Search results for: data fusion

Authors: Sara Ibn El Ahrache, Tajje-eddine Rachidi, Hassan Badir, Abderrahmane Sbihi

Abstract:

Recently, there has been increasing confidence for a favorable usage of big data drawn out from the huge amount of information deposited in a cloud computing system. Data kept on such systems can be retrieved through the network at the user’s convenience. However, the data that users send include private information, and therefore, information leakage from these data is now a major social problem. The usage of secret sharing schemes for cloud computing have lately been approved to be relevant in which users deal out their data to several servers. Notably, in a (k,n) threshold scheme, data security is assured if and only if all through the whole life of the secret the opponent cannot compromise more than k of the n servers. In fact, a number of secret sharing algorithms have been suggested to deal with these security issues. In this paper, we present a Mapreduce implementation of Shamir’s secret sharing scheme to increase its performance and to achieve optimal security for cloud data. Different tests were run and through it has been demonstrated the contributions of the proposed approach. These contributions are quite considerable in terms of both security and performance.

Keywords: cloud computing, data security, Mapreduce, Shamir's secret sharing

Procedia PDF Downloads 280

Authors: Sang Hun Park, Moung Young Lee, Su Min Yu, Hyun Sang Jo, Ji Hyeon Kim, Chul Gyu Song

Abstract:

A near-infrared light source used as a light source in the fluorescence imaging system is suitable for use in real-time during the operation since it has no interference in surgical vision. However, fluorescence images do not have depth information. In this paper, we configured the device with the research on molecular imaging systems for monitoring thrombus imaging using fluorescence and photoacoustic. Fluorescence imaging was performed using a phantom experiment in order to search the exact location, and the Photoacoustic image was in order to detect the depth. Fluorescence image obtained when evaluated through current phantom experiments when the concentration of the contrast agent is 25μg / ml, it was confirmed that it looked sharper. The phantom experiment is has shown the possibility with the fluorescence image and photoacoustic image using an indocyanine green contrast agent. For early diagnosis of cardiovascular diseases, more active research with the fusion of different molecular imaging devices is required.

Keywords: fluorescence, photoacoustic, indocyanine green, carotid artery

Procedia PDF Downloads 575

Authors: Farzad Vakili-Farahani, Joern Lungershausen, Kilian Wasmer

Abstract:

Wobbled laser beam welding, fast oscillations of a tiny laser beam within a designed path (weld geometry) during the laser pulse illumination, opens new possibilities to improve the marco-to micro-manufacturing process. The present work introduces the wobbled laser beam welding as a robust welding strategy for improving macro-to micro-fabrication process, e.g., the laser processing for gap-bridging and packaging industry. The typical requisites and relevant equipment for the development of a wobbled laser processing unit are addressed, including a suitable laser source, light delivery system, optics, proper beam deflection system and the design geometry. In addition, experiments have been carried out on titanium plate to compare the results of wobbled laser welding with conventional pulsed laser welding. As compared to the pulsed laser welding, the wobbled laser welding offers a much greater fusion area (i.e. additional molten material) while minimizing the HAZ and provides a better confinement of the material microstructural changes.

Keywords: wobbled laser beam welding, wobbling function, beam oscillation, micro welding

Procedia PDF Downloads 295

Authors: Kyle De Freitas, Margaret Bernard

Abstract:

Enabling data mining analysis among a wider audience of educators is an active area of research within the educational data mining (EDM) community. The paper proposes a framework for developing an environment that caters for educators who have little technical data mining skills as well as for more advanced users with some data mining expertise. This framework architecture was developed through the review of the strengths and weaknesses of existing models in the literature. The proposed framework provides a modular architecture for future researchers to focus on the development of specific areas within the EDM process. Finally, the paper also highlights a strategy of enabling analysis through either the use of predefined questions or a guided data mining process and highlights how the developed questions and analysis conducted can be reused and extended over time.

