WSEAS Transactions on Computer Research
Print ISSN: 1991-8755
E-ISSN: 2415-1521
Volume 4, 2016
Notice: As of 2014 and for the forthcoming years, the publication frequency/periodicity of WSEAS Journals is adapted to the 'continuously updated' model. What this means is that instead of being separated into issues, new papers will be added on a continuous basis, allowing a more regular flow and shorter publication times. The papers will appear in reverse order, therefore the most recent one will be on top.
Volume 4, 2016
Title of the Paper: Creating Words by Inflexion and Derivation in RDFCFL Graphs
Authors: Alena Lukasová, Martin Žáček
Abstract: This paper presents an approach to express morphology and syntax rules of specifications of basic terminology of English language. The article continues the previous work of these authors in English Morphology in RDFCFL graphic metalanguage following the methodology of Andres Carstairs-McCarthy. This article aims are creating words by inflexion and derivation.
Keywords: RDF model, Clausal Form Logic (CFL), RDFCFL formal system, morphological rules, Carstairs-McCarthyˈs, morphology
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #23, pp. 208-212
Title of the Paper: On the Impact of Using Mixed Real Number Representations in a Graphics Pipeline
Authors: Ovidiu Sicoe, Mircea Popa
Abstract: This paper aims to analyse how the accuracy of rendering 3D models on 2D surfaces is affected by using different real number representation inside different stages of a graphics pipeline. Also, an optimal format that would not alter the final product in a significant way was a target of our research. Additionally, a great number of 2D projections for each 3D Model is targeted. In order to achieve our goals, we have modified an own software implementation of the OpenGL ES 1.1 specification so that it could be easily specified which real number formats are used in each stage.
Keywords: Fixed-Point, Floating-Point, Graphics pipeline, instrumentation
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #22, pp. 202-207
Title of the Paper: Governmental Cloud Security Problems
Authors: Roumen Trifonov
Abstract: There are no guidelines to define a generic security framework that allows assessing and benchmarking Governmental Cloud security and the proper problems. The recent European studies testify that so far is no comprehensive analysis of the security frameworks of currently running or planned governmental Cloud deployments. The present article aims to outline the security problems and some key points for building Governmental Cloud Security Framework including Risk Assessment, Security Measures, Service Level Agreement, Security Certification and Incident Reporting.
Keywords: Governmental cloud, risk assessment, security measures, service level agreement, security certification, incident reporting
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #21, pp. 194-201
Title of the Paper: Cooperative Human-Machine Interfaces for Safety of Intelligent Transport Systems: Requirements, Development and Assessment
Authors: Aleksandr Orekhov, Anastasiia Orekhova, Vyacheslav Kharchenko
Abstract: In this work, the analysis into the available solutions regarding the cooperative intelligent transport systems (ITS) and their human-machine interfaces (HMI) has been conducted. The standards and recommendations in the area of transport systems interfaces design have been analysed. Profiling of the requirements to cooperative human-machine interface (CHMI) for such systems including requirements to usability and safety is based on a set of standards for ITSs. An approach and a design technique of human-machine interfaces for ITSs are suggested. This approach is based on designing the cooperative humanmachine interfaces for intelligent transport systems. The architecture of cloud-based CHMI for intelligent transport systems has been developed. The system consists of three main parts: client end, server end, and coreproject that includes data models for the communication protocol and the common utility functions. The prototype of software system for CHMI for ITS is described. The prototype has been examined in the laboratory conditions. The assessment using the GOMS method has allowed to calculate the efficiency to execute the task of user interaction with the system.
Keywords: cooperative human-machine interfaces, green human-machine interfaces, intelligent transport systems, cloud computing, GOMS method, GPS, safety, assessment
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #20, pp. 183-193
Title of the Paper: Assurance Case Driven Design for Internet of Things
Authors: Vladimir Sklyar, Vyacheslav Kharchenko
Abstract: Assurance (Security and Safety) Case is a proven-in-use methodology to demonstrate a system compliance with security and safety critical requirements. An advanced approach to improve Assurance Case is proposed in a view of Assurance Case Driven Design (AC DD). A practical using of AC DD lays in cost-effectiveness improvement of certification and licensing processes. We analyze basic mathematical models and methods to improve a known formal notation at the top level. As a result we develop Claim-Argument-Evidence-Criteria (CAEC) notation as well as Development-Verification&Validation-Assurance Case (DVA) notation for AC DD implementation. This approach is implemented for the Internet of Things (IoT). Low-energy informed assessment has to be added to the IoT Assurance Case. Assurance Case concept for the IoT safety critical applications is developed and demonstrated.
