Knowledge Digest For IT Community

2y ago
28 Views
3 Downloads
2.54 MB
7 Pages
Last View : 22d ago
Last Download : 2m ago
Upload by : Kian Swinton
Transcription

52 pages including coverVolume No. 41 Issue No. 97 DecemberOctober 2017201750/-www.csi-india.orgwww.csi-india.orgISSN ISSN0970-647X0970-647XKnowledge Digest for IT CommunityCOVER STORYCyber Physical Systems (CPS) andCOVER STORYits Implications 8CSI Nihilent eGovernance Awards 7TECHNICAL TRENDSMachine Learning inTECHNICAL TRENDSAdvanced Python 11Meri Sadak 2.0 :One step closer to SMART CITY 15RESEARCH FRONTEnterprise Information Security RiskRESEARCH FRONTManagement 20Remote Monitoring and Localizationusing Sensors: Tools for e-Governance 17ARTICLEApplication Security using Blockchain inCyber ARTICLEPhysical System 25OntologyModeling in E-Governance for aSECURITYCORNERSemanticDigital25 Systems 31SecurityIssues inCyberIndiaPhysical

CSI COMMUNICATIONSVOLUME NO. 41 ISSUE NO. 9 DECEMBER 2017Chief EditorS S AGRAWALKIIT Group, GurgaonEditorPRASHANT R. NAIRAmrita Vishwa Vidyapeetham, CoimbatorePublished byA. K. NAYAKHony. SecretaryFor Computer Society of IndiaContentsCover StoryCyber Physical Systems (CPS) and its ImplicationsS. Suseela and T. KavithaTechnical Trends8Machine Learning in Advanced PythonSuchithra M S and Maya L Pai11Blockchain: A Primer15Editorial Board:Durgesh Barwal, Rajat Kumar Behera and Abhaya Kumar SahooBhabani Shankar Prasad Mishra,Research FrontArun B Samaddar, NIT, SikkimKIIT University, BhubanewarDebajyoti Mukhopadhyay, MIT, PuneJ. Yogapriya, Kongunadu Engg. College, TrichyEnterprise Information Security Risk ManagementK. Srujan Raju and M. Varaprasad RaoM Sasikumar, CDAC, Mumbai,ArticlesR Subburaj, SRM University, ChennaiPoonam N. Railkar, Sandesh Mahamure and Dr. Parikshit N. MahalleR K Samanta, Siliguri Inst. of Tech., West BengalR N Behera, NIC, BhubaneswarSudhakar A M, University of MysoreSunil Pandey, ITS, GhaziabadShailesh K Srivastava, NIC, PatnaVishal Mehrotra, TCSApplication Security using Blockchain in Cyber Physical System25Cyber Physical Systems and Smart Cities29Nishtha Kesswani and Sanjay KumarSecurity CornerSecurity Issues in Cyber Physical Systems31Cyber Security and Human Rights34Swati Maurya and Anurag JainSubrata Paul, Anirban Mitra and Brojo Kishore MishraDesign, Print and Dispatch byPractitioner WorkbenchGP OFFSET PVT. LTD.Fun with Digital Image Processing in PHP on Windows and Linux PlatformPlease note:CSI Communications is published by ComputerSociety of India, a non-profit organization.Views and opinions expressed in the CSICommunications are those of individualauthors, contributors and advertisers and theymay differ from policies and official statementsof CSI. These should not be construed as legalor professional advice. The CSI, the publisher,the editors and the contributors are notresponsible for any decisions taken by readerson the basis of these views and opinions.Although every care is being taken to ensuregenuineness of the writings in this publication,CSI Communications does not attest to theoriginality of the respective authors’ content. 2012 CSI. All rights reserved.Instructors are permitted to photocopy isolatedarticles for non-commercial classroom usewithout fee. For any other copying, reprint orrepublication, permission must be obtainedin writing from the Society. Copying for otherthan personal use or internal reference, or ofarticles or columns not owned by the Societywithout explicit permission of the Society orthe copyright owner is strictly prohibited.20Baisa L. Gunjal36PLUSKnow Your CSIICANN 60CSI Patna Chapter ReportReport on CSI Student Conventions :Karnataka & Haryana State Level conventionState Student Convention 2017, West BengalLatex Workshop & Workshop on Python - Programming Tool for Data ScienceCSI ReportsStudent Branches NewsCSI Calendar 2017-182nd Cover6740The 2017 India-Africa ICT SummitBack Page414142443rd CoverPrinted and Published by Prof. A. K. Nayak on behalf of Computer Society of India, Printed at G.P. Offset Pvt. Ltd.269 / A2, Shah & Nahar Industrial Estate, Dhanraj Mill Compound, Lower Parel (W), Mumbai 400 013 and published fromComputer Society of India, Samruddhi Venture Park, Unit-3, 4th Floor, Marol Industrial Area, Andheri (East), Mumbai 400 093.Tel. : 022-2926 1700 Fax : 022-2830 2133 Email : hq@csi-india.org3CSI COMMUNICATIONS DECEMBER 2017