Keywords: educational data mining, learning management system, learning analytics, EDM framework

Procedia PDF Downloads 303

Authors: Vikrum Malhotra, Manpreet Kaur, Ayesha Ghotto

Abstract:

Background: Cardiac registries such as ACC Percutaneous Coronary Intervention Registry require high quality data to be abstracted, including data elements such as nuclear cardiology, diagnostic coronary angiography, and PCI. Introduction: The audit tool created is used by data abstractors to provide data audits and assess the accuracy and inter-rater reliability of abstraction performed by the abstractors for a health system. This audit tool solution has been developed across 13 registries, including ACC/NCDR registries, PCI, STS, Get with the Guidelines. Methodology: The data audit tool was used to audit internal registry abstraction for all data elements, including stress test performed, type of stress test, data of stress test, results of stress test, risk/extent of ischemia, diagnostic catheterization detail, and PCI data elements for ACC/NCDR PCI registries. This is being used across 20 hospital systems internally and providing abstraction and audit services for them. Results: The data audit tool had inter-rater reliability and accuracy greater than 95% data accuracy and IRR score for the PCI registry in 50 PCI registry cases in 2021. Conclusion: The tool is being used internally for surgical societies and across hospital systems. The audit tool enables the abstractor to be assessed by an external abstractor and includes all of the data dictionary fields for each registry.

Keywords: abstraction, cardiac registry, cardiovascular registry, registry, data

Procedia PDF Downloads 82

Authors: Seyed Esmail Seyedi Bariran, Laysheng Ewe, Amy Ling

Abstract:

Since supplier selection appears as a vital decision, selecting supplier based on the best and most accurate ways has a lot of importance for enterprises. In this study, a new Artificial Intelligence approach is exerted to remove weaknesses of supplier selection. The paper has three parts. First part is choosing the appropriate criteria for assessing the suppliers’ performance. Next one is collecting the data set based on experts. Afterwards, the data set is divided into two parts, the training data set and the testing data set. By the training data set the best structure of GEP and ANN are selected and to evaluate the power of the mentioned methods the testing data set is used. The result obtained shows that the accuracy of GEP is more than ANN. Moreover, unlike ANN, a mathematical equation is presented by GEP for the supplier selection.

Keywords: supplier selection, automotive supply chains, ANN, GEP

Procedia PDF Downloads 604

Authors: Ma Lei-Lei, Zhou You

Abstract:

Aiming at the problems of low detection accuracy of deep learning algorithm in traffic sign detection, this paper proposes improved EfficientDet based traffic sign detection algorithm. Multi-head self-attention is introduced in the minimum resolution layer of the backbone of EfficientDet to achieve effective aggregation of local and global depth information, and this study proposes an improved feature fusion pyramid with increased vertical cross-layer connections, which improves the performance of the model while introducing a small amount of complexity, the Balanced L1 Loss is introduced to replace the original regression loss function Smooth L1 Loss, which solves the problem of balance in the loss function. Experimental results show, the algorithm proposed in this study is suitable for the task of traffic sign detection. Compared with other models, the improved EfficientDet has the best detection accuracy. Although the test speed is not completely dominant, it still meets the real-time requirement.

Keywords: convolutional neural network, transformer, feature pyramid networks, loss function

Procedia PDF Downloads 77

Authors: Yasuyuki Takahashi, Maya Yamashita, Kyoko Saito

Abstract:

In general, dynamic SPECT data acquisition needs a few minutes for one rotation. Thus, the time-activity curve (TAC) derived from the dynamic SPECT is relatively coarse. In order to effectively shorten the interval, between data points, we adopted a 180-degree interpolation method. This method is already used for reconstruction of the X-ray CT data. In this study, we applied this 180-degree interpolation method to SPECT and investigated its effectiveness.To briefly describe the 180-degree interpolation method: the 180-degree data in the second half of one rotation are combined with the 180-degree data in the first half of the next rotation to generate a 360-degree data set appropriate for the time halfway between the first and second rotations. In both a phantom and a patient study, the data points from the interpolated images fell in good agreement with the data points tracking the accumulation of 99mTc activity over time for appropriate region of interest. We conclude that data derived from interpolated images improves the apparent time resolution of dynamic SPECT.

