3

I have had a problem of unexpected spacing below the object with tcolorbox inside the wrapfigure environment. You can see the attached snapshot.

enter image description here

Below is my MWE.

\documentclass[a4paper,12pt]{article}
\usepackage[scaled]{helvet}
\renewcommand\familydefault{\sfdefault}
\usepackage[utf8]{inputenc}
\usepackage{pifont}
\usepackage{wrapfig}
\usepackage[framemethod=tikz]{mdframed}
\usepackage{xcolor,tcolorbox}
%% https://tex.stackexchange.com/a/126741
\newenvironment{WrapText1}[1][r]
  {\wrapfigure{#1}{0.5\textwidth}\tcolorbox}
  {\endtcolorbox\endwrapfigure}
\newenvironment{WrapText2}[1][r] 
{\wrapfigure{#1}{0.5\textwidth}\mdframed[backgroundcolor=gray!20,skipabove=0pt,
skipbelow=0pt]}
  {\endmdframed\endwrapfigure}
%opening
\title{}
\author{}

\begin{document}

\maketitle

\section{Complex networks}

\subsection{Small-world networks}
\label{sec:small-world}
\index{Complex Network!Small-world network}
For many real world phenomena, the average path length $l$ of a network is much 
smaller than that network size $n$, that is $l \ll n$. Such networks are said 
to be characterising the small-world property [1,2]. In 
mathematics, physics and sociology a small-world network (SWN) is a category of 
networks in which most nodes are not neighbours of one another, but most nodes 
can be reached from every other by a small number of \textit{hops} or 
\textit{steps}. D. Watts and S. Strogatz introduced this terminology in 1998 
[5] (also called WS model) that was originated from the famous 
experiment made by Milgram in 1967 [3]. Milgram found that two 
US citizens chosen randomly were connected by an average of six acquaintances.

\subsubsection*{\ding{228} Small-world networks in real life}
small-world networks can be found in many real-world applications, including 
road maps, food chains, electric power grids, metabolite processing networks, 
networks of brain neurons, voter networks, telephone call graphs, and social 
influence networks. These systems comprise of many local links and fewer long 
range \textit{``shortcuts''}, often use with a high degree of local clustering 
but 
relatively small diameter (see more detail below). Networks found in many 
biological and man-made systems are ``small-world networks'', which are highly 
clustered, but the minimum distance between any two randomly chosen nodes in 
the 
graph is short. Thus, studies on SWNs have been interested by many 
researchers in a variety of fields such as mathematics, computer sciences, 
physics, social sciences, etc.

\paragraph{\ding{51}}In a study of Indian physicians [10], they 
have analysed and showed the structure of the Indian railway network (IRN). 
Identifying the stations as nodes of the network and a train which stops at any 
two stations as the edges between the nodes, Sen and co-authors measured the 
average distance between an arbitrary pair of stations and find that it 
depends logarithmically on the total number of stations in the country. While 
from the network point of view this implies the small-world nature of the 
railway 
network, in practice a traveller has to change only a few trains to reach an 
arbitrary destination. This implies that over the years, the railway network 
has 
evolved with the sole aim of becoming fast and economical; eventually its 
structure has become a SWN.
\begin{WrapText2}
 \vspace{-1.0mm}
 In Goyal's study [6], the principal conditions that a network 
$G$ exhibits \textit{small-world} properties are as the following:
\begin{enumerate}\itemsep1pt \parskip0pt \parsep0pt
\item The number of nodes is very large as compared to the average number 
of links (the average degree), i.e. $n \gg k$
\item The network is integrated; a giant component exists and covers a large 
share of the population.
\item The average distance between nodes $l$ (called characteristic path 
length) in the giant component is small, i.e. $l$ is of order $ln(n)$.
\item The global clustering coefficient is high, i.e. $C \gg k/n$ 
\end{enumerate}
\vspace{-1.0mm}
\end{WrapText2}
\paragraph{\ding{51}}Another investigation on Boston subway, Latora and his
collaborators [2] showed that the whole transportation system
MBTA\footnote{Boston underground transportation system} (consists of $n = 124$
stations and $k = 124$ tunnels) and bus turns out to be a small-world with a
slight increase in the cost. This paper showed that a generic closed
transportation system can exhibit the small-world behaviour, substantiating 
the
idea that, in the grand picture, the diffusion of small-world networks can be
interpreted as the need to create networks that are both globally and locally
efficient.

