Search results for: Consistency Subset Evaluation.
2495 A Consistency Protocol Multi-Layer for Replicas Management in Large Scale Systems
Authors: Ghalem Belalem, Yahya Slimani
Abstract:
Large scale systems such as computational Grid is a distributed computing infrastructure that can provide globally available network resources. The evolution of information processing systems in Data Grid is characterized by a strong decentralization of data in several fields whose objective is to ensure the availability and the reliability of the data in the reason to provide a fault tolerance and scalability, which cannot be possible only with the use of the techniques of replication. Unfortunately the use of these techniques has a height cost, because it is necessary to maintain consistency between the distributed data. Nevertheless, to agree to live with certain imperfections can improve the performance of the system by improving competition. In this paper, we propose a multi-layer protocol combining the pessimistic and optimistic approaches conceived for the data consistency maintenance in large scale systems. Our approach is based on a hierarchical representation model with tree layers, whose objective is with double vocation, because it initially makes it possible to reduce response times compared to completely pessimistic approach and it the second time to improve the quality of service compared to an optimistic approach.Keywords: Data Grid, replication, consistency, optimistic approach, pessimistic approach.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15632494 Combined Automatic Speech Recognition and Machine Translation in Business Correspondence Domain for English-Croatian
Authors: Sanja Seljan, Ivan Dunđer
Abstract:
The paper presents combined automatic speech recognition (ASR) of English and machine translation (MT) for English and Croatian and Croatian-English language pairs in the domain of business correspondence. The first part presents results of training the ASR commercial system on English data sets, enriched by error analysis. The second part presents results of machine translation performed by free online tool for English and Croatian and Croatian-English language pairs. Human evaluation in terms of usability is conducted and internal consistency calculated by Cronbach's alpha coefficient, enriched by error analysis. Automatic evaluation is performed by WER (Word Error Rate) and PER (Position-independent word Error Rate) metrics, followed by investigation of Pearson’s correlation with human evaluation.
Keywords: Automatic machine translation, integrated language technologies, quality evaluation, speech recognition.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 29022493 The Survey Research and Evaluation of Green Residential Building Based on the Improved Group Analytical Hierarchy Process Method in Yinchuan
Abstract:
Due to the economic downturn and the deterioration of the living environment, the development of residential buildings as high energy consuming building is gradually changing from “extensive” to green building in China. So, the evaluation system of green building is continuously improved, but the current evaluation work has the following problems: (1) There are differences in the cost of the actual investment and the purchasing power of residents, also construction target of green residential building is single and lacks multi-objective performance development. (2) Green building evaluation lacks regional characteristics and cannot reflect the different regional residents demand. (3) In the process of determining the criteria weight, the experts’ judgment matrix is difficult to meet the requirement of consistency. Therefore, to solve those problems, questionnaires which are about the green residential building for Ningxia area are distributed, and the results of questionnaires can feedback the purchasing power of residents and the acceptance of the green building cost. Secondly, combined with the geographical features of Ningxia minority areas, the evaluation criteria system of green residential building is constructed. Finally, using the improved group AHP method and the grey clustering method, the criteria weight is determined, and a real case is evaluated, which is located in Xing Qing district, Ningxia. A conclusion can be obtained that the professional evaluation for this project and good social recognition is basically the same.
Keywords: Evaluation, green residential building, grey clustering method, group AHP.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 8222492 A Validation Technique for Integrated Ontologies
Authors: Neli P. Zlatareva
Abstract:
Ontology validation is an important part of web applications’ development, where knowledge integration and ontological reasoning play a fundamental role. It aims to ensure the consistency and correctness of ontological knowledge and to guarantee that ontological reasoning is carried out in a meaningful way. Existing approaches to ontology validation address more or less specific validation issues, but the overall process of validating web ontologies has not been formally established yet. As the size and the number of web ontologies continue to grow, more web applications’ developers will rely on the existing repository of ontologies rather than develop ontologies from scratch. If an application utilizes multiple independently created ontologies, their consistency must be validated and eventually adjusted to ensure proper interoperability between them. This paper presents a validation technique intended to test the consistency of independent ontologies utilized by a common application.Keywords: Knowledge engineering, ontological reasoning, ontology validation, semantic web.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19222491 The Different Ways to Describe Regular Languages by Using Finite Automata and the Changing Algorithm Implementation
Authors: Abdulmajid Mukhtar Afat
Abstract:
This paper aims at introducing finite automata theory, the different ways to describe regular languages and create a program to implement the subset construction algorithms to convert nondeterministic finite automata (NFA) to deterministic finite automata (DFA). This program is written in c++ programming language. The program reads FA 5tuples from text file and then classifies it into either DFA or NFA. For DFA, the program will read the string w and decide whether it is acceptable or not. If accepted, the program will save the tracking path and point it out. On the other hand, when the automation is NFA, the program will change the Automation to DFA so that it is easy to track and it can decide whether the w exists in the regular language or not.
