch

Abstract:

Traditional religious and spiritual texts offer a surprising wealth of relevant theoretical and practical knowledge about human behavior. This wellspring may contribute significantly to expanding our current body of knowledge in the social sciences and criminology in particular. In Jewish religious texts, specifically by Maimonides, we can find profound analyses of human traits and guidelines for a normative way of life. Among other things, modern criminological literature attempts to link certain character traits and divergent behaviors. Using the hermeneutic phenomenological approach, we analyzed the writings of Maimonides, mainly Laws of Human Dispositions, in order to understand Moses ben Maimon's (1138–1204) view of character traits. The analysis yielded four themes: (1) Human personality between nature and nurture; (2) The complexity of human personality, imbalance and criminality; (3) Extremism as a way to achieve balance; and (4) The Middle Way, flexibility and common sense. These themes can serve therapeutic purposes, as well as inform a rehabilitation model. Grounded in a theoretical rationale about the nature of humans, this model is designed to direct individuals to balance their traits by self-reflection and constant practice of the Middle Way. The proposal we will present is that implementing this model may promote normative behavior and thus contribute to rehabilitating offenders.

Keywords: rehabilitation, traits, offenders, maimonides, middle way

Procedia PDF Downloads 46

Authors: Pradeep Singh, Shrish Verma

Abstract:

Software fault prediction models are created by using the source code, processed metrics from the same or previous version of code and related fault data. Some company do not store and keep track of all artifacts which are required for software fault prediction. To construct fault prediction model for such company, the training data from the other projects can be one potential solution. The earlier we predict the fault the less cost it requires to correct. The training data consists of metrics data and related fault data at function/module level. This paper investigates fault predictions at early stage using the cross-project data focusing on the design metrics. In this study, empirical analysis is carried out to validate design metrics for cross project fault prediction. The machine learning techniques used for evaluation is Naïve Bayes. The design phase metrics of other projects can be used as initial guideline for the projects where no previous fault data is available. We analyze seven data sets from NASA Metrics Data Program which offer design as well as code metrics. Overall, the results of cross project is comparable to the within company data learning.

Keywords: software metrics, fault prediction, cross project, within project.

Procedia PDF Downloads 315

Authors: Vesna Kirandziska, Nevena Ackovska, Ana Madevska Bogdanova

Abstract:

The problem of emotion recognition is a challenging problem. It is still an open problem from the aspect of both intelligent systems and psychology. In this paper, both voice features and facial features are used for building an emotion recognition system. A Support Vector Machine classifiers are built by using raw data from video recordings. In this paper, the results obtained for the emotion recognition are given, and a discussion about the validity and the expressiveness of different emotions is presented. A comparison between the classifiers build from facial data only, voice data only and from the combination of both data is made here. The need for a better combination of the information from facial expression and voice data is argued.

Keywords: emotion recognition, facial recognition, signal processing, machine learning

Procedia PDF Downloads 293

Authors: Sartaj Singh, Amar Singh, Ashok Sharma, Sandeep Kaur

Abstract:

With upcoming threats in a digital world, we need to work continuously in the area of security in all aspects, from hardware to software as well as data modelling. The rise in social media activities and hunger for data by various entities leads to cybercrime and more attack on the privacy and security of persons. Cryptography has always been employed to avoid access to important data by using many processes. Symmetric key and asymmetric key cryptography have been used for keeping data secrets at rest as well in transmission mode. Various cryptosystems have evolved from time to time to make the data more secure. In this research article, we are studying various cryptosystems in asymmetric cryptography and their application with usefulness, and much emphasis is given to Elliptic curve cryptography involving algebraic mathematics.

