Fingerprint Verification Methods Using

Delaunay Triangulations

Manuel Flores, Gualberto Torres, Gina García, and Miguel Licona

Instituto Politécnico Nacional, Sección de Estudios de Posgrado e Investigación, Unidad Culhuacán, México

Abstract: This paper presents a modification for robust minutiae based fingerprint verification methods that use Delaunay triangulations. The purpose of this modification is to decrease the number of comparison operations and the error rates within the matching process, by doing a full analysis of the Delaunay triangles. From this full analysis, a modified method was proposed. The identified minutiae represent nodes of a coZnnected graph composed of triangles. With this technique, the minimum angle over all triangulations is maximized, which gives local stability to the constructed structures against rotation and translation variations. Geometric thresholds and minutiae data were used to characterize the triangulations created from input and template fingerprint images. The effectiveness of the proposed modification is confirmed with calculations of  False Acceptance Rate (FAR), False Rejected Rate (FRR) and Equal Error Rate (EER) over FVC2002 databases compared to other approaches results.

Keywords: Angle of orientation, delaunay triangulation, EER, fingerprint, geometric thresholds.

Skyline Recommendation in Distributed Networks

Zhenhua Huang¹, Jiawen Zhang¹, Zheng Liu¹, Bo Zhang², and Dong Wang³

¹School of Electronics and Information, Tongji University, China

²College of Information Mechanical and Electrical Engineering, Shanghai Normal University, China

³College of Computer Science and Information Engineering, Shanghai Institute of Technology, China

Abstract: Skyline recommendation technology has recently received a lot of attention in the database community. However, the existing works mostly focus on how to obtain skyline objects from fine-grained data in centralized environments. And the time cost of skyline recommendation will increase exponentially as the number of data and skyline recommendation instructions increases, which will seriously influence the recommendation efficiency. Motivated by the above fact, this paper proposes an efficient algorithm Skyline Recommendation Algorithm in Distributed Networks (SRADN) in Super-Peer Architecture (SPA) distributed networks to handle multiple subspace skyline recommendations by prestoring the set of skyline snapshots under the cost constraint of maintenance and communication. The proposed SRADN algorithm fully considers the characteristic of storage and communication of SPA networks, and uses the map/reduce distributed computation model. The SRADN algorithm can quickly produce the optimal set of skyline snapshots through the following two phases: Heuristically constructing the initial set of snapshots, and adjusting the set of snapshots based on the genetic algorithm. The detailed theoretical analyses and extensive experiments demonstrate that the proposed SRADN algorithm is both efficient and effective.

Keywords: Skyline recommendation, distributed networks, map/reduce, genetic algorithm.

Received August 30, 2014; accepted December 16, 2014

Proficient Decision Making on Virtual Machine

Creation in IaaS Cloud Environment

Radhakrishnan Ayyapazham and Kavitha Velautham

Department of Computer Science, Anna University, India

Abstract: Cloud computing is a most fascinated technology that is being utilized by IT companies to reduce their infrastructure setup cost by outsourcing data and computation on demand. Cloud computing offer services in three basic models such as SaaS, PaaS and Infrastructure as a Service (IaaS). Where IaaS is one of the fundamental cloud service model in which cloud provider offers Virtual Machines (VMs) as resources to cloud customers through virtualization. The VMs act as dedicated computer system to consumers which are created on physical hosts of cloud provider. Making decision of physical host selection for VMs creation is a challenging task for cloud provider. Any deficiency of this selection causes VMs migration in middle of computation or restart computation from the scratch; these would sternly affect profit and trust of cloud provider. In this paper, we proposed a novel methodology to handle VMs creation and allocation for IaaS service. The proposed methodology employs a genetically weight optimized neural network component in each host to predict their near future availability during VMs creation. We analyses the host load prediction performance of various neural networks through real time host load values. Also we proposed a proficient decision making algorithm named Future Load Based Virtual machine Creation (FLBVC) to choose appropriate launching hosts for VMs. The performance of our methodology is validated using CloudAnalyst tool. The results demonstrated that our proposed approach reduces response time of cloud customers and rental cost of VMs. 

