INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
UK: Managing Editor International Journal of Innovative Technology and Creative Engineering 1a park lane, Cranford London TW59WA UK
USA: Editor International Journal of Innovative Technology and Creative Engineering Dr. Arumugam Department of Chemistry University of Georgia GA-30602, USA.
India: Editor International Journal of Innovative Technology & Creative Engineering 36/4 12th Avenue, 1st cross St, Vaigai Coliny Ashok Nagar Chennai , India 600083 Email: editor@ijitce.co.uk
www.ijitce.co.uk
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
IJITCE PUBLICATION
International Journal of Innovative Technology & Creative Engineering Vol.9 No.2 February 2019
www.ijitce.co.uk
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
From Editor's Desk Dear Researcher, Greetings!
The month of witnessed some of important events around the world like, Pondering a tablet screen displaying a town scene, a pre-K student tilts her head to the side and taps her lip thoughtfully. The girl uses her forefinger to pan around the scene. She eventually selects an image of a girl not wearing purple. The robot, which MIT researchers are testing with students in a Boston-area public school, tilts toward the girl, who leans in close so that her cheek is right next to Tega’s. This kind of tight connection is typical of child-robot interactions, says MIT social robotics and human-robot interaction researcher randomly assigned 24 students ages 10 to 12 to either two weeks of reading aloud alone or with Minnie. Robots can also encourage specific reasoning strategies, such as thinking aloud, which is supposed to help students craft more deliberate, organized plans for multistep problem-solving. More students who read aloud with a robot companion said that the activity motivated them to read and increased their reading comprehension than students who read aloud alone. Thanks, Editorial Team IJITCE
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Editorial Members Dr. Chee Kyun Ng Ph.D Department of Computer and Communication Systems, Faculty of Engineering,Universiti Putra Malaysia,UPMSerdang, 43400 Selangor,Malaysia. Dr. Simon SEE Ph.D Chief Technologist and Technical Director at Oracle Corporation, Associate Professor (Adjunct) at Nanyang Technological University Professor (Adjunct) at ShangaiJiaotong University, 27 West Coast Rise #08-12,Singapore 127470 Dr. sc.agr. Horst Juergen SCHWARTZ Ph.D, Humboldt-University of Berlin,Faculty of Agriculture and Horticulture,Asternplatz 2a, D-12203 Berlin,Germany Dr. Marco L. BianchiniPh.D Italian National Research Council; IBAF-CNR,Via Salaria km 29.300, 00015 MonterotondoScalo (RM),Italy Dr. NijadKabbaraPh.D Marine Research Centre / Remote Sensing Centre/ National Council for Scientific Research, P. O. Box: 189 Jounieh,Lebanon Dr. Aaron Solomon Ph.D Department of Computer Science, National Chi Nan University,No. 303, University Road,Puli Town, Nantou County 54561,Taiwan Dr. Arthanariee. A. M M.Sc.,M.Phil.,M.S.,Ph.D Director - Bharathidasan School of Computer Applications, Ellispettai, Erode, Tamil Nadu,India Dr. Takaharu KAMEOKA, Ph.D Professor, Laboratory of Food, Environmental & Cultural Informatics Division of Sustainable Resource Sciences, Graduate School of Bioresources,Mie University, 1577 Kurimamachiya-cho, Tsu, Mie, 514-8507, Japan Dr. M. Sivakumar M.C.A.,ITIL.,PRINCE2.,ISTQB.,OCP.,ICP. Ph.D. Project Manager - Software,Applied Materials,1a park lane,cranford,UK Dr. Bulent AcmaPh.D Anadolu University, Department of Economics,Unit of Southeastern Anatolia Project(GAP),26470 Eskisehir,TURKEY Dr. SelvanathanArumugamPh.D Research Scientist, Department of Chemistry, University of Georgia, GA-30602,USA. Dr. S.Prasath Ph.D Assistant Professor, Department of Computer Science, Nandha Arts & Science College, Erode , Tamil Nadu, India
Review Board Members Dr. Paul Koltun Senior Research ScientistLCA and Industrial Ecology Group,Metallic& Ceramic Materials,CSIRO Process Science & Engineering Private Bag 33, Clayton South MDC 3169,Gate 5 Normanby Rd., Clayton Vic. 3168, Australia Dr. Zhiming Yang MD., Ph. D. Department of Radiation Oncology and Molecular Radiation Science,1550 Orleans Street Rm 441, Baltimore MD, 21231,USA Dr. Jifeng Wang Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign Urbana, Illinois, 61801, USA Dr. Giuseppe Baldacchini
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
ENEA - Frascati Research Center, Via Enrico Fermi 45 - P.O. Box 65,00044 Frascati, Roma, ITALY. Dr. MutamedTurkiNayefKhatib Assistant Professor of Telecommunication Engineering,Head of Telecommunication Engineering Department,Palestine Technical University (Kadoorie), TulKarm, PALESTINE. Dr.P.UmaMaheswari Prof &Head,Depaartment of CSE/IT, INFO Institute of Engineering,Coimbatore. Dr. T. Christopher, Ph.D., Assistant Professor &Head,Department of Computer Science,Government Arts College(Autonomous),Udumalpet, India. Dr. T. DEVI Ph.D. Engg. (Warwick, UK), Head,Department of Computer Applications,Bharathiar University,Coimbatore-641 046, India. Dr. Renato J. orsato Professor at FGV-EAESP,Getulio Vargas Foundation,São Paulo Business School,RuaItapeva, 474 (8° andar),01332-000, São Paulo (SP), Brazil Visiting Scholar at INSEAD,INSEAD Social Innovation Centre,Boulevard de Constance,77305 Fontainebleau - France Y. BenalYurtlu Assist. Prof. OndokuzMayis University Dr.Sumeer Gul Assistant Professor,Department of Library and Information Science,University of Kashmir,India Dr. ChutimaBoonthum-Denecke, Ph.D Department of Computer Science,Science& Technology Bldg., Rm 120,Hampton University,Hampton, VA 23688 Dr. Renato J. Orsato Professor at FGV-EAESP,Getulio Vargas Foundation,São Paulo Business SchoolRuaItapeva, 474 (8° andar),01332-000, São Paulo (SP), Brazil Dr. Lucy M. Brown, Ph.D. Texas State University,601 University Drive,School of Journalism and Mass Communication,OM330B,San Marcos, TX 78666 JavadRobati Crop Production Departement,University of Maragheh,Golshahr,Maragheh,Iran VineshSukumar (PhD, MBA) Product Engineering Segment Manager, Imaging Products, Aptina Imaging Inc. Dr. Binod Kumar PhD(CS), M.Phil.(CS), MIAENG,MIEEE HOD & Associate Professor, IT Dept, Medi-Caps Inst. of Science & Tech.(MIST),Indore, India Dr. S. B. Warkad Associate Professor, Department of Electrical Engineering, Priyadarshini College of Engineering, Nagpur, India Dr. doc. Ing. RostislavChoteborský, Ph.D. Katedramateriálu a strojírenskétechnologieTechnickáfakulta,Ceskázemedelskáuniverzita v Praze,Kamýcká 129, Praha 6, 165 21 Dr. Paul Koltun Senior Research ScientistLCA and Industrial Ecology Group,Metallic& Ceramic Materials,CSIRO Process Science & Engineering Private Bag 33, Clayton South MDC 3169,Gate 5 Normanby Rd., Clayton Vic. 3168 DR.ChutimaBoonthum-Denecke, Ph.D Department of Computer Science,Science& Technology Bldg.,HamptonUniversity,Hampton, VA 23688 Mr. Abhishek Taneja B.sc(Electronics),M.B.E,M.C.A.,M.Phil., Assistant Professor in the Department of Computer Science & Applications, at Dronacharya Institute of Management and Technology, Kurukshetra. (India).
