GRAPH BASED ANALYSIS OF HUMAN COMMUNITIES

The objective of finding communities from human eProfiles is to

identify set of shortest path for group of people (community) having

individual eProfiles on some common parameters. Human eProfile

has different attributes like personal, professional, educational,

transactional, socio-economic, and other parameters. We will be

using Graph based Algorithm and Web Page Community

Identification Algorithm for the above said problem. We will consider

Email, SMS and Chatting graphs in order to analyze transactional /

interaction level. The benefit expected from our work is to categorize

human on the basis of their parameters by which search become

easier.

WEB-PROXY SERVER WITH CONTENT FILTER

Web-proxy servers act as the intermediary when a client requests for
resources from web servers. A web proxy server receives the
response, and then gives the response to the actual client which
requests for the resource. The idea is to block some particular
content from reaching the clients; this has been implemented by
adding functionalities to the web proxy server, since all the requests
are fulfilled through it. The web proxy server with content
management functionality would block the defined content from
reaching the client which requested it. Each client would have an
individual access-level. The proxy server would provide content
filtering based on following features:
a) Tier-level filtering: The web-proxy server provides three tier
filtering i.e., strict, medium, and normal.
b) Client-level filtering based on individual clients: Each client
would have an individual access-level, assigned by the
administrator of the web proxy server.
c) Multimedia-object filtering: Multimedia objects such as
streaming video and audio would be filtered based on title,
tags and links.
d) Text filtering: HTML pages would be filtered based on text
comparison.
e) Image-filtering: CBIR technique has been used for comparing
the images received as response from the remote server with
the ones defined by the administrator, in order to filter them.

JOBBER

The continued increase in the amount of content and the number for

information sources available online is making the need for effective

personalized content delivery more acute. This has resulted in a

renewed interest in Web personalization as an indispensable tool for

Web-based organizations. The portals developed up till now are

working on user based searches, which are somehow static in a way

that a person has to search each and every thing by himself/herself.

What If we provide them a way that automatically gets the

employment opportunity for them and they don’t have to start

searching all the way again and again. Our objective is to create a

job portal with an enhancement of web personalization to provide a

better help in automated searching and providing assistance, helping

the job seeker to find the best suited job. One more objective over

here is to provide the companies a platform to have their own Web

Portals to manage e-business. There will be a new type of job portal

that will provide unique set of implementation that will assist the user

in better ways based on individual’s interest, experience and skill

sets.

GESTURE RECOGNITION USING WII REMORTE

Gesture recognition is used in futuristic interfaces to enhance the

Human Computer Interaction. It also furthers the Ubiquitous

Computing vision, by removing the obstacles in directly interacting

with a computer. As humans, we frequently use gestures in everyday

tasks to communicate. Gesture recognition researchers are

investigating ways of making systems which can realize the vision of

using gestures to control computer systems. This research area

received a boost when Nintendo unveiled its Wii Console system

containing Wiimote (gesture making device). Since the success of

the Wii console, intensive research was started in this area, with

giants like Microsoft employing teams of researchers to create their

own gesture recognition technology. We propose to develop a

system (API) which will use the Wiimote to a) read and recognize

dynamic hand or finger 2D gestures, b) do real time tracking and c)

do pinch clicking, to perform operations on user applications. Such a

system could be integrated with new and existing applications, to be

controlled through the above mentioned type of gestures. A

demonstration shall be provided with some characteristic user

applications software. Additionally a Gesture Manager software will

be provided to easily do gesture management (creation, remove, add

etc.). And finally our work will provide a gesture system which will be

low-cost because of the low-cost and easy accessibility of the

Wiimote.

SUFFIX TREE BASED DOCUMENT CLUSTERING

Document clustering as an unsupervised approach is extensively
used to navigate, filter, summarize and manage large collection of
document repositories like Internet. Recently, focus in this domain
shifted from traditional vector based document similarity for clustering
to suffix tree based document similarity, as it offers more semantic
representation of the text. In this project, we compare and contrast
two recently introduced approaches to document clustering based on
suffix tree data model. The first is a phrase based document
clustering, which extracts phrases from documents to form a compact
document representation and uses a similarity measure based on the
phrases to cluster the documents. The second approach is a
frequent word/word meaning sequence based document clustering
which similarly extracts the common word sequences from the
document forms a compact representation and uses document
clustering approach to cluster the compact documents. These
algorithms are using agglomerative hierarchical document clustering
to perform the actual clustering step. The difference in these
approaches is mainly based on the model representation as a
compact document, and the similarity measures used for clustering.
This project investigates the computational aspect of the two
algorithms, and the quality of results they produce.

GENETIC ALGORITHM FOR WARFARIN DOSING

The daily dose of Warfarin, an anti-coagulant, typically takes 5 to 7days, a time period patients suffering from artificial valves, atrialfibration and permilary embolism can ill-afford. In the present study a
genetic algorithm, from the family of evolutionary algorithms, is integrated with artificial neural networks to develop an algorithm to
eliminate the uncertainty of dose and hasten the process of
prescription. The present prescription procedure only takes into account the patients clinical factors. Research shows that genetic factors are 33% responsible of prescription of warfarin dose. The
algorithm stimulates the behavior of a patient taking as input both
clinical and genetic factors of a patient. The algorithm suggests new
values of dose, with each iteration, until the optimum dose is
achieved. The algorithm reduces the one week period of dose
prescription to the matter of seconds.

HAND GESTURE RECOGNITION FOR DISABLED

The aim of this project is to develop an application capable of

recognizing a range of gestures from a single user through computer

vision techniques. A requirement of this project is to produce a low

cost, real time solution. Since we humans communicate and

exchange our thoughts, messages or information chiefly by vision

and sound, therefore a human computer interface would be more

intuitive if it made greater use of sound and vision recognition.

Another benefit is that, the user not need to have any physical

contact with the computer. However, unlike audio commands, a

visual system should be preferable in noisy environments or in

situations where sound would cause a disturbance. The visual

system we have chosen is the recognition of hand gestures. The

amount of computation required to process hand gestures is much

greater than that of the mechanical devices; however standard

desktop computers are now quick enough to make this project, hand

gesture recognition using computer vision, a viable proposition.

Potential Uses of Gesture Recognition

1) As an Input device for the disabled (deaf and dumb)

2) Human-Computer Interface

3) Security Purposes

DATA SECURITY WITH MULTIMEDIA OBJECTS

project is based on the concept of Steganography which is the

art of hiding/extracting information within cover medium contrary to

cryptography. Information is in the form of plaintext file. Cover

mediums may be an image or an audio file. The proposed system is

a website that hides the data inside the cover file. The system has

the capability of sending a StegoFile to a particular receiver, who can

have original data by receiving an email and by providing StegoFile

and key password to extract text from StegoFile using this system.

Detecting hidden message inside a StegoFile is difficult because

original message has been spread randomly inside cover file.