README
			----------------------
		Smart Phone based Risk Assessment V 1.0
		=======================================
CONTENTS
--------

1. Introduction
2. System requirements
3. tar file contents
4. Installation Instructions 
5. Instructions for Using the Software
6. Contact Information


1. Introduction
   ------------
   Risk assessment is an important step in Microfinance applications. 
   Here we provide a software for Risk assessment which can run on a Smart Phone.


2. System requirements
   -------------------
   Any J2ME compliant smartphone which supports 
   - MIDP 2.0
   - CLDC 1.1 
   - FileConnection APIs. 
   Nokia 6630, 6680, 6681 are some of such smart phones.


3. tar file contents
   -----------------
   Smart_Phone_based_Risk_Assessment.tz  file contains
      + report.pdf
      + code.tz  (tar file contains all components of code required for execution )
         |- src   (source code folder)
         |- RiskAssessor_device.jar  (jar file to be deployed on the smart phone)
         |- RiskAssessor_emulator.jar  (jar file required for the emulation)
      + Readme.txt  ( how to install and configure for the execution )
      + index.html 

4. Installation Instructions
   -------------------------
   Smart_Phone_based_Risk_Assessment.tz is distribution file for this software. It contains code 
   as well as jar files. Given below are instructions for installation -
   1) Extract files from Smart_Phone_based_Risk_Assessment.tz
   2) In the Smart_Phone_based_Risk_Assessment directory there is a pre-compiled jar file 
      named RiskAssessor_device.jar
   3) Install this jar file into the smartphone using cable or Bluetooth.
   4) Run the application named RiskAssessor


5. Instructions for Using the Software
   -----------------------------------
   1) There are 3 modes in which this software can run - Training Input mode, 
      Training Mode and Risk Assessment Mode.
   2) When running for the first time, the algorithm needs some training data. 
      So start with Training Input Mode. In case there is training data 
      already present you may skip this step.

      In Training Input mode, data is fed into the phone and is stored 
      in a file called at E:/trainingData. The data gets appended to the file 
      in case this file exists already.
   3) Once all the training data is fed into the specified file, 
      training can be performed. This involves classifying the given data 
      into 5 clusters. The centroids of the clusters are calculated using 
      K-means algorithm, and are displayed on screen for the user to 
      train the program. User is now required to decide which centroid 
      represents which type of cluster, namely - 
      Very Low Risk, Low Risk, Medium Risk, High Risk and Very High Risk. 
      User's response is stored into another file called E:/centroids
   4) Once training is done, Risk Assessment can be done for new inputs. 
      User is required to select Risk Assessment, fill details of Assessee 
      and then submit the data. Closeness of the given input with the 
      previously computed centroids is displayed as percentage. 
      Decision can be made based on the percentage displayed. 
      So if Very Low Risk gets maximum percentage, then it means that 
      the given input is closest to Very Low Risk cluster. 
      The user also has an option to enter the data into trainingData file.

6. Contact Information
   -------------------
   Developers can be contacted for support and enhancement on following address-
   girish@it.iit.ac.in , prathabk@it.iit.ac.in , radha@it.iit.ac.in