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Dr. Ing. RostislavChotěborský,ph.d, Katedramateriálu a strojírenskétechnologie, Technickáfakulta,Českázemědělskáuniverzita v Praze,Kamýcká 129, Praha 6, 165 21
Dr. AmalaVijayaSelvi Rajan, B.sc,Ph.d, Faculty – Information Technology Dubai Women’s College – Higher Colleges of Technology,P.O. Box – 16062, Dubai, UAE Naik Nitin AshokraoB.sc,M.Sc Lecturer in YeshwantMahavidyalayaNanded University Dr.A.Kathirvell, B.E, M.E, Ph.D,MISTE, MIACSIT, MENGG Professor - Department of Computer Science and Engineering,Tagore Engineering College, Chennai Dr. H. S. Fadewar B.sc,M.sc,M.Phil.,ph.d,PGDBM,B.Ed. Associate Professor - Sinhgad Institute of Management & Computer Application, Mumbai-BangloreWesternly Express Way Narhe, Pune - 41 Dr. David Batten Leader, Algal Pre-Feasibility Study,Transport Technologies and Sustainable Fuels,CSIRO Energy Transformed Flagship Private Bag 1,Aspendale, Vic. 3195,AUSTRALIA Dr R C Panda (MTech& PhD(IITM);Ex-Faculty (Curtin Univ Tech, Perth, Australia))Scientist CLRI (CSIR), Adyar, Chennai - 600 020,India Miss Jing He PH.D. Candidate of Georgia State University,1450 Willow Lake Dr. NE,Atlanta, GA, 30329 Jeremiah Neubert Assistant Professor,MechanicalEngineering,University of North Dakota Hui Shen Mechanical Engineering Dept,Ohio Northern Univ. Dr. Xiangfa Wu, Ph.D. Assistant Professor / Mechanical Engineering,NORTH DAKOTA STATE UNIVERSITY SeraphinChallyAbou Professor,Mechanical& Industrial Engineering Depart,MEHS Program, 235 Voss-Kovach Hall,1305 OrdeanCourt,Duluth, Minnesota 55812-3042 Dr. Qiang Cheng, Ph.D. Assistant Professor,Computer Science Department Southern Illinois University CarbondaleFaner Hall, Room 2140-Mail Code 45111000 Faner Drive, Carbondale, IL 62901 Dr. Carlos Barrios, PhD Assistant Professor of Architecture,School of Architecture and Planning,The Catholic University of America Y. BenalYurtlu Assist. Prof. OndokuzMayis University Dr. Lucy M. Brown, Ph.D. Texas State University,601 University Drive,School of Journalism and Mass Communication,OM330B,San Marcos, TX 78666 Dr. Paul Koltun Senior Research ScientistLCA and Industrial Ecology Group,Metallic& Ceramic Materials CSIRO Process Science & Engineering Dr.Sumeer Gul Assistant Professor,Department of Library and Information Science,University of Kashmir,India Dr. ChutimaBoonthum-Denecke, Ph.D Department of Computer Science,Science& Technology Bldg., Rm 120,Hampton University,Hampton, VA 23688
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Dr. Renato J. Orsato Professor at FGV-EAESP,Getulio Vargas Foundation,São Paulo Business School,RuaItapeva, 474 (8° andar)01332-000, São Paulo (SP), Brazil Dr. Wael M. G. Ibrahim Department Head-Electronics Engineering Technology Dept.School of Engineering Technology ECPI College of Technology 5501 Greenwich Road - Suite 100,Virginia Beach, VA 23462 Dr. Messaoud Jake Bahoura Associate Professor-Engineering Department and Center for Materials Research Norfolk State University,700 Park avenue,Norfolk, VA 23504 Dr. V. P. Eswaramurthy M.C.A., M.Phil., Ph.D., Assistant Professor of Computer Science, Government Arts College(Autonomous), Salem-636 007, India. Dr. P. Kamakkannan,M.C.A., Ph.D ., Assistant Professor of Computer Science, Government Arts College(Autonomous), Salem-636 007, India. Dr. V. Karthikeyani Ph.D., Assistant Professor of Computer Science, Government Arts College(Autonomous), Salem-636 008, India. Dr. K. Thangadurai Ph.D., Assistant Professor, Department of Computer Science, Government Arts College ( Autonomous ), Karur - 639 005,India. Dr. N. Maheswari Ph.D., Assistant Professor, Department of MCA, Faculty of Engineering and Technology, SRM University, Kattangulathur, Kanchipiram Dt - 603 203, India. Mr. Md. Musfique Anwar B.Sc(Engg.) Lecturer, Computer Science & Engineering Department, Jahangirnagar University, Savar, Dhaka, Bangladesh. Mrs. Smitha Ramachandran M.Sc(CS)., SAP Analyst, Akzonobel, Slough, United Kingdom. Dr. V. Vallimayil Ph.D., Director, Department of MCA, Vivekanandha Business School For Women, Elayampalayam, Tiruchengode - 637 205, India. Mr. M. Moorthi M.C.A., M.Phil., Assistant Professor, Department of computer Applications, Kongu Arts and Science College, India PremaSelvarajBsc,M.C.A,M.Phil Assistant Professor,Department of Computer Science,KSR College of Arts and Science, Tiruchengode Mr. G. Rajendran M.C.A., M.Phil., N.E.T., PGDBM., PGDBF., Assistant Professor, Department of Computer Science, Government Arts College, Salem, India. Dr. Pradeep H Pendse B.E.,M.M.S.,Ph.d Dean - IT,Welingkar Institute of Management Development and Research, Mumbai, India Muhammad Javed Centre for Next Generation Localisation, School of Computing, Dublin City University, Dublin 9, Ireland Dr. G. GOBI Assistant Professor-Department of Physics,Government Arts College,Salem - 636 007 Dr.S.Senthilkumar Post Doctoral Research Fellow, (Mathematics and Computer Science & Applications),UniversitiSainsMalaysia,School of Mathematical Sciences, Pulau Pinang-11800,[PENANG],MALAYSIA. Manoj Sharma Associate Professor Deptt. of ECE, PrannathParnami Institute of Management & Technology, Hissar, Haryana, India RAMKUMAR JAGANATHAN Asst-Professor,Dept of Computer Science, V.L.B Janakiammal college of Arts & Science, Coimbatore,Tamilnadu, India Dr. S. B. Warkad Assoc. Professor, Priyadarshini College of Engineering, Nagpur, Maharashtra State, India
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Dr. Saurabh Pal Associate Professor, UNS Institute of Engg. & Tech., VBS Purvanchal University, Jaunpur, India Manimala Assistant Professor, Department of Applied Electronics and Instrumentation, St Joseph’s College of Engineering & Technology, Choondacherry Post, Kottayam Dt. Kerala -686579 Dr. Qazi S. M. Zia-ul-Haque Control Engineer Synchrotron-light for Experimental Sciences and Applications in the Middle East (SESAME),P. O. Box 7, Allan 19252, Jordan Dr. A. Subramani, M.C.A.,M.Phil.,Ph.D. Professor,Department of Computer Applications, K.S.R. College of Engineering, Tiruchengode - 637215 Dr. SeraphinChallyAbou Professor, Mechanical & Industrial Engineering Depart. MEHS Program, 235 Voss-Kovach Hall, 1305 Ordean Court Duluth, Minnesota 558123042 Dr. K. Kousalya Professor, Department of CSE,Kongu Engineering College,Perundurai-638 052 Dr. (Mrs.) R. Uma Rani Asso.Prof., Department of Computer Science, Sri Sarada College For Women, Salem-16, Tamil Nadu, India. MOHAMMAD YAZDANI-ASRAMI Electrical and Computer Engineering Department, Babol"Noshirvani" University of Technology, Iran. Dr. Kulasekharan, N, Ph.D Technical Lead - CFD,GE Appliances and Lighting, GE India,John F Welch Technology Center,Plot # 122, EPIP, Phase 2,Whitefield Road,Bangalore – 560066, India. Dr. Manjeet Bansal Dean (Post Graduate),Department of Civil Engineering,Punjab Technical University,GianiZail Singh Campus,Bathinda -151001 (Punjab),INDIA Dr. Oliver Jukić Vice Dean for education,Virovitica College,MatijeGupca 78,33000 Virovitica, Croatia Dr. Lori A. Wolff, Ph.D., J.D. Professor of Leadership and Counselor Education,The University of Mississippi,Department of Leadership and Counselor Education, 139 Guyton University, MS 38677
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Contents Band concert at Thomas Middle School By Hemharsith
…………………..…..[623]
A Performance Comparison Of Authentication And Privacy Preserving Techniques For Secured Communication In Vanet ………………..…..[625] Analytical Measures for Detecting Fraud Using Classification Algorithms ……………. [636]
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
BAND CONCERT AT THOMAS MIDDLE SCHOOL By Hemharsith In Chicago, Thomas Middle School on the first day of February there was a fabulous sixthgrade band festival. Where sixth-grade band students cover the mind of the audience by playing four songs and the songs were penguins on parade, udalu’m, junk funk and Antagonist. Penguins on parade song are written by Fred Hubbell. He is a businessman and a politician who also writes kid’s song in the meantime. Penguins on parade are the first song we played. That song is a kids song with a lot of hard notes. It conveys about being untouchable Udalu’m is a Nigerian Folk Song arr. Written by Belwin Division. The song is traditional. Nigerian folk song that tells the story of Chinyere. The story has a happy ending that's reflected on this tuneful work for beginning bands. This song is about an African storyteller. Junk funk is a song written by Kevin Mixon. It is a song where the notes go high to low fares and it reverses to low to high vers. It was a song where my friend and I need to clap and stomp foot thought go and shhhhh and it was a fun song. This song tells us to just dance & be happy Antagonist is a song written by Larry Clark. This song is a 2-page long and has a very fast tempo. This song makes us be serious and temp's our mind to listen to it again and again. This song conveys a bold and aggressive spirit, this forceful piece is the highlight of our concert. The dynamics, chord progressions, and rhythms create an antagonizing character within the music.
623
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
I am also one of a performer in the concert playing Alto saxophone. my friends and I worked 3 hours straight on all the songs which we never have played before. All these songs were very hard at our stage. Before if we need to put up a performs we had a month to practice all the songs. But, on February 1 we just had three hours BTW which is a limited time to put up a show to all 150 audiences. We still did our best and accomplished our goal. This event was great practice for me and my friends because when we go to higher classes we will have some limited time to put up a show and also we need to learn new notes. And this song’s made us all feel that we are a better musician. On the same day in our town, Arlington Heights has an Elementary School named IVY Hill which held a fun fair. Where kids from that school go and play games, collect points, eat snacks and dance. using the points kids redeem some for himself or herself at the end of the fun fair and go home happily with their toys that they redeemed.
