/PlayingwPlaggie

Grabbed Plaggie off of SourceForge to understand ins and out...potentially massage the application

Primary LanguageJava

README for Plaggie

Program Code Plagiarism Detection Tool plag.parser.plaggie.Plaggie

  1. License information

Copyright (C) 2006 Aleksi Ahtiainen and Mikko Rahikainen.

See COPYING.PLAGGIE for Plaggie license. See COPYING.CUP for license on the CUP generated parts. See COPYING.JAVA15GRAMMAR for the license of the Java 1.5 grammar that was used in this software.

........................................................................ Copyright notice for Plaggie:

This program (Plaggie) is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License as published
by the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
02110-1301  USA

........................................................................ Copyright notice for CUP Parser Generator:

CUP Parser Generator Copyright Notice, License, and Disclaimer Copyright 1996-1999 by Scott Hudson, Frank Flannery, C. Scott Ananian

Permission to use, copy, modify, and distribute this software and its documentation for any purpose and without fee is hereby granted, provided that the above copyright notice appear in all copies and that both the copyright notice and this permission notice and warranty disclaimer appear in supporting documentation, and that the names of the authors or their employers not be used in advertising or publicity pertaining to distribution of the software without specific, written prior permission.

The authors and their employers disclaim all warranties with regard to this software, including all implied warranties of merchantability and fitness. In no event shall the authors or their employers be liable for any special, indirect or consequential damages or any damages whatsoever resulting from loss of use, data or profits, whether in an action of contract, negligence or other tortious action, arising out of or in connection with the use or performance of this software. This is an open source license. It is also GPL-Compatible (see entry for "Standard ML of New Jersey"). The portions of CUP output which are hard-coded into the CUP source code are (naturally) covered by this same license, as is the CUP runtime code linked with the generated parser.

Java is a trademark of Sun Microsystems, Inc. References to the Java programming language in relation to JLex are not meant to imply that Sun endorses this product.

........................................................................ Copyright notice for Java 1.5. Grammar:

/* Java 1.5 parser for CUP.

  • Based on a Java 1.5 pre-release version by C. Scott Ananian.
  • Extended to support annotations by Jeremy Brown. Original version
  • copyright (C) C. Scott Ananian, 2003; modifications copyright (C)
  • Jeremy H. Brown, 2006. This program is released under the terms
  • of the GPL; see the file COPYING for more details. There is NO
  • WARRANTY on this code.
    */
  1. Introduction

This tool was created to address the problem of program code plagiarism at the Helsinki University of Technology. The tool is a stand-alone command line Java application and can be used to check programming assignment submissions for possible plagiarism.

It is important to know that this is just a tool and the reported possible plagiarism cases still need to be checked by hand before taking further action.

  1. Current list of features

The current features of the tool are listed here. See the section 12 (Further Development) for information on suggesting new features.

  • Programming languages supported: Java 1.5
  • Configuration via a configuration file
  • Command line interface for running the program
  • Detection reports generated in graphical HTML format (using frames)
  • Exclusion of code from the detection according to the following criteria:
    • filename
    • subdirectory name
    • interface
  • Exclusion of template code (i.e. code distributed as a part of the assignment to the students) from the detection
  • Blacklist file for highlighting results of certain students
  1. Algorithm used

The basic algorithm used for comparing two source code files is tokenization followed by greedy string tiling. This algorithm is documented in [1]. The algorithm in the current implementation is not optimized with the optimizations documented also in [1].

The algorithm has been extended to support the exclusion of existing code (e.g. code given within the programming assignment description) from the comparison.

Two similarity values (similarity values A and B, corresponding to two submissions A and B) are counted both between two files and two submissions, which can consist of several files. The file similarity value is the percentage of matched tokens compared to the total number of tokens within the file. The submission similarity value A is calculated by taking the best match (i.e. the highest file similarity value A) of each file in submission A and counting the average of these values. The submission similarity value B is the average of the corresponding similarity values in submission B.

