Search results for: Data mining from uncertain data

Authors: Payal Abichandani, Rishi Prakash, Paras Nath Barwal, B. K. Murthy

Abstract:

Information generated from various computerization processes is a potential rich source of knowledge for its designated community. To pass this information from generation to generation without modifying the meaning is a challenging activity. To preserve and archive the data for future generations it’s very essential to prove the authenticity of the data. It can be achieved by extracting the metadata from the data which can prove the authenticity and create trust on the archived data. Subsequent challenge is the technology obsolescence. Metadata extraction and standardization can be effectively used to resolve and tackle this problem. Metadata can be categorized at two levels i.e. Technical and Domain level broadly. Technical metadata will provide the information that can be used to understand and interpret the data record, but only this level of metadata isn’t sufficient to create trustworthiness. We have developed a tool which will extract and standardize the technical as well as domain level metadata. This paper is about the different features of the tool and how we have developed this.  

Keywords: Digital Preservation, Metadata, OAIS, PDI, XML.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1819

Authors: Zhiyong Li, Zhigang Chen, Chao Fu, Shipeng Zhang

Abstract:

Load forecasting has always been the essential part of an efficient power system operation and planning. A novel approach based on support vector machines is proposed in this paper for annual power load forecasting. Different kernel functions are selected to construct a combinatorial algorithm. The performance of the new model is evaluated with a real-world dataset, and compared with two neural networks and some traditional forecasting techniques. The results show that the proposed method exhibits superior performance.

Keywords: combinatorial algorithm, data mining, load forecasting, support vector machines

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1637

Authors: Rami N. Khushaba, Adel Al-Jumaily

Abstract:

The myoelectric signal (MES) is one of the Biosignals utilized in helping humans to control equipments. Recent approaches in MES classification to control prosthetic devices employing pattern recognition techniques revealed two problems, first, the classification performance of the system starts degrading when the number of motion classes to be classified increases, second, in order to solve the first problem, additional complicated methods were utilized which increase the computational cost of a multifunction myoelectric control system. In an effort to solve these problems and to achieve a feasible design for real time implementation with high overall accuracy, this paper presents a new method for feature extraction in MES recognition systems. The method works by extracting features using Wavelet Packet Transform (WPT) applied on the MES from multiple channels, and then employs Fuzzy c-means (FCM) algorithm to generate a measure that judges on features suitability for classification. Finally, Principle Component Analysis (PCA) is utilized to reduce the size of the data before computing the classification accuracy with a multilayer perceptron neural network. The proposed system produces powerful classification results (99% accuracy) by using only a small portion of the original feature set.

Keywords: Biomedical Signal Processing, Data mining andInformation Extraction, Machine Learning, Rehabilitation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1730

Authors: İsmet Öztürk, İbrahim Soğukpınar

Abstract:

With the fast progression of data exchange in electronic way, information security is becoming more important in data storage and transmission. Because of widely using images in industrial process, it is important to protect the confidential image data from unauthorized access. In this paper, we analyzed current image encryption algorithms and compression is added for two of them (Mirror-like image encryption and Visual Cryptography). Implementations of these two algorithms have been realized for experimental purposes. The results of analysis are given in this paper.

Keywords: image encryption, image cryptosystem, security, transmission

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4949

Authors: Deepak Goswami, Padma Lochan Hazarika, Kandarpa Kumar Sarma

Abstract:

The customary practice of identifying industrial sickness is a set traditional techniques which rely upon a range of manual monitoring and compilation of financial records. It makes the process tedious, time consuming and often are susceptible to manipulation. Therefore, certain readily available tools are required which can deal with such uncertain situations arising out of industrial sickness. It is more significant for a country like India where the fruits of development are rarely equally distributed. In this paper, we propose an approach based on Artificial Neural Network (ANN) to deal with industrial sickness with specific focus on a few such units taken from a less developed north-east (NE) Indian state like Assam. The proposed system provides decision regarding industrial sickness using eight different parameters which are directly related to the stages of sickness of such units. The mechanism primarily uses certain signals and symptoms of industrial health to decide upon the state of a unit. Specifically, we formulate an ANN based block with data obtained from a few selected units of Assam so that required decisions related to industrial health could be taken. The system thus formulated could become an important part of planning and development. It can also contribute towards computerization of decision support systems related to industrial health and help in better management.