624
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
A PERFORMANCE COMPARISON OF AUTHENTICATION AND PRIVACY PRESERVING TECHNIQUES FOR SECURED COMMUNICATION IN VANET K . Nirmala Ph.D Research Scholar (Part-Time), Department of Computer Science, Nandha Arts & Science College, Erode, Tamil Nadu, India E-mail ID: nirmalabuasc@gmail.com Dr.S.Prasath Assistant Professor & Research Supervisor, Department of Computer Science, Nandha Arts & Science College, Erode, Tamil Nadu, India E-mail ID: softprasaths@gmail.com
Keywords: Vehicle ad hoc network, security, trust, vehicle authentication, privacy, communication.
Abstract- Vehicle adhoc network (VANET) is vital role in communication which is used for enhancing the traffic efficiency and safety through communicating one vehicle with other vehicles. Security is the key problems in VANETs and trust is an essential one that avoid the generic attacks on network. A misuse of information leads to the traffic accident and loss of human lives. Vehicle authentication is need for improving the security level in VANET. In the authentication, vehicle data like identity and location information are kept private. Privacy is an important one during communication in VANETs. The vehicle privacy information like current position, license number, drivers identity and travel route are maintained as confidential one for long time period. Many techniques were developed for secured communication in VANET. But the existing techniques have some drawbacks, there is a need to improve the authentication accuracy and privacy performance during communication in VANET. To improve the security level during communication, machine learning and cryptographic techniques is used.
1. INTRODUCTION Vehicular Ad-Hoc Network technology selected moving vehicles as nodes in network to create the mobile network. The movement of vehicles is limited by roads and traffic regulations use fixed infrastructure at critical locations. VANET presents road safety rules where the information concerning vehicle current speed and location coordinates is provided with or without exploitation. When vehicles goes beyond the signal range and drop out of the network, additional vehicles join, link the vehicles to another where the mobile internet are generated. Vehicle authentication is requirement for improving the security level in VANET. The authentication includes the identity authentication and message integrity to guarantee the security of VANETs. When identity authentication is not satisfied, malicious vehicle imitate as legal vehicle to broadcast messages for receiving the illegal benefits. The message integrity is not guaranteed and malicious vehicle broadcast
625
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
the falsified messages to disrupt the traffic for surrounding vehicles without being caught. This paper is structured as follows: Section 2 explains authentication and privacy preservation techniques for secured communication in VANET. Section 3 shows the analysis of the existing authentication and privacy preservation techniques for secured communication in VANET. Section 4 identifies the possible comparison between the techniques of VANET. Section 5 presents the discussion and limitations of authentication and privacy preserving techniques. Section 6 concludes the paper to improve the security level during communication in VANET.
depending on the cooperative authentication method. A two-layer pseudo-identity generation method built key update tree for efficient revocation. But, self-healing functionality is not used to protect the success of group key update when the vehicles miss update messages. An efficient randomized authentication protocol carried out the homomorphic encryption [5] to permit every individual vehicle for self-generating the number of authenticated identities to attain the anonymity in VANETs. However, authentication time is not minimized using efficient randomized authentication protocol. Event Based Reputation System (EBRS) proposed in [6] for dynamic reputation and trusted value. Every event reduces the spread of false messages. In automatic mode, trust relationship among the participant vehicles is not identified. An analytical framework developed [7] message dissemination process in vehicular network with malicious vehicles distributed in network. The probability with destination vehicle at fixed distance collected the message correctly from source vehicle. However, security level is not enhanced by analytical framework.
2. RELATED WORKS A local identity-based anonymous message authentication protocol (LIAP) proposed in [1] for VANETs where every vehicle and Road Side Unit (RSU) assigned long term certification from Certificate Authority (CA). Vehicle selected anonymous identity to sign safety message verified by single authentication technique. But, authentication overhead is not reduced using LIAP. A new detection approach called Greedy Detection for VANETs (GDVAN) is developed in [2] for greedy behavior attacks. The design of algorithm identified greedy behavior and established list of compromised nodes through defined metrics. However, reaction method against the greedy attack not considers the GDVAN approach to eliminate serious problems.
PassWord-based Conditional Privacy Preserving Authentication and Group-Key generAtion (PW-CPPA-GKA) protocol proposed [8] for VANETs. The design of protocol is lightweight during the computation and communication without the bilinearpairing and elliptic curve. But, random oracle model not used for security. A vehicular authentication protocol called distributed aggregate privacy-preserving authentication designed in [9]. The protocol depends on multiple trusted authority one-time identitybased aggregate signature method. However, DAPPA failed to improve the security level through privacy-preserving authentication.
A Novel and Efficient Conditional Privacy-Preserving Authentication (NECPPA) scheme developed in [3] for secure communications in VANET. The hacking single On-Board Unit (OBU) failed to threaten the network in Tamper Proof Device Based (TPDB) scheme that create whole vehicles to re-register and change secret keys. However, the privacy preserving rate is not improved using NECPPA. An anonymous authentication protocol designed [4]
626
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
3. AUTHENTICATION AND PRIVACY PRESERVATION TECHNIQUES FOR SECURED COMMUNICATION IN VANET
revocation list. With increase of revoked vehicle, size of CRL not scalable that leads to high computational complexity and communication overhead.
In huge development of wireless communication, adhoc networking, automotive and transportation industry, vehicular adhoc networks have attracted large attention from government, industry and academia because of potential to provide enhanced driving experience and road safety. It is essential to meet critical security needs of VANETs like data integrity, reputation management, privacy protection, etc. The methods without protection of security and privacy methods resulted in bad user experience results.
A Local Identity-based anonymous message Authentication Protocol (LIAP) for VANETs utilized the efficient revocation of PKI and authentication efficiency of identity. Every node attain unique long term certificate from CA in system initialization phase. When the vehicle entered the communication range of new RSU, it requested local master keys with its certificate. The validity of communication message between vehicle and vehicle are guaranteed through mutual authentication process. After collecting the valid master keys, vehicle created the localized anonymous identity to mark safetyrelated message. When node is cooperated, CA invalidated their unique certificate. The safety-related message is verified by single or batch manner to enhance the authentication efficiency. In Expedite Message Authentication Protocol (EMAP) and Anonymous Batch AutHentication (ABAH), certificate used to verify the validity of safety-related message. For every received message, it computed the Hash Message Authentication Code (HMAC) and authenticate certificate. The certificate selected to authenticate validity of the nodes. CRL checking and certificate verification were implemented on mutual authentication between RSU and vehicle. The validity of safety-related message is verified by identitybased signature. Vehicle created the pseudonym after collecting the master keys from RSU. EMAP and ABAH pseudonym allocated by the CA. RSU allocates and modernizes their local master keys separately in LIAP and ABAH.
3.1 Local Identity-based anonymous message Authentication Protocol in VANETs Vehicle communicates with the additional nodes through open wireless channel that increases the safety issues. The Public Key Infrastructure (PKI) and identity based authentication protocols addressed the security and privacy requirements of VANETs. The receiver verifies the Certificate Revocation List (CRL) before certificate and signature verification in PKI-base methods. CRL checking minimized the authentication efficiency. In identity-based schemes, every vehicle maintains the valid identities to preserve the privacy. The authentication technique is developed on the public key infrastructure and identity-based encryption technology. In PKI based systems, certificate authority allocated pseudonym certificates and public/private key pairs for each registered vehicle that are preloaded into storage unit. When vehicle transmits the safety-related message, it chooses the private key to generate signature and matching certificate is embedded in message. When vehicle is revoked, CA adds all of pseudonym certificates into certificate
\\Local Identity-based anonymous message Authentication Algorithm\\ Vehicle �� authenticates hello message �ℎ Require: Receive a hello message
627
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Step 1: Check đ?‘‡đ?‘’ Step 2: If đ?‘‡đ?‘’ is valid then Check đ?‘ƒđ??žđ?‘…đ??ź whether đ?‘…đ?‘– is a new RSU Step 3: If đ?‘ƒđ??žđ?‘…đ??ź = đ?‘ đ?‘’đ?‘¤ then Check đ??śđ?‘’đ?‘&#x;đ?‘Ąđ?‘…đ?‘– against the RCRL Step 4: If đ??śđ?‘’đ?‘&#x;đ?‘Ąđ?‘…đ?‘– ∉ đ?‘…đ??śđ?‘…đ??ż then Verify đ??śđ?‘’đ?‘&#x;đ?‘Ąđ?‘…đ?‘– and đ?œŽđ?‘…đ?‘–
3.2 A New Greedy Behavior Attack Detection Algorithm for VANETs Vehicular Ad hoc Networks is used to provide the road safety and enhance the driving conditions. VANET were exposed to many types of attacks like Denial of Service (DoS) attacks that affect the availability of given services for legitimate users. A new detection approach termed Greedy Detection for VANETs (GDVAN) developed for greedy behavior attacks in VANETs. The detection approach comprises two phases termed suspicion phase and the decision phase. The suspicion phase based on linear regression mathematical ideas while decision phase depending on fuzzy logic decision scheme. The designed algorithm identified the existence of greedy behavior and established list of compromised nodes by newly defined metrics.
