Transcript STOCK MARKET MANAGEMENT USING TEMPORAL DATA

STOCK MARKET MANAGEMENT USING
TEMPORAL DATA MINING
TEAM MEMBERS
STALIN ANTONY RAJ . A
ARAVIND . V
PROJECT GUIDE
Mr . ASHOK KUMAR
ABSTRACT
The stock market domain is a dynamic and
unpredictable environment. Traditional techniques,
such as fundamental and technical analysis can
provide investors with some tools for managing their
stocks and predicting their prices. However, these
techniques cannot discover all the possible relations
between stocks and thus there is a need for a different
approach that will provide a deeper kind of analysis.
Data mining can be used extensively in the financial
markets and help in stock-price forecasting. Therefore,
we propose in this paper a portfolio management
solution with business intelligence characteristics .
Objective
 Designing intelligent stock market assistant using
temporal data mining.
 The main aim of this system is to provide
valuable suggestions and results about details of
all stocks present in the market to the depositor.
 This software offers a wide range of standard
modules providing a highly effective stock
market assistant solution for all type of investors.
EXISTING SYSTEM
• The stocks are selling based on the company influence such like P/E factor,
• Volume, Business Sector, Historical Behavior, Rumors, Book (Net Asset)
Value, Stock Earnings
• Data mining can be used extensively in the financial markets and help in
stock-price forecasting
• In olden days stock maintenance is not accurate and efficiency.
• Stock maintenance doesn’t follow any exact algorithm and all.
PROPOSED SYSTEM
•
The capabilities of this tool are based on temporal data mining patterns, extracted
from stock market data
•
In the monitoring part, the user is able to define stock portfolios, to view stock
Price values over time of companies with similar characteristics (e.g. same
business sector, price range, P/E etc...)
•
In the prediction part, the tool helps users to decide on their stock trades. A
sequence mining algorithm is used in order to identify frequently occurring
sequences of stock fluctuations and thus recommend some good trading
opportunities, based on the extracted frequent patterns.
•
Temporal data mining was introduced to improve the accurate details of stock
maintenance
SOFTWARE REQUIREMENTS
o Front End: ASP.Net & C#
o Back End: SQL Server 2005
o Client side Scripting: JavaScript
HARDWARE REQUIREMENTS
Processor
RAM
Monitor/Panel
Hard Disk
Keyboard
Mouse
:
:
:
:
:
:
Intel Pentium IV
512 MB
Plug and Play Monitor
40 GB
109 Keys
PS/2 Compatible Mouse
DATA FLOW DIAGRAM
username
login
password
Level 0
db
Enter profile
db
registration
user
login
invalid
Valid user
Select option
View all share
db
Level 1
db
db
Enter profile
db
Release
share
registration
View report
Update share
User or
Company
login
Valid user
invalid
Select option
Predict share
View all share
Buy share
db
db
db
Level 2
db
Registration Process for both
Company and Investor
 For first entry with stock market management
every company should register with that.
 After completion of registration work only
company can release their various types of
shares.
 Shareholders have to register to view
suggestions and related details with stock
market management.
 The company registration includes all details
of the company and investor registration will
contains all details about an investor.
Share Release
• In this module company will able to release
it’s all types of shares with various names.
Investor can’t do anything in this share
release.
• Share release control will come under
company. Share release contains details like
company name, type of share released by that
company, quantity of released shares and like
information.
Share Updation
 After share released by a company it will be
updated by a company regularly.
 Share updations modules will contains details
like share type, price of the share, share
availability similar information of every company.
 The entire control of share updations will comes
under company; they won’t have permission for
share updating of other companies.
View Share
• Every company can see their own shares
based on change with index.
• One company not able to see other company
share status.
• Each company see the status of various type
of shares and status based on index.
Share Prediction
• This module will feel right to investor and an
investor can view different types of shares like
finance, accounts, IT, Steel etc, from various
companies.
• Every shareholder can see details about their
share updations from company.
• Investor can compare their share with other
company share.
Variation Status
• Day to day variation of stock is not only based
on stock price and it will also includes the
following factors like name of the company,
financial position of that company, price per
annual earnings etc.
• Investor can get idea about their share status
by comparison with other company shares.
Graphical Representation
• In this module investor by selecting his share
he can get the graphical representation of
variation of that share with occasion from its
releasing time.
Buy Share
• Shareholder can buy the number of shares he
wants to buy from the particular company.
• It will holds details like name of the company,
share type with its name ,availability of shares
etc.
SCREEN SHOTS
LOGIN FORM:
USER/AGENT REGISTRATION:
COMPANY REGISTRATION:
VIEW ALL SHARES:
PREDICT SHARE:
GRAPHICAL VIEW OF SHARE
VARIATION:
COMPANY:
RELEASE SHARE:
UPDATE TODAY SHARE:
VIEW SHARE BY INDEX:
REFERENCES
1.
2.
3.
4.
5.
6.
7.
Adriaans, P., Zantinge, D.: Data Mining. Addison Wesley Longman,
Harlow,England (1996) 1-10
Lin, W., Orgun, M., Williams, G.: An Overview of Temporal Data Mining. Proc. Of
the Australasian Data Mining Workshop, ADM02, Sydney, Australia (2002) 8390
Das, G., Gunopulos, D., Mannila, H.: Finding Similar Time Series. Proc. 3rd
Int.Conf. Knowledge Discovery and Data Mining (KDD ’97), Newport Beach,
California, USA (1997) 88-100
Agrawal, R., Faloutsos, C., Swami, A.: Efficient Similarity Search in Sequence
Databases. Proc. 4th Int. Conf. Foundations of Data Organizations and
Algorithms (FODO ’93), Chicago, IL USA (1993) 69-84
Chan, K.-P., Fu, A.W.-C.: Efficient Time Series Matching by Wavelets. Proc. 15th
Int. Conf. Data Engineering, Sydney, Australia (1999) 126-133
Agrawal, R., Srikant, R.: Mining sequential patterns. Proc. 11th Int. Conf. Data
Engineering (ICDE ’95), Taipei, Taiwan (1995) 3-14
Mannila, H., Toivonen, H., Verkano, I.: Discovery of frequent episodes in event
sequences. Data Mining and Knowledge Discovery, Vol. 1:3 (1997) 259-289
•
•
•
•
•
•
•
8. Bettini, C., Wang, X.S., Jajodia, S., Lin, J.L.: Discovering Temporal Relationships
with Multiple Granularities in Time Sequences. Technical Report, George Mason
University (1996)
9. Pediaditakis, K.: A Temporal Data Mining Approach for the Extraction of
Optimal Temporal Constraints of Sequential Patterns on Event Sequences. MPhil
Thesis, School of Informatics, Faculty of Humanities, University of Manchester
(2005)
10. Bettini, C., Wang, X.S., Jajodia, S.: Mining temporal relationships with
multiple granularities in time sequences. Data Engineering Bulletin (1998)
21:32-38
11. Manilla, H., Ronkainen, P.: Similarity of Event Sequences. Proc. 4th Int.
Workshop Temporal Representation and Reasoning (TIME ’97), Daytona Beach,
FL USA (1997) 136-139
12. Yahoo! Finance Web site: http://finance.yahoo.com
13. Weigend, A., Chen, F., Figlewski, S., Waterhouse, S.R.: Discovering Technical
Trades in the T-Bond Futures Market. Proc. 4th Int. Conf. Knowledge Discovery
and Data Mining (KDD ’98), New York, NY USA (1998) 354-358
14. Worcester Polytechnic Institute, Knowledge Discovery and Data Mining
Research