Keywords: Internet of Things, Assurance Case, Safety and Security Life Cycle
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #19, pp. 173-182
Title of the Paper: A Novel Design and Implementation of a Real-time Wireless Video and Audio Transmission Device
Authors: J. K. Saha, Md. M. A. Ghuri, Md. M. R. Mamur, M. J. Hossain, T. A. Chowdhury, B. Paul
Abstract: A wireless communication system consolidated with real-time video transmission and half duplex voice communication features is implemented in this project. The system disseminates live video with synchronized audio both via wired and wireless means. LAN, Wi-Fi and GSM network are the modes of communication employed in transmission of video and audio data. Video decoding, processing and altogether transcoding the captured video for transmission is executed by Raspberry Pi. Pi camera is incorporated with the Raspberry Pi to serve the purpose of capturing video and still images. The audio transmission system is designed utilizing the module SA828. Security is ensured by automatically generated random and undisclosed IP address. The system is drafted ensuring portability, ease of operation, reliability and cost optimization. Some specific applications of this project include remote monitoring and military tactical surveillance operations
Keywords: Wireless, Wi-Fi, GSM, Raspberry Pi, SA828, Transcoding
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #18, pp. 161-172
Title of the Paper: Clustering Utilization for Coverage Improvement in Wireless Border Surveillance Networks
Authors: Zhilbert Tafa
Abstract: In many border surveillance applications (such as military, homeland security, etc.), the wireless sensor networks cannot be deployed manually and the barrier coverage breaks can appear along a given surveillance line. This paper introduces a cluster-based algorithm and new metrics to determining the number and the positions of the additional nodes needed to be deployed by drones, robots, or moved in a network; in order to fill the gaps in a randomly deployed network. Simulation results show that the proposed algorithm optimizes the number of additional nodes and outperforms the alternative in 52,15% of cases while it performs similarly in the rest of the cases. The machine learning classification algorithms are used to show that the decision on choosing one or another algorithm is highly classifiable. Precisely, the proposed algorithm is shown to be the approach of choice in implementations where the sensing range is relatively small.
Keywords: algorithm, barrier coverage, clustering; machine learning, decision trees, neural networks, support vector machines, wireless sensor networks
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #17, pp. 153-160
Title of the Paper: A Novel Approach to Lifelong Learning for Robotic Response to Gesture
Authors: Paul M. Yanik, Anthony L. Threatt, Jessica Merino, Joe Manganelli, Johnell O. Brooks, Keith E. Green, Ian D. Walker
Abstract: For unskilled or impaired users, intuitive interfaces are seen as key to the wide adoption of assistive robotic agents. Interfaces based on gesture offer a natural communication modality for Human-Robot Interaction which facilitates ease of use for the target population. However, as the abilities of the user change with time, their capacity to properly perform a given command choreography may degrade. Therefore, it is critical to the longterm success of the interface that it learn the behavioral intentions of the user and adapt accordingly throughout the lifespan of the agent. This paper describes a clustering approach to an agent’s lifelong learning based on the Growing Neural Gas algorithm. A simulated user issues reinforcement feedback so as to progressively refine an agent’s actions toward a goal configuration. Several network distance metrics for neighborhood learning within connected clusters are compared for their relative burden to the user-trainer in terms of speed of convergence. It is shown that the proposed method learns new gestures while retaining past learning with few training iterations.
