jdemonasterio · jdemonasterio · Jul 28, 2018 · Jul 28, 2018 · Jul 30, 2018 · Jul 30, 2018
diff --git a/tex/abstract.tex b/tex/abstract.tex
@@ -1,10 +1,9 @@
+
 \chapter{Abstract}
 \label{cha:abstract}
 
 The Chagas disease continues to represent a global epidemiological problem, particularly for the South American continent.
 Current control methods focus solely on the vector-infested regions. 
-%Still, human mobility patterns represent an important factor in the congenital transmission of this disease. 
-%Its spread from high to low infested regions is strengthened by seasonal and long-term migrations. 
 Still, human mobility patterns represent an important factor in the geographical spread of this disease, since its dissemination from high to low infested regions is strengthened by seasonal and long-term migrations. 
 
 Using anonymized mobile phone data, and in collaboration with the Mundo Sano Foundation, the objective of this work is to assess the relationship of calling patterns and user behavior, with migrations. 

diff --git a/tex/appendix.tex b/tex/appendix.tex
diff --git a/tex/biblio.bib b/tex/biblio.bib
@@ -10,7 +10,6 @@ @inproceedings{sarraute2015descubriendo
 
 @inproceedings{deMonasterio2016analyzing,
 	doi = {10.1109/asonam.2016.7752298},
-	url = {https://doi.org/10.1109%2Fasonam.2016.7752298},
 	year = 2016,
 	month = {aug},
 	publisher = {{IEEE}},
@@ -22,15 +21,11 @@ @inproceedings{deMonasterio2016analyzing
 % machine learning citations
 
 @book{hastie-elemstatslearn,
-  added-at = {2009-12-09T12:44:18.000+0100},
   author = {Hastie, T. and Tibshirani, R. and Friedman, J.},
-  interhash = {39bdf0a76d889fa39deb5d1ac793ff4e},
-  intrahash = {0cd6d6683b38e7b5a15ee550266b5b9c},
+  title = {The Elements of Statistical Learning},
   keywords = {ml statistics},
   publisher = {Springer New York Inc.},
   series = {Springer Series in Statistics},
-  timestamp = {2008-08-16T16:17:43.000+0200},
-  title = {The Elements of Statistical Learning},
   year = 2009
 }
 
@@ -47,10 +42,7 @@ @Inproceedings { rasmussenGhahramani2001occamsRazor
    title = {Occam's Razor},
    year = {2001},
    month = {4},
-   pages = {294-300},
-   abstract = {The Bayesian paradigm apparently only sometimes gives rise to Occam's Razor; at other times very large models perform well. We give simple examples of both kinds of behaviour. The two views are reconciled when measuring complexity of functions, rather than of the machinery used to implement them. We analyze the complexity of functions for some linear in the parameter models that are equivalent to Gaussian Processes, and always find Occam's Razor at work.},
-   url = {http://www.kyb.tuebingen.mpg.de/fileadmin/user_upload/files/publications/pdf2215.pdf},
-   web_url = {http://books.nips.cc/nips13.html},
+   pages = {294--300},
    editor = {Leen, T.G. Dietterich, V. Tresp},
    publisher = {MIT Press},
    address = {Cambridge, MA, USA},
@@ -68,7 +60,7 @@ @InProceedings{ McKinney-pandas
   author    = { Wes McKinney },
   title     = { Data Structures for Statistical Computing in Python },
   booktitle = { Proceedings of the 9th Python in Science Conference },
-  pages     = { 51 - 56 },
+  pages     = { 51 -- 56 },
   year      = { 2010 },
   editor    = { Stefan van der Walt and Jarrod Millman }
 }
@@ -85,15 +77,13 @@ @article{graphlab
  issn = {2150-8097},
  pages = {716--727},
  numpages = {12},
- url = {http://dx.doi.org/10.14778/2212351.2212354},
- doi = {10.14778/2212351.2212354},
  acmid = {2212354},
  publisher = {VLDB Endowment},
 }
 