Keywords: Finite Automata, subset construction DFA, NFA.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19752490 Ensuring Consistency under the Snapshot Isolation
Authors: Carlos Roberto Valˆencio, F´abio Renato de Almeida, Thatiane Kawabata, Leandro Alves Neves, Julio Cesar Momente, Mario Luiz Tronco, Angelo Cesar Colombini
Abstract:
By running transactions under the SNAPSHOT isolation we can achieve a good level of concurrency, specially in databases with high-intensive read workloads. However, SNAPSHOT is not immune to all the problems that arise from competing transactions and therefore no serialization warranty exists. We propose in this paper a technique to obtain data consistency with SNAPSHOT by using some special triggers that we named DAEMON TRIGGERS. Besides keeping the benefits of the SNAPSHOT isolation, the technique is specially useful for those database systems that do not have an isolation level that ensures serializability, like Firebird and Oracle. We describe all the anomalies that might arise when using the SNAPSHOT isolation and show how to preclude them with DAEMON TRIGGERS. Based on the methodology presented here, it is also proposed the creation of a new isolation level: DAEMON SNAPSHOT.
Keywords: Data consistency, serialization, snapshot.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 21102489 Study of Natural Patterns on Digital Image Correlation Using Simulation Method
Authors: Gang Li, Ghulam Mubashar Hassan, Arcady Dyskin, Cara MacNish
Abstract:
Digital image correlation (DIC) is a contactless fullfield displacement and strain reconstruction technique commonly used in the field of experimental mechanics. Comparing with physical measuring devices, such as strain gauges, which only provide very restricted coverage and are expensive to deploy widely, the DIC technique provides the result with full-field coverage and relative high accuracy using an inexpensive and simple experimental setup. It is very important to study the natural patterns effect on the DIC technique because the preparation of the artificial patterns is time consuming and hectic process. The objective of this research is to study the effect of using images having natural pattern on the performance of DIC. A systematical simulation method is used to build simulated deformed images used in DIC. A parameter (subset size) used in DIC can have an effect on the processing and accuracy of DIC and even cause DIC to failure. Regarding to the picture parameters (correlation coefficient), the higher similarity of two subset can lead the DIC process to fail and make the result more inaccurate. The pictures with good and bad quality for DIC methods have been presented and more importantly, it is a systematic way to evaluate the quality of the picture with natural patterns before they install the measurement devices.
Keywords: Digital image correlation (DIC), Deformation simulation, Natural pattern, Subset size.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 27862488 Integral Domains and Their Algebras: Topological Aspects
Authors: Shai Sarussi
Abstract:
Let S be an integral domain with field of fractions F and let A be an F-algebra. An S-subalgebra R of A is called S-nice if R∩F = S and the localization of R with respect to S \{0} is A. Denoting by W the set of all S-nice subalgebras of A, and defining a notion of open sets on W, one can view W as a T0-Alexandroff space. Thus, the algebraic structure of W can be viewed from the point of view of topology. It is shown that every nonempty open subset of W has a maximal element in it, which is also a maximal element of W. Moreover, a supremum of an irreducible subset of W always exists. As a notable connection with valuation theory, one considers the case in which S is a valuation domain and A is an algebraic field extension of F; if S is indecomposed in A, then W is an irreducible topological space, and W contains a greatest element.Keywords: Algebras over integral domains, Alexandroff topology, valuation domains, integral domains.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4902487 The Study on Evaluation System and Method of Legacy System
Authors: Chao Qi, Fuyang Peng, Bo Deng, Xiaoyan Su
Abstract:
In the upgrade process of enterprise information systems, how to deal with and utilize those legacy systems affects the efficiency of construction and development of the new system. We propose an evaluation system, which comprehensively describes the capacity of legacy information systems in five aspects. Then we propose a practical legacy systems evaluation method. Base on the evaluation result, we can determine the current state of legacy system which was evaluated.Keywords: Legacy Information Systems, Evaluation IndexSystem, Evaluation Method, Evaluation Level
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16082486 Feature Subset Selection approach based on Maximizing Margin of Support Vector Classifier
Authors: Khin May Win, Nan Sai Moon Kham
Abstract:
Identification of cancer genes that might anticipate the clinical behaviors from different types of cancer disease is challenging due to the huge number of genes and small number of patients samples. The new method is being proposed based on supervised learning of classification like support vector machines (SVMs).A new solution is described by the introduction of the Maximized Margin (MM) in the subset criterion, which permits to get near the least generalization error rate. In class prediction problem, gene selection is essential to improve the accuracy and to identify genes for cancer disease. The performance of the new method was evaluated with real-world data experiment. It can give the better accuracy for classification.Keywords: Microarray data, feature selection, recursive featureelimination, support vector machines.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15332485 Current Issues on Enterprise Architecture Implementation Evaluation
