Mas e Deg ee P og am in
Da a Science and Ad anced Analy ics
Wha a e he s uc u al di e ences in cha ac e ne wo ks
wi hin Po uguese-language li e a y wo ks ac oss a ious
gen es?
Ca a ina Ribei o Dua e
Mas e Thesis
p esen ed as pa ial equi emen o ob aining a Mas e ’s Deg ee in Da a Science and Ad anced Analy ics
NOVA In o ma ion Managemen School
Ins i u o Supe io de Es a ís ica e Ges ão de In o mação
Uni e sidade No a de Lisboa
MDSAA
i
NOVA In o ma ion Managemen School
Ins i u o Supe io de Es a ís ica e Ges ão de In o mação
Uni e sidade No a de Lisboa
Wha a e he s uc u al di e ences in cha ac e ne wo ks wi hin Po uguese-language
li e a y wo ks ac oss a ious gen es?
by
Ca a ina Ribei o Dua e
Mas e Thesis p esen ed as pa ial equi emen o ob aining he Mas e ’s deg ee in Da a
Science and Ad anced Analy ics, wi h a specializa ion in Da a Science
Supe ised by
Flá io Pinhei o, PhD, NOVA In o ma ion Managemen School
João L. M. Pe ei a, PhD, Uni e sidade de É o a
July, 2024
i
STATEMENT OF INTEGRITY
I he eby decla e ha ing conduc ed his academic wo k wi h in eg i y. I con i m ha I ha e no
used plagia ism o any o m o undue use o in o ma ion o alsi ica ion o esul s along he
p ocess leading o i s elabo a ion. I u he decla e ha I ha e ully acknowledged he Rules
o Conduc and Code o Hono om he NOVA In o ma ion Managemen School.
Lisboa, 15 h July 2024
ii
DEDICATION
Pa a os meus Pais, Ana Isabel e João Paulo, que le o semp e comigo mesmo sem se Joana.

iii
ACKNOWLEDGEMENTS
I am deeply g a e ul o my p ima y supe iso , P o esso Flá io Pinhei o, o his con inuous
suppo , in aluable guidance, and e e -p esen good mood. My hanks also ex end o
P o esso João Pe ei a o joining as a co-ad iso and en iching his wo k wi h his insigh s and
knowledgeable poin o iew. Addi ionally, I would like o exp ess my hea el app ecia ion
o NOVA IMS o p o iding an excellen educa ional en i onmen h oughou my s udies.
A dis inc i e hanks o Tiago Caná io, an indispu able su p ise and a an as ic eamma e. No
only o he ha d wo k, mo i a ion, and companionship in ha d challenges bu also o he
music ecommenda ions ha became he sound ack o mos o my w i ing sessions.
I his expe ience we e a book, i would be a mix o gen es, bu mainly i was an ad en u e, a
ale o pe se e ance, and an ode o he academic jou ney ha b ough me he e. I’m so
g a e ul o he cha ac e s in his na a i e, whose ne wo k would be as and dense, and I
would like o hank he dea and impo an cha ac e s in his s o y.
Fi s ly, my whole amily, whose us and suppo I ha e always had and who passed on o me
a lo e o books. Mãe, Pai, Mano, and Magali, you ha e been my home whe e I could always
es and ind com o in he middle o he jou ney in so many ways. You suppo was i al o
his na a i e and kep he s o y mo ing o wa d.
Ma a, who has been p esen and essen ial since he P ologue, e en hough om a di e en
backg ound, celeb a ed so deeply e e y conques and s uggle wi h me, he only one able o
make 300 km seem so close. And Ca olina, so in insically p esen in e e y chap e , always
o e ing wo ds o encou agemen and eaching me in so many o ms he meaning o
'uncondi ional'.
I also ex end my g a i ude o all my closes iends who added a ouch o comedy o his ale;
i would be a mys e y how i could ha e been wi hou all and each o you.