 
-@article{python3.5,
- author = {Python Software Foundation},
+@misc{python3.5,
+ author = {{\relax Python} Software Foundation},
  title = {Python Language Reference, version 3.5},
  issue_date = {June 2015},
  month = jun,
@@ -119,7 +109,6 @@ @book{breiman-cart84
   publisher     = {Wadsworth and Brooks},
   address       = {Monterey, CA},
   year          = {1984},
-  remarks       = {cited in \cite{cslu:esca98mm,cslu:icslp98cronk,cstr:unitsel97} for CART, clustering, and decision trees}
 }
 
 
@@ -139,8 +128,6 @@ @article{breiman-randomforests
 number="1",
 pages="5--32",
 issn="1573-0565",
-doi="10.1023/A:1010933404324",
-url="http://dx.doi.org/10.1023/A:1010933404324"
 }
 
 @article{breiman-arcingclassifiers,
@@ -156,7 +143,6 @@ @article{breiman-arcingclassifiers
 
 @article{friedman-gradientBoosting2001,
 author = "Friedman, Jerome H.",
-doi = "10.1214/aos/1013203451",
 fjournal = "The Annals of Statistics",
 journal = "Ann. Statist.",
 month = "10",
@@ -165,7 +151,6 @@ @article{friedman-gradientBoosting2001
 publisher = "The Institute of Mathematical Statistics",
 title = "Greedy function approximation: A gradient boosting
        machine.",
-url = "http://dx.doi.org/10.1214/aos/1013203451",
 volume = "29",
 year = "2001"
 }
@@ -187,11 +172,11 @@ @book{Mitchell-MLearning
 
 @article{decisionTreesNP,
   title = {{Constructing Optimal Binary Decision Trees is NP-complete}},
-  author = {L Hyafil, and R. L. Rivest},
+  author = {L Hyafil  and R. L. Rivest},
   journal = {Information Processing Letters},
   volume = {5},
   number = {1},
-  pages = {15-17},
+  pages = {15--17},
   year = {1976},
   publisher = {Elsevier},
 }
@@ -204,7 +189,6 @@ @article{HoFirstRandomForest
     year = {1995},
     isbn = {0-8186-7128-9},
     pages = {278--},
-    url = {http://dl.acm.org/citation.cfm?id=844379.844681},
     publisher = {IEEE Computer Society},
 
 }
@@ -218,8 +202,6 @@ @Article{james-biasVarianceGeneral
 number="2",
 pages="115--135",
 issn="1573-0565",
-doi="10.1023/A:1022899518027",
-url="http://dx.doi.org/10.1023/A:1022899518027"
 }
 
 
@@ -229,10 +211,6 @@ @article{dworkFeldman-adaptiveDataAnalysis2015
   journal   = {CoRR},
   volume    = {abs/1506.02629},
   year      = {2015},
-  url       = {http://arxiv.org/abs/1506.02629},
-  timestamp = {Wed, 07 Jun 2017 14:42:33 +0200},
-  biburl    = {http://dblp.uni-trier.de/rec/bib/journals/corr/DworkFHPRR15},
-  bibsource = {dblp computer science bibliography, http://dblp.org}
 }
 
 
@@ -263,32 +241,29 @@ @article{bengio-unbiasedCvEstimator
   volume    = {5},
   pages     = {1089--1105},
   year      = {2004},
-  url       = {http://www.jmlr.org/papers/volume5/grandvalet04a/grandvalet04a.pdf},
-  biburl    = {http://dblp.uni-trier.de/rec/bib/journals/jmlr/BengioG04},
 }
 
 
 @article{mason-rocAucRelationship,
-  author    = {S. J. Mason, N. E. Graham},
+  author    = {S. J. Mason and N. E. Graham},
   title     = {{Areas beneath the relative operating characteristics (ROC) and relative operating levels (ROL) curves: Statistical significance and interpretation}},
   journal   = {Q.J.R. Meteorol. Soc.},
   volume    = {128},
   number     = {584},
   year      = {2002},
-  pages     = {2145-2166},
-  doi    = {10.1256/003590002320603584},
+  pages     = {2145--2166},
 }
 
 
-@article{schapire-adaBoost,
+@inproceedings{schapire-adaBoost,
   author =       {Freund, Yoav and Schapire, Robert E.},
   title =        {{Experiments with a New Boosting Algorithm}},
   booktitle =    {Proceedings of the Thirteenth International Conference on Machine Learning (ICML 1996)},
   year =         {1996},
   ISBN =         {1-55860-419-7},
   editor =    {Saitta, Lorenza},
   publisher = {Morgan Kaufmann},
-  pages =     {148-156},
+  pages =     {148--156},
 }
 