Authors: Fatemeh Nikpay, Rodina Binti Ahmad, Babak Darvish Rouhani
Abstract:
Enterprise Architecture (EA) is employed by enterprises for providing integrated Information Systems (ISs) in order to support alignment of their business and Information Technology (IT). Evaluation of EA implementation can support enterprise to reach intended goals. There are some problems in current evaluation methods of EA implementation that lead to ineffectiveness implementation of EA. This paper represents current issues on evaluation of EA implementation. In this regard, we set the framework in order to represent evaluation’s issues based on their functionality and structure. The results of this research not only increase the knowledge of evaluation, but also could be useful for both academics and practitioners in order to realize the current situation of evaluations.
Keywords: Current issues on EA, implementation evaluation, Evaluation, Enterprise Architecture, Evaluation of Enterprise Architecture Implementation.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 40032484 Preparation and Characterization of M. × Piperita L. Oil Based Gel Formulation
Authors: Peeyush Kumar, Sapna Mishra, Anushree Malik, Santosh Satya
Abstract:
The essential oil of M. × piperita L. was formulated into a topical gel. The prepared gel was characterized for its pH, viscosity, spreadiblity, consistency and extrudiblity, while its stability was evaluated under different temperature conditions. The prepared M. × piperita oil gel was clear and transparent. The pH value of developed gel was 6.6, while its viscosity was 1200 cP. Spreadability and consistency of the M. × piperita oil gel was 10.7 g.cm/sec and 7 mm, respectively. The prepared gel showed good extrudiblity. During the stability studies, no significant change in pH and viscosity as a function of time for gel was observed, indicating stability of prepared formulation. The gel developed in this study is expected to forward the usage of M. × piperita essential towards commercial application.Keywords: M. × piperita L., formulation, gel, characterization, stability
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 31112483 Feature Selection for Breast Cancer Diagnosis: A Case-Based Wrapper Approach
Authors: Mohammad Darzi, Ali AsgharLiaei, Mahdi Hosseini, HabibollahAsghari
Abstract:
This article addresses feature selection for breast cancer diagnosis. The present process contains a wrapper approach based on Genetic Algorithm (GA) and case-based reasoning (CBR). GA is used for searching the problem space to find all of the possible subsets of features and CBR is employed to estimate the evaluation result of each subset. The results of experiment show that the proposed model is comparable to the other models on Wisconsin breast cancer (WDBC) dataset.Keywords: Case-based reasoning; Breast cancer diagnosis; Genetic algorithm; Wrapper feature selection
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 28512482 A Novel Microarray Biclustering Algorithm
Authors: Chieh-Yuan Tsai, Chuang-Cheng Chiu
Abstract:
Biclustering aims at identifying several biclusters that reveal potential local patterns from a microarray matrix. A bicluster is a sub-matrix of the microarray consisting of only a subset of genes co-regulates in a subset of conditions. In this study, we extend the motif of subspace clustering to present a K-biclusters clustering (KBC) algorithm for the microarray biclustering issue. Besides minimizing the dissimilarities between genes and bicluster centers within all biclusters, the objective function of the KBC algorithm additionally takes into account how to minimize the residues within all biclusters based on the mean square residue model. In addition, the objective function also maximizes the entropy of conditions to stimulate more conditions to contribute the identification of biclusters. The KBC algorithm adopts the K-means type clustering process to efficiently make the partition of K biclusters be optimized. A set of experiments on a practical microarray dataset are demonstrated to show the performance of the proposed KBC algorithm.Keywords: Microarray, Biclustering, Subspace clustering, Meansquare residue model.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16002481 Symbolic Model Checking of Interactions in Sequence Diagrams with Combined Fragments by SMV
Authors: Yuka Kawakami, Tomoyuki Yokogawa, Hisashi Miyazaki, Sousuke Amasaki, Yoichiro Sato, Michiyoshi Hayase
Abstract:
In this paper, we proposed a method for detecting consistency violation between state machine diagrams and a sequence diagram defined in UML 2.0 using SMV. We extended a method expressing these diagrams defined in UML 1.0 with boolean formulas so that it can express a sequence diagram with combined fragments introduced in UML 2.0. This extension made it possible to represent three types of combined fragment: alternative, option and parallel. As a result of experiment, we confirmed that the proposed method could detect consistency violation correctly with SMV.