Las ly, hanks o A ó Celes e, whose belie in me endu es.
i
ABSTRACT
This s udy in es iga es he s uc u al di e ences in cha ac e ne wo ks wi hin Po uguese-
language li e a y wo ks ac oss a ious gen es, including Fan asy, Mys e y, and Romance. By
explo ing how cha ac e in e ac ions and ela ionships di e ac oss gen es, we aimed o
unde s and he unde lying na a i e s uc u es ha de ine each gen e. Th ough Social
Ne wo k Analysis, speci ic me ics we e ex ac ed o each gen e, analyzed, and a P incipal
Componen Analysis (PCA) was conduc ed, wi h eal-li e ne wo ks added o compa ison and
e e ence. Ou analysis e ealed ha , al hough no ex emely p ominen me ic di e ences
we e obse ed among he gen es, he e we e no able s uc u al endencies wi hin each
gen e. Fan asy na a i es end o ocus on a single main cha ac e ha in e ac s wi h many
o he cha ac e s, exhibi ing lowe densi y and clus e ing han o he gen es, and ha e less
ocus on ela ionship building. Mys e y na a i es ea u e mode a ely sized, dense ne wo ks
wi h key b idging cha ac e s and he highes cen ali y on a ela ionship, ansla ing in o a
balance o ela ionship and cha ac e building. In con as , Romance na a i es p esen he
mos a iable ne wo k size, emphasizing mul iple cen al cha ac e s and cohesi e subg oups,
ocusing especially on ela ionship building. This makes Romance and Fan asy he mos
di e en gen es, while Mys e y ac s as a mix o hese s uc u e s a egies and is
simul aneously close o wha is obse ed in eal-li e ne wo ks. The s udy acknowledges
limi a ions ela ed o sample size and da a p ocessing, highligh ing he need o u u e
esea ch wi h la ge and mo e di e se da ase s o alida e hese indings.
KEYWORDS
Social Ne wo ks Analysis (SNA); NLP; Gen e Theo y; Po uguese Li e a u e
TABLE OF CONTENTS
1. In oduc ion .................................................................................................................. 1
2. Rela ed wo k ................................................................................................................. 3
3. Me hodology ................................................................................................................ 6
4. Resul s and discussion ................................................................................................ 16
5. Conclusions and u u e wo ks .................................................................................... 24
Bibliog aphical Re e ences .............................................................................................. 26
Appendix A ...................................................................................................................... 31
i
LIST OF FIGURES
Figu e 1 - O e iew o he gene ic cha ac e ne wo k ex ac ion me hodology ...................... 7
Figu e 2 - Cha ac e Ne wo k o "O Filho de Mil Homens" .................................................... 10
Figu e 3 - Va iance explained by each p incipal componen and cumula i e a iance .......... 19
Figu e 4 - No els PC sco es ep esen ed in 2D o 2 PC ........................................................... 21
Figu e 5 - Cha ac e Ne wo k o "A Cos a dos Mu mú ios" ................................................... 31
Figu e 6 - Cha ac e Ne wo k o "A Cidade do Medo" .......................................................... 36
Figu e 7 - Cha ac e Ne wo k o "Limbo" ............................................................................... 41
6
3. METHODOLOGY
This sec ion ou lines he me hodology adop ed o analyze he s uc u al di e ences in
cha ac e ne wo ks wi hin Po uguese-language li e a y wo ks ac oss a ious gen es. This
wo k necessi a es a me iculous selec ion o gen es and li e a y wo ks, which o m he
ounda ional amewo k o ansla ing ex ual elemen s in o cha ac e ne wo ks. These
ne wo ks will hen be measu ed using ele an me ics chosen o se e as compa ison poin s
among gen es. The p ocesses de ailed below aim o cap u e he unique na a i e dynamics
ha cha ac e ize each gen e and highligh hei di e ences.
We s a by ou line he gen es ha we e chosen o compa ison. The ocus was on subgen es
o Fic ional Na a i es o c ea e iche ne wo ks and se mo e cohe en expec a ions o hese
subgen es. Sou ces di e in ca ego izing he same no els; some no els a e labeled wi h
mul iple gen es, while o he s a e no ca ego ized a all. To add ess his challenge and enable
a mo e obus and dis inc compa ison, we selec ed gen es whe e he eade 's expec a ions
a e clea ly de ined. Fo example, eade s an icipa e a lo e s o y in a omance no el o a c ime
o be sol ed in a mys e y no el. This con as s wi h gen es like ad en u e and an asy, which
may ha e mo e o e lapping hemes and less dis inc eade expec a ions (Fos e , J. n.d.).