Keywords: cryptography, symmetric key cryptography, asymmetric key cryptography

Procedia PDF Downloads 94

Authors: W. Du, X. Wang, Jun Cao, H. F. Wang

Abstract:

Due to the randomness and uncertainty of wind energy, modern power systems integrating large-scale wind generation will be significantly impacted in terms of system performance and technical challenges. System inertia with high wind penetration is decreasing when conventional thermal generators are gradually replaced by wind turbines, which do not naturally contribute to inertia response. The power imbalance caused by wind power or demand fluctuations leads to the instability of system frequency. Accordingly, the need to attach the supplementary virtual inertia control to wind farms (WFs) strongly arises. When multi-wind farms are connected to the grid simultaneously, the selection of which critical WFs to install the virtual inertia control is greatly important to enhance the stability of system frequency. By building the small signal model of wind power systems considering frequency regulation, the installation locations are identified by the geometric measures of the mode observability of WFs. In addition, this paper takes the impacts of grid topology and selection of feedback control signals into consideration. Finally, simulations are conducted on a multi-wind farms power system and the results demonstrate that the designed virtual inertia control method can effectively assist the frequency regulation.

Keywords: frequency regulation, virtual inertia control, installation locations, observability, wind farms

Procedia PDF Downloads 374

Authors: Rong-Terng Juang

Abstract:

Autonomous vehicles offer the possibility of significant benefits to social welfare. However, fully automated cars might not be going to happen in the near further. To speed the adoption of the self-driving technologies, many governments worldwide are passing laws requiring data recorders for the testing of autonomous vehicles. Currently, the self-driving vehicle, (e.g., shuttle bus) has to be monitored from a remote control center. When an autonomous vehicle encounters an unexpected driving environment, such as road construction or an obstruction, it should request assistance from a remote operator. Nevertheless, large amounts of data, including images, radar and lidar data, etc., have to be transmitted from the vehicle to the remote center. Therefore, this paper proposes a data compression method of in-vehicle networks for remote monitoring of autonomous vehicles. Firstly, the time-series data are rearranged into a multi-dimensional signal space. Upon the arrival, for controller area networks (CAN), the new data are mapped onto a time-data two-dimensional space associated with the specific CAN identity. Secondly, the data are sampled based on differential sampling. Finally, the whole set of data are encoded using existing algorithms such as Huffman, arithmetic and codebook encoding methods. To evaluate system performance, the proposed method was deployed on an in-house built autonomous vehicle. The testing results show that the amount of data can be reduced as much as 1/7 compared to the raw data.

Keywords: autonomous vehicle, data compression, remote monitoring, controller area networks (CAN), Lidar

Procedia PDF Downloads 138

Authors: Adriana B. Ribeiro, Camila B. Araujo, Frank L. Bueno, Luiz Eduardo V. Silva, Caroline V. Fortes, Helio C. Salgado, Rubens Fazan Jr., Claudia H. L. da Silva

Abstract:

Denture-related stomatitis (DRS) has recently been associated with deleterious cardiovascular effects, including hypertension. This study evaluated salivary parameters, blood pressure (BP) and heart rate variability (HRV), before and after DRS treatment in edentulous patients (n=14). Collection of unstimulated and stimulated saliva, as well as blood pressure (BP) measurements and electrocardiogram recordings were performed before and after 10 days of DRS treatment. The salivary flow (mL/min) was found similar at both times while pH was smaller (more neutral) after treatment (7.3 ± 2.2 vs. 7.1 ± 0.24). Systolic BP (mmHg) showed a trend, but not a significant reduction after DRS treatment (158 ± 25.68 vs. 148 ± 16,72, p=0,062) while diastolic BP was found similar in both times (86 ± 13.93 and 84 ± 9.38). Overall HRV, measured by standard deviation of RR intervals was not affected by DRS treatment (24 ± 4 vs 18 ± 2 ms), but differences of successive RR intervals (an index of parasympathetic cardiac modulation) increased after the treatment (26 ± 4 vs 19 ± 2 ms). Moreover, another index of vagal modulation of the heart, the power of RR interval spectra at high-frequency, was also markedly higher after DRS treatment (236 ± 63 vs 135 ± 32 ms²). Such findings strongly suggest that DRS is linked to an autonomic imbalance with sympathetic overactivity, which is markedly deleterious, increasing cardiovascular risk and the incidence of diseases such as hypertension. Acknowledgment: This study is supported by FAPESP, CNPq.