Step 5: If đ??śđ?‘’đ?‘&#x;đ?‘Ąđ?‘…đ?‘– = đ?‘Ąđ?‘&#x;đ?‘˘đ?‘’ and đ?œŽđ?‘…đ?‘– = đ?‘Ąđ?‘&#x;đ?‘˘đ?‘’ then Step 6: Store (đ?‘…đ?‘ƒđ??žđ?‘–1 , đ?‘…đ?‘ƒđ??žđ?‘–2 ) Step 7: Store 1 2 1 2 (đ?‘…đ?‘ƒđ??ˇđ?‘–−1 , đ?‘…đ?‘ƒđ??ˇđ?‘–−1 , đ?‘…đ?‘ƒđ??ˇđ?‘–+1 , đ?‘…đ?‘ƒđ??ˇđ?‘–+1 ) Step 8: Send a request message đ?‘€đ?‘&#x; to đ?‘…đ?‘– End if End if End if End if The above algorithm explains the algorithmic process of local identity-based anonymous message authentication algorithm. A hybrid authentication protocol proposed on PKI and identity-based signature that meet the needs of security and conditional privacy in VANETs. Every node comprises long term PKI-based certificate to authenticate the node validity. For safetyrelated message, vehicle creates the localized anonymous identity to mark it. The mutual authentication between the RSU and vehicle guarantee that vehicle communicates with the unrevoked RSU and valid vehicle gathers the local master keys. For increasing the authentication efficiency, node verify safety-related message by single or batch authentication manner. CA manages revoked certificates by the RSU Certificate Revocation List (RCRL) and the Vehicle Certificate Revocation List (VCRL). When the node gets compromised, CA revokes their certificate. The time-consuming CRL checking is executed in mutual authentication process avoided in message authentication of V2V communication.
GDVAN used three defined metrics for greedy detection in high mobile environment like VANET. The connection is short and nodes not have sufficient amount of time to execute the adaptive manipulation of backoff parameters. It comprised both suspicion and decision phases to enhance linear regression and fuzzy logic ideas. In through monitoring the network traffic traces, the algorithm affirms the existence. In affirmative case, it identified the dependable nodes. GDVAN are passive, non-resource-intensive and failed to need variations in MAC layer. It is transparent to the users and it executed by any node of the network. \\ Greedy Detection Algorithm\\ INPUT: T= Monitoring speed, State_Greedy=False OUTPUT: Annonce_Greedy(State_Greedy) Begin Repeat Step 1: Collect traffic traces during T Step 2: Calculate correlation coefficient ‘đ?œŒâ€™ Step 3: If đ?œŒ is close to 1 then Goto (8)
628
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Else Goto (13) end Step 4: Calculate slope of the linear regression straight Step 5: If the slope is close to 1 then State_Greedy= FALSE else run (13) end Step 6: Greedy behavior is suspected: Return and run watchdog supervision tool Step 7: Return; Annonce_Greedy(State_Greedy) Step 8: until no existing communication End
3.3 A Novel and Efficient Conditional Privacy- Preserving Authentication scheme for VANET (NECPPA) Vehicular Ad-hoc Networks are developing one in recent years for providing real-time communication between vehicles safer and more comfortable driving. The key objective of VANET is to broadcast the adhoc messages like traffic incidents and emergency events. VANETs are safe and commercialized. It connects the central stations or internet through VANETs to exchange the data. VANETs are one of the key components of intelligent transportation systems. The main objective of direct communication is vehicle safety and traffic minimization. VANETs are unique type of MANETs where the vehicles in VANET represent the nodes. Vehicles detect the additional vehicles to form the network by connecting them and perform suitable communications. The node movement is selected property of networks that allows them to vary their pattern immediately. A Novel Efficient Conditional Privacy Preserving Authentication (NECPPA) scheme developed for VANETs which is mixture of Tamper Proof Device Based (TPDB) and Road Side Unit Based (RSUB) techniques. In this scheme, authentication of vehicles carried out without the group signature. The signature verification of this scheme not increases linearly with number of revoked vehicles. The revocation process scheme is efficient than anonymous certificate schemes when the revocation of vehicles not increases CRL suddenly. The verifier not verifies the CRL for every signature. The sensitive information and master key of Trusted Authority (TA) were not in tamper proof device of vehicles but they were stored in RSU tamper proof devices. RSUs include the direct, fast and secure communicational link with TA that make easier and faster to modernize the system parameters and revoke vehicles. The compromise of single vehicle failed to imply varying parameters of whole network and reregistration of all vehicles. The vehicles
The above algorithm describes the greedy detection process in VANET. A new decision scheme developed for identifying the greedy behavior for VANETs. The designed scheme identified the nodes to violate the Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) protocol rules to increase their bandwidth at expense of the well-behaving nodes. The newly defined metrics are convenient to high mobile networks and employed during the short monitoring periods. In watchdog detection software, three newly defined metrics for each node in VANET used are: ➢ Number of connection attempts ➢ Average of connection duration ➢ Average of waiting times between connection From the fuzzy logic, existence of greedy behavior from certain value of parameter is imagined. In between two threshold values, suspicion is slow. The idea is depending on use of tools presented by fuzzy logic theory.
629
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
require contacting the RSUs when they enter area covered by them. The signature verification is carried out by vehicles without any online RSUs. 4. COMPARISON OF AUTHENTICATION AND PRIVACY PRESERVING TECHNIQUES FOR SECURED COMMUNICATION In order to compare the authentication and privacy preserving techniques for secured communication in VANET, number of vehicular nodes is considered to perform the experiment. Various parameters are used in authentication and privacy preserving techniques for secured communication in VANET. The coding output screenshots of LIAP, GDVAN and NECPPA scheme is described in fig.1.1, fig. 1.2 and fig. 1.3 respectively.
Fig. 1.2 Output Screenshot for GDVAN In fig. 1.2, the output for the GDVAN is given. The number of nodes that are accurately classified, ending time and number of nodes that are incorrectly classified values are obtained for GDVAN in the simulation outputs. These values are substituted in equ. (1), (2) and (3) to obtain the authentication accuracy, authentication time and false positive rate of GDVAN. In fig. 1.3, the coding result of NECPPA scheme is provided.
Fig. 1.1 Output Screenshot for LIAP In fig. 1.1, the output for LIAP is given. The number of nodes that are accurately classified, ending time and number of nodes that are incorrectly classified values are obtained for LIAP in the simulation outputs. These values are substituted in equ. (1), (2) and (3) to obtain the authentication accuracy, authentication time and false positive rate of LIAP. In fig. 1.2, the coding result of GDVAN is illustrated.
Fig. 1.3 Output Screenshot for NECPPA scheme In fig. 1.3, the output for NECPPA scheme is described. The number of nodes that are accurately classified, ending time and number of nodes that are incorrectly classified values are obtained in simulation outputs for NECPPA scheme. These values are substituted in equ. (1), (2) and (3) to obtain the authentication accuracy, authentication time and false positive rate of GDVAN. In fig. 1.3, the coding result of NECPPA scheme is provided. Based on the
630
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
values obtained, the table values are given in table 1.1, table 1.2 and table 1.3.
scheme. The graphical representation of authentication accuracy as shown in fig. 1.4.
4.1 Authentication Accuracy Authentication accuracy is defined as the ratio of number of vehicular nodes that are accurately authenticated to the total number of vehicular nodes. It is measured in terms of percentage (%) using the equ. (1). đ??´đ?‘˘đ?‘Ąâ„Žđ?‘’đ?‘›đ?‘Ąđ?‘–đ?‘?đ?‘Žđ?‘Ąđ?‘–đ?‘œđ?‘› đ?‘Žđ?‘?đ?‘?đ?‘˘đ?‘&#x;đ?‘Žđ?‘?đ?‘Ś = đ?‘ đ?‘˘đ?‘šđ?‘?đ?‘’đ?‘&#x; đ?‘œđ?‘“ đ?‘Łđ?‘’â„Žđ?‘–đ?‘?đ?‘˘đ?‘™đ?‘Žđ?‘&#x; đ?‘›đ?‘œđ?‘‘đ?‘’đ?‘ đ?‘Ąâ„Žđ?‘Žđ?‘Ą đ?‘Žđ?‘&#x;đ?‘’ đ?‘Žđ?‘?đ?‘?đ?‘˘đ?‘&#x;đ?‘Žđ?‘Ąđ?‘’đ?‘™đ?‘Ś đ?‘Žđ?‘˘đ?‘Ąâ„Žđ?‘’đ?‘›đ?‘Ąđ?‘–đ?‘?đ?‘Žđ?‘Ąđ?‘’đ?‘‘ đ?‘‡đ?‘œđ?‘Ąđ?‘Žđ?‘™ đ?‘›đ?‘˘đ?‘šđ?‘?đ?‘’đ?‘&#x; đ?‘œđ?‘“ đ?‘Łđ?‘’â„Žđ?‘–đ?‘?đ?‘˘đ?‘™đ?‘Žđ?‘&#x; đ?‘›đ?‘œđ?‘‘đ?‘’đ?‘
100 equ... (1) From the equ.(1), the authentication accuracy is calculated. When the authentication accuracy is higher, that method is more efficient.