Keywords: Industrial, Health, Classification, ANN, MLP, MSE.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1688

Authors: Taras Agryzkov, Almudena Nolasco-Cirugeda, Jos´e L. Oliver, Leticia Serrano-Estrada, Leandro Tortosa, Jos´e F. Vicent

Abstract:

This paper aims to represent the commercial activity of a city taking as source data the social network Foursquare. The city of Murcia is selected as case study, and the location-based social network Foursquare is the main source of information. After carrying out a reorganisation of the user-generated data extracted from Foursquare, it is possible to graphically display on a map the various city spaces and venues especially those related to commercial, food and entertainment sector businesses. The obtained visualisation provides information about activity patterns in the city of Murcia according to the people‘s interests and preferences and, moreover, interesting facts about certain characteristics of the town itself.

Keywords: Social networks, Foursquare, spatial analysis, data visualization, geocomputation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2672

Authors: Chatchai Pesee

Abstract:

This paper examines long-range dependence or longmemory of financial time series on the exchange rate data by the fractional Brownian motion (fBm). The principle of spectral density function in Section 2 is used to find the range of Hurst parameter (H) of the fBm. If 0< H <1/2, then it has a short-range dependence (SRD). It simulates long-memory or long-range dependence (LRD) if 1/2< H <1. The curve of exchange rate data is fBm because of the specific appearance of the Hurst parameter (H). Furthermore, some of the definitions of the fBm, long-range dependence and selfsimilarity are reviewed in Section II as well. Our results indicate that there exists a long-memory or a long-range dependence (LRD) for the exchange rate data in section III. Long-range dependence of the exchange rate data and estimation of the Hurst parameter (H) are discussed in Section IV, while a conclusion is discussed in Section V.

Keywords: Fractional Brownian motion, long-rangedependence, memory, short-range dependence.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1875

Authors: Praveen Boinee, Alessandro De Angelis, Gian Luca Foresti

Abstract:

Leo Breimans Random Forests (RF) is a recent development in tree based classifiers and quickly proven to be one of the most important algorithms in the machine learning literature. It has shown robust and improved results of classifications on standard data sets. Ensemble learning algorithms such as AdaBoost and Bagging have been in active research and shown improvements in classification results for several benchmarking data sets with mainly decision trees as their base classifiers. In this paper we experiment to apply these Meta learning techniques to the random forests. We experiment the working of the ensembles of random forests on the standard data sets available in UCI data sets. We compare the original random forest algorithm with their ensemble counterparts and discuss the results.

Keywords: Random Forests [RF], ensembles, UCI.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2699

Authors: Mohamed A. Mahfouz, M. A. Ismail

Abstract:

This paper introduces new algorithms (Fuzzy relative of the CLARANS algorithm FCLARANS and Fuzzy c Medoids based on randomized search FCMRANS) for fuzzy clustering of relational data. Unlike existing fuzzy c-medoids algorithm (FCMdd) in which the within cluster dissimilarity of each cluster is minimized in each iteration by recomputing new medoids given current memberships, FCLARANS minimizes the same objective function minimized by FCMdd by changing current medoids in such away that that the sum of the within cluster dissimilarities is minimized. Computing new medoids may be effected by noise because outliers may join the computation of medoids while the choice of medoids in FCLARANS is dictated by the location of a predominant fraction of points inside a cluster and, therefore, it is less sensitive to the presence of outliers. In FCMRANS the step of computing new medoids in FCMdd is modified to be based on randomized search. Furthermore, a new initialization procedure is developed that add randomness to the initialization procedure used with FCMdd. Both FCLARANS and FCMRANS are compared with the robust and linearized version of fuzzy c-medoids (RFCMdd). Experimental results with different samples of the Reuter-21578, Newsgroups (20NG) and generated datasets with noise show that FCLARANS is more robust than both RFCMdd and FCMRANS. Finally, both FCMRANS and FCLARANS are more efficient and their outputs are almost the same as that of RFCMdd in terms of classification rate.

Keywords: Data Mining, Fuzzy Clustering, Relational Clustering, Medoid-Based Clustering, Cluster Analysis, Unsupervised Learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2392

Authors: N C Santhosh Kumar, N K Kishore

Abstract:

Using mini modules of Tmotes, it is possible to automate a small personal area network. This idea can be extended to large networks too by implementing multi-hop routing. Linking the various Tmotes using Programming languages like Nesc, Java and having transmitter and receiver sections, a network can be monitored. It is foreseen that, depending on the application, a long range at a low data transfer rate or average throughput may be an acceptable trade-off. To reduce the overall costs involved, an optimum number of Tmotes to be used under various conditions (Indoor/Outdoor) is to be deduced. By analyzing the data rates or throughputs at various locations of Tmotes, it is possible to deduce an optimal number of Tmotes for a specific network. This paper deals with the determination of optimum distances to reduce the cost and increase the reliability of the entire sensor network with Wireless Local Loop (WLL) capability.