Keywords: UML, model checking, SMV, sequence diagram.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14542480 Relative Radiometric Correction of Cloudy Multitemporal Satellite Imagery
Authors: Seema Biday, Udhav Bhosle
Abstract:
Repeated observation of a given area over time yields potential for many forms of change detection analysis. These repeated observations are confounded in terms of radiometric consistency due to changes in sensor calibration over time, differences in illumination, observation angles and variation in atmospheric effects. This paper demonstrates applicability of an empirical relative radiometric normalization method to a set of multitemporal cloudy images acquired by Resourcesat1 LISS III sensor. Objective of this study is to detect and remove cloud cover and normalize an image radiometrically. Cloud detection is achieved by using Average Brightness Threshold (ABT) algorithm. The detected cloud is removed and replaced with data from another images of the same area. After cloud removal, the proposed normalization method is applied to reduce the radiometric influence caused by non surface factors. This process identifies landscape elements whose reflectance values are nearly constant over time, i.e. the subset of non-changing pixels are identified using frequency based correlation technique. The quality of radiometric normalization is statistically assessed by R2 value and mean square error (MSE) between each pair of analogous band.Keywords: Correlation, Frequency domain, Multitemporal, Relative Radiometric Correction
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19702479 Gene Selection Guided by Feature Interdependence
Authors: Hung-Ming Lai, Andreas Albrecht, Kathleen Steinhöfel
Abstract:
Cancers could normally be marked by a number of differentially expressed genes which show enormous potential as biomarkers for a certain disease. Recent years, cancer classification based on the investigation of gene expression profiles derived by high-throughput microarrays has widely been used. The selection of discriminative genes is, therefore, an essential preprocess step in carcinogenesis studies. In this paper, we have proposed a novel gene selector using information-theoretic measures for biological discovery. This multivariate filter is a four-stage framework through the analyses of feature relevance, feature interdependence, feature redundancy-dependence and subset rankings, and having been examined on the colon cancer data set. Our experimental result show that the proposed method outperformed other information theorem based filters in all aspect of classification errors and classification performance.
Keywords: Colon cancer, feature interdependence, feature subset selection, gene selection, microarray data analysis.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 21272478 Generator of Hypotheses an Approach of Data Mining Based on Monotone Systems Theory
Authors: Rein Kuusik, Grete Lind
Abstract:
Generator of hypotheses is a new method for data mining. It makes possible to classify the source data automatically and produces a particular enumeration of patterns. Pattern is an expression (in a certain language) describing facts in a subset of facts. The goal is to describe the source data via patterns and/or IF...THEN rules. Used evaluation criteria are deterministic (not probabilistic). The search results are trees - form that is easy to comprehend and interpret. Generator of hypotheses uses very effective algorithm based on the theory of monotone systems (MS) named MONSA (MONotone System Algorithm).Keywords: data mining, monotone systems, pattern, rule.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 12492477 A Formal Suite of Object Relational Database Metrics
Authors: Justus S, K Iyakutti
Abstract:
Object Relational Databases (ORDB) are complex in nature than traditional relational databases because they combine the characteristics of both object oriented concepts and relational features of conventional databases. Design of an ORDB demands efficient and quality schema considering the structural, functional and componential traits. This internal quality of the schema is assured by metrics that measure the relevant attributes. This is extended to substantiate the understandability, usability and reliability of the schema, thus assuring external quality of the schema. This work institutes a formalization of ORDB metrics; metric definition, evaluation methodology and the calibration of the metric. Three ORDB schemas were used to conduct the evaluation and the formalization of the metrics. The metrics are calibrated using content and criteria related validity based on the measurability, consistency and reliability of the metrics. Nominal and summative scales are derived based on the evaluated metric values and are standardized. Future works pertaining to ORDB metrics forms the concluding note.Keywords: Measurements, Product metrics, Metrics calibration, Object-relational database.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16532476 Effect of Hull-Less Barley Flakes and Malt Extract on Yoghurt Quality
Authors: Ilze Beitane, Evita Straumite
Abstract:
The aim of the research was to evaluate the influence of flakes from biologically activated hull-less barley grain and malt extract on quality of yoghurt during its storage.
