iXtract

desertcockatooΔιαχείριση Δεδομένων

20 Νοε 2013 (πριν από 3 χρόνια και 4 μήνες)

68 εμφανίσεις







Specification Document

for

iXtract
.






























Prepared by

Center for Software Development

BITS Pilani

TABLE OF CONTENTS




1

INTRODUCING
IXTRACT
.

--------------------------------
--------------------------------
----------------------

3

1.1

W
HO
B
ENEFITS
F
ROM THIS
?

--------------------------------
--------------------------------
-------------------------

3


2

FUNCTIONALITY ELUCID
ATION

--------------------------------
--------------------------------
--------

3

2.1

D
ATA
S
EARCH
E
NGINE

--------------------------------
--------------------------------
-----------------------------

3

2.2

D
ATA
M
INING
E
NGINE

--------------------------------
--------------------------------
------------------------------

3

2.3

M
ETA
-
D
ATA
E
NGINE

--------------------------------
--------------------------------
-------------------------------

4

2.4

P
RESENT
ATION
E
NGINE

--------------------------------
--------------------------------
----------------------------

4


3

SOME NON USEFUL REQU
IREMENTS

--------------------------------
--------------------------------
--------

4






































Specification document for
IXtract



Center for Software Development

3

1

I
ntroduc
ing

i
Xtract
.


This document provide
s

some basic
requirements for
iXtract, a distributed
search and data m
ining engine. The need of the hour is to design a tool
that is used for retrieval of data
from

distributed knowledge repositories
and to mine through the same to achieve clustering, classification and
grouping of semi and unstructured data.



The system sh
all provide the following functionalities in the bare minimum
setup:



Data Search Engine



Data Mining Engine



Meta
-
Data Engine



Presentation Engine


1.1
Who Benefits From this?

The system should provide advances in “
Knowledge Worker
” productivity
and network P
erformance. Anyway who benefits from this?



Knowldege seeker / client
benefits

from

iXtract because
he/she is able to have access to data from various
distributed sources
with a high degree of accuracy and
relecvance.



Managers of knowledge repositories

will

help them in
mining information which in turn would help in the building
of knowledge bases from distributed repositories.

2

Functionality Elucidation


2.1

Data Search Engine


The Data Search Engine is used to carry out a distributed search for
retrieval of dat
a that is spread across the internet. The engine is
expected to use various search algorithms and should have the capability
of determining the choice of algorithm as the case demands.


2.2

Data Mining Engine


The data mining engine is used for data classifi
cation on the basis of user
parameters. As mentioned earlier, the data mining engine would classify
data into various categories so that a more effective search can be carried
out. Agent based technology would be used here and the agents would be
capable o
f self activation depending on system parameters.





Specification document for
IXtract



Center for Software Development

4

2.3

Meta
-
Data Engine


The Meta
-
Data Engine is used to moderate and specify meta
-
data
standards for the data retrieval process. This engine
will be used in the
building of knowledge bases from distributed repo
sitories. The Meta
-
Data
Engine would maintain meta
-
data about these various repositories that
are in various corners of the world.



2.4

Presentation Engine


The Presentation Engine is used for looking into the user interface aspects
of
iXtract.
This engine is

associated with the presentation of the results to
the end user of the system. Issues of ease of use and aesthetics will be
dealt with here.


3

Some Non useful requirements

1.

The system i
nvolves remoting and distributed programming
.

Though the
first choice o
f language would be C#, students are encouraged to do
the same in any object oriented language that they are comfortable
with.


2.

We would appreciate if the project designers meet frequently with the
CSD

team

to understand the
system (
Please be a bit lenien
t
with

us if
we are
late

at times

)
.