New user registration
Event organizer registration
Login via your institution account
Login via your organization account
Login via your publisher account
Login using your new user credentials
Forgot username or password?
Subscribe to SciArchives
Business, Management and Accounting >
Management Information Systems
Indian Journal of Education and Information Management
| ISSN:2277 - 5374
Publisher:Indian society for Education and Environment
Editor in chief:Prof.Natarajan Gajendran
Your selection(s) could not be saved due to an internal error. Please try again.
Browse by Volumes
Sentiment Analysis Using Crf And Optimal Temporal Boundary
Volume: 5 | Issue no: 4-2016 | Pagination: 1-8
Objective: To extract and categorize aspects based on Conditional Random Field to improve the sentiment classification accuracy and to determine the optimal time boundary for temporal sentiment analysis. Methods: There are several methods developed for sentiment analysis. In order to improve the sentiment classification accuracy and to identify sentiment variation Conditional Random Field model, firefly algorithm are proposed. Findings: Sentiment analysis is a part of data mining technique where we know users attitude, judgment, opinion and emotions about a particular product or event or a system. In an existing sentiment analysis technique the aspect terms were extracted using bag words. Then the polarities of aspects were determined through the Stanford parser; and lexicon based approach is used to classify the sentiments. In this approach spatial relationship among the aspect terms were ignored and it fails to concentrate on the temporal trends of sentiments. Application/improvements: To increase sentiment classification accuracy and to identify the sentiment variation conditional random field and firefly are proposed.
Improved Top K Similarity Join With Data Dependent Hashing In Heterogeneous Information Network
Author:C.P. Rushida, V.R. Nagarajan.
Volume: 5 | Issue no: 4-2016 | Pagination: 1-7
Objectives: To reduce the length of hash codes in Local Sensitive Hashing (LSH) Methods: Heterogeneous information network is a network where computers and other devices with different operating system are connected together. Today heterogeneous information network gets more attention in a network. But data mining becomes more difficult in heterogeneous information network. Similarity join is more important for many applications like online advertising, friend recommendation etc., similarity join is a measure of relationship between any two objects or strings or nodes. In this paper we considered the semantic meaning behind the paths to give top k similar pairs through Path-based Similarity join (PS-join) method. Then the expensive computations are removed by using bucket based data dependent hashing while the Local Sensitive Hashing is more expensive and involves more tedious process like to hold lengthier hash codes and approximate near neighbor problem. Findings: The proposed data dependent hashing reduced the computation cost, memory and storage cost of hash codes and also overcome the problem of approximate near neighbor. The experimental results prove that the proposed technique works more efficiently than the existing technique in terms of recall, running time, and error ratio. Application/Improvements: To increase the recall and to reduce the computation time and error ratio PS join with data dependent hashing is proposed.
A Novel Machine Learning Approach For The Prediction Of Subarachnoid Hemorrhage
Author:C. Dheeba, S.Vidhya.
Volume: 5 | Issue no: 3-2016 | Pagination: 1-8
Objectives: To predict outcome of patients with Subarachnoid Hemorrhage effectively by using novel ensemble classification method. Methods: The different machine learning approaches are used to improve the outcome of patients with SAH prediction. One of such approach utilizes random forest classifier which is used for enhancing the prediction accuracy. Findings: The outcome of patients with Subarachnoid Hemorrhage (SAH) prediction is helpful for guiding and caring patients. Such type of prediction is the most important in medical research area. Mostly SAH prediction is achieved by classification techniques such as decision rules, naive Bayesian classifiers, support vector machines, nearest neighbor classifiers and etc. However, these classifiers are not efficient for higher number of training cases. Application/Improvements: In this paper, we propose a novel ensemble classification technique for effective classification. In which, a random forest classifier is introduced for providing efficient classification by integrating various machine learning algorithms. The algorithms used are C4.5, REPTree, and PART. The experimental results show that the best ensemble classifier and effectiveness of the random forest algorithm. Keywords: Subarachnoid hemorrhage, Decision Tree classifier, Support vector machine, Naive Bayesian classifier, nearest neighbor classifier, Random forest algorithm.
Certificate Revocation Based On Weighted Voting Game And Secure Way Routing Protocol (Cr-Wvg-Swr)
Author:R.R. Pavithra, V.R. Nagarajan.
Volume: 5 | Issue no: 3-2016 | Pagination: 1-7
Objectives: To improve the security of MANET by effectively removing the malicious nodes through certificate revocation mechanism based on game theory approach. Methods: Weighted Voting Game (WVG) approach is most widely used game theory approach to find and remove the certificate of malicious nodes from the network. it used the voting mechanism in which each node can detect its nearby nodes behavior with the help of detection mechanism. Findings: Mobile Adhoc Network contains mobile nodes which are deployed randomly in the network. These mobile nodes should cooperate each other for successful communication. But due to the frequent changes in network topology, the infrastructure for communication is not available and also presence of some malicious nodes cause attacks in the networks routing path. Hence, security is the major concern in this type of network. Certificate revocation mechanism based on weighted voting game (CR-WVG) approach was introduced in which malicious nodes were identified based on the voting mechanism with respect to weights and quota of each node. However, security in routing paths was still a problem in this approach. Application/Improvements: To overcome this security issue, Certificate revocation based on weighted voting game and Secure Way Routing protocol (CR-WVG-SWR) is developed in which certificate from certificate authority (CA), shortest route discovered and data packets are encrypted using various encryption algorithms. Thus the performance of CR-WVG-SWR scheme is improved and compared with CR-WVG in terms of false revocation, revocation of malicious nodes, accuracy ratio of revocation and normalised time to revocation.