  1. Installation

  • Prerequisites and external software

  • Extraction of the detection tool (plaggie.tar.gz)

    • Create an installation directory for the plagiarism tool
    • Extract the package in it
    • Make sure that the installation directory is in your CLASSPATH

  • Installation of the prerequisite software and packages

    • if JavaCUP is not installed:
      • Create an installation directory for the JavaCUP files and extract the package in it
      • Make sure that the JavaCUP directory is in your CLASSPATH
      • Compile the JavaCUP files in the installation directory: javac java_cup/.java java_cup/runtime/.java

  • Installation of the plagiarism detection tool

    • The detection tool is provided as Java source files, which need to be compiled in the installation directory:
      • First, edit Makefile and set variables CUP_PATH and PLAGGIE_PATH to point to the locations of CUP and Plaggie directories.
      • Then run: make all

  • Documentation

    • Documentation is provided in this README file. Additional configuration documentation is provided in the configuration file template, plaggie.properties.
  1. Configuration of the detection tool

Configuration parameters are documented within the configuration file (plaggie.properties). This file needs to exist in the directory where the plagiarism detection is executed. Before configuring and running the tool, the following issues need to be known about the programming exercise to be checked:

  • Are the students using some common code base or templates distributed along with the assignment? If so, then consider the following:

    • Those files not modified by the students should be added to the configuration parameter plag.parser.plaggie.excludeFiles
    • If the students are adding their code within these files, the base files have to be added to the configuration parameter plag.parser.plaggie.excludeCodeFiles.

  • If certain subdirectories within the submissions are not to be checked (e.g. if entire packages are distributed), they can be added to the plag.parser.plaggie.excludeSubdirectories configuration parameter.

  1. Running the detection tool

The tool is executed from the command line:

% java plag.parser.plaggie.Plaggie [-s] [-nohtml] directory1|file1 [directory2|file2]

At least one directory containing the submissions is required. If two directories or files are specified, they are assumed to be submissions of two students and are compared to each other.

Optional parameters:

-s A new value for the minimum submission similarity value. Overrides the value specified in the configuration file. -nohtml No HTML report generated, a textual report is always generated on the standard output

During execution the tool displays statistics of its progress every 5 second. After the detection has process has completed, the tool generates a report of the detected possible plagiarism cases. The format of the report can be specified in the configuration file; usually it is a set of HTML pages.

  1. The detection report
  • Viewing the report

The report can be viewed by any HTML browser, which supports frames.

  • Contents of the report

** Configuration

Information about the main properties of the configuration are included in the report.

** Statistics

Several statistics are generated of all the results (even the ones not included in the report separately). Distributions of the different similarity values are included. The different kinds of similarity values are documented below.

Also a distribution of the number of files in submissions is reported as are the final values of the counters.

** Report

Top results are reported as separate set of HTML pages. The results can be sorted according to different similarity values. The similarity values are:

  • Submission similarity Product of values A and B. High value in this is the best indicator of possible plagiarism case.
  • Submission similarity A and B Indicates the percentage of code in one submission, that was also found in the other submission.
  • Maximum file similarity The maximum similarity between a pair of source files in the two submissions.

By clicking the student identification numbers, an extensive report of the corresponding similarity comparison between two students is opened. In that report, the following similarity values between the students' files are shown:

  • File similarity A and B Indicates the percentage of tokens matched in the identical token sequences between the two files.
  1. Known problems

The detection algorithm is not at its best when using it on GUI code or some code generated using automated development tools. The problem with GUI code is, that it is common to have many (member) variable declarations or method calls in one section. In that case the algorithm might detect a match even if the variables and called methods are completely different.

  1. Known unsuccessful attacks

The following commonly used attacks are not usable againts this plagiarism detector, since the detection is based completely on structural analysis of the code:

  • Changing the comments

  • Changing the intendation

  • Method and variable name changes

The following attack does not work, because the files are compared to each other without taking the file names into account:

  • Renaming the classes
  1. Known successful attacks

Because the detection algorithm is based on finding similar token sequences of some minimum length, an attack can be made by altering the program code so much, that the similar sequences are shorter than the configured minimum match length. The minimum match length can be configured to be shorter, but then the possiblity for false matches increases.

The following attacks can be successful against the algorithm:

  • Moving inline code to separate methods and vice versa

  • Inclusion of redundant program code

  • Changing the order of if-else blocks and case-blocks

  1. Further development

If you have ideas about further development of the tool, please contact the author, Aleksi Ahtiainen, at Aleksi.Ahtiainen@iki.fi.

  1. List of references

[1] Prechelt Lutz, Malpohl Guido, Phlippsen Michael: JPlag: Finding plagiarisms among a set of programs, Fakultät für Informatik, Universität Karlsruhe. 28.3.2000, http://wwwipd.ira.uka.de/~prechelt/Biblio/jplagTR.pdf