The results showed that the concentration of added malt extract and storage time influenced the changes of pH and lactic acid in yoghurt samples. Sensory properties – aroma, taste, consistency and appearance – of yoghurt enriched with flakes from biologically activated hull-less barley grain and malt extract changed significantly (p<0.05) during storage. Yoghurt with increased proportion of malt extract had sweeter taste and more flowing consistency. Sensory properties (taste, aroma, consistency and appearance) of yoghurt samples enriched with 5% flakes from biologically activated hull-less barley grain (YFBG 5%) and 5% flakes from biologically activated hull-less barley grain and 2% malt extract (YFBG 5% ME 2%) did not change significantly during one week of storage.
Keywords: Barley flakes, malt extract, yoghurt, sensory analysis.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 21272475 A Note on the Minimum Cardinality of Critical Sets of Inertias for Irreducible Zero-nonzero Patterns of Order 4
Authors: Ber-Lin Yu, Ting-Zhu Huang
Abstract:
If there exists a nonempty, proper subset S of the set of all (n+1)(n+2)/2 inertias such that S Ôèå i(A) is sufficient for any n×n zero-nonzero pattern A to be inertially arbitrary, then S is called a critical set of inertias for zero-nonzero patterns of order n. If no proper subset of S is a critical set, then S is called a minimal critical set of inertias. In [Kim, Olesky and Driessche, Critical sets of inertias for matrix patterns, Linear and Multilinear Algebra, 57 (3) (2009) 293-306], identifying all minimal critical sets of inertias for n×n zero-nonzero patterns with n ≥ 3 and the minimum cardinality of such a set are posed as two open questions by Kim, Olesky and Driessche. In this note, the minimum cardinality of all critical sets of inertias for 4 × 4 irreducible zero-nonzero patterns is identified.
Keywords: Zero-nonzero pattern, inertia, critical set of inertias, inertially arbitrary.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 11902474 Minimal Critical Sets of Inertias for Irreducible Zero-nonzero Patterns of Order 3
Authors: Ber-Lin Yu, Ting-Zhu Huang
Abstract:
If there exists a nonempty, proper subset S of the set of all (n + 1)(n + 2)/2 inertias such that S Ôèå i(A) is sufficient for any n × n zero-nonzero pattern A to be inertially arbitrary, then S is called a critical set of inertias for zero-nonzero patterns of order n. If no proper subset of S is a critical set, then S is called a minimal critical set of inertias. In [3], Kim, Olesky and Driessche identified all minimal critical sets of inertias for 2 × 2 zero-nonzero patterns. Identifying all minimal critical sets of inertias for n × n zero-nonzero patterns with n ≥ 3 is posed as an open question in [3]. In this paper, all minimal critical sets of inertias for 3 × 3 zero-nonzero patterns are identified. It is shown that the sets {(0, 0, 3), (3, 0, 0)}, {(0, 0, 3), (0, 3, 0)}, {(0, 0, 3), (0, 1, 2)}, {(0, 0, 3), (1, 0, 2)}, {(0, 0, 3), (2, 0, 1)} and {(0, 0, 3), (0, 2, 1)} are the only minimal critical sets of inertias for 3 × 3 irreducible zerononzero patterns.
