IJAIR- Volume 7 Issue 4 (April 2018)

 


Sr/No/ Title Pdf
1 AUGMENTING VOCABULARY THROUGH NEWSPAPERS.
Dr.S.V.Karthiga

Abstract—English language plays a very important role in developing an individual personality. Without knowledge of English vocabularies one cannot play or work easily in this competitive world. Language is a stem of total English language, branches like speaking, writing and other skills, moreover, the vocabulary is considered as leaves. Vocabulary plays main role in communication; communication tells us about a person’s ability and knowledge the way of using the vocabulary in a particular situation and how a person makes use of the apt vocabularies during conversations. Newspaper plays a major role in education field in order to develop student’s knowledge towards English language. This will be very useful for the students whose medium of instruction is not English. When Newspaper is read the learner’s vocabulary can be improved tremendously.
Page No. 01 - 03.


Download
2 DESIGN OF MODIFIED ICA CLASSIFIER BASED EFFICIENT SOUND-SOURCE LOCALIZATION
Dheeraj kumar yadav, prof. shadma pragi

Abstract—Proposed paper utilize a general source separation technique, Independent Component Analysis. Particularly, basic ICA was applied to separate mixtures for low frequency, narrow band, non-Gaussian signals by using closely spaced uni-directional microphones. localization routine worked with an average condition number for test cases. routine was tested on data collected form MATLAB standard audio files. Localizing sounds with different frequency & time domain characteristics in a dynamic listening environment is a challenging task that has not been explored in field for robotics as much as other perceptual tasks. This thesis presents an integrated auditory system localization method which can be used in humanoid robot, Sounds with different frequency components & time domain characteristics have to be localized.
Page No. 05 - 09.


Download
3 Design of Efficient Spectrum Sensing using modified Energy Detection in cognitive Radio system
Nandini Upadhyay, Prof Papiya dutta

Abstract—Range detecting includes acquiring the range use qualities over different measurements, for example, time, space, recurrence, and code and figuring out what kind of signs are involving the range. In this venture, OFDM based Cognitive Radio and Spectrum Sensing strategies specifically Energy Detection Based Spectrum Sensing with Wavelet bundle change and Cyclostationary Spectrum Sensing are examined. In numerous zones subjective radio structures exist together with other radio systems, using a similar range yet without making undue impedance. The most straightforward and simple to actualize detecting system is vitality location. Since, it doesn't require any earlier data of the flag exhibit in the recurrence band under perception. In this proposition, a few range detecting strategy is presented yet the fundamental work is conveyed for vitality identification strategies. Vitality recognition method is executed utilizing MATLAB for pseudo arbitrary succession generator.
Page No. 10 - 16.


Download
4 NEW GLCM & THRESHOLDING BASED SKIN DISEASE DETECTION WITH HIGH ACCURACY RATE
Sonam Vishwakarma, Prof. Aditya Shama

Abstract—Skin surface examination is one for than testing issues in field for helpful conclusion. Diverse sorts for skin ailments are impacting human life like skin dryness, living being, & ominously helpless reactions. objective for this paper is to separate skin ailment using surface examination for skin picture & by standing out test picture from a described pictures or reference pictures. Reinforce Vector Machine (SVM) based GLCM has been associated as impression for features. Treatment for SKIN TEXTURE pictures is one for parts for this field. MATLAB is been used as blueprint proposed system, to make arrangement simple to utilize a GUI is been created as front end use & outside for anyone's ability to see proposed estimation works, results found are incredible to extent precision & time delay.



Download
5 A Discovery Knowledge Extraction Hybrid approach using the Synaptic data discovery Model
Surbhi Tiwari , Asst. Prof. Nitesh Gupta

Abstract—Web Mining is an extraction of knowledge from the web data. A number of data get generate while working with web usage. An analysis of such data and finding the usable entity to provide a better user experience can be a advantage of algorithms. Thus an knowledge discovery and providing a quick solutions to the input query can be performed. Many approaches for the data analysis, weight analysis and data processing is performed by previous author. TF-IDF, Semantic, FP growth algorithm and such other techniques are used by previous research for knowledge analysis. In this paper, an advance synaptic data discovery model for the web data extraction and analysis is performed. The proposed algorithm work with the tree architecture based discovery and enable finding the relevant terminology. Thus finding a better solution for the prediction and finding a better knowledge query output is performed. The experiment result shows the effectiveness of proposed approach over the traditional algorithm.



Download
6 A Document Pattern Analysis over the document stream using Enhance NN Architecture
Pooja Singh , Asst. Prof. Nitesh Gupta

Abstract—Document annotation and its availability for the user is important in many application area. Application such as research labs, student labs where continuous document availability and release perform need document investigation. Document writing style gives different pattern with it. Pattern analysis gives understanding of document and its knowledge extraction. There are two type of patterns are available with document which is topical and sequential. Majorly document follows topical pattern which identify in easy manner. Rare document do follows sequential pattern, thus the detection techniques are not available. A research field to detect sequential pattern document is need to investigate. In this paper a survey related to document pattern detection is discussed. Various techniques and pattern analysis over the document stream performed is discussed. Existing author performed user aware rare sequential pattern approach. Related work performed research over twitter real time & synthetic dataset and thus shows the efficiency of their approach. Our further work is to find optimal technique for sequential pattern analysis for document stream in real time dataset.



Download
7 An Enhance Energy optimised Path Selections and Efficient Routing Approach
Shruti Dubey, Abhishek Pandey

Abstract—The networking node help in participation and transferring the data packets between the source & destination node. Mobile adhoc network help in secure and fast processing of data packets. Efficient routing and energy efficient path selection is always required in any of the communication area. Many algorithms are performed for optimizing the communication. In this paper, an optimized energy efficient path selection algorithm is proposed. This algorithm helps in finding an optimal route with the short node selection with maximizing the energy routing. This algorithm help in achieving high data packet delivery ratio and less packet drop. Thus the implementation is performed using NS2 and result observed are efficient while comparing with traditional OLSR approach.



Download
8 A HIGH SECURITY APPROACH WITH FAST INDEXING OVER CLOUD STORAGE HLMS
Varsha Sadhwani, Sunil Malviya

Abstract—In the field of cloud computing the approach for sharing of data and storage between a groups of people requires efficient approach with cost cutting and low maintenance procedure. Sharing of outsourced files assures security and ensures data veracity according to the frequent changes in shared data files. Issues like preserving privacy of data among the group stored data shared in common cloud storage. Prominent attacks within the cloud requires scheme of defense mechanisms use key distribution mechanism in secured communication channels. Some of the existing system use fine grain access control with revoked user that protects prominent attacks for original data achieving security. In our proposed paper we inculcate a Cloud Storage Controller (CSC) that manages the allocation of group of data stored and it can only referenced and cannot download any database from cloud. The data will be blended together into one of two sets for example if two users storing data of 2 megabyte and 3 megabyte each then that will be merged together as two datasets with 1 megabyte of first user and 1 megabyte of second user. In the second data set it merges 1 megabyte of first user and 2 megabyte of second user using Fusing Data Technique (FDT). Every dataset will be bookmarked which symbolizes the user that dataset belongs to and that will be indexed in the Cloud Storage controller (CSC). It helps in fast and highly secured datasets storage management in cloud storage system. It achieves efficient database privacy and revokes user data swiftly and securely.



Download