Uprava spolecnych casti ohledne text alignment

[pan13-paper.git] / pan13-paper / pan13-notebook.tex

diff --git a/pan13-paper/pan13-notebook.tex b/pan13-paper/pan13-notebook.tex
index 3febdcc6c2cee3222d7038b9e19f165b4878bb69..a6a711656247111e8eabf7c95e05c38c1fbf4a99 100755 (executable)
--- a/pan13-paper/pan13-notebook.tex
+++ b/pan13-paper/pan13-notebook.tex
@@ -7,7 +7,7 @@
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 \begin{document}
 
-\title{Improving plagiarism detection}
+\title{Neco Simonovo and Feature Type Selection for Pairwise Document Comparison}
 %%% Please do not remove the subtitle.
 \subtitle{Notebook for PAN at CLEF 2013}
 
@@ -21,22 +21,26 @@
 This paper describes approaches used for the Plagiarism Detection task in PAN 2013 international competition
 on uncovering plagiarism, authorship, and social software misuse.  
 We present modified three-way search methodology for Source Retrieval subtask and analyse snippet similarity performance.
-Next, we show changes in selected feature for text alignement which led to plagdet score improvement.
-The results of source retrieval show, that presented approach is adaptable in real-world plagiarism situations.
-Improved results for text alignment achieved in the competition overall third place.
+The results show, that presented approach is adaptable in real-world plagiarism situations.
+For the detailed comparison task, we discuss feature type selection,
+global postprocessing. We have significantly improved the pairwise comparison
+results with even further optimizations possible.
 \end{abstract}
 
 
 \section{Introduction}
 In PAN 2013 competition on plagiarism detection we participated in both the Source Retrieval
-and the Text Alignment subtask. In both tasks we adapted methodology used in PAN 2012.
+and the Text Alignment subtask. In both tasks we adapted methodology used in PAN 2012\footnote{%
+See \cite{pan2012} for an overview of PAN 2012 plagiarism detection campaign.} \cite{suchomel_kas_12}.
 Section~\ref{source_retr} describes querying approach for source retrieval, where we used three different 
 types of queries. We present a new type of query based on text paragraphs.
 The query execution were controled by its type and by preliminary similarities
 discovered during the searches. 
-In section~\ref{text_alignment} we present modified common text feature fot text alignment.
-We also compare performance of both the previous and the modified algorithms. 
-
+In Section~\ref{text_alignment} we describe our approach for the text alignment
+(pairwise comparison) subtask. We briefly introduce our system,
+and then we discuss the feature types, which are usable for pairwise comparison,including the evaluation of their feasibility for this purpose. We then describe
+the global (corpus-wide) optimizations used, and finally we discuss
+the results achieved and further development.
 
 \input{simon-source_retrieval}
 \input{yenya-text_alignment}
@@ -47,6 +51,13 @@ We also compare performance of both the previous and the modified algorithms.
 Unfortunately the ChatNoir search engine does not support phrasal search, therefore it
 is possible that evaluated results may be quite distorted in this manner.
 
+In the text alignment subtask, we have achieved a significant improvement
+with respect to our system from PAN 2012. Further development in this
+area is still possible. For a real-world system, however, a completely
+different set of parameters and heuristics needs to be used, as a result
+of plagdet score together with the structure of the competition corpus
+being too different from the real world.
+
 \bibliographystyle{splncs03}
 \begin{raggedright}
 \bibliography{pan13-notebook}