∗ Fig. 1.4 Measure of Authentication Accuracy
Table. 1.1 Comparison of Authentication Accuracy Number of Authentication Accuracy (%) Vehicular NECPP Nodes LIAP GDVAN A (Number) Scheme 10 85 78 69 20 87 80 71 30 88 82 73 40 91 79 77 50 89 77 75 60 86 75 70 70 88 78 72 80 90 81 74 90 93 83 76 100 95 85 78 From the table. 1.1 describes the authentication accuracy with respect to number of vehicular nodes ranging from 10 to 100. Authentication accuracy compares with local identity-based anonymous message authentication protocol, Greedy Detection for VANETs and novel efficient conditional privacy preserving authentication scheme. From the table 1.1, it is observed that the authentication accuracy using local identitybased anonymous message authentication protocol is higher when compared to Greedy Detection for VANETs and novel efficient conditional privacy preserving authentication
From the fig. 1.4, authentication accuracy based on different number of vehicular node is explained. From the fig. 1.4, it is observed that the authentication accuracy using local identity-based anonymous message authentication protocol is higher when compared to Greedy Detection approach for VANETs and novel efficient conditional privacy preserving authentication scheme. Hybrid authentication protocol based on PKI and identity-based signature meet requirements of security and conditional privacy in VANETs. Every node includes the long term PKI-based certificate to authenticate node validity. For safetyrelated message, vehicle creates localized anonymous identity to mark it. This local identity-based anonymous message authentication protocol provides 12% higher authentication accuracy than GDVAN approach and consumes 21% higher authentication accuracy than novel efficient conditional privacy preserving authentication scheme. 4.2Authentication Time Authentication time is defined as amount of time taken to perform the authentication for secured communication in VANET. The difference of starting time and ending time during the authentication for secured communication is called
631
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
authentication time. It is measured in terms of milliseconds (ms) using the equ. (2). đ??´đ?‘˘đ?‘Ąâ„Žđ?‘’đ?‘›đ?‘Ąđ?‘–đ?‘?đ?‘Žđ?‘Ąđ?‘–đ?‘œđ?‘› đ?‘‡đ?‘–đ?‘šđ?‘’ = đ??¸đ?‘›đ?‘‘đ?‘–đ?‘›đ?‘” đ?‘Ąđ?‘–đ?‘šđ?‘’ − đ?‘ đ?‘Ąđ?‘Žđ?‘&#x;đ?‘Ąđ?‘–đ?‘›đ?‘” đ?‘Ąđ?‘–đ?‘šđ?‘’ đ?‘œđ?‘“ đ?‘›đ?‘œđ?‘‘đ?‘’ đ?‘Žđ?‘˘đ?‘Ąâ„Žđ?‘’đ?‘›đ?‘Ąđ?‘–đ?‘?đ?‘Žđ?‘Ąđ?‘–đ?‘œđ?‘› equ‌..(2) From the equ. (2) authentication time is calculated. When the authentication time is low, the method is more efficient. Table 1.2 Comparison of Authentication Time Number Authentication Time (ms) of LIAP GDVAN NECPPA Vehicular scheme Nodes (Number) 10 32 24 45 20 35 27 47 30 38 30 51 40 42 33 53 50 46 36 57 60 43 34 55 70 40 31 52 80 44 35 56 90 47 38 60 100 50 41 64
Fig. 1.5 Measure of Authentication Time In fig. 1.5, authentication time based on different number of vehicular node is described. From the fig. 1.5, authentication time using Greedy Detection approach for VANETs is less when compared to local identity-based anonymous message authentication protocol and novel efficient conditional privacy preserving authentication (NECPPA) scheme. Because, GDVAN have suspicion phase and decision phase. Suspicion phase depend on linear regression mathematical ideas while decision phase depend on fuzzy logic decision scheme. The algorithm is identified the existence of greedy behavior and established list of compromised nodes through newly defined metrics. GDVAN approach consumes 21% low authentication time than local identity-based anonymous message authentication protocol and consumes 39% less authentication time than novel efficient conditional privacy preserving authentication scheme.
Table 1.2 discusses the authentication time with respect to number of vehicular nodes in the range 10 to 100. The authentication time compared with local identity-based anonymous message authentication protocol, Greedy Detection for VANETs and Conditional privacy preserving authentication scheme. From the table 1.2, it shows the authentication time using Greedy Detection for VANETs is less when compared to local identity-based anonymous message authentication protocol and novel efficient conditional privacy preserving authentication scheme. The graphical representation of authentication time is illustrated in fig. 1.5.
4.3 False Positive Rate False positive rate is defined as ratio of number of vehicular nodes that are incorrectly classified to the total number of vehicular nodes. It is measured in terms of percentage (%) using the equ. (3). đ??šđ?‘Žđ?‘™đ?‘ đ?‘’ đ?‘?đ?‘œđ?‘ đ?‘–đ?‘Ąđ?‘–đ?‘Łđ?‘’ đ?‘…đ?‘Žđ?‘Ąđ?‘’ = đ?‘ đ?‘˘đ?‘šđ?‘?đ?‘’đ?‘&#x; đ?‘œđ?‘“ đ?‘Łđ?‘’â„Žđ?‘–đ?‘?đ?‘˘đ?‘™đ?‘Žđ?‘&#x; đ?‘›đ?‘œđ?‘‘đ?‘’đ?‘ đ?‘Ąâ„Žđ?‘Žđ?‘Ą đ?‘Žđ?‘&#x;đ?‘’ đ?‘–đ?‘›đ?‘?đ?‘œđ?‘&#x;đ?‘&#x;đ?‘’đ?‘?đ?‘Ąđ?‘™đ?‘Ś đ?‘?đ?‘™đ?‘Žđ?‘ đ?‘ đ?‘–đ?‘“đ?‘–đ?‘’đ?‘‘ đ?‘‡đ?‘œđ?‘Ąđ?‘Žđ?‘™ đ?‘›đ?‘˘đ?‘šđ?‘?đ?‘’đ?‘&#x; đ?‘œđ?‘“ đ?‘Łđ?‘’â„Žđ?‘–đ?‘?đ?‘˘đ?‘™đ?‘Žđ?‘&#x; đ?‘›đ?‘œđ?‘‘đ?‘’đ?‘
equ‌. ..(3)
632
www.ijitce.co.uk
∗ 100
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
From the equ. (3), the false positive rate is calculated. When the false positive rate is low, the method is more efficient.
From the fig.1.6, false positive rate based on different number of vehicular node is explained. From the fig. 1.6, it describes the false positive rate using novel efficient conditional privacy preserving authentication scheme is less when compared to local identity-based anonymous message authentication protocol and Greedy Detection for VANETs. Because of authentication vehicles are carried out without group or ring signature. The signature verification of designed scheme not increases linearly with number of revoked vehicles. The revocation process of design the scheme efficient than anonymous certificate schemes when revocation of vehicles not enhances CRL suddenly. A novel efficient conditional privacy preserving authentication scheme has 43% low false positive rate than local identity-based anonymous message authentication protocol and consumes 29% low false positive rate than Greedy Detection for VANETs.
Table 1.3 Comparison of False Positive Rate Number False Positive Rate (%) of LIAP GDVAN NECPPA Vehicular scheme Nodes (Number) 10 32 25 14 20 35 27 16 30 37 30 19 40 41 32 22 50 43 35 26 60 39 31 23 70 42 34 25 80 45 36 28 90 47 39 31 100 50 42 34 Table 1.3 shows the false positive rate with number of vehicular nodes varies from 10 to 100. False positive rate comparison takes place on local identity-based anonymous message authentication protocol, Greedy Detection for VANETs and novel efficient conditional privacy preserving authentication scheme. The graphical representation of false positive rate is shown in fig. 1.6.
5. DISCUSSION ON LIMITATION OF AUTHENTICATION AND PRIVACY PRESERVING TECHNIQUES FOR SECURED COMMUNICATION LIAP was developed for VANETs where every vehicle and Road Side Unit (RSU) assigned with distinctive long term certification. Hybrid authentication protocol depending on PKI and identity-based signature enhances the security level and conditional privacy level in VANETs. The valid vehicle collected local master keys from RSU to produce the localized anonymous identity. Node authenticates the safetyrelated message through single authentication manner for increasing the authentication efficiency. But, the authentication overhead is not reduced using LIAP. Greedy Detection for VANETs developed for identifying the greedy behavior attacks in VANETs. The algorithm identified the existence of greedy behavior and
Fig. 1.6 Measure of False Positive Rate
633
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
established list of compromised nodes. GDVAN are passive, non-resource-intensive and failed to need variations in Medium Access Control (MAC) layer. The designed approach executed by any node of network and failed to require any modification of IEEE 802.11p standard. But, the reaction method against the greedy attacks not used in GDVAN approach to eliminate serious impacts. NECPPA employed the keys and essential parameters of system in Tamper Proof Device (TPD) of Road Side Units (RSUs). A secure and fast communicational link between TA and RSU insert TPD in RSU is efficient. The designed scheme for cost efficient than other online RSUB scheme it failed to need establishment of on-line RSUs in whole roads. The privacy preserving rate not improved using NECPPA. 6. CONCLUSION The comparison of different existing authentication and privacy preserving techniques for secured communication in VANET is studied. From the study, it is observed that existing techniques failed to reduce the authentication overhead using LIAP. The existing reaction method against the greedy attacks was not used in GDVAN approach to eliminate serious impacts. Further, the privacy preserving rate not improved using NECPPA. The wide range of experiments on existing methods computes the performance of many authentication and privacy preserving techniques for secured communication in VANET with limitations. In this research, work can be carried out using machine learning and cryptographic techniques for enhancing authentication and privacy preservation performance during communication in VANET.