Keywords: biofilm, denture stomatitis, HRV, blood pressure

Procedia PDF Downloads 207

Authors: Samar M. Alqhtani, Suhuai Luo, Brian Regan

Abstract:

Data fusion technology can be the best way to extract useful information from multiple sources of data. It has been widely applied in various applications. This paper presents a data fusion approach in multimedia data for event detection in twitter by using Dempster-Shafer evidence theory. The methodology applies a mining algorithm to detect the event. There are two types of data in the fusion. The first is features extracted from text by using the bag-ofwords method which is calculated using the term frequency-inverse document frequency (TF-IDF). The second is the visual features extracted by applying scale-invariant feature transform (SIFT). The Dempster - Shafer theory of evidence is applied in order to fuse the information from these two sources. Our experiments have indicated that comparing to the approaches using individual data source, the proposed data fusion approach can increase the prediction accuracy for event detection. The experimental result showed that the proposed method achieved a high accuracy of 0.97, comparing with 0.93 with texts only, and 0.86 with images only.

Keywords: data fusion, Dempster-Shafer theory, data mining, event detection

Procedia PDF Downloads 380

Authors: Ioannis Iglezakis, Theodoros D. Trokanas, Panagiota Kiortsi

Abstract:

This paper aims to explore major legal issues arising from the collection and processing of Health Big Data in the light of the new European secondary legislation for the protection of personal data of natural persons, placing emphasis on the General Data Protection Regulation 679/2016. Whether Big Health Data can be characterised as ‘personal data’ or not is really the crux of the matter. The legal ambiguity is compounded by the fact that, even though the processing of Big Health Data is premised on the de-identification of the data subject, the possibility of a combination of Big Health Data with other data circulating freely on the web or from other data files cannot be excluded. Another key point is that the application of some provisions of GPDR to Big Health Data may both absolve the data controller of his legal obligations and deprive the data subject of his rights (e.g., the right to be informed), ultimately undermining the fundamental right to the protection of personal data of natural persons. Moreover, data subject’s rights (e.g., the right not to be subject to a decision based solely on automated processing) are heavily impacted by the use of AI, algorithms, and technologies that reclaim health data for further use, resulting in sometimes ambiguous results that have a substantial impact on individuals. On the other hand, as the COVID-19 pandemic has revealed, Big Data analytics can offer crucial sources of information. In this respect, this paper identifies and systematises the legal provisions concerned, offering interpretative solutions that tackle dangers concerning data subject’s rights while embracing the opportunities that Big Health Data has to offer. In addition, particular attention is attached to the scope of ‘consent’ as a legal basis in the collection and processing of Big Health Data, as the application of data analytics in Big Health Data signals the construction of new data and subject’s profiles. Finally, the paper addresses the knotty problem of role assignment (i.e., distinguishing between controller and processor/joint controllers and joint processors) in an era of extensive Big Health data sharing. The findings are the fruit of a current research project conducted by a three-member research team at the Faculty of Law of the Aristotle University of Thessaloniki and funded by the Greek Ministry of Education and Religious Affairs.

Keywords: big health data, data subject rights, GDPR, pandemic

Procedia PDF Downloads 107

Authors: Yong-Kyu Jung

Abstract:

The fast growth in information technology has led to de-mands to access/process data. CPSs heavily depend on the time of hardware/software operations and communication over the network (i.e., real-time/parallel operations in CPSs (e.g., autonomous vehicles). Since data processing is an im-portant means to overcome the issue confronting data management, reducing the gap between the technological-growth and the data-complexity and channel-bandwidth. An adaptive perpetual data approximation method is intro-duced to manage the actual entropy of the digital spectrum. An ADAC implemented as an accelerator and/or apps for servers/smart-connected devices adaptively rescales digital contents (avg.62.8%), data processing/access time/energy, encryption/decryption overheads in AI/ML applications (facial ID/recognition).