Keywords: Average throughput, data rate, multi-hop routing, optimum data transfer, throughput, Tmotes, wireless local loop.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1354

Authors: Supunmali Ahangama, Danny Chiang Choon Poo

Abstract:

Health analytics (HA) is used in healthcare systems for effective decision making, management and planning of healthcare and related activities. However, user resistances, unique position of medical data content and structure (including heterogeneous and unstructured data) and impromptu HA projects have held up the progress in HA applications. Notably, the accuracy of outcomes depends on the skills and the domain knowledge of the data analyst working on the healthcare data. Success of HA depends on having a sound process model, effective project management and availability of supporting tools. Thus, to overcome these challenges through an effective process model, we propose a HA process model with features from rational unified process (RUP) model and agile methodology.

Keywords: Agile methodology, health analytics, unified process model, UML.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2322

Authors: Suraiya Jabin, Kamal K. Bharadwaj

Abstract:

This research presents a system for post processing of data that takes mined flat rules as input and discovers crisp as well as fuzzy hierarchical structures using Learning Classifier System approach. Learning Classifier System (LCS) is basically a machine learning technique that combines evolutionary computing, reinforcement learning, supervised or unsupervised learning and heuristics to produce adaptive systems. A LCS learns by interacting with an environment from which it receives feedback in the form of numerical reward. Learning is achieved by trying to maximize the amount of reward received. Crisp description for a concept usually cannot represent human knowledge completely and practically. In the proposed Learning Classifier System initial population is constructed as a random collection of HPR–trees (related production rules) and crisp / fuzzy hierarchies are evolved. A fuzzy subsumption relation is suggested for the proposed system and based on Subsumption Matrix (SM), a suitable fitness function is proposed. Suitable genetic operators are proposed for the chosen chromosome representation method. For implementing reinforcement a suitable reward and punishment scheme is also proposed. Experimental results are presented to demonstrate the performance of the proposed system.

Keywords: Hierarchical Production Rule, Data Mining, Learning Classifier System, Fuzzy Subsumption Relation, Subsumption matrix, Reinforcement Learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1450

Authors: Amir Azizi, Amir Yazid B. Ali, Loh Wei Ping

Abstract:

Throughput is an important measure of performance of production system. Analyzing and modeling of production throughput is complex in today-s dynamic production systems due to uncertainties of production system. The main reasons are that uncertainties are materialized when the production line faces changes in setup time, machinery break down, lead time of manufacturing, and scraps. Besides, demand changes are fluctuating from time to time for each product type. These uncertainties affect the production performance. This paper proposes Bayesian inference for throughput modeling under five production uncertainties. Bayesian model utilized prior distributions related to previous information about the uncertainties where likelihood distributions are associated to the observed data. Gibbs sampling algorithm as the robust procedure of Monte Carlo Markov chain was employed for sampling unknown parameters and estimating the posterior mean of uncertainties. The Bayesian model was validated with respect to convergence and efficiency of its outputs. The results presented that the proposed Bayesian models were capable to predict the production throughput with accuracy of 98.3%.

Keywords: Bayesian inference, Uncertainty modeling, Monte Carlo Markov chain, Gibbs sampling, Production throughput

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2138

Authors: S. Pattanapairoj, D. Chetchotsak

Abstract:

This paper proposed classification models that would be used as a proxy for hard disk drive (HDD) functional test equitant which required approximately more than two weeks to perform the HDD status classification in either “Pass" or “Fail". These models were constructed by using committee network which consisted of a number of single neural networks. This paper also included the method to solve the problem of sparseness data in failed part, which was called “enforce learning method". Our results reveal that the constructed classification models with the proposed method could perform well in the sparse data conditions and thus the models, which used a few seconds for HDD classification, could be used to substitute the HDD functional tests.