634
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
REFERENCES Shibin Wang and Nianmin Yao, “LIAP: A local identity-based anonymous message authentication protocol in VANETs”, Computer Communications, Elsevier, Vol. 112, Pp. no.154–164, 2017. Mohamed Nidhal Mejri and Jalel BenOthman, “GDVAN: A New Greedy Behavior Attack Detection Algorithm for VANETs”, Journal of IEEE Transaction on Mobile Computing, Vol. 16, Iss.3, Pp. no 759–771, 2017. Seyed Morteza Pournaghi, Behnam Zahednejad, Majid Bayat and Yaghoub Farjami “NECPPA: A novel and efficient conditional privacypreserving authentication scheme for VANET”, Computer Networks, Springer, Vol. 134, Pp. no. 78–92, 2018. Hyo Jin Jo, In Seok Kim and Dong Hoon Lee “Reliable Cooperative Authentication for Vehicular Networks” IEEE Transactions on Intelligent Transportation Systems, IEEE Transactions on Intelligent Transportation Systems, Vol. 19, Iss. 4, Pp. no. 1065–1079, 2018. Jian Kang, Dan Lin, Wei Jiang and Elisa Bertino, “Highly efficient randomized authentication in VANETs”, Pervasive and Mobile Computing, Elsevier, Vol. 44, Pp. no 31–44, 2018. Xia Feng, Chun-yan Li, De-xin Chen and Jin Tang, “A method for defensing against multi-source Sybil attacks in VANET”, Peer-to-Peer Networking and Applications, Springer, Vol. 10, Iss. 2, Pp. no. 305–314, 2017 Jieqiong Chen and Guoqiang Mao “On the security of warning message dissemination in vehicular Ad hoc networks”, Journal of Communications and Information Networks, Springer, Vol. 2, Iss. 2, Pp. no. 46–58, 2017 SK Hafizul Islam, Mohammad S. Obaidat, Pandi Vijayakumarc, Enas
[9]
635
Abdulhayd, Fagen Lie, M Krishna Chaitanya Reddyf, “A Robust and Efficient Password-based Conditional Privacy Preserving Authentication and Group-Key Agreement Protocol for VANETs”, Future Generation Computer Systems, Elsevier, Vol. 84, Pp. no. 216-227, 2018 Lei Zhang, QianhongWu, Josep Domingo-Ferrer, Bo Qin, and Chuanyan Hu, “Distributed Aggregate Privacy-Preserving Authentication in VANETs”, IEEE Transactions on Intelligent Transportation Systems, Vol. 18, Iss. 3, Pp. no. 516 – 526, 2017
www.ijitce.co.uk
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
ANALYTICAL MEASURES FOR DETECTING FRAUD USING CLASSIFICATION ALGORITHMS D .Vimal Kumar1 Associate Professor, Dept. Computer Science, Nehru Arts and Science College, Coimbatore, Tamil Nadu, India. Email: drvimalcs@gmail.com M.V.Jisha2 Ph.D Scholar, Dept. Computer Science, Nehru Arts and Science College, Coimbatore, Tamil Nadu, India. 2Email: jisharudhra@gmail.com ABSRACT- Abundant proliferation in usage of credit, debit and ATM card transactions, their use has become increasingly rampant in recent years. The proposed paper work investigates the efficiency of applying classifying algorithms to detect frauds prevailing in its usage. There exists various factors to analyze plentiful classification algorithms ,like KNN, Logistic Regression, Support Vector Machine and Random Forest. The proposed work analyzed that the performance of Random Forest is the efficient algorithm to detect the fraud transaction in terms of different factors . Keywords : Credit cards, KNN, Logistic Regression, Support Vector Machine and Random Forest I. INTRODUCTION Data mining is the process of discovering patterns in large data sets involving methods at the intersection of machine learning ,statistics, and database systems. Modern techniques based on Data mining, Machine learning , Sequence Alignment technique,Fuzzy Logic , Genetic Programming, Artificial Intelligence etc., has been introduced for detecting and preventing credit /ATM card ,cheque book type of fraudulent transactions(A.Shen,et.al,2007). Fraud detection is generally viewed as a
classification problem in data mining , where the objective is to correctly classify the credit card transactions as fraudulent. In the present scenario , when the term fraud comes into discussion ,credit card follows in the banks and the financial frauds done by the cash card cloning and various fraud clicks. With the increase in credit cards, ATM cards and Etransactions ,fraud has increasing excessively in recent years. Fraud detection includes analyzing of the spending behavior of users or customer order purpose, uncovering or escaping of undesirable behavior. As credit card becomes the most general mode of payment or both online as well as regular purchase ,fraud relate with it are also accelerate(S.Benson Edwin Raj,et al.2011). Fraud is a millions dough business and it is rising every year. Fraud presents significant cost to our financial prudence measure world wide. In this study, we evaluate to advanced data mining approaches, Random Forest and Support Vector Machines ,together with the well known logistic regression, K nearest neighbors, as part of an attempt to better detect credit card fraud .The study used the dataset from large data transactions. Statistical fraud detection methods have been divided into two broad categories: Supervised
636
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
and unsupervised. In supervised fraud detection methods , models are estimated based on the samples of fraudulent and legitimate transactions, to classify new transaction as fraudulent or legitimate. In unsupervised fraud detection , outliers or unusual transactions are identified as potential cases of fraudulent transactions. Both these fraud detection methods predict the probability of fraud in any given transaction. The remainder of this paper is organized as follows. Section II gives a brief description on credit card fraud ,the next section describes the methodology. Section IV gives the comparison of classification algorithms using various algorithms, followed by conclusion and references. II. CREDIT CARD FRAUD Credit card fraud has been divided into two types: Offline fraud and On-line fraud. Offline fraud is committed by using a stolen physical card at call center or any other place (D.W.Hosmer,et al.,2000). On-line fraud is committed via internet, phone ,shopping ,web, or in absence of card holder.Authorized users are permitted for credit card transactions by using the parameters such as credit card number, signatures, card holders address, expiry date etc . The unlawful use of card or card information without the knowledge of the owner itself and thus in an act of criminal deception refers to credit card fraud. Credit card fraud detection is quite confidential and is not much disclosed publicly as in Fig.1 . Commonly used detection methods are Rule-based techniques, Random Forest, Decision trees, Support Vector machines ,Logistic regression ,ANNs and metaheuristics such as k-means clustering , Genetic algorithms and nearest neighbors algorithms. Fraud is some kind of human behavior that
relate to larceny, misinterpretation, misrepresent, unethical , craftiness false suggestions etc. Many companies deals with millions of external parties, it is cost prohibitive to check the majority of the external parties activity and identity manually. Certainly, for investigating each suspicious transaction, they incur a direct overload cost for each of them. If in case, transaction amount is smaller than overhead cost, investigating is not worthwhile. Transaction involve among banking institutions offering financial transaction services, logistics companies offering various kind of transportation services. These transactions contain sensitive information in the form of data so there must be a technique which is applied on these financial transactions. The basic goals of information security, • Privacy : Information must be secret from unauthorized parties. • Integrity : Assurance that received data not contains any kind of alteration, addition ,scoring through or replay. • Authentication :The assurance that the communication entity is one that is claimed to be. • Non-repudiation : Sender and receiver prove their identities to each other. • Access control : This service controls that have access to a resource and under what condition access can occurs. The frequent use of plastic cash card is the most sensitive and vulnerable part of transaction system. It leads to the violation of all the above security issues and attracts skimmers. But the most important and prominent part is that when the customer access the ATM machine for transaction.
637
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Transaction Products Credit & Debit cards Relationanship to accounts
Technologies: ATM & Internet Utilization Counter Foit
Types of frauds
Handling of Transaction
Business Processes : Applications & Transactions
Manner & Timing
Fig.1. Types of Frauds
III. METHODOLOGY The methodology describes the different data mining techniques used, the dataset,the tool used and the analytical measures used to evaluate the performance of the mentioned classification algorithms. A. DATAMINING TECHNIQUES AND THE CONFUSION MATRIX We investigated the performance of four techniques in predicting fraud: Logistic regression (LR), Support Vector Machines (SVM), Random Forest (RF) and K nearest neighbours (KNN).In the paragraph below, we briefly describe the four techniques employed in this study( Siddhartha Bhattacharya, et al,2010). 1.) Logistic Regression:Qualitative response models are appropriate when dependent is categorical. In this study, our dependent variable fraud is binary and logistic regression is widely used in such problems(D.W. Hosmer et al.2000).For example, used binary choice models in the case of insurance frauds to predict the likelihood of a claim being fraudulent . Prior work in related areas has estimated logit
models of fraudulent claims insurance, food stamp programs and so forth (V. Vapnik,1998). The confusion matrix for Logistic Regression algorithm is given in Fig.2 ,which represents the actual detection of 100 % fraud from the dataset.We have the True Positive Value as 21,True Negative value as 633,False Positive value as 51 and False Negative as 557. Predicted class Actual class Row total Yes No Yes 557 21 578 No 51 684 633 Column total 608 654 1262 Fig.2. CONFUSION MATRIX FOR LR 2.) Support Vector Machines:Support Vector Machines are statistical learning techniques that have been found to be very successful in a variety of classification tasks. Several unique features of these algorithms make them suitable for binary classification programs like fraud detection SVMs are linear classifiers that works in a high dimensional feature space that is a non-linear mapping of the input space of the problem at hand. This simplicity of linear classifiers and the capability to work in a feature-rich space make SVMs attractive for fraud detection task where highly unbalanced nature of data (fraud and non-fraud cases) make extraction of meaningful features critical to the detection of fraudulent transaction is difficult to achieve . Applications of SVMs include informatics, machine vision, text categorization and time series analysis . The confusion matrix for Support Vector Machine algorithm is given in Fig.3 ,which represents the actual detection of 100 % fraud from the dataset.We have the True Positive Value as 41,True Negative value as 634,False Positive value as 20 and False Negative as 567.