Keywords: dynamic SPECT, time resolution, 180-degree interpolation method, 99mTc-GSA.

Procedia PDF Downloads 479

Authors: Inigo de Miguel Beriain

Abstract:

Genetic data of deceased persons are of great interest for both biomedical research and clinical use. This is due to several reasons. On the one hand, many of our diseases have a genetic component; on the other hand, we share genes with a good part of our biological family. Therefore, it would be possible to improve our response considerably to these pathologies if we could use these data. Unfortunately, at the present moment, the status of data on the deceased is far from being satisfactorily resolved by the EU data protection regulation. Indeed, the General Data Protection Regulation has explicitly excluded these data from the category of personal data. This decision has given rise to a fragmented legal framework on this issue. Consequently, each EU member state offers very different solutions. For instance, Denmark considers the data as personal data of the deceased person for a set period of time while some others, such as Spain, do not consider this data as such, but have introduced some specifically focused regulations on this type of data and their access by relatives. This is an extremely dysfunctional scenario from multiple angles, not least of which is scientific cooperation at the EU level. This contribution attempts to outline a solution to this dilemma through an alternative proposal. Its main hypothesis is that, in reality, health data are, in a sense, a rara avis within data in general because they do not refer to one person but to several. Hence, it is possible to think that all of them can be considered data subjects (although not all of them can exercise the corresponding rights in the same way). When the person from whom the data were obtained dies, the data remain as personal data of his or her biological relatives. Hence, the general regime provided for in the GDPR may apply to them. As these are personal data, we could go back to thinking in terms of a general prohibition of data processing, with the exceptions provided for in Article 9.2 and on the legal bases included in Article 6. This may be complicated in practice, given that, since we are dealing with data that refer to several data subjects, it may be complex to refer to some of these bases, such as consent. Furthermore, there are theoretical arguments that may oppose this hypothesis. In this contribution, it is shown, however, that none of these objections is of sufficient substance to delegitimize the argument exposed. Therefore, the conclusion of this contribution is that we can indeed build a general framework on the processing of personal data of deceased persons in the context of the GDPR. This would constitute a considerable improvement over the current regulatory framework, although it is true that some clarifications will be necessary for its practical application.

Keywords: collective data conceptual issues, data from deceased people, genetic data protection issues, GDPR and deceased people

Procedia PDF Downloads 137

Authors: Abdullah I. Alkraiji, Thomas W. Jackson, Ian Murray

Abstract:

The proliferation of health data standards today is somewhat overlapping and conflicting, resulting in market confusion and leading to increasing proprietary interests. The government role and support in standardization for health data are thought to be crucial in order to establish credible standards for the next decade, to maximize interoperability across the health sector, and to decrease the risks associated with the implementation of non-standard systems. The normative literature missed out the exploration of the different steps required to be undertaken by the government towards the development of national health data standards. Based on the lessons learned from a qualitative study investigating the different issues to the adoption of health data standards in the major tertiary hospitals in Saudi Arabia and the opinions and feedback from different experts in the areas of data exchange and standards and medical informatics in Saudi Arabia and UK, a list of steps required towards the development of national health data standards was constructed. Main steps are the existence of: a national formal reference for health data standards, an agreed national strategic direction for medical data exchange, a national medical information management plan and a national accreditation body, and more important is the change management at the national and organizational level. The outcome of this study can be used by academics and practitioners to develop the planning of health data standards, and in particular those in developing countries.