Keywords: IaaS, VMs, jordon neural network, genetic algorithm, service level agreements, FLBVC.

Recieved October 20, 2013; accepted July 2, 2014

Online Approach to Handle Concept Drifting Data Streams using Diversity

Parneeta Sidhu and Mohinder Bhatia

Division of CoE, Netaji Subhas Institute of Technology, University of Delhi, India

Abstract: Concept drift is the trend observed in almost all real time applications. Many online and offline algorithms were developed in the past to analyze this drift and train our algorithms. Different levels of diversity are required before and after a drift to get the best generalization accuracy. In our paper, we present a new online approach Extended Dynamic Weighted Majority with diversity (EDWM) to handle various types of drifts from slow gradual to abrupt drifts. Our approach is based on the Weighted Majority(WM) vote of the ensembles containing different diversity levels. Experiments on the various artificial and real datasets proved that our proposed ensemble approach learns drifting concepts better than the existing online approaches in a resource constrained environment.

Keywords: Online learning, ensemble, concept drift, data streams, diversity.

Received October 29, 2013; accepted December 16, 2014

A Decision Support System Using Demographic

Issues: A Case Study in Turkey

Suat Secgin and Gokhan Dalkilic

Department of Computer Engineering Department, Dokuz Eylul University, Turkey

Abstract: The demographic distribution of people by cities is an important parameter to address the people’s behaviour. To distinguish people behaviour is useful for companies to understand the customer behaviour. In this article, a case study covering all 81 cities in Turkey and measuring 35 topics for each of them is handled. By using these topics and cities, it is investigated that how the cities are clustered. Because its efficiency, the Agglomerative hierarchical clustering and the K-medoids clustering methods in rapidminer data mining software are used to cluster the data. To measure the efficiency of the agglomerative clustering algorithm, the Cophenetic Correlation Coefficient (CPCC) is used. After clustering, the results are inserted into a geographic information system to depict the results in a Turkey map. The results show that, the cities distributed in the same geographical areas are in the same clusters with some exempts. On the other hand, some cities those are in different provinces show the same behaviour. The results of the study can also be used as a decision support system for a customer relations management. 

Keywords: Agglomerative clustering, customer behaviour, data mining, decision support.

Revived July 24, 2014; accepted August 16, 2015

Full Text

Semantic Similarity based Web Document Classification Using Support Vector Machine

Kavitha Chinniyan, Sudha Gangadharan, and Kiruthika Sabanaikam

Department of Computer Science and Engineering, PSG College of Technology, India

Abstract: With the rapid growth of information on the World Wide Web (WWW), classification of web documents has become important for efficient information retrieval. Relevancy of information retrieved can also be improved by considering semantic relatedness between words which is a basic research area in fields of natural language processing, intelligent retrieval, document clustering and classification, word sense disambiguation etc. The web search engine based semantic relationship from huge web corpus can improve classification of documents. This paper proposes an approach for web document classification that exploits information, including both page count and snippets. To identify the semantic relations between the query words, a lexical pattern extraction algorithm is applied on snippets. A sequential pattern clustering algorithm is used to form clusters of different patterns. The page count based measures are combined with the clustered patterns to define the features extracted from the word-pairs. These features are used to train the Support Vector Machine (SVM), in order to classify the web documents. Experimental results demonstrate 5% and 9% improvement in F1 measure for Reuters 21578 and 20 Newsgroup datasets in the classifier performance.

Keywords: Document classification, text mining, SVM, latent semantic indexing.

A Feature Model Metrics-Based Approach to

Developa Software Product Line

Yacine Djebar¹, Mohamed Kimour², and Nouredine Guersi³

¹Department of Computer Science, University of 08 Mai 1945, Algeria

²Department of Computer Science, University of Badji Mokhtar, Algeria

³Department of Electronics, University of Badji Mokhtar, Algeria

Abstract: In recent years, the Software Product Line (SPL) is becoming a mainstream strategy in software development. The high reusability and the great derivability by modelling common and variable artefacts are undoubtedly its significant strengths. Taking advantage of these strengths requires a design of efficient product line. Often, most existing SPL design approaches build on feature modelling by analysis of existing similar products. However, existing feature-based modelling techniques lack analysis support for building SPL with regard to different stakeholder’s views. In this paper, we propose an approach based on analyzing and assessing process for creating expressive structure of an SPL. Such a process provides stakeholders with a set of optimal structures of SPL in different models and a set of metrics. In doing so, we argue that we facilitate the selection of appropriate predefined products structures under the form of a set of configuration-views.