Keywords: Human-robot interaction, Growing Neural Gas, Lifelong learning, Q-learning
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #16, pp. 138-152
Title of the Paper: Multiclassifier Systems for Sequential Recognition Based on the Concept of Meta-Bayes Classifier
Authors: Marek Kurzynski, Marcin Majak, Karol Puchala
Abstract: In the sequential recognition there exist dependencies among the successive objects to be classified. In this study two original multiclassifier (MC) systems for the sequential recognition are developed. In the first MC systems base classifiers are defined for particular steps of sequential recognition independently, whereas in the second MC system base classifiers classify an object at the current step on the base of its features and features of previous objects. Both MC systems in combining procedure uses original concept of meta-Bayes classifier and produces decision according to the Bayes rule. The performance of both MC systems were evaluated experimentally and compared with six state-of-the-art sequential recognition methods using computer generated data. Results obtained in experiments imply that MC system is effective approach, which improves recognition accuracy in sequential decision scheme.
Keywords: Multiclassifier system, Sequential recognition, Probabilistic model, Meta-Bayes classifier
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #15, pp. 132-137
Title of the Paper: Healthcare, Big Data and Cloud Computing
Authors: Adriana Alexandru, Cristina Adriana Alexandru, Dora Coardos, Eleonora Tudora
Abstract: The rapidly expanding field of Big Data is a result of the data deluge generated by the rapid increase in the use of mobile devices and social media. It has provided tools to collect, store, manage, and analyze huge volumes of data (either structured, semi-structured or unstructured) produced by current healthcare systems. Big Data Analytics has started to play a crucial role in the evolution of healthcare practices and research, being now applied towards aiding the process of care delivery and disease exploration. Cloud computing enables all Big Data operations through the provision of large storage and processing power. This paper introduces all of these notions in the context of healthcare. It discusses the advantages, but also challenges brought by Big Data to this field. It also proposes a Big Data Architecture for healthcare. Finally, it presents some of the areas of application of Big Data to healthcare in Romania.
Keywords: Big Data, Big Data Analytics, Cloud Computing, Healthcare, semi-structured data, unstructured data, NoSQL databases, Hadoop
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #14, pp. 123-131
Title of the Paper: Methodology of the Use of Meteorological Radar MMR50 and Other Meteorological Systems for Purpose Determination Risk of Flash Floods
Authors: David Šaur
Abstract: This article aims to determine the resulting risk of flash floods on the basis of outputs of meteorological radar MMR50, numerical weather prediction model GFS, aerological radiosonde Prostejov, Flash Flood Guidance and Integrated Warning Service System of Czech Hydrometeorological Institute. This methodology is used for the emergency management of individual municipalities with extended powers of the Zlín Region. This article also includes a description of the basic principles of radar rainfall measurements and all radar products that are part of software equipment of the meteorological radar MMR50. The first chapter focuses on basic description of meteorological radar, the principle of transfer of radar reflectivity on the intensity of precipitation, including a description of radar outputs. Use of radar outputs for measurement and prediction of convective precipitation is demonstrated on historical weather conditions of July 24, 2015, when local flash floods were in the central part of the Zlín region. In the last chapter, the theory of methodology is described with use of meteorological sources for measuring and predicting thunderstorms. The theory is verified in the same situation as in the previous case study. Outputs of methodology are compared with the station and other meteorological data in the Zlin Region.
Keywords: Radar meteorology, flash floods, thunderstorms, convective precipitation, crisis management, meteorological systems, numerical weather prediction models
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #13, pp. 109-122
Title of the Paper: Risk Management Quality in Selected Critical Facilities
Authors: Dana Procházková, Jan Procházka
Abstract: Series of events from recent years connected with failure of critical facilities show their high importance. The critical facilities under account represent multistage mutually overlapping systems, i.e. great complex systems, the type of which is a system of systems. Present paper shows the very advanced procedure of work with risks ensuring the safety of facilities under account that is based on the system safety management; i.e. on the combination of principles: All-Hazard-Approach and Defence-In-Depth; safety culture; and on special management procedure of work with risks in time; i.e., it shows: the concept of critical facility safety; the way of work with risks; the method of critical complex facility safety management model building; and the model for critical complex facility safety management in time. In the second part the paper summarizes the results of inspections directed to judgement of consistency of real facilities safety performance with demands of ideal model formed on the facility integral safety concept.
