Professional Documents
Culture Documents
Mini Project On Machine Learning: Support Vector Machine
Mini Project On Machine Learning: Support Vector Machine
Members:
Ekta Bhosale (Roll No-153 and PRN no. 71627738H)
Virangni Gaikwad (Roll No-156 and PRN no. 71627571G)
Ankita Istalkar (Roll No-160 and PRN no. 71627750G)
Rajashri Tapkir (Roll No-161 and PRN no. 71627815E)
Title:
Support Vector Machine for classification
Problem Statement:
Apply the support vector machine for classification on a data set
obtained from UCI ML Repository.
Dataset used:-
IRIS dataset.
Requirements:-
1)Ubuntu
2)Python
THEORY:
Hyperplanes:
1)Hyperplanes are decision boundaries that are used to classify the
data points.
2)The dimension of the hyperplane depends upon the number of
features.
B) Margin
It is the shortest distance from hyperplane to one side of its margin is
equal to the shortest distance from the hyperplane to the other side
of the margin, where the “sides” of the margin are parallel to the
hyperplane.
3.POLYNOMIAL KERNEL:
In machine learning, the polynomial kernel is a kernel function
commonly used with support vector machine (SVMs) and other
kernalizedmodels, that represents the similarity of vectors (training
samples) in a feature space over polynomials of the original variables,
allowing learning of non-linear models.
Conclusion:
Hence we have implemented a mini project on Support Vector Machine.
Members:
Ekta Bhosale (Roll No-153 and PRN no. 71627738H)
Virangni Gaikwad (Roll No-156 and PRN no. 71627571G)
Ankita Istalkar (Roll No-160 and PRN no. 71627750G)
Rajashri Tapkir (Roll No-161 and PRN no. 71627815E)
Title :
Data sharing over the cloud framework using Security Algorithms .
Problem definition:
For proposing an online resource management frame-work that helps
in providing more security to the data shared / communicated among
people using it.
Prerequisite:
Concepts of Cloud Computing and some security algorithms , Web
Technologies and Java.
Software Requirements:
1. Operating System: Linux
2. IDE: Eclipse (Juno) / NetBeans 8.2
3. Programming Language: JAVA, JSP
4. Java Version: JDK 1.7 or Above
5. Database: MySQL 5.5
Outcome:
Proposed system can be used to enhanced privacy preservation in
cloud service.
Theory :
Cloud computing:
Cloud computing is a strategy for conveying data innovation (IT)
benefits in which assets are recovered from the Web through online
instruments and applications, rather than an immediate association with a
server. Instead of keeping documents on a restrictive hard drive or nearby
stockpiling gadget, cloud-based capacity makes it conceivable to spare
them to a remote database. For whatever length of time that an electronic
gadget approaches the web, it approaches the information and the product
projects to run it. It's called cloud computing in light of the fact that the
data stored in the cloud and does not require a client to be in a particular
place to access it. This kind of framework enables representatives to work
remotely.
Organizations giving cloud administrations empowers clients to store
records and applications on remote servers, and after that entrance every
one of the information by means of the web. Information and Data security
and access control is a testing research work in cloud computing. Cloud
bene t clients transfer their private and secret information over the cloud.
Security must be given to such outsourced information, with the goal that
client is not stressed while transferring their classified information.
Proposed System:
Conclusion:
To address privacy issue of sensitive data on cloud, proposed a
crypto-system for secure sharing of data over the cloud. This work shows a
security framework that can give protection and uprightness to trading
delicate data through the cloud or the correspondence systems, in view of
the utilization of the combination of Attribute-Based Encryption and Byte
Rotation Encryption Algorithm.
So, the proposed scheme is possibly applied for enhanced data
protection in cloud application and reduce the encryption and decryption
time using multi-core parallel processing. This indicates that the proposed
system can be used to enhanced privacy preservation in cloud service.