Keywords: Permutation digraph, zero-nonzero pattern, irreducible pattern, critical set of inertias, inertially arbitrary.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 12302473 Least-Squares Support Vector Machine for Characterization of Clusters of Microcalcifications
Authors: Baljit Singh Khehra, Amar Partap Singh Pharwaha
Abstract:
Clusters of Microcalcifications (MCCs) are most frequent symptoms of Ductal Carcinoma in Situ (DCIS) recognized by mammography. Least-Square Support Vector Machine (LS-SVM) is a variant of the standard SVM. In the paper, LS-SVM is proposed as a classifier for classifying MCCs as benign or malignant based on relevant extracted features from enhanced mammogram. To establish the credibility of LS-SVM classifier for classifying MCCs, a comparative evaluation of the relative performance of LS-SVM classifier for different kernel functions is made. For comparative evaluation, confusion matrix and ROC analysis are used. Experiments are performed on data extracted from mammogram images of DDSM database. A total of 380 suspicious areas are collected, which contain 235 malignant and 145 benign samples, from mammogram images of DDSM database. A set of 50 features is calculated for each suspicious area. After this, an optimal subset of 23 most suitable features is selected from 50 features by Particle Swarm Optimization (PSO). The results of proposed study are quite promising.
Keywords: Clusters of Microcalcifications, Ductal Carcinoma in Situ, Least-Square Support Vector Machine, Particle Swarm Optimization.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 18012472 Loss Function Optimization for CNN-Based Fingerprint Anti-Spoofing
Authors: Yehjune Heo
Abstract:
As biometric systems become widely deployed, the security of identification systems can be easily attacked by various spoof materials. This paper contributes to finding a reliable and practical anti-spoofing method using Convolutional Neural Networks (CNNs) based on the types of loss functions and optimizers. The types of CNNs used in this paper include AlexNet, VGGNet, and ResNet. By using various loss functions including Cross-Entropy, Center Loss, Cosine Proximity, and Hinge Loss, and various loss optimizers which include Adam, SGD, RMSProp, Adadelta, Adagrad, and Nadam, we obtained significant performance changes. We realize that choosing the correct loss function for each model is crucial since different loss functions lead to different errors on the same evaluation. By using a subset of the Livdet 2017 database, we validate our approach to compare the generalization power. It is important to note that we use a subset of LiveDet and the database is the same across all training and testing for each model. This way, we can compare the performance, in terms of generalization, for the unseen data across all different models. The best CNN (AlexNet) with the appropriate loss function and optimizers result in more than 3% of performance gain over the other CNN models with the default loss function and optimizer. In addition to the highest generalization performance, this paper also contains the models with high accuracy associated with parameters and mean average error rates to find the model that consumes the least memory and computation time for training and testing. Although AlexNet has less complexity over other CNN models, it is proven to be very efficient. For practical anti-spoofing systems, the deployed version should use a small amount of memory and should run very fast with high anti-spoofing performance. For our deployed version on smartphones, additional processing steps, such as quantization and pruning algorithms, have been applied in our final model.
Keywords: Anti-spoofing, CNN, fingerprint recognition, loss function, optimizer.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4052471 The Flashbulb Memory of the Positive and Negative Events: Wenchuan Earthquake and Acceptance to College
Authors: Aiping Liu, Xiaoping Ying, Jing Luo
Abstract:
53 college students answered questions regarding the circumstances in which they first heard about the news of Wenchuan earthquake or the news of their acceptance to college which took place approximately one year ago, and answered again two years later. The number of details recalled about their circumstances for both events was high and didn-t decline two years later. However, consistency in reported details over two years was low. Participants were more likely to construct central (e.g., Where were you?) than peripheral information (What were you wearing?), and the confidence of the central information was higher than peripheral information, which indicated that they constructed more when they were more confident.
Keywords: flashbulb memory, consistency, reconstructive error, confidence
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 24242470 Integrated Subset Split for Balancing Network Utilization and Quality of Routing
Authors: S. V. Kasmir Raja, P. Herbert Raj
Abstract:
The overlay approach has been widely used by many service providers for Traffic Engineering (TE) in large Internet backbones. In the overlay approach, logical connections are set up between edge nodes to form a full mesh virtual network on top of the physical topology. IP routing is then run over the virtual network. Traffic engineering objectives are achieved through carefully routing logical connections over the physical links. Although the overlay approach has been implemented in many operational networks, it has a number of well-known scaling issues. This paper proposes a new approach to achieve traffic engineering without full-mesh overlaying with the help of integrated approach and equal subset split method. Traffic engineering needs to determine the optimal routing of traffic over the existing network infrastructure by efficiently allocating resource in order to optimize traffic performance on an IP network. Even though constraint-based routing [1] of Multi-Protocol Label Switching (MPLS) is developed to address this need, since it is not widely tested or debugged, Internet Service Providers (ISPs) resort to TE methods under Open Shortest Path First (OSPF), which is the most commonly used intra-domain routing protocol. Determining OSPF link weights for optimal network performance is an NP-hard problem. As it is not possible to solve this problem, we present a subset split method to improve the efficiency and performance by minimizing the maximum link utilization in the network via a small number of link weight modifications. The results of this method are compared against results of MPLS architecture [9] and other heuristic methods.