Keywords: complex critical facility, provision of territory services, security, safety, concept of facility safety formation, model for facility safety management in time, results of inspections
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #12, pp. 96-108
Title of the Paper: Applications of Chaotic Maps and Coding for Secure Transmission of Images over Wireless Channels
Authors: Mona F. M. Mursi, Ayman H. Abd El-Aziem
Abstract: In this paper we present the applications of image encryption techniques and channel coding techniques, to design scheme effective for secure image transmission over wireless channels, we present three proposed scheme as follow: First scheme apply image encryption by a combination of hybrid chaotic maps using Baker map and our proposed H?non chaotic map 3 in three different modes of operations, beside using fractional Fourier transform (FRFT). Second image encryption scheme is combine the main advantages of FRFT, Arnold cat map for confusion and our proposed H?non chaotic map 3 for diffusion. Third scheme based on combination of hybrid chaotic encryption and low density parity check (LDPC) to secure transmission of image over wireless channel. Our proposed scheme improve throughput. The proposed scheme enhanced the performance parameters and achieved both security and reliability of image transmission over wireless channels.
Keywords: Image encryption, Chaotic maps, ECC
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #11, pp. 86-95
Title of the Paper: Recognition of Metonymy by Tagging Named Entities
Authors: H. Burcu Kupelioglu, Tankut Acarman, Tassadit Amghar, Bernard Levrat
Abstract: Metonymy is referential phenomenon in which one entity is referred by another one on the base of the existence of a relation between the two entities. It occurs very often in texts and so its recognition and resolution is required to be fulfilled for a lot of Natural Language Processing Applications. Among the methodologies and domains implied in the achievement of these tasks we can cite semantic classifiers, discourse understanding methodology, unsupervised and statistical methods. In this paper we propose to expand existing approaches by a preliminary tagging of named entities of the text with the Stanford NER program and using the argument structure of predicates as they figure in the WordNet thesaurus. We show how we can so eliminate a of lot of work which without this should have been made by human.
Keywords: Metonymy Recognition, Named Entity, Natural Language Processing, Stanford CoreNLP, WordNet
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #10, pp. 81-85
Title of the Paper: An Integrated Framework of Multi Software Agent and Cloud Forensics
Authors: Arwa Ibrahim Ahmed
Abstract: Cloud computing is one of the most prominent technological trends as it provides a number of digital storage and services. However, evidence acquisition and investigation of violations occurring in cloud environment is still a critical issue in cloud forensics. This paper develops an integrated framework that consists two main layers are Cloud Forensics Layer (Cloud Service Provider, law enforcement, forensics investigators and cloud users), and Multi Agent System (MAS) architecture layer that includes two agents: Cloud Acquiring Agent (CAA) and Cloud Forensics Agent (FCA). The model is then tested based on the Cloud Forensic Capability Matrix (CFCM) using a sample covered the cloud actors who are cloud forensic practitioners and experts from cloud users, CSPs, academia, cloud broker, cloud investigators, and law enforcement. The proposed framework was supported, providing preliminary evidence that the two layers can and should be integrated. The integrated framework helps build a conceptual bridge to the cloud forensics.
Keywords: Cloud Computing Forensics, Multi Agent System, Cloud Forensic Capability Matrix, Digital Forensics, Integrated Framework, and Cloud Forensic Investigative Architecture
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #9, pp. 72-80
Title of the Paper: Architecture and Simulation for the Fault Tolerance of Information Technology Difficulties
Authors: Milena Janakova
Abstract: The topic of this paper is focused on better information technology (IT) implementation and easier work with data. IT products have an important place in people’s everyday activities. IT users work with data via a number of applications. They use computers, mobile devices, tablets as well as cloud solutions. Information technology includes verified methodologies, methods and tools for optimal IT implementation. Architecture plays the key role, but there are still controversies. The problems are budgets, time, aims and quality; therefore, many IT users run spreadsheets, such as MS Excel, for data analyses via sensitivity analyses, contingency tables or summaries. Implemented software must mirror the high level of diversity and variety of relations in the global information society, and also offer an intuitive interface for immediate use. In this situation, actively merged architecture and simulation help. Contribution this paper is the design of complex work with data based on a simulation via available intelligences. Specified innovation is focused on searching unexpected relations between data. There are created objects working with available intelligences as artificial, business, competitive, customer, swarm and computational. Designed solution was applied to “Open Case Duration”. Such an approach offers a better support of fault tolerance of mistakes and errors in IT implementation and data processing. Simulations and intelligences show hidden changes in stored data and IT designers with data scientists better understand new relations.