Keywords: interoperabilty, medical data exchange, health data standards, case study, Saudi Arabia

Procedia PDF Downloads 318

Authors: SangWon Han, MuWook Pyeon, Sujung Moon, DaeKyo Seo

Abstract:

Recently, technologies based on three-dimensional (3D) space information are being developed and quality of life is improving as a result. Research on real-time digital map (RDM) is being conducted now to provide 3D space information. RDM is a service that creates and supplies 3D space information in real time based on location/shape detection. Research subjects on RDM include the construction of 3D space information with matching image data, complementing the weaknesses of image acquisition using multi-source data, and data collection methods using big data. Using RDM will be effective for space analysis using 3D space information in a U-City and for other space information utilization technologies.

Keywords: RDM, multi-source data, big data, U-City

Procedia PDF Downloads 409

Authors: Nieto Bernal Wilson, Carmona Suarez Edgar

Abstract:

The organizations have structured and unstructured information in different formats, sources, and systems. Part of these come from ERP under OLTP processing that support the information system, however these organizations in OLAP processing level, presented some deficiencies, part of this problematic lies in that does not exist interesting into extract knowledge from their data sources, as also the absence of operational capabilities to tackle with these kind of projects.  Data Warehouse and its applications are considered as non-proprietary tools, which are of great interest to business intelligence, since they are repositories basis for creating models or patterns (behavior of customers, suppliers, products, social networks and genomics) and facilitate corporate decision making and research. The following paper present a structured methodology, simple, inspired from the agile development models as Scrum, XP and AUP. Also the models object relational, spatial data models, and the base line of data modeling under UML and Big data, from this way sought to deliver an agile methodology for the developing of data warehouses, simple and of easy application. The methodology naturally take into account the application of process for the respectively information analysis, visualization and data mining, particularly for patterns generation and derived models from the objects facts structured.

Keywords: data warehouse, model data, big data, object fact, object relational fact, process developed data warehouse

Procedia PDF Downloads 386

Authors: Go Bong Choi, Shin Je Lee, Sung Jin Yoo, Gibaek Lee, Jong Min Lee

Abstract:

In South Korea, it is difficult to obtain data for statistical pipe assessment. In this paper, to address these issues, we find that various statistical model presented before is how data mixed with noise and are whether apply in South Korea. Three major type of model is studied and if data is presented in the paper, we add noise to data, which affects how model response changes. Moreover, we generate data from model in paper and analyse effect of noise. From this we can find robustness and applicability in Korea of each model.

Keywords: proportional hazard model, survival model, water main deterioration, ecological sciences

Procedia PDF Downloads 711

Authors: Anusha Konduru, Zhiyong Shan, Preethi Santhanam, Vinod Namboodiri, Rajiv Bagai

Abstract:

Mobile applications are increasing in a significant amount, each to address the requirements of many users. However, the quick developments and enhancements are resulting in many underlying defects. Android apps create and handle a large variety of 'instance' data that has to persist across runs, such as the current navigation route, workout results, antivirus settings, or game state. Due to the nature of Android, an app can be paused, sent into the background, or killed at any time. If the instance data is not saved and restored between runs, in addition to data loss, partially-saved or corrupted data can crash the app upon resume or restart. However, it is difficult for the programmer to manually test this issue for all the activities. This results in the issue of data loss that the data entered by the user are not saved when there is any interruption. This issue can degrade user experience because the user needs to reenter the information each time there is an interruption. Automated testing to detect such data loss is important to improve the user experience. This research proposes a tool, DroidDL, a data loss detector for Android, which detects the instance data loss from a given android application. We have tested 395 applications and found 12 applications with the issue of data loss. This approach is proved highly accurate and reliable to find the apps with this defect, which can be used by android developers to avoid such errors.