Keywords: Sparse data, Classifications, Committee network

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1729

Authors: Z. Ezzouine, A. Nakheli

Abstract:

This paper presents a high position electromagnetic sensor system (HPESS) that is applicable for moving object detection. The authors have developed a high-performance position sensor prototype dedicated to students’ laboratory. The challenge was to obtain a highly accurate and real-time sensor that is able to calculate position, length or displacement. An electromagnetic solution based on a two coil induction principal was adopted. The HPESS converts mechanical motion to electric energy with direct contact. The output signal can then be fed to an electronic circuit. The voltage output change from the sensor is captured by data acquisition system using LabVIEW software. The displacement of the moving object is determined. The measured data are transmitted to a PC in real-time via a DAQ (NI USB -6281). This paper also describes the data acquisition analysis and the conditioning card developed specially for sensor signal monitoring. The data is then recorded and viewed using a user interface written using National Instrument LabVIEW software. On-line displays of time and voltage of the sensor signal provide a user-friendly data acquisition interface. The sensor provides an uncomplicated, accurate, reliable, inexpensive transducer for highly sophisticated control systems.

Keywords: Electromagnetic sensor, data acquisition, accurately, position measurement.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 954

Authors: Adil AL-Rammahi

Abstract:

When we prefer to make the data secure from various attacks and fore integrity of data, we must encrypt the data before it is transmitted or stored. This paper introduces a new effective and lossless image encryption algorithm using a natural logarithmic function. The new algorithm encrypts an image through a three stage process. In the first stage, a reference natural logarithmic function is generated as the foundation for the encryption image. The image numeral matrix is then analyzed to five integer numbers, and then the numbers’ positions are transformed to matrices. The advantages of this method is useful for efficiently encrypting a variety of digital images, such as binary images, gray images, and RGB images without any quality loss. The principles of the presented scheme could be applied to provide complexity and then security for a variety of data systems such as image and others.

Keywords: Linear Systems, Image Encryption, Calculus.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2390

Authors: P. Parvizi, S. Mohammadi

Abstract:

an intelligent BRT system is necessary when communities looking for new ways to use high capacity rapid transit at a reduced cost.This paper will describe the intelligent control system that works with Datacenter. With the help of GPS system, the data center can monitor the situation of each bus and bus station. Through RFID technology, bus station and traffic light can transfer data with bus and by Wimax communication technology all of parts can talk together; data center learns all information about the location of bus, the arrival of bus in each station and the number of passengers in station and bus.Finally, the paper presents the case study of those theories in Tehran BRT.

Keywords: TehranBRT, RFID, Intelligent Transportation

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2443

Authors: Tirtha S. Das, Ayan K. Sau, Subir K. Sarkar

Abstract:

Digital watermarking is a way to provide the facility of secure multimedia data communication besides its copyright protection approach. The Spread Spectrum modulation principle is widely used in digital watermarking to satisfy the robustness of multimedia signals against various signal-processing operations. Several SS watermarking algorithms have been proposed for multimedia signals but very few works have discussed on the issues responsible for secure data communication and its robustness improvement. The current paper has critically analyzed few such factors namely properties of spreading codes, proper signal decomposition suitable for data embedding, security provided by the key, successive bit cancellation method applied at decoder which have greater impact on the detection reliability, secure communication of significant signal under camouflage of insignificant signals etc. Based on the analysis, robust SS watermarking scheme for secure data communication is proposed in wavelet domain and improvement in secure communication and robustness performance is reported through experimental results. The reported result also shows improvement in visual and statistical invisibility of the hidden data.

Keywords: Spread spectrum modulation, spreading code, signaldecomposition, security, successive bit cancellation

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2771

Authors: Ion-Cosmin Dita, Vasile Gui, Franz Quint, Marius Otesteanu

Abstract:

In automatic manufacturing and assembling of mechanical, electrical and electronic parts one needs to reliably identify the position of components and to extract the information of these components. Data Matrix Codes (DMC) are established by these days in many areas of industrial manufacturing thanks to their concentration of information on small spaces. In today’s usually order-related industry, where increased tracing requirements prevail, they offer further advantages over other identification systems. This underlines in an impressive way the necessity of a robust code reading system for detecting DMC on the components in factories. This paper compares two methods for estimating the angle of orientation of Data Matrix Codes: one method based on the Hough Transform and the other based on the Mean Shift Algorithm. We concentrate on Data Matrix Codes in industrial environment, punched, milled, lasered or etched on different materials in arbitrary orientation.

Keywords: Industrial data matrix code, Hough transform, mean shift.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1330

Authors: Chee Siong Teh, Chee Peng Lim

Abstract:

This paper proposes a novel architecture for developing decision support systems. Unlike conventional decision support systems, the proposed architecture endeavors to reveal the decision-making process such that humans' subjectivity can be incorporated into a computerized system and, at the same time, to preserve the capability of the computerized system in processing information objectively. A number of techniques used in developing the decision support system are elaborated to make the decisionmarking process transparent. These include procedures for high dimensional data visualization, pattern classification, prediction, and evolutionary computational search. An artificial data set is first employed to compare the proposed approach with other methods. A simulated handwritten data set and a real data set on liver disease diagnosis are then employed to evaluate the efficacy of the proposed approach. The results are analyzed and discussed. The potentials of the proposed architecture as a useful decision support system are demonstrated.