Rega ding book selec ion, he sou ces o gen e agging included “P oje o Adamas o ”
3
and
he Good eads
4
websi e.
“P oje o Adamas o ” is a digi al lib a y o public domain Po uguese-language wo ks and i
has been u ilized in Sil a e al. (2023) esea ch. Each book is agged wi h i s gen e, allowing
o ex ac ion and gen e classi ica ion o use in his s udy. This pla o m is con inuously
upda ed, and i is no ewo hy no only o con e ing ex s in o digi al o ma bu also o i s
me iculous e iew o each a ailable wo k o minimize e o s and adhe e o he cu en
Po uguese Spelling Re o m. This is c ucial o his s udy as i p o ides a base ex ha is he
leas e o -p one and hus will p oduce he mos accu a e cha ac e ne wo ks. Acco ding o
he “Collabo a o Guide o P oje o Adamas o ” by Lou enço, R.F. (2014), each collabo a o
iden i ies a gen e du ing he selec ion and book ea men p ocess, which is hen alida ed by
a pee e iew in ol ing a leas one collabo a o no in ol ed in he ini ial e iew. They
conduc a comp ehensi e eading o he ile o ensu e he quali y o he book and he accu acy
o gen e agging.
“Good eads” is a book-based social websi e whe e membe s sha e, e iew, and a e li e a y
wo ks and connec wi h o he eade s (Thelwall & Kousha, 2017). On his pla o m, gen es a e
a ibu ed in a manne ha conside s se e al ac o s bu emains la gely undisclosed o use s.
I is known, howe e , ha use in luence plays a ole due o gen e ags e lec ing he sel -
c ea ed names by i s communi y. Shel es a e a me hod o use s o o ganize and g oup hei
3
h ps://p ojec oadamas o .o g
4
h ps://www.good eads.com

7
books, o en named a e gen es, such as omance o c ime. These names in luence gen e
assignmen s o he books, as he gen e lis s on Good eads book home pages include up o
en o he mos popula assigned book gen es (Thelwall, 2019).
Fo he book selec ion wi hin each gen e, i was ensu ed ha he book, whe he on “P oje o
Adamas o ” o among he i s i e gen e ags on Good eads, was iden i ied wi h ha gen e.
Al hough no gen e assignmen s a egy can pe ec ly ca ego ize gen es and ensu e a book i s
only one ca ego y, hese p ecau ions and me hodological app oaches we e adop ed o his
s udy.
The chosen gen es, all wi hin he ealm o ic ion, in he Po uguese language, including
B azilian Po uguese, a e p esen ed in able 1 wi h hei espec i e de ini ions om
Good eads. This alignmen p o ides a p e iew o eade expec a ions and he sample
collec ed ocused on p esen ing mo e con empo a y books.
Rega ding he p e-p ocessing equi ed o he book ex be o e passing i in o he pipeline o
ex ac he ne wo ks, only a simple cleanup is needed o emo e non-na a i e ex om he
digi al e sions o he books. The e o e, o co ec his issue, he ini ial o ending unnecessa y
chunks we e manually emo ed om he TXT iles o he books.
The pipeline used o his wo k, TAGGUS (Caná io & Dua e,2024) was speci ically designed o
ex ac cha ac e ne wo ks om Po uguese no els, add essing and o e coming all he
impedimen s and cons ain s associa ed wi h he language ha ha e been p e iously
men ioned. De ailed in o ma ion abou he en i e pipeline p ocess and he decisions made
can be ound in i ’s Gi hub eposi o y
5
.
Ne e heless, Laba u & Bos (2020), in hei widely ci ed wo k, ou lined he ex ac ion
p ocess, which also se ed as he ounda ion o he pipeline's c ea ion and i ’s ep esen ed
in igu e 1.