Keywords: Android, automated testing, activity, data loss

Procedia PDF Downloads 213

Authors: James Moir

Abstract:

The mainstay of Big Data is prediction in that it allows practitioners, researchers, and policy analysts to predict trends based upon the analysis of large and varied sources of data. These can range from changing social and political opinions, patterns in crimes, and consumer behaviour. Big Data has therefore shifted the criterion of success in science from causal explanations to predictive modelling and simulation. The 19th-century science sought to capture phenomena and seek to show the appearance of it through causal mechanisms while 20th-century science attempted to save the appearance and relinquish causal explanations. Now 21st-century science in the form of Big Data is concerned with the prediction of appearances and nothing more. However, this pulls social science back in the direction of a more rule- or law-governed reality model of science and away from a consideration of the internal nature of rules in relation to various practices. In effect Big Data offers us no more than a world of surface appearance and in doing so it makes disappear any context-specific conceptual sensitivity.

Keywords: big data, appearance, disappearance, surface, epistemology

Procedia PDF Downloads 389

Authors: Stepan Papacek, Jiri Jablonsky, Radek Kana, Ctirad Matonoha, Stefan Kindermann

Abstract:

FRAP (Fluorescence Recovery After Photobleaching) is a widely used measurement technique to determine the mobility of fluorescent molecules within living cells. While the experimental setup and protocol for FRAP experiments are usually fixed, data processing part is still under development. In this paper, we formulate and solve the problem of data selection which enhances the processing of FRAP images. We introduce the concept of the irrelevant data set, i.e., the data which are almost not reducing the confidence interval of the estimated parameters and thus could be neglected. Based on sensitivity analysis, we both solve the problem of the optimal data space selection and we find specific conditions for optimizing an important experimental design factor, e.g., the radius of bleach spot. Finally, a theorem announcing less precision of the integrated data approach compared to the full data case is proven; i.e., we claim that the data set represented by the FRAP recovery curve lead to a larger confidence interval compared to the spatio-temporal (full) data.

Keywords: FRAP, inverse problem, parameter identification, sensitivity analysis, optimal experimental design

Procedia PDF Downloads 256

Authors: Nabilah Filzah Mohd Radzuan, Zalinda Othman, Azuraliza Abu Bakar, Abdul Razak Hamdan

Abstract:

Uncertain data is believed to be an important issue in building up a prediction model. The main objective in the time series uncertainty analysis is to formulate uncertain data in order to gain knowledge and fit low dimensional model prior to a prediction task. This paper discusses the performance of a number of techniques in dealing with uncertain data specifically those which solve uncertain data condition by minimizing the loss of compression properties.

Keywords: compression properties, uncertainty, uncertain time series, mining technique, weather prediction

Procedia PDF Downloads 408

Authors: Maram Saleh

Abstract:

Knowledge has become an essential resource in today’s economy and become the most important asset of maintaining competition advantage in organizations. The importance of knowledge has made organizations to manage their knowledge assets and resources through all multiple knowledge management stages such as: Knowledge Creation, knowledge storage, knowledge sharing and knowledge use. Researches on data mining are continues growing over recent years on both business and educational fields. Data mining is one of the most important steps of the knowledge discovery in databases process aiming to extract implicit, unknown but useful knowledge and it is considered as significant subfield in knowledge management. Data miming have the great potential to help organizations to focus on extracting the most important information on their data warehouses. Data mining tools and techniques can predict future trends and behaviors, allowing businesses to make proactive, knowledge-driven decisions. This review paper explores the applications of data mining techniques in supporting knowledge management process as an effective knowledge discovery technique. In this paper, we identify the relationship between data mining and knowledge management, and then focus on introducing some application of date mining techniques in knowledge management for some real life domains.

Keywords: Data Mining, Knowledge management, Knowledge discovery, Knowledge creation.