Keywords: Interactive evolutionary computation, multivariate data projection, pattern classification, topographic map.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1446

Authors: Jiho Park, Yong-Gyu Lee, Gilwon Yoon

Abstract:

Data security in u-Health system can be an important issue because wireless network is vulnerable to hacking. However, it is not easy to implement a proper security algorithm in an embedded u-health monitoring because of hardware constraints such as low performance, power consumption and limited memory size and etc. To secure data that contain personal and biosignal information, we implemented several security algorithms such as Blowfish, data encryption standard (DES), advanced encryption standard (AES) and Rivest Cipher 4 (RC4) for our u-Health monitoring system and the results were successful. Under the same experimental conditions, we compared these algorithms. RC4 had the fastest execution time. Memory usage was the most efficient for DES. However, considering performance and safety capability, however, we concluded that AES was the most appropriate algorithm for a personal u-Health monitoring system.

Keywords: biosignal, data encryption, security measures, u-health

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2115

Authors: Kristina Georgoulakis

Abstract:

A new blind symbol by symbol equalizer is proposed. The operation of the proposed equalizer is based on the geometric properties of the two dimensional data constellation. An unsupervised clustering technique is used to locate the clusters formed by the received data. The symmetric properties of the clusters labels are subsequently utilized in order to label the clusters. Following this step, the received data are compared to clusters and decisions are made on a symbol by symbol basis, by assigning to each data the label of the nearest cluster. The operation of the equalizer is investigated both in linear and nonlinear channels. The performance of the proposed equalizer is compared to the performance of a CMAbased blind equalizer.

Keywords: Blind equalization, channel equalization, cluster based equalisers

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1426

Authors: Y. N. Phang, E. F. Loh

Abstract:

The zero inflated models are usually used in modeling count data with excess zeros where the existence of the excess zeros could be structural zeros or zeros which occur by chance. These type of data are commonly found in various disciplines such as finance, insurance, biomedical, econometrical, ecology, and health sciences which involve sex and health dental epidemiology. The most popular zero inflated models used by many researchers are zero inflated Poisson and zero inflated negative binomial models. In addition, zero inflated generalized Poisson and zero inflated double Poisson models are also discussed and found in some literature. Recently zero inflated inverse trinomial model and zero inflated strict arcsine models are advocated and proven to serve as alternative models in modeling overdispersed count data caused by excessive zeros and unobserved heterogeneity. The purpose of this paper is to review some related literature and provide a variety of examples from different disciplines in the application of zero inflated models. Different model selection methods used in model comparison are discussed.

Keywords: Overdispersed count data, model selection methods, likelihood ratio, AIC, BIC.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4518

Authors: Sherif S. Nafee, Amir K. Al-Ramady, Salem S. Shaheen

Abstract:

Evaluated nuclear decay data for the 217Po nuclide is presented in the present work. These data include recommended values for the half-life T1/2, α-, β-- and γ-ray emission energies and probabilities. Decay data from 221Rn α and 217Bi β—decays are presented. Q(α) has been updated based on the recent published work of the Atomic Mass Evaluation AME2012. In addition, the logft values were calculated using the Logft program from the ENSDF evaluation package. Moreover, the total internal conversion electrons and the K-shell to L-shell and L-shell to M-shell and to N-shell conversion electrons ratios K/L, L/M and L/N have been calculated using Bricc program. Meanwhile, recommendation values or the multi-polarities have been assigned based on recently measurement yield a better intensity balance at the 254 keV and 264 keV gamma transitions.

Keywords: Atomic Mass Evaluation, Nuclear Data Evaluation, Total Electron Conversion Electrons.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2251

Authors: Shoujia Fang, Guoqing Ding, Xin Chen

Abstract:

The quality of press-fit assembly is closely related to reliability and safety of product. The paper proposed a keypoint detection method based on convolutional neural network to improve the accuracy of keypoint detection in press-fit curve. It would provide an auxiliary basis for judging quality of press-fit assembly. The press-fit curve is a curve of press-fit force and displacement. Both force data and distance data are time-series data. Therefore, one-dimensional convolutional neural network is used to process the press-fit curve. After the obtained press-fit data is filtered, the multi-layer one-dimensional convolutional neural network is used to perform the automatic learning of press-fit curve features, and then sent to the multi-layer perceptron to finally output keypoint of the curve. We used the data of press-fit assembly equipment in the actual production process to train CNN model, and we used different data from the same equipment to evaluate the performance of detection. Compared with the existing research result, the performance of detection was significantly improved. This method can provide a reliable basis for the judgment of press-fit quality.