Taxofinder With Optimal Number Of Concepts And Word2vector For Efficient Taxonomy Learning
Author:Sritha S, B Mathumathi.
Volume: 5 | Issue no: 3-2016 | Pagination: 1-7
Objective: To find optimal number of concepts for taxofinder to learn taxonomy by using an efficient technique. The word2vector representation is presented to define the relationship among the concepts that improves the efficiency of taxonomy learning. Methods: There are several approaches were developed for taxonomy learning. Taxofinder is an approach that learns taxonomy based on graph representation. In this approach the concepts in text corpus were extracted and the concepts were represented in graph representation to define the associative strength between the concepts. Thus fixed number of concepts was given as input to the taxofinder to learn taxonomy it degrades the performance of taxofinder. Findings: Taxonomies represent the relation among the concepts within a domain. Taxofinder is an approach that learns taxonomy based on graph representation. In this approach the concepts in text corpus were extracted and the concepts were represented in graph representation to define the associative strength between the concepts. Thus fixed number of concepts was given as input to the taxofinder to learn taxonomy it degrades the performance of taxofinder. In this paper, an optimal number of concepts are determined and it is fed as input to the taxofinder. Then the optimal concepts are ranked and build CGraph based on optimal number of concepts and associative strength between the concepts. The associative strength is determined by using Word2vector model. Application/Improvements: To increase the precision, recall and f-measure for taxonomy construction taxofinder with optimal number of concepts and word2vector is proposed.
Issues And Challenges In Engineering College Libraries- A Study
Author:R. Arumugam Dr.K.Nithyanandam.
Volume: 5 | Issue no: 2-2016 | Pagination: 1-7
Objectives: To identify the issues and challenges in engineering college libraries, first we should identify the user needs. To identify the resources available in engineering college libraries. To find out the user needs of engineering college libraries. Methods: Data collected from 60 users of engineering college libraries by questionnaire method. This study was conducted in Jeppiaar SRR Engineering College Library users i.e. students teaching and non teaching members of the institution. Findings: Almost all the users are satisfied with Textbooks/Book bank resources and Reference books. Majority of the users agree those prescribed text books/book banks are available in library. Many users agree that all reference books for one subject is available, some of the users of engineering college library disagree with the resource. Almost all the users agree that project reports, CDs collections, back volumes of printed journals & question bank are not given outside the library. Majority of the users agree that limited number of computers with internet facility is available in the library. Improvements/Applications: It was observed from the study number of text books given to the users only by central library is impossible; it is possible by providing text books by book bank or department library. User satisfaction as well as resource utilization could be reached by issuing Journals, Magazines, Reference books, Project Reports, CDs Collections, and Back Volumes of the Journals etc can be issued to the users for one day by collecting their id card. Providing extension services ie evening library hours and shelf catalogue will save the time of the user.
Privacy Preserving Protocol Using K-Nearest Neighbor Algorithm For Cloud Based E-Healthcare Systems
Author:Mohanraj Govindaraj, Saritha Karthik.
Volume: 5 | Issue no: 2-2016 | Pagination: 1-8
Background/Objectives: To develop a kNN-privacy preserving model for preserving the privacy of the patients in a cloud assisted e-healthcare system as the sensitive information is needed to be maintained confidential and should not be revealed to public users other than the physicians. Methods/Statistical analysis: PPDM uses a privacy-preserving fully Homomorphic data aggregation as the basic scheme. The outsourcing of disease modeling and the early intervention is performed by developing privacypreserving function correlation matching PPDM1 from dynamic medical text mining and also a privacy-preserving medical image feature extraction PPDM2. Both PPDM1 and PPDM2 provides higher security level with reduced cipher text attach possibility and minimal overhead. Though the computational and communication overhead are reduced in PPDM, the use of correlation function threshold in PPDM1 can further be improved by utilizing an efficient machine learning algorithm. Hence, the simplest and efficient machine learning algorithm, k-nearest neighbor is utilized to develop kNN-PPP model. Findings: In kNN-PPP model, instead of using correlation function threshold based matching, secure squared Euclidean distance of encrypted personal data and encrypted physician template is determined and then matched with better probability. Secure squared Euclidean distance protocol and secure multiplication protocols are the most prominent protocols among those utilized in kNN-PPP model. Improvements/Applications: Using kNN-PPP protocols, the computation and communication overheads are also reduced considerably than the PPDM model for the better health status determination of the patients. Experimental results also show that the kNN-PPP model has minimized overheads and higher matching probability.