Procedia PDF Downloads 183

Authors: Norbert Csizmadia

Abstract:

The 21st Century's main power centers are the cities. More than 52% of the world’s population lives in cities, in particular in the megacities which have a population over 10 million people and is still growing. According to various research and forecasts, the main economic concentration will be in 40 megacities and global centers. Based on various competitiveness analyzes and indices, global city centers, and city networks are outlined, but if we look at other aspects of urban development like complexity, connectivity, creativity, technological development, viability, green cities, pedestrian and child friendly cities, creative and cultural centers, cultural spaces and knowledge centers, we get a city competitiveness index with quite new complex indicators. The research shows this result. In addition to the megacities and the global centers, with the investigation of functionality, we got 64 so-called ‘fusiononopolis’ (i.e., fusion-polis) which stand for the most decisive economic power centers of the 21st century. In this city competition Asian centers considerably rise, as the world's functional city competitiveness index is being formed.

Keywords: economic geography, human geography, technological development, urbanism

Procedia PDF Downloads 337

Authors: Murat Yazici

Abstract:

Anomalies are irregularities found in data that do not adhere to a well-defined standard of normal behavior. The identification of outliers or anomalies in data has been a subject of study within the statistics field since the 1800s. Over time, a variety of anomaly detection techniques have been developed in several research communities. The cluster analysis can be used to detect anomalies. It is the process of associating data with clusters that are as similar as possible while dissimilar clusters are associated with each other. Many of the traditional cluster algorithms have limitations in dealing with data sets containing categorical properties. To detect anomalies in categorical data, fuzzy clustering approach can be used with its advantages. The fuzzy k-Mode (FKM) clustering algorithm, which is one of the fuzzy clustering approaches, by extension to the k-means algorithm, is reported for clustering datasets with categorical values. It is a form of clustering: each point can be associated with more than one cluster. In this paper, anomaly detection is performed on two simulated data by using the FKM cluster algorithm. As a significance of the study, the FKM cluster algorithm allows to determine anomalies with their abnormality degree in contrast to numerous anomaly detection algorithms. According to the results, the FKM cluster algorithm illustrated good performance in the anomaly detection of data, including both one anomaly and more than one anomaly.

Keywords: fuzzy k-mode clustering, anomaly detection, noise, categorical data

Procedia PDF Downloads 27

Authors: Victor Onomza Waziri, John K. Alhassan, Idris Ismaila, Noel Dogonyara

Abstract:

This paper describes the problem of building secure computational services for encrypted information in the Cloud. Computing without decrypting the encrypted data; therefore, it meets the yearning of computational encryption algorithmic aspiration model that could enhance the security of big data for privacy or confidentiality, availability and integrity of the data and user’s security. The cryptographic model applied for the computational process of the encrypted data is the Fully Homomorphic Encryption Scheme. We contribute a theoretical presentations in a high-level computational processes that are based on number theory that is derivable from abstract algebra which can easily be integrated and leveraged in the Cloud computing interface with detail theoretic mathematical concepts to the fully homomorphic encryption models. This contribution enhances the full implementation of big data analytics based on cryptographic security algorithm.

Keywords: big data analytics, security, privacy, bootstrapping, Fully Homomorphic Encryption Scheme

Procedia PDF Downloads 454

Authors: Sarisa Pinkham, Kanyarat Bussaban

Abstract:

The research aims to approximate the amount of daily rainfall by using a pixel value data approach. The daily rainfall maps from the Thailand Meteorological Department in period of time from January to December 2013 were the data used in this study. The results showed that this approach can approximate the amount of daily rainfall with RMSE=3.343.

Keywords: daily rainfall, image processing, approximation, pixel value data

Procedia PDF Downloads 367

Authors: Mohammad Javad Mollakazemi, Farhad Asadi, Aref Ghafouri

Abstract:

In this paper, we considered and applied parametric modeling for some experimental data of dynamical system. In this study, we investigated the different distribution of output measurement from some dynamical systems. Also, with variance processing in experimental data we obtained the region of nonlinearity in experimental data and then identification of output section is applied in different situation and data distribution. Finally, the effect of the spanning the measurement such as variance to identification and limitation of this approach is explained.