Figu e 1 - O e iew o he gene ic cha ac e ne wo k ex ac ion me hodology
5
h ps://gi hub.com/ icadu/ aggus.
8
Table 1 - Chosen Gen es and selec ed books o each
Book
Gen e
Good eads de ini ion
Au ho , Book i le and yea
Romance6
"Two basic elemen s comp ise e e y omance
no el: a cen al lo e s o y and an emo ionally
sa is ying and op imis ic ending." Bo h he
con lic and he climax o he no el should be
di ec ly ela ed o ha co e heme o de eloping
a oman ic ela ionship, al hough he no el can
also con ain subplo s ha do no speci ically
ela e o he main cha ac e s' oman ic lo e”
Val e Hugo Mãe,
“O Filho de Mil Homens” (2011)7
Lídia Jo ge, “A Cos a dos
Mu mú ios” (1988)8
Miguel Sousa Ta a es,
“Equado ” (2003)9
João Rica do Ped o, “O eu
Ros o se á o úl imo” (2012) 10
Fan asy11
Fan asy is a gen e ha uses magic and o he
supe na u al o ms as a p ima y elemen o plo ,
heme, and/o se ing. Fan asy is gene ally
dis inguished om science ic ion and ho o by
he expec a ion ha i s ee s clea o
echnological and macab e hemes, espec i ely
[…].
Thiago d'E ecque,
“Limbo” (2015) 12
Edua do Spoh , “A ba alha do
Apocalipse” (2007) 13
Sand a Ca alho,
“A Úl ima Fei icei a” (2005) 14
Filipe Fa ia, “Oblí io” (2011)15
Mys e y16
The mys e y gen e is a gen e o ic ion ha
ollows a c ime […] om he momen i is
commi ed o he momen i is sol ed. Mys e y
no els o en […] u n he eade in o a de ec i e
ying o igu e ou he who, wha , when, and
how o a pa icula c ime.
Ped o Ga cia Rosado,
“A Cidade do Medo” (2015) 17
Nuno Nepomuceno,
“A Célula Ado mecida” (2007) 18
Raphael Mon es,
“Dias pe ei os” (2005)19
F ancisco Moi a Flo es, “O
Bai o da Es ela Pola ” (2012)20
6
h ps://www.good eads.com/gen es/ omance
7
h ps://www.good eads.com/book/show/12395665-o- ilho-de-mil-homens
8
h ps://www.good eads.com/book/show/3371637-a-cos a-dos-mu m- ios
9
h ps://www.good eads.com/book/show/1140942.Equado
10
h ps://www.good eads.com/book/show/13597497-o- eu- os o-se -o-l imo
11
h ps://www.good eads.com/gen es/ an asy
12
h ps://www.good eads.com/book/show/25844730-limbo
13
h ps://good eads.com/book/show/18299638-a-ba alha-do-apocalipse
14
h ps://www.good eads.com/book/show/6333188-a-l ima- ei icei a
15
h ps://www.good eads.com/book/show/10474666-obl- io
16
h ps://www.good eads.com/gen es/mys e y
17
h ps://www.good eads.com/book/show/12742439-a-cidade-do-medo
18
h ps://www.good eads.com/book/show/32574062-a-c-lula-ado mecida
19
h ps://www.good eads.com/book/show/21405408-dias-pe ei os
20
h ps://www.good eads.com/book/show/16155430-o-bai o-da-es ela-pola
9
The p ocess begins wi h he p elimina y da ase , a single li e a u e wo k, which is ed in o he
model's pipeline. The i s s ep is cha ac e iden i ica ion, di ided in o wo sequen ial s ages.
Ini ially, all he di e en cha ac e names a e iden i ied and s o ed. Howe e , no e e y name
ep esen s a unique cha ac e , as na a i es, like eal li e, o en use a ious names o e e o
he same cha ac e . Fo example, in he wo k P ojec Hail Ma y
21
, Ryland G ace may also be
e e ed o in he same s o y as M . G ace, ep esen ing he same cha ac e bu by ano he
exp ession. The e o e, a co- e e ence esolu ion s ep is employed o uni y cha ac e
occu ences. The pipeline used in his wo k akes p ecau ions o ensu es accu a e uni ica ion
by ma ching gende , nicknames, likelihood o he mos popula cha ac e , and anapho ic
esolu ion.