638
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Actual class Yes No Column total
Predicted class Yes No 567 41 20 634
Row total 608 654
587
1262
675
Predicted class Row total Yes No 401 209 Yes 610 207 445 No 652 Column total 608 654 1262 Fig.3. CONFUSION MATRIX FOR KNN Actual class
Fig.3. CONFUSION MATRIX FOR SVM 3.) K-nearest neighbor:K-nearest neighbor algorithm is a non-parametric method used for classification and regression .In both case ,the input consist of the K closest training examples in the feature space. The output depends on whether K-NN is used for classification or regression.Both for classification and regression, a useful technique can be used to assign weight to the contribution of the neighbor, so that the closest neighbors contribute more to the average than the more distant ones. For example, a common weighting scheme consist in giving each neighbor a weight of 1/d where d is the distance to the neighbor. The neighbors are taken from a set of objects for which the class( for k-NN classification ) or the object property value (kNN regression) is known. This can be thought of as training set for the algorithm, though no explicit training step is required. It is used in application for face detection mean shift tracking analysis and typical computer vision problems. The confusion matrix for K-nearest neighbours algorithm is given in Fig.4 ,which represents the actual detection of 100 % fraud from the dataset.We have the True Positive Value as 209,True Negative value as 445,False Positive value as 207and False Negative as 401.
4.) Random Forests:The popularity of decision tree models in data mining arises from their ease of use, flexibility interms of handling various data attribute types, and interpretability . Single tree models , however, can be unstable and overly sensitive to specific data . Ensemble methods seek to address this problems by developing a set of models and aggregating their predictions in determining the class label for a data point . A Random Forest(Raghavendra Patidar et,al.,2011) model is an ensemble of classification (or regression) trees. Ensembles perform well when individual members are dissimilar , and Random Forests obtain variation among individual tress using two sources for randomness : first , each tree is build on bootstrapped samples of the training data; secondly, only a randomly selected subset of data attributes is considered at each node in building the individual trees. Random Forest thus combine the concept of bagging , where individual models in an ensemble are developed through sampling with replacement from the training data, and the random subspace method, where each tree in an ensemble is build from a random subset of attributes. Random Forest are computationally efficient since each tree is build independently of the others . Theyhave been applied in recent years across varied domains from crediting customer churn(Jisha.m.v et al,2018), image classification, to various bio-medical problems.
639
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
The confusion matrix for Random Forest algorithm is given in Fig.4 ,which represents the actual detection of 100 % fraud from the dataset.We have the True Positive Value as 4,True Negative value as 650,False Positive value as 37 and False Negative as 571. Predicted Row class total Actual class Yes No Yes 571 4 575 No 37 650 687 Column 608 654 1262 total Fig. 4. CONFUSION MATRIX FOR RANDOM FOREST B. Dataset A Comparison of the classification algorithms – Random Forest , KNN, Support Vector Machine and Logistic Regression by using the dataset containing 31 attributes and 3 lakhs (approx.) transactions for detecting the frauds .Analysis is done by highlighting the 16 factors ,to detect the best fraud detecting classification algorithm using the R tool( Jisha.M.V,2018). C. Tools Used R studio software is used in our work for the analysis of various algorithms. In Rsudio it is very easy to install required packages because of its user friendly behavior. It is an open source integrated development environment (IDE) for R programming language. R language provides a wide variety of statistical and modern graph techniques. It is very easy to understand and implanting a code with this tool. D. Analytical measures The performance of the proposed system was evaluated using Precision, Recall, F-Measure and Kappa Statistics. Precision, Recall and F-
Measure were calculated using the result of a confusion matrix. • Precision = TP/(TP=FP) • Recall = TP/(TP=FN) • F-Measure = 2 *[(Precision * Recall) / (Precision + Recall)] • Sensitivity= TP/FP • Specificity=TN/FN Along with the above factors ,the other evaluated factors are Accuracy, Kappa, Prevalence ,detection rate, detection prevalence, P -value, Positive Pred value ,Negative Pred value , Mcnemar’s Test PValue, AUC etc. The above mentioned factors are used for evaluating the performance of the four classifiers. 1) Comparison of the classifiers using precision:Precision measures the number of true positives divided by the number of true positives and false positives. In other words, precision is the measure of a classifier exactness. Table 3 presents the precision values of the four classifiers. It was observed that the Random forest has higher precision values than the other classifiers. A lower precision indicates large number of false positives. It is therefore inferred that other classifiers has more non-fraudulent transactions labeled as fraudulent. 2) Comparison of the classifiers using recall:Recall measures the number of the true positives divided by the number of true positives and the number of false negatives. In essence, recall can be thought of as a measure of the classifier completeness. A low recall indicates many false negatives(Oluwafolake Ayano, et al,2017). 3) Comparison of the classifiers using FMeasure:The F-Measure indicates the balance between the recall and precision values.
640
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
4) Comparison of the classifiers using Kappa statistics:Kappa statistics represent the extent to which the data collected correctly represents the variables measured. 5) Comparison of the classifiers using Sensitivity and Specificity:Sensitivity compares the amount of items correctly identified as fraud to the amount incorrectly listed as fraud ,also known as the ratio of true positives to false positives. Specificity refers the same concept with legitimate transactions, or the comparison of true negatives to false negatives. 6) Comparison of the classifiers using P-Value: A small p-value (typically <=0.05) indicates strong evidence against the null hypothesis, so we reject the null hypothesis. A large p-value (>0.05)indicates weak evidence against the null hypothesis, so we fail to reject the null hypothesis. 7) Comparison of the classifiers using Mc Nemarâ&#x20AC;&#x2122;s test: It is a statistical test used on paired normal data. It is applied to 2*2 contingency tables with a dichotomous trait, with matched pairs of subjects, to determine whether the row and column marginal frequencies are equal. It is named after Quinn McNemar, who introduced it in 1947. 8) Comparison of the classifiers using Prevalence:Prevalence is a term which means being widespread and it is distinct from incidence. Prevalence is a measurement of all individuals affected by the disease at a particular time, whereas incidence is a measurement of the number of new individuals who contract a disease during a particular period of time. 9) Comparison of the classifiers using Area under the curve: AUC is the abbreviation for area under the curve. It is used in classification analysis in order to determine which of the used
models predicts the classes best. An example of its application are ROC curves. Here, the True positive rates are plotted against False positive rates. 10) Comparison of the classifiers using Detection rate: Detection rate is mainly reflected in confusion matrix. It is a parameter that will vary according to the dataset. Detection rate =TP/(TP+FP+FN+TN). 11) Confusion Matrix Representation for Classification Algorithms: Confusion Matrix is a table that is often used to describe the performance of a classification model on a set of test data for which the true values are known. This matrix itself is relatively simple to understand , but the related terminology can be confusing.Each confusion matrix for each classification algorithm is given above. IV. Comparision Of Classification Algorithms Using Various Factors The Dataset is divided ,to detect the fraud from the large data transactions.Table1 shows the performance of the algorithms on detecting the 25 % fraud, we could analyse that for a small data set ,SVM is having high accuracy ,followed by Random Forest ,Logistic Regression and KNN.
641
FACTORS Accuracy 95% CI No Information Rate P-Value [Acc > NIR] Kappa Mcnemar's Test PValue Sensitivity Specificity
Table 1 Dataset With 25 % Fraud Data LOGISTIC RANDOM KNN REGRESSION FOREST 0.6518 0.9393 0.9521 0.5961, 0.9222, 0.9068, 0.9631 0.7045 0.9729
0.9617 0.934, 0.98
0.5335
SVM
0.5335
0.5335
0.5463
< 2e-16
< 2.2e-16
<2e-16
0.8777
0.9033
0.9228
0.3382
0.1687
0.009823
0.3865
0.6644 0.6407
0.9110 0.9641
0.9110 0.9880
0.9718 0.9532
1.49e05 0.3037
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Pos Pred Value Neg Pred Value Prevalence Detection Rate Detection Prevalence Balanced Accuracy 'Positive' Class AUC Precision F-measure
0.6178
0.9568
0.9852
0.9452
‘Positive’ Class
0.6859
0.9253
0.9270
0.9760
AUC
0.4665
0.4665
0.4665
0.4537
0.3099
0.4249
0.4249
0.4409
Precision F-Measure
0.5016
0.4441
0.4313
0.4665
0.6526
0.9375
0.9495
0.9625
Yes
Yes
Yes
Yes
0.652 0.5512 0.6025
0.941 0.9446 0.9274
0.956 0.9547 0.9323
0.963 0.9383 0.9547
Table 2 gives the comparison of classification algorithms for detecting 50% fraud from the given dataset.It is viewed that Random Forest is having high accuracy compared to other algorithms.From this point ,We could view that as the size of the dataset increases, the performance of SVM decreases.