Keywords: Architecture, data science, information technology, intelligence, simulation
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #8, pp. 64-71
Title of the Paper: Transfer and Validation of Web-Based Solutions in the Area of Transport and Logistics
Authors: Viktorija Ponomarenko, Leonids Novickis, Mihails Sotnichoks, Antanas Mitasiunas
Abstract: Technology transfer is a bridge between research organizations and industry. It is an important task to find right way how to present a new technology for the audience. In this research new technology is web-based solutions in the area of transport and logistics. The goal of the paper is to validate the readiness of developed IT solution to be transferred to the market. To achieve this goal was applied technology transfer model developed during the Baltic Sea Region INTERREG Program project. The web-based solution was validated by technology developer and potential end-users. First results of validation are presented in this paper.
Keywords: information technology transfer, Web-based solutions, transport and logistics, validation
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #7, pp. 56-63
Title of the Paper: Training Video on Low-Skilled Workers
Authors: Rula Sharqi, Ammar Kaka
Abstract: Accident in construction sites are a result of several reasons such as the worker’s training, lack of awareness, site working conditions, and the design of the project. An understanding of what is the best tool to enhance the worker health and safety awareness is a needed especially for low-skilled worker. Since the steel structures construction continues to be one of the most hazard jobs in the construction industry, the visual training process to enhance the workers understanding becomes essential. The objective of this paper, which is part of my Ph.D. is to presents the development of a system that can demonstrate the hazards of site activities in an advanced visualization tool, the system can determine whether this activity exist in a given digital image by processing digital color information. For the development of the image processor, color image processing is employed instead of grayscale image processing commonly used in previous researches.
Keywords: Health and Safety, Construction Industry, Activity and Hazards
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #6, pp. 46-55
Title of the Paper: A Computer Vision Application to Detect Unwanted Weed in Early Stage Crops
Authors: Beatriz Nathalia Serrato Panqueba, Carlos Arturo Castillo Medina
Abstract: This article shows an application for computer vision to detect unwanted weed in crops from one region with more agricultural impact in Colombia. We took images in different crops, wanting estimate population degree from arvenses species. An Image processing was developed to obtain regions of interest were finally processed through neural networks.
Keywords: machine vision, weed identification, classical classifier, neuronal networks, unwanted weed, weed in crops
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #5, pp. 41-45
Title of the Paper: An Open and Lightweight Internet of Things Platform to Facilitate the Development of Context-Aware Applications
Authors: Fernando Ribeiro, José Metrôlho
Abstract: The Internet of Things (IoT) has encouraged the development of innovative applications and services that make use of the enormous amount and diversity of data provided by such things to provide novel and useful services. However, to be successful, these platforms have to meet the expectations of different stakeholders, namely application developers and end-users. In this work we analyze some previous works with similar goals and we discuss some issues that need to be addressed when developing an IoT platform. Based on this knowledge, we proposed and developed an open and lightweight IoT platform that allows application developers and end-users to integrate IoT objects, to create new sensor data models monitoring and managing objects and that allows developers to access and use, in their applications, high level APIs to handle the information related to the IoT objects. In this approach there is particular focus on two characteristics: simplicity and flexibility. Simplicity, allowing application developers to easily integrate their own objects using their own semantics, and allowing end-users to visualize and manage their objects through the web portal. Flexibility, due its open nature that supports the integration of new types of objects and data formats, it allows developers to customize their own devices and models. The first experimental results indicate that this is a viable approach to manage Things on the Internet and to simplify the development of context-aware applications.