Keywords: adaptive codec, AI, ML, HPC, cyber-physical, cybersecurity

Procedia PDF Downloads 58

Authors: Kemka C. Ihemelandu, Chukwuemeka U. Ihemelandu

Abstract:

The incidence of melanoma has been increasing rapidly over the past two decades, making melanoma a current public health crisis. Unfortunately, even as screening efforts continue to expand in an effort to ameliorate the death rate from melanoma, there is a need to improve diagnostic accuracy to decrease misdiagnosis. Artificial intelligence (AI) a new frontier in patient care has the ability to improve the accuracy of melanoma diagnosis. Convolutional neural network (CNN) a form of deep neural network, most commonly applied to analyze visual imagery, has been shown to outperform the human brain in pattern recognition. However, there are noted limitations with the accuracy of the CNN models. Our aim in this study was the optimization of convolutional neural network algorithms for the automated diagnosis of melanoma. We hypothesized that Optimal selection of the momentum and batch hyperparameter increases model accuracy. Our most successful model developed during this study, showed that optimal selection of momentum of 0.25, batch size of 2, led to a superior performance and a faster model training time, with an accuracy of ~ 83% after nine hours of training. We did notice a lack of diversity in the dataset used, with a noted class imbalance favoring lighter vs. darker skin tone. Training set image transformations did not result in a superior model performance in our study.

Keywords: melanoma, convolutional neural network, momentum, batch hyperparameter

Procedia PDF Downloads 80

Authors: Sašo Pečnik, Borut Žalik

Abstract:

This paper presents a real-time visualization technique and filtering of classified LiDAR point clouds. The visualization is capable of displaying filtered information organized in layers by the classification attribute saved within LiDAR data sets. We explain the used data structure and data management, which enables real-time presentation of layered LiDAR data. Real-time visualization is achieved with LOD optimization based on the distance from the observer without loss of quality. The filtering process is done in two steps and is entirely executed on the GPU and implemented using programmable shaders.

Keywords: filtering, graphics, level-of-details, LiDAR, real-time visualization

Procedia PDF Downloads 279

Authors: Hasik Lee, Seung-Young Kho

Abstract:

Nowadays, automatic fare collection (AFC) system is widely used in many countries. However, smart card data from many of cities does not contain alighting information which is necessary to build OD matrices. Therefore, in order to utilize smart card data, destinations of passengers should be estimated. In this paper, kernel density estimation was used to forecast probabilities of alighting stations of bus passengers and applied to smart card data in Seoul, Korea which contains boarding and alighting information. This method was also validated with actual data. In some cases, stochastic method was more accurate than deterministic method. Therefore, it is sufficiently accurate to be used to build OD matrices.

Keywords: destination estimation, Kernel density estimation, smart card data, validation

Procedia PDF Downloads 327

Authors: Jae Won Shin

Abstract:

We develop an evaluated nuclear data based photonuclear reaction model of GEANT4 for a more accurate simulation of photon-induced neutron production. The evaluated photonuclear data libraries from the ENDF/B-VII.1 are taken as input. Incident photon energies up to 140 MeV which is the threshold energy for the pion production are considered. For checking the validity of the use of the data-based model, we calculate the photoneutron production cross-sections and yields and compared them with experimental data. The results obtained from the developed model are found to be in good agreement with the experimental data for (γ,xn) reactions.