\paragraph{\ding{51}}The World Wide Web has a small-world topology as well 
[12]. In this paper, Adamic made a comparison between SWNs
 for sites, and the corresponding random graphs, the subset of \textit{.edu} 
sites was considered. Because the \textit{.edu} subset is significantly 
smaller,
 distances between every node could be computed. $3,456$ of the $11,000$
\textit{.edu} sites formed the largest SCC. In
summary, the largest SCCs of both sites in general
and the subset of \textit{.edu} sites are SWNs with small
average minimum distances.

\paragraph{\ding{51}}In fact, rich-species food webs with a good taxonomic 
resolution display the properties of small-world behaviour [1]. 
Montoya and Sol\'e analysed the four large food webs and compared between real 
webs and randomly generated webs. Consequently, they approved that the 
clustering coefficient of both types is the same average number of links per 
species. One important result is that in all cases, the clustering coefficient 
is clearly larger than the one of the random networks. For the characteristic 
path length, the difference between the random and real case is almost very 
small.
\end{document}

I referred the way to create box as an image proposed in this answer, or even the similar way was proposed two days ago via my question at this answer.

The question is: How to remove the unexpected spaces below or above the box?

  • You don't use the optional first argument of \wrapfig[...] which specifies the number of lines to be used – user31729 Dec 28 '14 at 23:06
  • 1
    Please, don't use images at Dropbox, but upload them using the interface (click on the image button). – egreg Dec 28 '14 at 23:16
  • @egreg: I do not have a look carefully the tool box. I'll sure to avoid duplicating the issue. – Tung Dec 28 '14 at 23:20
4

The wrapfig environment has an optional first argument, which is used for the number of lines to be used (this is some trial and error way, however, to find the 'correct' value) (I do not remember right now, what the default value is, but I believe, it's 12)

I changed the definition of the \WrapText1 environments and added a mandatory 2nd argument to be specified as the line numbers.

I did not delete text etc. and did not look for other issues in OP's MWE

\documentclass[a4paper,12pt]{article}
\usepackage[scaled]{helvet}
\renewcommand\familydefault{\sfdefault}
\usepackage[utf8]{inputenc}
\usepackage{pifont}
\usepackage{wrapfig}
\usepackage[framemethod=tikz]{mdframed}
\usepackage{xcolor,tcolorbox}
%% http://tex.stackexchange.com/a/126741
\newenvironment{WrapText1}[2][r]
  {\wrapfigure[#2]{#1}{0.5\textwidth}\tcolorbox}
  {\endtcolorbox\endwrapfigure}
\newenvironment{WrapText2}[2][r] 
{\wrapfigure[#2]{#1}{0.5\textwidth}\mdframed[backgroundcolor=gray!20,skipabove=0pt,
skipbelow=0pt]}
  {\endmdframed\endwrapfigure}
%opening
\title{}
\author{}

\begin{document}

\maketitle

\section{Complex networks}

\subsection{Small-world networks}
\label{sec:small-world}
\index{Complex Network!Small-world network}
For many real world phenomena, the average path length $l$ of a network is much 
smaller than that network size $n$, that is $l \ll n$. Such networks are said 
to be characterising the small-world property [1,2]. In 
mathematics, physics and sociology a small-world network (SWN) is a category of 
networks in which most nodes are not neighbours of one another, but most nodes 
can be reached from every other by a small number of \textit{hops} or 
\textit{steps}. D. Watts and S. Strogatz introduced this terminology in 1998 
[5] (also called WS model) that was originated from the famous 
experiment made by Milgram in 1967 [3]. Milgram found that two 
US citizens chosen randomly were connected by an average of six acquaintances.

