- Call For Paper 2016 | IJSRD Journal
- Impact Factor: 2.39 | ISSN (Online) : 2321-0613
- Publication charges : 750 INR
- Useful website for Engineering (M.tech,B.E/M.E/Ph.D) Students….
- We publish original and high quality papers.
- Leading E-Journal of INDIA
- Indexed in : Academia,ISSUU,Google Scholar,Research Bible,Scridb, Slideshare and many more
- Contact Number : 08866191212/22
- like us : http://www.facebook.com/ijsrd
- Submit Papers Online : http://www.ijsrd.com/SubmitManuscript
- visit : http://www.ijsrd.com
As we all know that cloud computing is one of the trending technology for data storage and sharing. Although its beneficial, it has to evolve more to be efficient in various models. One of the article submitted to IJSRD journal by Vishakha Om.Gupta and Prof. Yogesh Rai from Shree Institute of Science & Technology, RGPV University, Bhopal is focusing on Achieving CPEL Criterion in Cloud with High Bandwidth and Cost Reduction.
They briefly introduce their work as under…
Cloud computing implementation is fast and simple with benefits like, Cost reduction by gaining a worldclass IT infrastructure without the hardware, secure access to critical information 24/7 using only a Web browser, ability to stay mobile and many more. According to the survey 85 per cent of public sector organizations are using public cloud services, while on the other side data owner having concern about data confidentiality and query privacy due to the lost control of the infrastructure. To address these issues a criterion called CPEL is constructed; Data confidentiality, query privacy, efficient query processing, and low in-house processing cost. Satisfying these requirements will dramatically increase the complexity of constructing query services in the Cloud. Some related approaches have been developed to address some aspects of the problem. However, they do not satisfactorily address all of these aspects. We proposed the random space perturbation (RASP) approach with traffic redundancy elimination techniques, in which practical range query is constructed by using k-nearest neighbor (kNN) and range query algorithm, and to reduce redundancy in process predictive acknowledgement algorithm and redundancy aware routing algorithm is used by this way we can achieve CPEL criterion with high bandwidth of data processing with reduced cost.
RASP is combination order preserving encryption, dimensionality expansion, random noise injection, and random project because of order preserving encryption confidentiality and indexing preserved by dimensionality expansion range query processing can be achieve , Random projection is mainly used to process the high dimensional data into low dimensional data representations which provides good scaling potential and performance , random noise injection used for adding noise which in result gives proper output .
In random space perturbation, the word perturbation is used to do collapsing this process will happen according to the key value that is given by the owner. In this module the data owner have to register as owner and have to give owner name and key value. And then the user have register and get the key value and data owner name from the owner to do access in the cloud. Here user can submit their query as range query or kNN query and get their answer. We analyze and show the result with encrypted and also in decrypted format of the data for the query construct by the user. kNN query cannot be directly impose on RASP perturbated data, so we have designed a kNN query processing algorithm based on range queries .We use square ranges, instead of spherical ranges, to find the approximate kNN results, so that the RASP range query service can be used.
As shown in below fig.1, working of RASP with kNN and Range query service, but here authorized user can upload, download, updates the similar data which increases data processing cost, to address this problem we are using predictive acknowledgement approach which is based on byte stream processing, Byte stream means it store the file as 1 byte that is 8 bit as a sequence of file, because of these increases the file download speed while transaction between the cloud and its end user’s. As shown in below fig. 2.
From their study shown in the article published at IJSRD, they concluded that “The key vantage of this scheme is the ability to get confidential and efficient query services with high bandwidth and reduced cost.”
To know more about such articles with innovative ideas please visit ijsrdscholarticles website.
To download more of such type of articles helping you in your research work please visit www.ijsrd.com
nearest neighbor (kNN) and range query algorithm, and to reduce redundancy in process predictive acknowledgement algorithm and redundancy aware routing algorithm is used by this way we can achieve CPEL criterion with high bandwidth of data processing with reduced cost
HP isn’t content to just 3D print in plastic. A job ad for a “robotics scientist for 3D printing” sounds normal enough until you dive into the text and read this:
“HP Labs’ research into printing of inorganic materials is working towards hybrid printing of glass (and other inorganic materials) onto items that are already mass produced,” the ad reads.
3D printing is generally reserved for working with plastic and metal. Glass is unusual. But HP has its reasons. According to a 2012 HP Labs paper:
HP is due to release its first-ever 3D printer in October. We don’t know much about it except that it will be aimed at businesses and a potential boost for the ailing company. It’s unlikely it’s a glass printer though. This is a project that still lives within HP Labs.
Reference : https://gigaom.com/2014/09/26/hp-labs-is-working-on-a-glass-3d-printer/
See our new video…
Publication Charges :- 750 INR
Impact Factor : 1.26
India’s Fastest Growing E-Journal Year 2013….
Collection of More Then 850 Research Papers….
Visit : www.ijsrd.com
|back-propagation algorithm, supervised learning method.|
The standard back-propagation algorithm is one of the most widely used algorithm for training feed-forward neural networks. One major drawback of this algorithm is it might fall into local minima and slow convergence rate. Natural gradient descent is principal method for solving nonlinear function is presented and is combined with the modified back-propagation algorithm yielding a new fast training multilayer algorithm. This paper describes new approach to natural gradient learning in which the number of parameters necessary is much smaller than the natural gradient algorithm. This new method exploits the algebraic structure of the parameter space to reduce the space and time complexity of algorithm and improve its performance.
Full Paper Link :http://www.ijsrd.com/Article.php?manuscript=IJSRDV1I6007
For More Papers Visit : www.ijsrd.com
Call for Paper volume 2:
SUBMIT YOUR RESEARCH ARTICLE VIA FOLLOWING LINK…