 
@@ -345,7 +320,6 @@ @article{manne2013barriers
   journal={PLoS Negl Trop Dis},
   volume={7},
   number={10},
-  pages={e2488},
   year={2013},
   publisher={Public Library of Science}
 }
@@ -356,7 +330,6 @@ @article{hotez2012texas
   journal={PLoS Negl Trop Dis},
   volume={6},
   number={3},
-  pages={e1497},
   year={2012},
   publisher={Public Library of Science}
 }
@@ -369,7 +342,6 @@ @article{hotez2013unfolding
   journal={PLoS Negl Trop Dis},
   volume={7},
   number={10},
-  pages={e2300},
   year={2013},
   publisher={Public Library of Science}
 }
@@ -380,7 +352,6 @@ @article{navarro2012chagas
   journal={PLoS Negl Trop Dis},
   volume={6},
   number={12},
-  pages={e1962},
   year={2012},
   publisher={Public Library of Science}
 }
@@ -430,7 +401,7 @@ @article{enns2013human
 
 @article{OPS2006chagas,
   title = {{Estimacion Cuantitiva de la Enfermedad del Chagas en las Americas}},
-  author = {OPS},
+  author = {{\relax OPS}},
   journal = {Organizacion Panamericana de la Salud/HDM/CD},
   volume = {425-06},
   pages = {1--6},
@@ -441,7 +412,7 @@ @article{OPS2006chagas
 
 @article{OPS2014mapa,
   title = {{Mapa de Transmision vectorial del Mal de Chagas}},
-  author = {OPS},
+  author = {{\relax OPS}},
   journal = {Organizacion Panamericana de la Salud},
   year = {2014},
   publisher = {OPS},
@@ -503,7 +474,6 @@ @article{tizzoni2014use
   journal = {PLoS Comput Biol},
   volume = {10},
   number = {7},
-  pages = {e1003716},
   year = {2014},
   publisher = {Public Library of Science},
 }
@@ -523,7 +493,7 @@ @article{wesolowski2012quantifying
 
 @article{who2016,
   title = {{Chagas disease (American trypanosomiasis)}},
-  author = {WHO},
+  author = {{\relax World Health Organization}},
   journal = {{World Health Organization} Fact sheets},
   year = {2016},
   url = {http://www.who.int/mediacentre/factsheets/fs340/en/},
@@ -532,7 +502,7 @@ @article{who2016
 
 
 @misc{plan_nacional_chagas,
-  author = {Ministerio de Salud Argentina},
+  author = {{\relax Ministerio de Salud Argentina}},
   title = {{Plan Nacional de Chagas}},
   url = {http://www.msal.gob.ar/chagas/},
   year = {2016},
@@ -542,7 +512,6 @@ @misc{plan_nacional_chagas
 
 @inproceedings{deMonasterio-analyzingSpread,
   doi = {10.1109/asonam.2016.7752298},
-  url = {https://doi.org/10.1109%2Fasonam.2016.7752298},
   year = {2016},
   month = {aug},
   publisher = {{IEEE}},
@@ -555,7 +524,7 @@ @inproceedings{deMonasterio-analyzingSpread
 % mobile phone records and mobility
 
 @misc{netmob,
-  author = {NetMob},
+  author = {{\relax NetMob}},
   title = {{Main conference on the scientific analysis of mobile phone datasets}},
   url = {http://netmob.org/},
   year = {2016},
@@ -564,7 +533,7 @@ @misc{netmob
 
 
 @misc{d4d,
-  author = {Orange D4D},
+  author = {{\relax Orange D4D}},
   title = {{Data for Development (D4D) Challenge}},
   url = {http://www.d4d.orange.com/},
   year = {2016},
@@ -617,7 +586,6 @@ @inproceedings{oliveira2015measurement
 
 @article{Ponieman2015mobility,
 	doi = {10.3233/aic-150687},
-	url = {https://doi.org/10.3233%2Faic-150687},
 	year = 2015,
 	month = {sep},
 	publisher = {{IOS} Press},
@@ -772,7 +740,7 @@ @article{wang2012understanding
 