Keywords: Gaussian process, nonlinearity distribution, particle filter, system identification

Procedia PDF Downloads 491

Authors: Jaya Mathew

Abstract:

Many organizations are faced with the challenge of how to analyze and build Machine Learning models using their sensitive telemetry data. In this paper, we discuss how users can leverage the power of R without having to move their big data around as well as a cloud based solution for organizations willing to host their data in the cloud. By using ScaleR technology to benefit from parallelization and remote computing or R Services on premise or in the cloud, users can leverage the power of R at scale without having to move their data around.

Keywords: predictive maintenance, machine learning, big data, cloud based, on premise solution, R

Procedia PDF Downloads 357

Authors: Sven Gehrke, Johannes Ruhland

Abstract:

Data is the oil of our time, without them progress would come to a hold [1]. On the other hand, the mistrust of data mining is increasing [2]. The paper at hand shows different aspects of the concept of trust and describes the information asymmetry of the typical stakeholders of a data mining project using the CRISP-DM phase model. Based on the identified influencing factors in relation to trust, problematic aspects of the current approach are verified using various interviews with the stakeholders. The results of the interviews confirm the theoretically identified weak points of the phase model with regard to trust and show potential research areas.

Keywords: trust, data mining, CRISP DM, stakeholder management

Procedia PDF Downloads 77

Authors: K. Thiagarajan, K. Saranya, A. Veeraiah, B. Sudha

Abstract:

This paper presents a novel algorithm for secure, reliable and flexible transmission of big data in two hop wireless networks using cooperative jamming scheme. Two hop wireless networks consist of source, relay and destination nodes. Big data has to transmit from source to relay and from relay to destination by deploying security in physical layer. Cooperative jamming scheme determines transmission of big data in more secure manner by protecting it from eavesdroppers and malicious nodes of unknown location. The novel algorithm that ensures secure and energy balance transmission of big data, includes selection of data transmitting region, segmenting the selected region, determining probability ratio for each node (capture node, non-capture and eavesdropper node) in every segment, evaluating the probability using binary based evaluation. If it is secure transmission resume with the two- hop transmission of big data, otherwise prevent the attackers by cooperative jamming scheme and transmit the data in two-hop transmission.

Keywords: big data, two-hop transmission, physical layer wireless security, cooperative jamming, energy balance

Procedia PDF Downloads 460

Authors: Yunbin Deng

Abstract:

Most existing applications of automatic speech recognition relies on cooperative subjects at a short distance to a microphone. Standoff speech recognition using microphone arrays can extend the subject to sensor distance somewhat, but it is still limited to only a few feet. As such, most deployed applications of standoff speech recognitions are limited to indoor use at short range. Moreover, these applications require air passway between the subject and the sensor to achieve reasonable signal to noise ratio. This study reports long range (50 feet) automatic speech recognition experiments using a Laser Doppler Vibrometer (LDV) sensor. This study shows that the LDV sensor modality can extend the speech acquisition standoff distance far beyond microphone arrays to hundreds of feet. In addition, LDV enables 'listening' through the windows for uncooperative subjects. This enables new capabilities in automatic audio and speech intelligence, surveillance, and reconnaissance (ISR) for law enforcement, homeland security and counter terrorism applications. The Polytec LDV model OFV-505 is used in this study. To investigate the impact of different vibrating materials, five parallel LDV speech corpora, each consisting of 630 speakers, are collected from the vibrations of a glass window, a metal plate, a plastic box, a wood slate, and a concrete wall. These are the common materials the application could encounter in a daily life. These data were compared with the microphone counterpart to manifest the impact of various materials on the spectrum of the LDV speech signal. State of the art deep neural network modeling approaches is used to conduct continuous speaker independent speech recognition on these LDV speech datasets. Preliminary phoneme recognition results using time-delay neural network, bi-directional long short term memory, and model fusion shows great promise of using LDV for long range speech recognition. To author’s best knowledge, this is the first time an LDV is reported for long distance speech recognition application.