EditorialDear Fellow CSI Members,The theme for the Computer Society of India (CSI) Communications (The Knowledge Digest for ITCommunity) December 2017 issue is Cyber Physical Systems.Prof. (Dr.) S. S. Agrawal“Cyber-Physical Systems or “smart” systems are co-engineered interacting networks of physical andcomputational components. These systems will provide the foundation of our critical infrastructure,form the basis of emerging and future smart services, and improve our quality of life in many areas.”National Institute of Standard & Technology (NIST), USAChief EditorProf. Prashant R. NairEditorAfter a series of thematic issues focusing on ICT in applications such as education, governance,agriculture and health, CSI Communications is focusing on cyber physical systems in this issue afteran issue on the research topic of machine learning. The next issue is also based on research theme,Machine Intelligence.Cyber Physical Systems (CPS) is poised to bring advances in personalized health care, emergencyresponse, traffic flow management, and electric power generation and delivery. This technologybuilds on embedded systems, computers and software embedded in devices whose principle missionis not computation, such as cars, toys, medical devices, and scientific instruments. CPS integratesthe dynamics of the physical processes with those of the software and networking, providingabstractions and modeling, design, and analysis techniques for the integrated wholeThe Cover story in this issue is “Cyber Physical Systems (CPS) and its Implications” by S. Suseela &T. Kavitha. In the cover story, the authors have traced the evolution and described the architecture,applications, platforms and functions of CPS.The technical trends showcased are “Machine Learning in Advanced Python” by Suchithra M.S. &Maya L Pai and “Blockchain: A Primer” by Durgesh Barwal Rajat Kumar Behera & Abhaya KumarSahooIn Research front, we have “Enterprise Information Security Risk Management” by K. SrujanRaju & M. Varaprasad Rao, who throw light upon current research and approaches for enterpriseinformation security risk management.Other articles in this issue on CPS provide us information on its applications in smart cities byNishtha Kesswani & Sanjay Kumar and Application Security using Blockchain in CPS by Poonam N.Railkar Sandesh Mahamure & Parikshit N. MahalleThe Security Corner has 2 contributions, “Security Issues in Cyber Physical Systems” by SwatiMaurya & Anurag Jain and “Cyber Security and Human Rights” by Subrata Paul, Anirban Mitra &Brojo Kishore Mishra.We have revived the Practitioner’s Workbench in this issue with “Fun with Digital Image Processingin PHP on Windows and Linux Platform” by Baisa L. GunjalThis issue also contains collage of ICANN 60 participation by CSI, MoU with Cisco, CSI activity reportsfrom chapters & student branches and calendar of eventsWe are thankful to entire ExecCom for their continuous support in bringing this issue successfully.We wish to express our sincere gratitude to the CSI publications committee, editorial board, authorsand reviewers for their contributions and support to this issue.We look forward to receive constructive feedback and suggestions from our esteemed membersand readers at csic@csi-india.org.With kind regards,Prof. (Dr.) S. S. Agrawal, Chief EditorProf. Prashant R. Nair, Editor4CSI COMMUNICATIONS DECEMBER 2017www.csi-india.org