Following cha ac e iden i ica ion, he nex s ep is o de ec in e ac ions be ween cha ac e s.
In his wo k, co-occu ence is used o coun in e ac ions. This me hod decomposes he
na a i e in o smalle uni s. These uni s may a y om p ojec o p ojec , conside ing wo
cha ac e s o in e ac when hey appea oge he in he same uni (Laba u & Bos , 2020).
Al hough his echnique is widely used due o i s simplici y and ease o implemen a ion, i has
he disad an age o po en ially labeling simply wo cha ac e s men ioned in he same uni as
in e ac ions, leading o alse posi i e in e ac ion coun s. Howe e , his emains as he mos
widesp ead app oach in he li e a u e o his issue (Laba u & Bos , 2020). In his pipeline he
na a i e uni chosen o de ec in e ac ion was wo cha ac e s men ion in he same sen ence.
Finally, a e uni ying he cha ac e s and de ec ing hei in e ac ions, a g aph is cons uc ed.
The g aph ha ep esen s a cha ac e ne wo k is buil wi h nodes and edges, each ha ing i s
own de ini ion. The nodes ep esen he cha ac e s, and he edges ep esen he ela ionships
among hem. Mo eo e , hese g aphs can also inco po a e empo al in eg a ion, esul ing in
ei he a s a ic ne wo k o a dynamic ne wo k (Laba u & Bos , 2020).
The s a ic ne wo k, which is he app oach mos commonly p esen ed in he li e a u e,
cap u es in e ac ions be ween cha ac e s o e he en i e na a i e pe iod. In con as , he
dynamic ne wo k in eg a es he p og ession o in e ac ions o e ime o ac oss di e en
sec ions o he s o y. While he s a ic ne wo k allows o a be e isualiza ion o he no el, i
does no cap u e he e olu ion o cha ac e s h oughou he na a i e (Laba u & Bos , 2020).
Fo his s udy, gi en he goal o unde s anding and analyzing he global s o y and he lack o
necessi y o empo al dimension analysis, s a ic ne wo ks we e used.
The simple elemen , nodes, can ep esen indi idual cha ac e s o g oups o cha ac e s and
also p esen cha ac e cha ac e is ics, such as cha ac e ype (e.g., human, dog, e c.), gende ,
ace, abili ies, e c. Mo eo e , weigh ed nodes can gi e us an idea o he ele ance o he
cha ac e s h oughou he s o y. On he o he hand, he edges ep esen he in e ac ions and
connec ions among cha ac e s and can con ain mo e in o ma ion. In e ac ion in ensi y can be
shown h ough weigh ed nodes. Addi ionally, he edges can ha e di ec ions, c ea ing a
21
h ps://www.good eads.com/book/show/54493401-p ojec -hail-ma y
10
di ec ed g aph ha ansmi s he speake /add essee in he in e ac ion. Fu he mo e,
assigned g aph p ope ies can p esen he in e ac ion pola i y ha con eys he sen imen o
he in e ac ion/ ela ionship (Laba u & Bos , 2020).
In his wo k, he s a ic ne wo ks will ea u e weigh ed nodes, whe e each node ep esen s a
dis inc cha ac e . While he nodes a e weigh ed o indica e he ele ance o each cha ac e ,
he edges in ou ne wo k will be weigh ed as well, undi ec ed, unsigned, and una ibu ed. An
example can be seen in igu e 2 and a ep esen a i e ne wo k o each gen e can be ound in
Appendix A as well as i s espec i e adjacency ma ix.