0.624 0.6545 0.6292
Yes
Yes
Yes
0.969
0.968
0.974
0.9513 0.9604
0.9729 0.9731
0.9965 0.9797
Table 3 gives the comparison of algorithms with 75% fraud data from the dataset .It is viewed that that Random Forest is having high accuracy of 0.9776 ,followed by SVM with 0.97.There is a variable difference between the two algorithms .But comparing other factors, Random Forest algorithm is good in its performance than the other algorithms. Table 3 Dataset With 75 % Fraud Data
0.5259
<2e-16 0.9464 0.332 0.9636 0.9821 0.9798 0.9676 0.4741 0.4568 0.4662 0.9728
642
RANDOM FOREST 0.9776 0.9659, 0.9861
0.5197
0.5304
0.5197
0.5197
4.604e15
< 2e-16
< 2e-16
< 2e-16
0.2888
0.9401
0.9187
0.955
0.2068
0.08897
0.01496
0.00225
0.6044 0.6838
0.9795 0.9618
0.9400 0.9774
0.9600 0.9938
0.6385
0.9578
0.9747
0.9931
0.6517
0.9815
0.9463
0.9641
0.4803
0.4696
0.4803
0.4803
0.2903
0.4600
0.4514
0.4610
0.4546
0.4803
0.4632
0.4642
0.6441
0.9707
0.9587
0.9769
Yes
Yes
Yes
Yes
0.645 0.6312 0.6173
0.971 0.9088 0.9427
0.960 0.9810 0.9599
0.979 0.9952 0.9771
KNN
SVM
Accuracy
0.6457 0.6141, 0.6763
No Information Rate P-Value [Acc > NIR] Kappa Mcnemar's Test PValue Sensitivity Specificity Pos Pred Value Neg Pred Value Prevalence Detection Rate Detection Prevalence Balanced Accuracy ‘Positive’ Class AUC Precision F-Measure
RANDOM FOREST 0.9733 0.9576, 0.9844
0.9701 0.9571, 0.9801
LOGISTIC REGRESSION 0.9594 0.9448, 0.9711
FACTORS
95% CI Table.2 Dataset With 50 % Fraud Data LOGISTIC FACTORS KNN SVM REGRESSION Accuracy 0.6248 0.9686 0.9686 0.5859, 0.9519, 0.9519, 95% CI 0.6625 0.9807 0.9807 No Information 0.5259 0.529 <2e-16 Rate P-Value 2.995e[Acc > <2e-16 0.9371 07 NIR] Kappa 0.2477 0.937 0.5259 Mcnemar's 0.6917 Test P0.8231 0.5023 Value Sensitivity 0.6060 0.9700 0.9735 Specificity 0.6418 0.9674 0.9642 Pos Pred 0.6040 0.9636 0.9608 Value Neg Pred 0.6437 0.9731 0.9758 Value Prevalence 0.4741 0.4710 0.4741 Detection 0.2873 0.4568 0.4615 Rate Detection 0.4757 0.4741 0.4804 Prevalence Balanced 0.6239 0.9687 0.9688 Accuracy
Yes
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
Table 4 gives the comparison of the algorithm for detecting 100% fraud data from the large dataset.It is viewed that that Random Forest is having high accuracy of 0.9675 ,followed by SVM with 0.9517.Finally considering all other factors, Random Forest algorithm is good in its performance than the other algorithms in detecting fraud from a large data transactions
FACTORS Accuracy 95% CI No Information Rate P-Value [Acc > NIR] Kappa Mcnemar's Test PValue Sensitivity Specificity Pos Pred Value Neg Pred Value Prevalence Detection Rate Detection Prevalence Balanced Accuracy â&#x20AC;&#x2DC;Positiveâ&#x20AC;&#x2122; Class AUC Precision F-measure
Table 4 Dataset With 100 % Fraud Data LOGISTIC KNN SVM REGRESSION 0.6704 0.9517 0.9429 0.6437, 0.9383, 0.9287, 0.9551 0.6963 0.9628
RANDOM FOREST 0.9675 0.9562, 0.9766
0.5349
0.5182
0.5182
<2e-16
< 2e-16
< 2.2e-16
< 2.2e-16
0.3399
0.9031
0.8855
0.9348
0.9609
0.01045
0.0006316
5.806e-07
0.6595 0.6804
0.9659 0.9393
0.9161 0.9679
0.9391 0.9939
0.6574
0.9326
0.9637
0.9930
0.6825
0.9694
0.9254
0.9461
0.4818
0.4651
0.4818
0.4818
0.3177
0.4493
0.4414
0.4525
0.4834
0.4818
0.4580
0.4556
0.6700
0.9526
0.9420
0.9665
Yes
Yes
Yes
Yes
0.670 0.6672 0.6633
0.953 0.9523 0.9590
0.945 0.9643 0.9394
0.970 0.9915 0.9645
IV.
CONCLUSION
In this paper , we have brief description discussion on the credit card fruad detection using four classifiers. The classifiers detects the fraud, from a large data transactions , using various factors The confusion matrix shows that among the classifiers, the Random Forest algorithm is efficient with high accuracy of 0.9675. The relative studies and our results arrive to the fact of detecting a proposed and efficient algorithm than random forest which is our next proposed work.
ACKNOWLEDGEMENT We Authors would like to thank all the contributors of the journal.All the authors whose articles helped us for our work, for reference , listed in reference . Wish our paper will be a reference for future scholars.
643
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
REFERENCES
[1]. Srivastava, Aman, et al , “Credit card fraud detection at merchant side using neural networks.” Computing for sustainable global development (INDIACom),2016 3rd International Conference on ,IEEE, 2016. [2]. Dal Pozzolo , Andrea, et al. “Credit card fraud detection and concept-drift adaptation with delayed supervised information .”Neural Networks (IJCNN),2015 International Joint Conference on, IEEE,2015. [3].A.Shen, R.Tong, and Y.Deng, “Application of classification models on redit card card fraud detection”,June 2007. [4].Abhinav Srivastava, Amlan Kundu, Shamik Sural and Arun K. Majumdar, “Creditcard Fraud Detection Using Hidden Markov model”IEEE, Transactions on Dependable and Secure Computing, vol.5, No 1., January-March 2008. [5]. S.Benson Edwin Raj, A. Annie Portia “ Analysis on Credit Card Fraud Detection Methods.”IEEE- International Conference on Computer, Communication and Electrical Technology; (2011).(152-156). [6]. Raghavendra Patidar, Lokesh Sharma, “Credit Card Fraud Detection Using Neural Networks ” , International Journal of Soft Computing and Engineering (IJSCE),2011.,Volume -1 Issue ,(32,38). [7] .Bhattacharya.S., et al. (2011). Data mining for credit card fraud :a comparative study. Decision support systems, vol.50, no.3, pp. 602613. [8]. Prajel Save, Pranali Tiwarekar , Ketan N.Jain , Neha Mahyavanshi , “ A Novel Idea for credit card fraud detection using decision tree”,International Journal of Computer Applications.,volume 161-no13,March-2017.
[9]. Pratiksha .L. Meshram, Parul Bhanarkar, “Credit and ATM Card Fraud Detection Using Genetic Approach”,International Journal of Engineering Research & Technology (IJERT),Vol.1 Issue 10, December-2012. [10].D.W. Hosmer, S. Lemeshow, Applied Logistic Regression, 2nd Ed,WileyInterscience,2000. [11]. V. Vapnik, Statistical Learning Theory, Wiley, New York., 1998. [12]. MJ Kim and T.S Kim, “A Neutral classifier with Fraud density Map for Effective Credit Card Fraud Detection”, Proc Int “I Conf.Intelligent Data Eng and Automated Learning. Pp. 378383,2002. [13] K. RamaKalyani, D.UmaDevi, “Fraud Deyection of Credit Card Payment System by Genetic Algorithm”,International Journal of Scientific & Engineering Research Volume 3, Issue 7, July-2012. [14] Ekrem Duman, M.Hamdi Ozcelik “ Detecting credit card fraud by genetic algorithm and scatter search”, Elsvier, Expert Systems with Applications, (2011). 38; (13057-13063) [15]. Jisha.M.V, Dr.D. Vimal Kumar, “ An Efficient Credit Card Fraud Classifier of the four data mining classification algorithmsA Comparative Analysis.”(JETIR)Journal of Emerging Technologies and Innovative Research. Nov.2018. [16]. Siddhartha Bhattacharya, et al, “Data mining for credit card fraud : A comparative study”,Decision Support Systems ,(2010). [17]. Oluwafolake Ayano, et al,“ A multialgorithm data mining classification approach for bank fraudulent transactions.” Academic journals African journal of mathematics and comuter science research,2017.
644
INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGY AND CREATIVE ENGINEERING (ISSN:2045-8711) VOL.9 NO.2 FEBURAY 2019
645