TECHNICAL TRENDSMachine Learning in Advanced PythonSuchithra M SMaya L PaiSchool of Arts & Sciences, Amrita University, Kochi, India.Email: suchithrams194@gmail.comSchool of Arts & Sciences, Amrita University, Kochi, India.Email: mayalpai@gmail.comMachine learning is a growing field and a motivated developer can quickly learn it up and start makingvery real and useful contributions. Machine learning algorithms are a big part of machine learning.Machine learning algorithms contain a lot of mathematics and theory. But we do not need to knowabout algorithm’s work to be able to implement them and apply them to achieve real and valuableresults. This is achieved through different machine learning tools. In this study, we explain aboutmachine learning and machine learning algorithms. The usage of machine learning tools like Weka, Rand Python and a review on recent trends of machine learning is also given due attention.Index Terms - machine learning, algorithms, tools, python.I.IntroductionA machine learning developeris a developer that built machinelearning systems. These systemscontain algorithms that could learnfrom data. Applied machine learningcan be overwhelming. There are somany things to try and explore on agiven problem. The developer can usea structured process, just like using astructured process to develop software[1]. The template for a multi-stepprocess when using machine learningto address a complex problem is1. Define the problem.2.Prepare the data.3.Spot checkalgorithms.variouslearning4.Tune well-performingalgorithms.learning5.Visualize the results.To speed up the process,understand the problem a little bit frommany different perspectives. What is the problem? Why does the problem need to besolved? How would I solve the problem?This last step helps us tounderstand why the problem is complexand requires a machine learning basedsolution. To get the best results, wemust understand how algorithms work.Mathematics plays an important role inunderstanding algorithms. There is amuch easier way by using the languageand methods that developers alreadyknow: Simple and clear algorithmdescriptions.might be worth spending some time ontuning. Test Harness algorithm is usedto evaluate different methodologies onthe same problem by comparing theresults from different techniques. Code examples without libraries.We can build up functions toevaluate predictions, estimate theskill of models and even implementthe learning algorithms themselves.A machine learning professional usesmachine learning to solve real-worldproblemsII.Applied machine learningUnderstanding of the following fourareas are needed for designing appliedmachine learning projects [2].1.Data Preparation:In this method, the developer loadsthe data from standard CSV file formatfor manipulation and prepares the datafor machine learning algorithms. Theperformance of algorithm on testingdata can be estimated using algorithmevaluation techniques. To evaluatethe efficiency of predictions made onunseen data the scoring methods areused. The best worse case results areanalyzed through Baseline Modelingtechniques to improve on a problem.Once we have a test harness that wecan trust, select and evaluate 5 to10 standard workhorse algorithms.This gives us an idea of how difficultour problem is and which algorithms2.Linear Algorithms:Simple Linear Regression [3]:It is used for numerical valueprediction and the dataset contains onlya single input. Multivariate Linear Regression:It is also used for numerical valueprediction and the dataset containsmore than one input. It is trained byusing Stochastic Gradient Descent. Logistic Regression:This method is used for class valueprediction on two class problems andit is trained by Stochastic GradientDescent. Perceptron:The easiest model of neuralnetwork for classification problems isperceptron and it is trained by usingStochastic Gradient Descent.3.Nonlinear Algorithms Regression and ClassificationTrees:These are decision trees andthat are applied to regression andclassification problems. Naive Bayes:It is an application of Bayes’Theorem for classification problems.11CSI COMMUNICATIONS DECEMBER 2017