Keywords: SPL, feature diagram, multi-criteria analysis, multi-view configuration.

Received June 9, 2014; accepted January 27, 2015

Shearing Invariant Texture Descriptor from a Local

Binary Pattern and its Application in Paper Fingerprinting

Omar Wahdan, Mohammad Nasrudin, and Khairuddin Omar

Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia, Malaysia

Abstract: In this paper, a Shearing Invariant Texture Descriptor (SITD) is proposed, which is a theoretically and computationally simple method based on the Rotation invariant Local Binary Pattern (Rot-LBP) descriptor. In real-world applications using flatbed scanners, such as paper texture fingerprinting, it’s common for a sheet of paper to rotate during the image acquisition process. Because the rotation is usually not based on the paper’s geometrical centre pivot, the produced image is deformed with irregular rotation resulting in shearing transforms. To tackle the shearing problem, the proposed SITD selects a few patterns from the conventional Rot-LBP to achieve either horizontal or vertical invariance. This paper presents the construction of the SITD operators and their performance in recognizing self-developed and standard image datasets, including real paper texture and Outex images, as well as those with distinctive shapes. The images were distorted with only a shearing transform. The self-developed images were distorted manually, while the standard images were distorted by software. The proposed description method achieved up to 100% correctly recognition rate in all the tested datasets based on the horizontal shear invariant operator. In addition to the accurate performance in all the conducted experiments, the operator significantly outperformed the Rot-LBP and another benchmark method, the Shearing Moment Invariant (SMI). The superiority of the descriptor in recognizing different types of patterns demonstrate its ability to be used in applications where the shearing transform is present.

Keywords: Shear invariant descriptor, texture fingerprint, image acquisition, local binary pattern, outex framework.

Received September 6, 2014; accepted February 4, 2015

Weighted Delta Factor Cluster Ensemble Algorithm for Categorical Data Clustering in Data Mining

Sarumathi Sengottaian¹, Shanthi Natesan², and Sharmila Mathivanan³

¹Department of Information Technology, K.S.R College of Technology, India

²Department of Computer Science and Engineering, Nandha Engineering College, India

³Department of Information Technology, M.Kumarasamy College of Engineering, India

Abstract: Though many cluster ensemble approaches came forward as a potential and dominant method for enhancing the robustness, stability and the quality of individual clustering systems, it is intensely observed that this approach in most cases generate a final data partition with deficient information. The primary ensemble information matrix generated in the traditional cluster ensemble approaches results only the cluster data point relations with unknown entries. This paper mainly denotes the improved analysis of the Link based Cluster Ensemble (LCE) approach which overcomes the problem of degrading the quality of clustering result and in particular it presents an efficient novel Weighted Delta Factor Cluster Ensemble algorithm (WDFCE) which enhances the refined matrix by augmenting the values of similitude measures between the clusters formed in the Bipartite cluster graph. Subsequently to obtain the final ultimate cluster result, the pairwise-similarity consensus method is used in which K-means clustering technique is applied over the similarity measures that are formulated from the Refined Similitude Matrix (RSM). Experimental results on few UCI datasets and synthetic dataset reveals that this proposed method always outperforms the traditional cluster ensemble techniques and individual clustering algorithms.

Keywords: Clustering, cluster ensembles, consensus function, data mining, refined matrix, similitude measures.

