Software Engineering Of Web Applications
Web Based Stock Forecasters
Group 5
Amarinder Cheema
Ateet Vora
Chetan Jain
Puneet Kataria
Ronak Shah
Siddharth Wagh

Our system will require you to install the following softwares initially
1. PHP v 5.2.5
2. IIS Server 
3. Mysql Server v 5.0
4. Matlab v7.1
5. Visual Studio 2005
 
The user also needs to interface PHP with MySQL server and IIS server. The following link provides a tutorial to interface them :-
 http://www.atksolutions.com/articles/install_php_mysql_iis.html


The user then also needs to interface Matlab with MySQL Database. The followin link provides a tutorial to interface the two:-
http://torque.oncloud8.com/archives/000111.html

1. Create 4 tables in Mysql. 
Organizationtable with columns OrganizationName (varchar) and OrganizationID (integer)
Organizationhistory with columns Date, OrganizationID (integer), Open (Float), High (Float), Low (Float), Close (Float), Volume (Integer)
Organizationpredict with columns OrganizationID (integer), currentprice (Float), predictsignal (Integer), predictprice (Float), demopredictsignal (integer), demopredictprice (Float), hotstock (Integer)  
User Details with columns Userame (varchar) and Password (varchar)

2. Datamining
For this you need to put the StockAssignment.php file in the root folder and then enter the following url in the browser
http://localhost/StockAssignment1.php

3. Webservices
Create a new webservice project in ASP.Net and copy paste the file 'service.cs' which is in StockPredictor WS folder. Then run the code in Visual Studio.

4. For neural network stock prediction
You need to give the command mypredict at the command line and code runs. It will take atleast 20 mins for the training and testing phase. The processed values will be stored in the database with the help of the code. Hence webservices can pick these values from database when client initiates any webservice like get prediction ( ).

5. All html pages are provided in the source code submission. All these should be put in the root folder. 

If you have any difficulties in running the code, please feel free to contact us at 732-306-0849.