Keywords: Keypoint detection, curve feature, convolutional neural network, press-fit assembly.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 926

Authors: Nahla Barakat, Joachim Diederich

Abstract:

Support vector machines (SVMs) have shown superior performance compared to other machine learning techniques, especially in classification problems. Yet one limitation of SVMs is the lack of an explanation capability which is crucial in some applications, e.g. in the medical and security domains. In this paper, a novel approach for eclectic rule-extraction from support vector machines is presented. This approach utilizes the knowledge acquired by the SVM and represented in its support vectors as well as the parameters associated with them. The approach includes three stages; training, propositional rule-extraction and rule quality evaluation. Results from four different experiments have demonstrated the value of the approach for extracting comprehensible rules of high accuracy and fidelity.

Keywords: Data mining, hybrid rule-extraction algorithms, medical diagnosis, SVMs

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1697

Authors: Martin Juhas, Bohuslava Juhasova, Igor Halenar, Andrej Elias

Abstract:

This article deals with the possibility of increasing efficiency, reliability and safety of the system for teledosimetric data collection management and their evaluation as a part of complex study for activity “Research of data collection, their measurement and evaluation with mobile and autonomous units” within project “Research of monitoring and evaluation of non-standard conditions in the area of nuclear power plants”. Possible weaknesses in existing system are identified. A study of available cluster solutions with possibility of their deploying to analysed system is presented

Keywords: Teledosimetric data, efficiency, reliability, safety, cluster solution.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1550

Authors: Gökhan Silahtaroğlu

Abstract:

Clustering large populations is an important problem when the data contain noise and different shapes. A good clustering algorithm or approach should be efficient enough to detect clusters sensitively. Besides space complexity, time complexity also gains importance as the size grows. Using hierarchies we developed a new algorithm to split attributes according to the values they have and choosing the dimension for splitting so as to divide the database roughly into equal parts as much as possible. At each node we calculate some certain descriptive statistical features of the data which reside and by pruning we generate the natural clusters with a complexity of O(n).

Keywords: Clustering, tree, split, pruning, entropy, gini.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1543

Authors: T. Thulare, M. Herselman, A. Botha

Abstract:

Poor system use, including inappropriate design of health information systems, causes difficulties in communication with patients and increased time spent by healthcare professionals in recording the necessary health information for medical records. System features like pop-up reminders, complex menus, and poor user interfaces can make medical records far more time consuming than paper cards as well as affect decision-making processes. Although errors associated with health information and their real and likely effect on the quality of care and patient safety have been documented for many years, more research is needed to measure the occurrence of these errors and determine the causes to implement solutions. Therefore, the purpose of this paper is to identify data integrity challenges in hospital information systems through a scoping review and based on the results provide recommendations on how to manage these. Only 34 papers were found to be most suitable out of 297 publications initially identified in the field. The results indicated that human and computerized systems are the most common challenges associated with data integrity and factors such as policy, environment, health workforce, and lack of awareness attribute to these challenges but if measures are taken the data integrity challenges can be managed.

Keywords: Data integrity, data integrity challenges, hospital information systems, South Africa.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1344

Authors: Sean Paulsen, Michael Casey

Abstract:

In this work, we present a self-supervised pretraining framework for transformers on functional Magnetic Resonance Imaging (fMRI) data. First, we pretrain our architecture on two self-supervised tasks simultaneously to teach the model a general understanding of the temporal and spatial dynamics of human auditory cortex during music listening. Our pretraining results are the first to suggest a synergistic effect of multitask training on fMRI data. Second, we finetune the pretrained models and train additional fresh models on a supervised fMRI classification task. We observe significantly improved accuracy on held-out runs with the finetuned models, which demonstrates the ability of our pretraining tasks to facilitate transfer learning. This work contributes to the growing body of literature on transformer architectures for pretraining and transfer learning with fMRI data, and serves as a proof of concept for our pretraining tasks and multitask pretraining on fMRI data.

Keywords: Transfer learning, fMRI, self-supervised, brain decoding, transformer, multitask training.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 128