Keywords: ambient intelligence, context-awareness, internet of things, open platform, smart spaces, ubiquitous computing, web of things
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #4, pp. 31-40
Title of the Paper: Simulation-Based Study of Distributed Denial of Service Attacks Counteract in the Cloud Services
Authors: Wael Alosaimi, Mazin Alshamrani, Khalid Al-Begain
Abstract: Network availability is threatened by the traditional Denial of Service (DoS) and Distributed Denial of Service (DDoS) attacks. The risk is much increased with the emergence of the new computing paradigm of cloud computing. In this era, DDoS attacks can threaten the cloud sustainability by hitting its pricing model exploiting the cloud scalability feature. Therefore, a new phenomenon is emerged as a result of launching DDoS attacks against the cloud customers. It is called Economic Denial of Sustainability (EDoS). It is no more than an economic version of DDoS attack sharing its mechanism but different in the final aim. In order to defeat DDoS and EDoS attacks, the filtering firewalls can play main role in this regard. This paper is an extended version of a previous work that invented by the authors which introduced a new technique to mitigate the impacts of such attacks depending on the firewall features in managing a verification process to maintain the targeted system. The proposed framework is known as Enhanced DDoS- Mitigation System (Enhanced DDoS-MS). The firewalls characteristics are evaluated using OPNET simulation tool. The results showed that the firewall is effective in mitigating the DDoS impacts by limiting the response time, throughput, server load, and the traffic sent and received under attack. The paper also suggests using an active test bed for evaluating the proposed framework in a real manner.
Keywords: cloud computing, Firewall, Distributed Denial of Service attacks, DDoS, Economic Denial of Sustainability, EDoS
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #3, pp. 19-30
Title of the Paper: Multi-Source Video Summarization in the Internet of Things: A Social Computing Approach
Authors: Klimis Ntalianis
Abstract: In this paper a multi-source video summarization scheme is proposed, which is focuses on novel concepts of the Internet of Things. The proposed scheme assumes several different cameras asynchronously recording the same event from different angles, positions, heights etc. One of its main novelties is in the area of content evaluation. More specifically a social computing approach is proposed in order to find out which of the recordings attracts more attention. Attention is estimated by considering a social network environment where each user that records the event can post his/her material and let his/her friends to interact with it. Simulation results are presented, to denote the full potential of the proposed scheme, its advantages as well as important issues for future work.
Keywords: Internet of Things, Social Computing, Video Summarization, Multiple Sources
WSEAS Transactions on WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #2, pp. 11-18
Title of the Paper: Frequent Pattern Mining Under Multiple Support Thresholds
Authors: Sadeq Darrab, Belgin Ergenç
Abstract: Traditional methods use a single minimum support threshold to find out the complete set of frequent patterns. However, in real word applications, using single minimum item support threshold is not adequate since it does not reflect the nature of each item. If single minimum support threshold is set too low, a huge amount of patterns are generated including uninteresting patterns. On the other hand if it is set too high, many of interesting patterns (called rare items) may be lost. Recently, several methods have been studied to tackle the rare item problem by avoiding using single minimum item support threshold. The nature of each item is considered where different items are specified with different Minimum Item Support thresholds (MIS) instead of using single support threshold. By this, the complete set of frequent patterns is generated without creating uninteresting patterns and losing substantial patterns. In this paper, we propose an efficient method, Multiple Item Support Frequent Pattern growth algorithm, MISFP-growth, to mine the complete set of frequent patterns with multiple item support thresholds. In this method, Multiple Item Support Frequent Pattern Tree, MISFP-Tree, is constructed to store all crucial information to mine frequent patterns. Since the construction of the MISFP-Tree is done with respect to minimum of MIS; post pruning and reconstruction phases are not required. In order to show the efficiency of the proposed method, it is compared with a recent tree-based algorithm, CFP-growth++ and various experiments are conducted on both real and synthetic datasets. Experimental results reveal that MISFP-growth outperforms in terms of execution time and memory space while we vary MIS values of items.
Keywords: Association rule mining, Frequent patterns, Rare itemsets, Multiple support thresholds
WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #1, pp. 1-10