Keywords: Constraint based routing, Link Utilization, Subsetsplit method and Traffic Engineering.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 13852469 Performance Evaluation of Distributed Deep Learning Frameworks in Cloud Environment
Authors: Shuen-Tai Wang, Fang-An Kuo, Chau-Yi Chou, Yu-Bin Fang
Abstract:
2016 has become the year of the Artificial Intelligence explosion. AI technologies are getting more and more matured that most world well-known tech giants are making large investment to increase the capabilities in AI. Machine learning is the science of getting computers to act without being explicitly programmed, and deep learning is a subset of machine learning that uses deep neural network to train a machine to learn features directly from data. Deep learning realizes many machine learning applications which expand the field of AI. At the present time, deep learning frameworks have been widely deployed on servers for deep learning applications in both academia and industry. In training deep neural networks, there are many standard processes or algorithms, but the performance of different frameworks might be different. In this paper we evaluate the running performance of two state-of-the-art distributed deep learning frameworks that are running training calculation in parallel over multi GPU and multi nodes in our cloud environment. We evaluate the training performance of the frameworks with ResNet-50 convolutional neural network, and we analyze what factors that result in the performance among both distributed frameworks as well. Through the experimental analysis, we identify the overheads which could be further optimized. The main contribution is that the evaluation results provide further optimization directions in both performance tuning and algorithmic design.
Keywords: Artificial Intelligence, machine learning, deep learning, convolutional neural networks.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 12372468 Generalized Mean-field Theory of Phase Unwrapping via Multiple Interferograms
Authors: Yohei Saika
Abstract:
On the basis of Bayesian inference using the maximizer of the posterior marginal estimate, we carry out phase unwrapping using multiple interferograms via generalized mean-field theory. Numerical calculations for a typical wave-front in remote sensing using the synthetic aperture radar interferometry, phase diagram in hyper-parameter space clarifies that the present method succeeds in phase unwrapping perfectly under the constraint of surface- consistency condition, if the interferograms are not corrupted by any noises. Also, we find that prior is useful for extending a phase in which phase unwrapping under the constraint of the surface-consistency condition. These results are quantitatively confirmed by the Monte Carlo simulation.
Keywords: Bayesian inference, generalized mean-field theory, phase unwrapping, statistical mechanics.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16492467 Interactive Model Based On an Extended CPN
Authors: Shuzhen Yao, Fengjing Zhao, Jianwei He
Abstract:
The UML modeling of complex distributed systems often is a great challenge due to the large amount of parallel real-time operating components. In this paper the problems of verification of such systems are discussed. ECPN, an Extended Colored Petri Net is defined to formally describe state transitions of components and interactions among components. The relationship between sequence diagrams and Free Choice Petri Nets is investigated. Free Choice Petri Net theory helps verifying the liveness of sequence diagrams. By converting sequence diagrams to ECPNs and then comparing behaviors of sequence diagram ECPNs and statecharts, the consistency among models is analyzed. Finally, a verification process for an example model is demonstrated.
Keywords: Consistency, liveness, Petri Net, sequence diagram.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15992466 Terrain Evaluation Method for Hexapod Robot
Authors: Tomas Luneckas, Dainius Udris
Abstract:
In this paper a simple terrain evaluation method for hexapod robot is introduced. This method is based on feet coordinate evaluation when all are on the ground. Depending on the feet coordinate differences the local terrain evaluation is possible. Terrain evaluation is necessary for right gait selection and/or body position correction. For terrain roughness evaluation three planes are plotted: two of them as definition points use opposite feet coordinates, third coincides with the robot body plane. The leaning angle of body plane is evaluated measuring gravity force using three-axis accelerometer. Terrain roughness evaluation method is based on angle estimation between normal vectors of these planes. Aim of this work is to present a simple method for embedded robot controller, allowing to find the best further movement settings.Keywords: Hexapod robot, pose estimation, terrain evaluation, terrain roughness.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1788