Figu e 2 - Cha ac e Ne wo k o "O Filho de Mil Homens"
When compa ing cha ac e ne wo ks, we ocused on measu able aspec s ha could be
ansla ed o na a i e signi icance and analyzed o gen e-speci ic di e ences. Fo each
ex ac ed cha ac e ne wo k, we compu ed he ne wo k size (N) ha in o ms on he numbe
o nodes o he ne wo k, i s size. In ou con ex i gi es us he numbe o cha ac e s o each
book (Sil a e al., 2023; Suen e al., 2013). Sil a e al. (2023) analysis showed ha he ne wo k
size a ies li le be ween Po uguese li e a y wo ks, wi h an a e age o nine cha ac e s in he
sample. In he case o igu e 2 i would be 12 nodes so, 12 cha ac e s.
Fu he mo e, he numbe o componen s coun s he subse o he e ices o a ne wo k ha
a e connec , his is, ha he e exis s a leas one pa h om each membe o ha subse o
each o he membe (Newman, 2010). In he con ex o cha ac e ne wo ks, analyzing he
numbe o componen s can e eal how agmen ed o cohesi e he na a i e s uc u e is. Fo
11
example, a high numbe o componen s migh indica e mul iple isola ed subplo s o cha ac e
g oups ha do no in e ac , while a low numbe o componen s, and especially a single one,
sugges s a mo e in eg a ed and in e connec ed s o y.
Nex , maximum deg ee shows he highes numbe o connec ions a single cha ac e has, his
is he highes deg ee o any node in he ne wo k (Sil a e al., 2023; Laba u & Bos , 2020). This
me ic helps iden i y he mos cen al o in luen ial cha ac e wi hin he na a i e and how
connec ed i is, p o iding insigh s in o he s uc u e and dynamics o he cha ac e
in e ac ions. Sudhaha & C is ianini (2013) ound in hei wo k ha he he o o a na a i e
always had he highes deg ee in a ne wo k, showing hei cen al posi ion in he na a i e. In
igu e 2 i ’s clea ha hese would be he numbe o connec ions Ma ilde has, due o being
he mos connec ed cha ac e .
In ou wo k, we p opose a new me ic, “Numbe o Main Cha ac e s” (NºMC) which aims o
coun , besides he cha ac e wi h he highes deg ee, how many o he cen al cha ac e s exis .
This is done by coun ing any o he node wi h a numbe o in e ac ions ( 𝑘𝑖) close o he highes
deg ee (𝑘𝑚𝑎𝑥), wi h a h eshold o a leas 80% and abo e a s a ic lowe bound o 10
in e ac ions. This me ic iden i ies bo h cen al and “seconda y” main cha ac e s, e lec ing
na a i es ha ea u e mul iple signi ican cha ac e s a he han ocusing solely on one. The
lowe bound o 10 in e ac ions ensu es ha cha ac e s conside ed as main ha e a meaning ul
le el o engagemen in he s o y. While me ics like maximum deg ee and single cha ac e
cen ali y (SCC) highligh he mos cen al cha ac e , hey may o e look o he impo an
cha ac e s. The “Numbe o Main Cha ac e s” me ic conside s cha ac e s wi h subs an ial
in e ac ions, p o iding a mo e comp ehensi e iew o he na a i e’s s uc u e and
ecognizing he impo ance o key seconda y cha ac e s alongside he p ima y one.
𝑁º𝑀𝐶 = ∑[𝑘𝑖≥0.8 𝑘𝒎𝒂𝒙 ∧𝑘𝑖≥10
𝒊](𝟏)
S ill in he opic o connec i i y, densi y measu es how closely he cha ac e s a e connec ed
in he ne wo k. This is possible by calcula ing he p opo ion be ween exis ing edges (E) in he
ne wo k and all possible ne wo k connec ions (Sil a e al., 2023; Laba u & Bos , 2020; Dekke
e al., 2019). The possible connec ions alue is he coun o pai o nodes ha can o m an
edge. The densi y alue anges om 0 o 1, whe e a alue close o 1 indica es a highly
in e connec ed ne wo k, and a alue close o 0 sugges s a spa sely connec ed ne wo k. Fo
example, A danuy & Spo lede (2015) no e ha in he case o he Ha y Po e
22
books, he
ea lie books a e conside ably dense han he la e ones, as he communi y ep esen ed in
hem becomes b oade and less igh ly kni wi h e e y new book. This shows how he densi y
22
h ps://www.good eads.com/se ies/45175-ha y-po e

12
me ic can e lec di e ence in he s uc u e and ocus o a na a i e, in his case in a se ies.
