
Before you use this code, you should know
=======================
	1. This code was used for the empirical study of the journal paper 
	"On Good and Fair Paper-Reviewer Assignment".
	2. This code is written in C/C++.
	3. This code runs under Unix/Linux.
	4. In case that you encounter any problems when using this code,
	please figure out the problem by yourself 
	(The code in fact is easy to read and you can modify it for your own purpose).

Usage
=======================

Step 1: specify the data input information.
<This could be done by editing the "Config" file
whose format is explained in Appendix I.>

[Note]: Config File Must be in the same directory with the executable file.

Step 2: compile the source code
make

Step 3: Run the code
./PRA $(Config File Path) $(Data File Directory) $(COI_File_Path)

[Example]: ./PRA ./Config ./Data/VaryP/P_496 ./COI/COI0
[Note]: No slash after the $(Data File Directory) !!!

[Note]: All the COI files are provided in the COI Folder. The information for each COI file is provided at Appendix III

Step 4: Check result, statistics and running time

You can find the result files in the "Result" folder in $(Data File Directory)
Those information will be explained in Appendix II


Appendix

I. The format of Config.txt
=============================

(1)<# of paper>
(2)<# of reviewer>
(3)<# of paper demand constraint>
(4)<# of reviewer workload>

[Note]: Make sure that (1)*(3) <= (2)*(4) or there will be no answer since the total workforce is less than the demand !!!

II. The format of <stat.txt>
=============================
1. The stat.txt file contains information about the topic coverage of each paper and the average topic coverage percentage.
The format is like following.

a) For the topic coverage of each paper, the format is
------------------------------------------------------------------------
PaperID, # of Total Topics, # of Covered Topics, Topic Cover Percentage
------------------------------------------------------------------------

b) For the average topic coverage percentage, the format is
------------------------------------------------------------------------------------------------------------
# of Paper, # of Reviewer, Paper Demand Constraint, Reviewer Workload, Average Topic Coverage, COIBreakNo
------------------------------------------------------------------------------------------------------------

[Note:] Since there might be cases in which COI must be broke so that there is a solution, COIBreakNo records the total
number of COI breaks.

III. About files in COI Folder
===============================
First of all, the COI information is represented by matrix in the following file. Each row stands for one paper and each column stands for one reviewer. The row number and column number is the ID of the paper and reviewer, respectively.And the (i,j)th entry of the matrix indicates whether there exists a conflict in this kind of COI. (1 for yes and 0 for no)

The type(s) of COI information each file contains is listed as follows. 
-------------------
COI0 : No COI
COI1 : COI Type 1
COI2 : COI Type 2
COI3 : COI Type 3
COI4 : COI Type 4
COI123 : COI Type 1, 2, and 3
COI1234 : COI Type 1, 2, 3, and 4

[Note]: All the files includes the case of paper-reviewer pair (p, r) where r is the author of p