 @booklet{enmodo2010,
   title={{ENMODO} (2009-2010). {R}esultados de la Encuesta Origen Destino. {M}ovilidad en el Area Metropolitana de {B}uenos {A}ires},
-  author={Secretaría de Transporte. Ministerio del Interior y Transporte.},
+  author={{\relax Secretaría de Transporte. Ministerio del Interior y Transporte.}},
   year={2010}
 }
 
@@ -781,13 +749,13 @@ @inproceedings{ponieman2013
   title={Human Mobility and Predictability enriched by Social Phenomena Information},
   author={Nicolás Ponieman and Alejos Salles and Carlos Sarraute},
   booktitle={IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)},
-  pages={1331-1336},
+  pages={1331--1336},
   year={2013},
   organization={IEEE}
 }
 
 @book{censo2010,
-  author    = "{I}nstituto {N}acional de {E}stadística y {C}ensos ({INDEC})",
+  author    = "{\relax {I}nstituto {N}acional de {E}stadística y {C}ensos ({INDEC})}",
   title     = "{C}enso {N}acional de {P}oblación, {H}ogares y {V}iviendas 2010",
   publisher = "{INDEC}",
   volume   = "1",

diff --git a/tex/ch_conclusions.tex b/tex/ch_conclusions.tex
@@ -96,7 +96,7 @@ \chapter{Conclusions}\label{ch:conclusions}
 
 
 
-\section{Discussion on the methodology}
+\section{Discussion on the Methodology}
 
 There are several points that can be stated to point to weak spots on the methodology used.
 
@@ -138,7 +138,7 @@ \section{Discussion on the methodology}
 % discriminated in the construction of the features, yet there were no long-term time
 
 
-\section{ Lines of Future Work }
+\section{Lines of Future Work }
 
 
 The mobility and social information extracted from CDRs analysis has been shown to be of practical use for long-term human migrations and for Chagas disease research.

diff --git a/tex/ch_descriptions_risk.tex b/tex/ch_descriptions_risk.tex
@@ -65,12 +65,12 @@ \section{Mobile Phone Data Sources}
 The same argument can be used to note that not all \textit{real} population movements will be captured by the logs.
 However, these limitations are offset by the huge datasets' sizes, from which we think we can safely assume that the amount of users observed in each set is sufficient to correlate CDR usage with human mobility or social links between different areas.
 
-\section{ Risk Maps Generation}\label{section:risk_maps}
+\section{Risk Maps Generation}\label{section:risk_maps}
 
 In this section we describe how the Chagas disease risk maps were built for Argentina and Mexico, and we give an overview of their uses.
 As an overview, we use the risk maps to hypothesize on the possibility of locating specific communities of higher disease prevalence outside of the endemic regions.
 They are based on the assumption that when we have stronger communication ties from one community to the endemic region, we will find a higher chance of disease risk.
-Earlier versions of this project were presented by \citep{sarraute2015descubriendo} at \emph{Simposio Internacional sobre Enfermedades Desatendidas}, and\citep{deMonasterio2016analyzing} at \emph{International Conference on Advances in Social Networks Analysis and Mining}.
+Earlier versions of this project were presented by \citep{sarraute2015descubriendo} at \emph{Simposio Internacional sobre Enfermedades Desatendidas}, and \citep{deMonasterio2016analyzing} at \emph{International Conference on Advances in Social Networks Analysis and Mining}.
 These were based on the results contained in this chapter.
 
 The generated heatmaps display a geographical representation with the TelCo's antennas situated on the map.
@@ -89,7 +89,7 @@ \subsection{Home Detection}\label{subsection:home_detection}
 This will enable us later to detect long range migrations by looking at changes in these home antennas.
 
 This \textit{home} characterization is based on the assumption that on any given day, users will be located at home during night time.
-The implications of this assumption for CDRs are explored by \citep{sarraute2015socialevents} and by\citep{csaji2012exploring}.
+The implications of this assumption for CDRs are explored by \citep{sarraute2015socialevents} and by \citep{csaji2012exploring}.
 There, the authors explain that given the large user base, this assumption proves helpful when trying to predict migration patterns at large scale.
 For our case we need not detect specific agent movements but are more interested in movement of large amount of people.