The equa ion (2) used o calcula e i was e ie ed om Bha acha ya e al. (2023).
𝑑 = 2𝐸
𝑁(𝑁−1)(𝟐)
The a e age pa h leng h measu es he a e age numbe o s eps o achie e he sho es pa hs
o all possible pai s o nodes in he ne wo k (Sil a e al., 2023; Laba u & Bos , 2020; Dekke
e al., 2019; Newman, 2010). This me ic p o ides insigh in o how connec ed he nodes,
cha ac e s, a e wi hin he ne wo k, e lec ing he o e all e iciency o in o ma ion o
in e ac ion low wi hin he na a i e. A lowe a e age pa h leng h sugges s a close communi y
whe e any cha ac e can be eached om any o he cha ac e h ough a ew in e ac ions.
Con e sely, a highe a e age pa h leng h indica es a mo e sp ead ou o agmen ed ne wo k.
In ou s udy, in he case o when we ha e a agmen ed ne wo k, we ha e in conside a ion
he a e age pa h leng h o he la ges componen since i ’s he bes and ai es me ic o ha
na a i e. The eal wo lds concep o six deg ees o sepa a ion, which sugges s ha any wo
people a e, on a e age, six o ewe social connec ions apa (Milg am, 1967), can illus a e
a e age pa h leng h usage in social ne wo ks. The de ined equa ion (3) was e ie ed om
Newman (2010) and 𝑑𝑖𝑗 is he sho es dis ance be ween cha ac e s 𝑖 and 𝑗.
𝐴𝑃𝐿 = 1
𝑁(𝑁−1) ∑𝑑𝑖𝑗
𝑖≠𝑗 (𝟑)
Now segueing in o he cen ali y measu es, which help iden i y he mos impo an o
in luen ial nodes wi hin he ne wo k, he single cha ac e cen ali y (SCC) measu e o
indi idual cha ac e s was p oposed by Suen e al. (2013) o see how much a s o y is ocused
on a single cha ac e . This is achie ed by measu ing he dispa i y be ween he cha ac e wi h
he highes weigh ed deg ee (𝑠𝑖) and he second highes , no malized, and unde s anding he
di e ence. While maximum deg ee p o ides a di ec measu e o a cha ac e ’s in e ac ions,
SCC quan i ies he dominance o dispa i y in cen ali y wi hin he ne wo k, indica ing how
much he s o y ocuses on one cha ac e abo e all o he s. The SCC alue anges be ween 0
and 1, whe e a alue close o 1 indica es high dominance o a single cha ac e , and a alue
close o 0 sugges s a mo e e enly dis ibu ed cha ac e ne wo k
13
𝑆𝐶𝐶 = 𝑚𝑎𝑥𝑖(𝑠𝑖)−𝑛𝑒𝑥𝑡_𝑚𝑎𝑥𝑖(𝑠𝑖)
∑𝑠𝑖𝑖 (𝟒)
Also p oposed by Suen e al. (2013), single ela ionship cen ali y (SRC) measu es he
p ominence o he mos cen al ela ionship ela i e o o he s in he ne wo k. This me ic
quan i ies how much a single ela ionship s ands ou in e ms o in e ac ions compa ed o he
second mos signi ican ela ionship. I helps iden i y he mos dominan pai o cha ac e s,
his is, he edge wi h mo e weigh 𝑤𝑖𝑗, and hei in luence on he na a i e. While single SCC
ocuses on he dominance o a single cha ac e , single ela ionship SRC highligh s he
impo ance o a speci ic in e ac ion be ween wo cha ac e s ollowing he same scale logic.