Keywords: covert speech acquisition, distant speech recognition, DSR, laser Doppler vibrometer, LDV, speech intelligence surveillance and reconnaissance, ISR

Procedia PDF Downloads 156

Authors: Romeo Botes, Imelda Smit

Abstract:

In today’s modern age of technology vast amounts of data needs to be processed in real-time to keep users satisfied. This data comes from various sources and in many formats, including electronic and mobile devices such as GPRS modems and GPS devices. They make use of different protocols including TCP, UDP, and HTTP/s for data communication to web servers and eventually to users. The data obtained from these devices may provide valuable information to users, but are mostly in an unreadable format which needs to be processed to provide information and business intelligence. This data is not always current, it is mostly historical data. The data is not subject to implementation of consistency and redundancy measures as most other data usually is. Most important to the users is that the data are to be pre-processed in a readable format when it is entered into the database. To accomplish this, programmers build processing programs and scripts to decode and process the information stored in databases. Programmers make use of various techniques in such programs to accomplish this, but sometimes neglect the effect some of these techniques may have on database performance. One of the techniques generally used,is to pull data from the database server, process it and push it back to the database server in one single step. Since the processing of the data usually takes some time, it keeps the database busy and locked for the period of time that the processing takes place. Because of this, it decreases the overall performance of the database server and therefore the system’s performance. This paper follows on a paper discussing the performance increase that may be achieved by utilizing array lists along with a pull-process-push data processing technique split in three steps. The purpose of this paper is to expand the number of clients when comparing the two techniques to establish the impact it may have on performance of the CPU storage and processing time.

Keywords: performance measures, algorithm techniques, data processing, push data, process data, array list

Procedia PDF Downloads 221

Authors: Nkongho Ayuketang Arreyndip, Ebobenow Joseph

Abstract:

In this paper, temperature extremes are forecast by employing the block maxima method of the generalized extreme value (GEV) distribution to analyse temperature data from the Cameroon Development Corporation (CDC). By considering two sets of data (raw data and simulated data) and two (stationary and non-stationary) models of the GEV distribution, return levels analysis is carried out and it was found that in the stationary model, the return values are constant over time with the raw data, while in the simulated data the return values show an increasing trend with an upper bound. In the non-stationary model, the return levels of both the raw data and simulated data show an increasing trend with an upper bound. This clearly shows that although temperatures in the tropics show a sign of increase in the future, there is a maximum temperature at which there is no exceedance. The results of this paper are very vital in agricultural and environmental research.

Keywords: forecasting, generalized extreme value (GEV), meteorology, return level

Procedia PDF Downloads 441

Authors: Abdulrahman K. Alshegaifi, Chun-Hsi Huang

Abstract:

Treating data based on its location in memory has received much attention in recent years due to its different properties, which offer important aspects for cache utilization. Stack data and non-stack data may interfere with each other’s locality in the data cache. One of the important aspects of stack data is that it has high spatial and temporal locality. In this work, we simulate non-unified cache design that split data cache into stack and non-stack caches in order to maintain stack data and non-stack data separate in different caches. We observe that the overall hit rate of non-unified cache design is sensitive to the size of non-stack cache. Then, we investigate the appropriate size and associativity for stack cache to achieve high hit ratio especially when over 99% of accesses are directed to stack cache. The result shows that on average more than 99% of stack cache accuracy is achieved by using 2KB of capacity and 1-way associativity. Further, we analyze the improvement in hit rate when adding small, fixed, size of stack cache at level1 to unified cache architecture. The result shows that the overall hit rate of unified cache design with adding 1KB of stack cache is improved by approximately, on average, 3.9% for Rijndael benchmark. The stack cache is simulated by using SimpleScalar toolset.

Keywords: hit rate, locality of program, stack cache, stack data

Procedia PDF Downloads 282