TECHNICAL TRENDSThe theory of probability is the base forNaïve Bayes.Backpropagation:The commonly used method ofartificial neural network and it is widelyapplicable to supervised learning orclassification that roots the broaderfield of deep learning.user. That is by giving an utterancefrom a user, it identifies the specificrequest made. k-Nearest Neighbors (KNN):These algorithms are used forpredicting categorical or numericaloutputs directly from the training data. LearningVectorQuantization(LVQ):A widely used method of neuralnetwork is LVQ which is more efficientthan KNN.4.Ensemble Algorithms Bootstrap Aggregation:It involves an ensemble of decisiontrees and also known as bagging. Random Forest:This is an extension of baggingwhich results in faster training andbetter performance. Stacked Aggregation:This method learns how to combinethe predictions from multiple models inan efficient method. It is an ensemblemethod and also known as blending orstacking.Many complex machine learningproblems can be reduced to one offour core problem types: Classification,Regression, Clustering and Ruleextraction. If we can map everydayproblems to one of these problems,we can then find and start testingalgorithms that can address thoseproblems. Examples of machinelearning problems:1. Spam Detection: To identify thegiven email message in a mailinbox as spam or not.2. Credit Card Fraud Detection: Toidentify the credit card transactionsthat were not made by the customerby the giving the transactions for acustomer in a month.3. Digit Recognition: To identifythe digit for each handwrittencharacter by giving the handwrittenzip codes on envelopes.4. Speech Understanding: To identifythe specific request made by theIV. Machine learning algorithmsMachine learning is closely relatedto many fields, i.e., it is a multidisciplinaryfield. It is very difficult to differentiatemachine learning from related fields.Machine Learning is built on the fieldof Computer Science and mathematics.Knowing these foundational fieldscan help us to understand why certainmathematical language is used whendescribing algorithms, such as vectors,matrices, functions and distributions.Three specific foundational fieldsinclude: Probability: It is the study ofcharacterizing the possibility ofrandom events. Statistics: It is the study ofprocesses to collect, analyzes,explain and present data. ArtificialIntelligence:Itisthe construction and study ofcomputational intelligent systems.Machine learning also has siblingfields that sit alongside. These specialfields give context to machine learningalgorithms. These include: Computational Intelligence: Itis the study and construction ofcomplex systems. Data Mining: It is the constructionand study of computational systemsthat discover useful relationshipsand patterns from large data sets.A useful way to group algorithmsis by their similarity in structure orlearning style [4]. The five classes ofmachine learning algorithm that can beused to group algorithms by structureand learning style are:1. Regression: linear regression,logistic regression and stepwiseregression.2. I n s t a n c e - b a s e d M e t h o d s :k-nearest neighbor, learning vectorquantization and self-organizingmap.3. Decision Tree Learning: C4.5, CARTand ID3.4. Kernel Methods: support vectormachine, radial basis network andlinear discriminant analysis.5. ArtificialNeuralNetworks:Perceptron, Hopfield and back-propagation.Our goal is to effectively use timeto process algorithms. That is to builda robust test harness so that we canthrow algorithms in and very quicklylearn what works and what doesn’t.There are 2 concerns when buildinga test harness: What is the performance measuresused to evaluate algorithms? What data to use to train and testour algorithm? Once we have a test harness thatwe can trust, select and evaluate5-to-10standardworkhorsealgorithms. This gives us an ideaof how difficult our problem is andwhich algorithms might be worthspending some time on tuning. Thistechnique is called spot-checking.There are two main tactics thatwe can use to get the most outof machine learning algorithms:Algorithm tuning and Ensembles.Generally,machinelearningalgorithms can be explained aslearning a output function (f) thatperfectly maps input variables (P)to an output variable (Q).Q f (P)Our goal in evaluating differentalgorithmsandevendifferentconfigurations of an algorithm is to finda good approximation for the outputfunction (f) to get really good predictions(Q) [5].We can often get a boost wellperforming models. These techniquesare called ensemble machine learningalgorithms and are often internallysimpler than we first think. Wheninvestigating how machine learningalgorithms work, there are twoensemble methods I would recommendlooking into:1. Bagging (e.g.: Random forest)2. Boosting (e.g.: Adaboost)These are two very simplefoundations of very powerful ensemblemachine learning algorithms [6].V.Machine Learning Tools1.Weka ToolThe best machine learning tool forbeginners is Weka. There are three mainreasons to use Weka for beginners:12CSI COMMUNICATIONS DECEMBER 2017www.csi-india.org