\subsubsection*{\ding{228} Small-world networks in real life}
small-world networks can be found in many real-world applications, including 
road maps, food chains, electric power grids, metabolite processing networks, 
networks of brain neurons, voter networks, telephone call graphs, and social 
influence networks. These systems comprise of many local links and fewer long 
range \textit{``shortcuts''}, often use with a high degree of local clustering 
but 
relatively small diameter (see more detail below). Networks found in many 
biological and man-made systems are ``small-world networks'', which are highly 
clustered, but the minimum distance between any two randomly chosen nodes in 
the 
graph is short. Thus, studies on SWNs have been interested by many 
researchers in a variety of fields such as mathematics, computer sciences, 
physics, social sciences, etc.

\paragraph{\ding{51}}In a study of Indian physicians [10], they 
have analysed and showed the structure of the Indian railway network (IRN). 
Identifying the stations as nodes of the network and a train which stops at any 
two stations as the edges between the nodes, Sen and co-authors measured the 
average distance between an arbitrary pair of stations and find that it 
depends logarithmically on the total number of stations in the country. While 
from the network point of view this implies the small-world nature of the 
railway 
network, in practice a traveller has to change only a few trains to reach an 
arbitrary destination. This implies that over the years, the railway network 
has 
evolved with the sole aim of becoming fast and economical; eventually its 
structure has become a SWN.
\begin{WrapText2}{20}
% \vspace{-1.0mm}
 In Goyal's study [6], the principal conditions that a network 
$G$ exhibits \textit{small-world} properties are as the following:
\begin{enumerate}\itemsep1pt \parskip0pt \parsep0pt
\item The number of nodes is very large as compared to the average number 
of links (the average degree), i.e. $n \gg k$
\item The network is integrated; a giant component exists and covers a large 
share of the population.
\item The average distance between nodes $l$ (called characteristic path 
length) in the giant component is small, i.e. $l$ is of order $ln(n)$.
\item The global clustering coefficient is high, i.e. $C \gg k/n$ 
\end{enumerate}
\vspace{-1.0mm}
\end{WrapText2}
\paragraph{\ding{51}}Another investigation on Boston subway, Latora and his
collaborators [2] showed that the whole transportation system
MBTA\footnote{Boston underground transportation system} (consists of $n = 124$
stations and $k = 124$ tunnels) and bus turns out to be a small-world with a
slight increase in the cost. This paper showed that a generic closed
transportation system can exhibit the small-world behaviour, substantiating 
the
idea that, in the grand picture, the diffusion of small-world networks can be
interpreted as the need to create networks that are both globally and locally
efficient.

\paragraph{\ding{51}}The World Wide Web has a small-world topology as well 
[12]. In this paper, Adamic made a comparison between SWNs
 for sites, and the corresponding random graphs, the subset of \textit{.edu} 
sites was considered. Because the \textit{.edu} subset is significantly 
smaller,
 distances between every node could be computed. $3,456$ of the $11,000$
\textit{.edu} sites formed the largest SCC. In
summary, the largest SCCs of both sites in general
and the subset of \textit{.edu} sites are SWNs with small
average minimum distances.

\paragraph{\ding{51}}In fact, rich-species food webs with a good taxonomic 
resolution display the properties of small-world behaviour [1]. 
Montoya and Sol\'e analysed the four large food webs and compared between real 
webs and randomly generated webs. Consequently, they approved that the 
clustering coefficient of both types is the same average number of links per 
species. One important result is that in all cases, the clustering coefficient 
is clearly larger than the one of the random networks. For the characteristic 
path length, the difference between the random and real case is almost very 
small.
\end{document}

enter image description here

  • You're really amazing! Thanks so much! But more an extra comment: If I want to shift enumerate items to the leftmost. How can we do that? – Tung Dec 28 '14 at 23:19
  • 1
    @Tung: That's an extra question, but this has been asked a lot of times ... see the enumitem package for quite an easy way. I think, you want to change the \leftmargin*** lengths? – user31729 Dec 28 '14 at 23:20
  • That's exact what I mean. The items look having more spaces in the left. I want to shift left a little. – Tung Dec 28 '14 at 23:25
  • 1
    @Tung: Try \begin{enumerate}[leftmargin=12pt,itemsep=1pt,parsep=0pt] \parskip=0pt instead and \usepackage{enumitem} – user31729 Dec 28 '14 at 23:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.