An Effective And Accurate Fusion Result From Multi Class Ensemble Classification
Author:C. Gayathri, R.Umarani.
Volume: 5 | Issue no: 1-2016 | Pagination: 1-7
Background/Objectives: Financial fraud detection is the most challenging task in an online transaction oriented applications which concern more to provide the secured environment for the users. Various researches has been conducted previously that focus on providing the most secured environment to the users by finding and preventing the malicious patterns. Methods/Statistical analysis: Classification is one of the most proved techniques for detecting the most malicious patterns that resides in the financial database by using which the malicious patterns can be identified. In our previous research work Optimal Ensemble Architecture Selection using Firefly and the dempster shafer theory based Ensemble is done for finding the fraudulent behaviour in the accurate manner. The ensemble classifier fusion approach used in the previous methodology called dempster shafer theory retrieves the fusion result as classifier output with more confidence value. This approach is computationally inefficient and doesn't concentrate on interrelation between different classifier results due to its additive measure property. Findings: This problem is resolved in this work by introducing the fuzzy integral measure based ensemble fusion using sugeno integral (FIM-EFSSI) and the fuzzy integral measure based ensemble fusion using Choquet integral (FIMEFSCI). These approaches can find the better and accurate Ensemble result by considering the relation between the different classifier results. Improvements/Applications: The experimental tests conducted were proves that the proposed approach provides better result than the existing approach in terms of improved classification accuracy in the matlab simulation environment.
Morphology And Taxonomy Of Oscillatoria Princeps Vaucher Ex Gomont (Oscillatoriales, Oscillatoriaceae)
Author:V. Uma Rani, U. Elaya Perumal , S. Palanivel.
Volume: 5 | Issue no: 1-2016 | Pagination: 1-5
Background: Oscillatoria princeps has been reported from various parts of India for its diversity but not studied in detail about morphology and reproduction hence we try to fill that lacuna. Methods: Samples were collected from a stream near Kanyakumari district, Tamil Nadu. One part was preserved in 4% formalin, while the other part was brought to unialgal culture by streaking method. The cultures were grown and maintained in ASN III (-NaCl) medium at the photoperiodic culture racks with 8 hours light and 16 hours dark. Semipermanent slides were prepared and Leica EC3 Microsystems was used for observation and documentation. Findings: The alga was dark blue green in colour, growing in clusters at the bottom of the stream. Individual filaments were blue green to olive green in colour. Mature trichome straight, cells much broader than long with distinct cross walls. Apical cells hemispherical with keritomized content. Individual cells were round in shape and the size varies from 57.60Ám to 69.05Ám in width and 5.20Ám to 9.55Ám in length thus the ratio of length and breadth as 1:8. When such cells were stacked one above the other it gives "stack of poker chips" appearance, characteristic feature of the species. Notches were found at the inner side of the cell corresponding to the slit on the trichome. When such cells were stacked one above the other, it results in the crack like appearance on the trichome. One to three notches / slits were observed in the present study. Reproduction both by fragmentation and hormogonia (formation of separation disc) was observed. Application/Improvements: The new finding like cracks/ slit like structure on trichome can be taken for further studies. The microphotographs will enable the researchers to have better understanding.
A Study On Factors Affecting Need Recognition For Life Insurance Policies In Chennai
Author:U. Manikumar, S.Premkumar.
Volume: 5 | Issue no: 1-2016 | Pagination: 1-6
Background/Objectives: This study examines the factors influencing need recognition of Life Insurance Policies on purchase of different types of plan and role of family life cycle in purchase decision of insurance policies. Methods/Statistical analysis: The 500 life insurance policy holders of Life Insurance Corporation (LIC) are chosen as test sample through multi stage random sampling technique. The frequency and percentage analysis are carried out to understand the demographic profile. The exploratory factor analysis, ANOVA (Analysis of Variance) and the multiple regression analysis are done to examine the influence of factors affecting need recognition for life insurance policies on purchase of types of plan. Findings: The study reveals that majority of life insurance policy holders are males and most of life insurance policy holders belong to the family life cycle of Bachelor Type. Most of life insurance policy holders are graduates and majority of life insurance policy holders are employed in private sector. Most of life insurance policy holders belong to the monthly income group of Rs. 20001 - 30000 and majority of life insurance policy holders are married. The exploratory factor analysis shows that social security, peer influence, prosperity and safety are the factors affecting need recognition for life insurance policies. There is significant difference between demographic profile of life insurance policy holders and factors affecting need recognition for life insurance policies except gender and factors affecting need recognition for life insurance policies. The regression analysis indicates that social security, prosperity and safety are positively and significantly influencing the purchase of type of plan. Application/Improvements: The insurance products of LIC should be designed and promoted as per the needs of the valuable policy holders. Since the family life cycle is the most important factor deciding the purchasing of life insurance policies, LIC should develop and market separate insurance policies for different stages of life cycle of policy holders.
Contact and support
Terms and conditions
Frequently asked questions
Copyright © 2015
All rights reserved. SciArchives.