Cuckoo Search with Mutation for Biclustering of Microarray Gene Expression Data

Balamurugan Rengeswaran, Natarajan Mathaiyan, and Premalatha Kandasamy

Department of Computer Science and Engineering, Bannari Amman Institute of Technology, India

Abstract: DNA microarrays have been applied successfully in diverse research fields such as gene discovery, disease diagnosis and drug discovery. The roles of the genes and the mechanisms of the underlying diseases can be identified using microarrays. Biclustering is a two dimensional clustering problem, where we group the genes and samples simultaneously. It has a great potential in detecting marker genes that are associated with certain tissues or diseases. The proposed work finds the significant biclusters in large expression data using the Cuckoo Search with Mutation (CSM). The cuckoo imitates its egg similar to host bird’s egg using a mutation operator. Mutation is used for exploration of search space, more precisely to allow candidates to escape from local minima. It focuses on finding maximum biclusters with lower Mean Squared Residue (MSR) and higher gene variance. A qualitative measurement of the formed biclusters with a comparative assessment of results is provided on four benchmark gene expression dataset. To demonstrate the effectiveness of the proposed method, the results are compared with the swarm intelligence techniques Binary Particle Swarm Optimization (BPSO), Shuffled Frog Leaping (SFL), and Cuckoo Search with Levy flight (CS) algorithm. The results show that there is significant improvement in the fitness value.

Keywords: Biclustering, CS, BPSO, SFL, levy flight, gene expression data, mutation.

Received January 1, 2014, accepted July 22, 2014

Feature Selection Algorithm Based on Correlation between Muti Metric Network Traffic Flow Features

Yongfeng Cui^1,2, Shi Dong^1,2,3, and Wei Liu²

¹School of Computer Science and Technology, Huazhong Universtiy of Science and Technology, China

²School of Computer Science and Technology, Zhoukou Normal University, China

³Department of Computer Science and Engineering, Washington University in St Louis, USA

Abstract: Traffic identification is a hot issue in recent years, in order to overcome shortcomings of port-based and Deep Packet Inspection (DPI), machine learning algorithm has gained wide attention, but nowadays research focus on traffic identification based on full packets dataset, which would be a great challenge to identify online traffic flow. It is a way to overcome this shortcoming by considering the sampled flow records as identification object. In this paper, flow records NOC_SET is constructed as dataset, and inherent NETFLOW and extended flow metrics are regarded as features. This paper proposes feature selection algorithm MSAS to select features with high correlation. And classical machine learning algorithms are used to identify traffic. Experimental results show that machine learning flow identification algorithm based on sampled flow records has almost the same identification results as method based on full packets dataset, and the proposed feature selection algorithm MSAS can improve the result of application identification.

Keywords: Port identification, deep packet inspection, netflow flow, machine learning.

Received Febrauary 5, 2014; accepted April 2, 2015

Mapping XML to Inverted Indexed Circular Linked Lists

Teng Lv¹, Ping Yan², and Weimin He³

¹School of Information Engineering, Anhui Xinhua University, China

²School of Science, Anhui Agricultural University, China

³Department of Computing and New Media Technologies, University of Wisconsin-Stevens Point, USA

Abstract: Extensible Markup Language (XML) has become the de facto standard for data exchange on the World Wide Web and is widely used in many fields, so it is urgent to develop some efficient methods to manage, store, and query XML data. Traditional methods use relational databases to store XML data which take advantage of mature technologies of relational databases. But it needs to map XML schemas to relational schemas, then rewrite XML queries to SQL queries, and finally, transform returned SQL-style results to XML-style results again. One possible solution to this is to store XML data directly and query it directly by XML query languages. In this paper, we research the problem of how to map XML data so that storing and querying it can be efficient. We propose the following framework to gain the goal: Firstly, we map a given XML data tree to a set of inverted indexed circular list, in which the relationships between parent and child nodes (and also ancestor and descendent nodes) are preserved. Then, an XML schema tree is used to guide and improve the efficiency of querying the corresponding XML data tree, which is generated from the given XML data tree. Finally, an efficient algorithm is given to query the XML data tree by using the corresponding set of inverted indexed circular list and its schema. The algorithms analysis and experiments prove the efficiency of our method over naïve method.

Keywords: XML, mapping, query, schema.

Received August 12, 2014; accepted December 23, 2014