Keywords: ENDF/B-VII.1, GEANT4, photoneutron, photonuclear reaction

Procedia PDF Downloads 253

Authors: Rashi Bhalla, Russel Pears, M. Asif Naeem

Abstract:

In this 'Information Explosion Era' analyzing data 'a critical commodity' and mining knowledge from vertically distributed data stream incurs huge communication cost. However, an effort to decrease the communication in the distributed environment has an adverse influence on the classification accuracy; therefore, a research challenge lies in maintaining a balance between transmission cost and accuracy. This paper proposes a method based on Bayesian inference to reduce the communication volume in a heterogeneous distributed environment while retaining prediction accuracy. Our experimental evaluation reveals that a significant reduction in communication can be achieved across a diverse range of dataset types.

Keywords: big data, bayesian inference, distributed data stream mining, heterogeneous-distributed data

Procedia PDF Downloads 138

Authors: Benny Sand, Yotam Lurie, Shlomo Mark

Abstract:

Medical Internet of Things (MIoT), AI, and data privacy are linked forever in a gordian knot. This paper explores the conflicts of interests between the stakeholders regarding data privacy in the MIoT arena. While patients are at home during healthcare hospitalization, MIoT can play a significant role in improving the health of large parts of the population by providing medical teams with tools for collecting data, monitoring patients’ health parameters, and even enabling remote treatment. While the amount of data handled by MIoT devices grows exponentially, different stakeholders have conflicting understandings and concerns regarding this data. The findings of the research indicate that medical teams are not concerned by the violation of data privacy rights of the patients' in-home healthcare, while patients are more troubled and, in many cases, are unaware that their data is being used without their consent. MIoT technology is in its early phases, and hence a mixed qualitative and quantitative research approach will be used, which will include case studies and questionnaires in order to explore this issue and provide alternative solutions.

Keywords: MIoT, data privacy, stakeholders, home healthcare, information privacy, AI

Procedia PDF Downloads 78

Authors: Deepak Singh, Rail Kuliev

Abstract:

The abstract highlights the critical importance of optimizing data integration and management strategies in the upstream oil and gas industry. With its complex and dynamic nature generating vast volumes of data, efficient data integration and management are essential for informed decision-making, cost reduction, and maximizing operational performance. Challenges such as data silos, heterogeneity, real-time data management, and data quality issues are addressed, prompting the proposal of several strategies. These strategies include implementing a centralized data repository, adopting industry-wide data standards, employing master data management (MDM), utilizing real-time data integration technologies, and ensuring data quality assurance. Training and developing the workforce, “reskilling and upskilling” the employees and establishing robust Data Management training programs play an essential role and integral part in this strategy. The article also emphasizes the significance of data governance and best practices, as well as the role of technological advancements such as big data analytics, cloud computing, Internet of Things (IoT), and artificial intelligence (AI) and machine learning (ML). To illustrate the practicality of these strategies, real-world case studies are presented, showcasing successful implementations that improve operational efficiency and decision-making. In present study, by embracing the proposed optimization strategies, leveraging technological advancements, and adhering to best practices, upstream oil and gas companies can harness the full potential of data-driven decision-making, ultimately achieving increased profitability and a competitive edge in the ever-evolving industry.

Keywords: master data management, IoT, AI&ML, cloud Computing, data optimization

Procedia PDF Downloads 43

Authors: Farhad Asadi, Mohammad Javad Mollakazemi, Aref Ghafouri

Abstract:

Recent research in neural networks science and neuroscience for modeling complex time series data and statistical learning has focused mostly on learning from high input space and signals. Local linear models are a strong choice for modeling local nonlinearity in data series. Locally weighted projection regression is a flexible and powerful algorithm for nonlinear approximation in high dimensional signal spaces. In this paper, different learning scenario of one and two dimensional data series with different distributions are investigated for simulation and further noise is inputted to data distribution for making different disordered distribution in time series data and for evaluation of algorithm in locality prediction of nonlinearity. Then, the performance of this algorithm is simulated and also when the distribution of data is high or when the number of data is less the sensitivity of this approach to data distribution and influence of important parameter of local validity in this algorithm with different data distribution is explained.

Keywords: local nonlinear estimation, LWPR algorithm, online training method, locally weighted projection regression method

Procedia PDF Downloads 471