%\documentclass{article}
\documentclass[preprint, 1p,authoryear]{elsarticle}
%\textwidth 6.0in
%\usepackage{savetrees}
%\usepackage[left=1in,right=1in, top=1in, bottom=1in]{geometry}

% for double space
%\usepackage{setspace}  
%\doublespacing

%bibilography
%\usepackage[colon]{natbib}
%\bibpunct{(}{)}{;}{a}{}{,}

%for images
\usepackage{graphicx}  
\graphicspath{{images/}}
\usepackage[table]{xcolor}
% *************** Enable hyperlinks in PDF documents with these colors
\usepackage{color}
\definecolor{sepia}         {cmyk}{0   , 0.83, 1   , 0.70}
\definecolor{red}           {cmyk}{0   , 1   , 1   , 0   }
\definecolor{maroon}        {cmyk}{0   , 0.87, 0.68, 0.32}

 \usepackage[plainpages=false,pdfpagelabels,bookmarksnumbered,%
 colorlinks=true,%
 linkcolor=sepia,%
 citecolor=sepia,%
 filecolor=maroon,%
 urlcolor=sepia,%
 pdftex,%
pdftitle={Learning Patterns of University Student Retention},
pdfauthor={Ashutosh R. Nandeshwar,Tim Menzies,Adam Nelson},
pdfkeywords={Data Mining, Student Retention, Higher Education},
 unicode]{hyperref} 


\usepackage{multirow}

\usepackage{rotating}
\usepackage{lscape}
\usepackage{longtable}

\usepackage{amsmath}

\usepackage[nolists]{endfloat}
\usepackage{endfloatsideways} 


\begin{document}

\begin{frontmatter}


\title{Learning Patterns of University Student Retention}


%\author{Ashutosh Nandeshwar \footnote{Ashutosh Nandeshwar, Kent State University, 126 Lowry Hall,  Kent, OH 44242, E-mail: \href{mailto:anandesh@kent.edu}{anandesh@kent.edu}, Phone: 330-672-82222, Web-site: \url{www.nandeshwar.info}} 
% \and Tim Menzies \footnote{Tim Menzies, West Virginia University, 841a Engineering Sciences Building, Morgantown, WV 26505, E-mail: \href{mailto:tim@menzies.us}{tim@menzies.us}, Web-site: \url{www.menzies.us}} \and Adam Nelson \footnote{Adam Nelson, West Virginia University, Engineering Sciences Building, Morgantown, WV 26505, E-mail: \href{mailto:rabituckman@gmail.com}{rabituckman@gmail.com} }}



\author[rvt]{Ashutosh Nandeshwar\corref{cor1}}
\ead{anandesh@kent.edu}
\ead[url]{www.nandeshwar.info}

\author[focal]{Tim Menzies}
\ead{tim@menzies.us}
\ead[url]{www.menzies.us}

\author[els]{Adam Nelson}
\ead{rabituckman@gmail.com}


\cortext[cor1]{Corresponding author}



\address[rvt]{Kent State University, 126 Lowry Hall,  Kent, OH 44242 Phone: 330-672-82222}
\address[focal]{West Virginia University, 841a Engineering Sciences Building, Morgantown, WV 26505}
\address[els]{West Virginia University, Engineering Sciences Building, Morgantown, WV 26505}





\begin{abstract}
Learning predictors for student retention is very difficult. After reviewing the literature,
it is  evident that there is considerable room for improvement in the current state of the art.
As  shown in this paper,
 improvements are possible if we (a)~explore a wide range of learning 
methods; (b)~take care when selecting attributes;  (c)~assess the efficacy of the
learned theory not just by its {\em median} performance, but also by the
{\em variance} in that performance; 
(d)~study the {\em delta} of  student factors
{\em between} those who stay and those who are retained.
Using these techniques,
for the goal of predicting if students will remain for the first
three years of an undergraduate degree, the following factors were
found to be informative:  family background and family's social-economic status, high school GPA and test scores.

\end{abstract}




\begin{keyword} 
data mining \sep student retention \sep predictive modeling \sep financial aid
\end{keyword}



\end{frontmatter}


\input{intro}
\input{litreview}
\input{data}
\input{experiment}
\input{results}
\input{conclusions}

% *************** Bibliography ***************
\bibliographystyle{apa-good}
\bibliography{paperrefs}
\end{document}