TECHNICAL TRENDS It has a graphical interface,which means that there is noprogramming. It offers a suite of state-of-theart machine learning algorithms,including ensemble methods. It is free and open source software.Weka platform allows us to quicklydesign and run experiments. We mustexperiment to discover how to get goodresults. The Weka experimenter allowsus to do this.1. Start Weka2. Design a new experiment Select a Dataset Select one or more algorithmsor algorithm configurations3. Run the experiment4. Review the results and usestatistics to check for significanceWith a few clicks we can quicklydesign experiments to test our ideasand intuitions on our problem. It is avery powerful feature that few machinelearning platforms offer.2.R ToolR is a platform that is used bysome of the best data scientists in theworld. The reason is not the strangescripting language. It is because of thevast number of techniques available.Academics that develop new machinelearning algorithms use R, meaningthat often new algorithms appearon R platform before any other. Withpackages like caret, we can accesshundreds of the top machine learningalgorithms in R through a consistentinterface, ideal for spot checkingtechniques on our dataset.1.PythonPython cannot be ignored inmachine learning. It is rapidly catchingup to platforms like R in terms ofcapability and adoption. The cause is thescikit-learn Python library for machinelearning that is built on top of the SciPystack, harnessing the speed and powerof Python libraries such as Numpy forfast data manipulation at C-like speeds.The scikit-learn library is fully featured,offering a suite of algorithms to choosefrom as well as data preparationscheme and clever Pipelines that allowus to design how data flows from oneelement to the next.Python is the fastest-growingplatform for applied machine learningamong experts of data scientists.We cannot get started with machinelearning in Python until we have accessto the platform. We must downloadand install the Python 2.7 platform onour computer. We also need to installthe SciPy platform and the scikitlearn library. We can install everythingat once with Anaconda. Anaconda isrecommended for beginners. We canload our own data from CSV files.The general structure for workingthrough a machine learning problemin Python with Pandas and scikit-learncan be divided into 6 steps:1. Install the Python and SciPyplatform.2. Load a standard dataset.3. Summarizethedatausingstatistical functions in Pandas.4. Visualize the data using plottingfunction in Pandas.5. Evaluatemachinelearningalgorithms in scikit-learn.6. Develop a final model and makesome predictions on new data.The better we can understand ourdata, the better and more accurate themodels that we can build. The first stepto understanding our data is to usedescriptive statistics. To learn how touse descriptive statistics to understandour data, the helper functions providedon the Pandas Data Frame. A secondway to improve

The usage of machine learning tools like Weka, R and Python and a review on recent trends of machine learning is also given due attention. Index Terms - machine learning, algorithms, tools, python. I. Introduction: A machine learning developer : is a developer that built

Related Documents:

Bruksanvisning för bilstereo . Bruksanvisning for bilstereo . Instrukcja obsługi samochodowego odtwarzacza stereo . Operating Instructions for Car Stereo . 610-104 . SV . Bruksanvisning i original

10 tips och tricks för att lyckas med ert sap-projekt 20 SAPSANYTT 2/2015 De flesta projektledare känner säkert till Cobb’s paradox. Martin Cobb verkade som CIO för sekretariatet för Treasury Board of Canada 1995 då han ställde frågan

service i Norge och Finland drivs inom ramen för ett enskilt företag (NRK. 1 och Yleisradio), fin ns det i Sverige tre: Ett för tv (Sveriges Television , SVT ), ett för radio (Sveriges Radio , SR ) och ett för utbildnings program (Sveriges Utbildningsradio, UR, vilket till följd av sin begränsade storlek inte återfinns bland de 25 största

Hotell För hotell anges de tre klasserna A/B, C och D. Det betyder att den "normala" standarden C är acceptabel men att motiven för en högre standard är starka. Ljudklass C motsvarar de tidigare normkraven för hotell, ljudklass A/B motsvarar kraven för moderna hotell med hög standard och ljudklass D kan användas vid

LÄS NOGGRANT FÖLJANDE VILLKOR FÖR APPLE DEVELOPER PROGRAM LICENCE . Apple Developer Program License Agreement Syfte Du vill använda Apple-mjukvara (enligt definitionen nedan) för att utveckla en eller flera Applikationer (enligt definitionen nedan) för Apple-märkta produkter. . Applikationer som utvecklas för iOS-produkter, Apple .

Real Property Value/Acre 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 2010 2014 2016 2019 Net M&O Digest Residential Digest Commercial Digest Industrial Digest Combined Near West Air Cities Based on data from Georgia Dept. of Revenue Tax Digest Consolidated Summary 20

Extracting the RC4 secret key of the Open Smart Grid Protocol (OSGP) 9 OSGP data integrity For each message, generate a digest (hash value) using the secret "Open Media Access Key" (OMAK): OSGP-Digest-plaintext message Algorithm 12-byte OMAK 8-byte digest Data Concentrator (DC) Device Transmit message and its digest: plaintext digest OMAK OMAK

Reader [s Digest maintain a strong connection with the Asian community. Reader [s Digest currently and will continue to circulate educational institutes throughout Asia, helping shape the young minds of Asias youth. Reader [s Digest is proud to be one of the only media brands in the world, to be a core part of the education system. Helping shape