𝑆𝑅𝐶 =𝑚𝑎𝑥𝑖,𝑗(𝑤𝑖,𝑗)−𝑛𝑒𝑥𝑡_𝑚𝑎𝑥𝑖,𝑗(𝑤𝑖,𝑗)
∑𝑤𝑖,𝑗𝑖,𝑗 (𝟓)
The be weenness cen ali y (BC) measu es he ex en o which a ce ain node is on he
sho es pa hs be ween o he nodes (Hansen e al., 2020; Laba u & Bos , 2020; Suen e al.,
2013; He inge e al., 2015). In o he wo ds, i helps iden i y cha ac e s who play a “b idge”
ole in a ne wo k. Fi s , calcula e he be weenness cen ali y o each node by coun ing he
numbe o imes i lies on he sho es pa h be ween o he nodes (𝜎𝑠𝑡(𝑖)). Then, add up all
he be weenness cen ali y alues o he nodes in he ne wo k. Nex , no malize each alue
by di iding i by he maximum possible alue o any node in he ne wo k, ensu ing all alues
all be ween 0 and 1. Finally, compu e he a e age be weenness cen ali y by di iding he
o al sum o he no malized alues by he numbe o nodes in he ne wo k. By a e aging i o
he whole ne wo k, a e age be weenness cen ali y (ABC), we can cap u e he o e all
impo ance o in e media y nodes wi hin he en i e ne wo k. A alue o 0 o a e age
be weenness cen ali y means ha , on a e age, nodes do no ac as in e media ies; his could
be due o a comple ely disconnec ed ne wo k o one whe e all he sho es pa hs a e di ec .
In con as , a high posi i e alue means ha many cha ac e s play a c i ical ole in acili a ing
in e ac ions o he low o in o ma ion be ween nodes.
𝐴𝐵𝐶 = 1
𝑁 ∑2
(𝑁−1)∗(𝑁−2) ∑𝜎𝑠𝑡(𝑖)
𝜎𝑠𝑡
𝑠≠𝑖≠𝑗
𝑁
𝑖=1 (𝟔)
The deg ee asso a i i y measu es how co ela ed he deg ees o connec ed e ices a e, in
o he wo ds, i measu es he endency o nodes o connec o simila nodes wi h espec o
hei deg ee (Sil a e al., 2023; Laba u & Bos , 2020). This me ic helps cha ac e ize he
ela ionship be ween p ima y and mino cha ac e s. Being a coe icien , asso a i i y is
calcula ed as he a io o co a iance o a iance. To calcula e he co a iance, we look a he
14
connec ions, deg ee 𝑘, o e e y node and hen de e mine he join deg ee dis ibu ion, 𝑒𝑗𝑘.
This join deg ee dis ibu ion indica es how o en nodes wi h a ce ain deg ee 𝑘 a e connec ed
o nodes wi h deg ee 𝑗. In simple e ms, i measu es he equency o edges exis ing be ween
nodes o di e en deg ees. Then, his co a iance is di ided by he a iance o he deg ee
dis ibu ion, 𝜎2=∑𝑘2𝑞𝑘−[∑𝑘2𝑞𝑘]
𝑘2
𝒌
.
The esul akes a alue be ween 1, asso a i e,
whe e nodes wi h simila deg ees end o connec wi h each o he , and -1, disasso a i e,
whe e nodes wi h di e ing deg ees end o connec , o 0 (neu al) in he case o non-
co ela ion (Bha acha ya e al., 2023). In Sil a e al. (2023), he ne wo ks we e gene ally
disasso a i e, e lec ing he di e si y o social in e ac ions in he analysed wo ks much like in
eal-wo ld ne wo ks. The used o mula was collec ed om Massey (2016).
𝑟 =∑(𝑒𝑗𝑘 −𝑞𝑗𝑞𝑘)
𝑗𝑘 𝜎𝑞
2(𝟖)
Las ly, he a e age clus e ing coe icien (ACC) measu es he deg ee o which nodes in a
ne wo k end o clus e oge he , p o iding insigh in o he o e all endency o cha ac e s o
o m cohesi e and closed g oups (Newman, 2010; Suen e al., 2013; A danuy e al., 2015;
Dekke e al., 2019; Laba u & Bos , 2020). This me ic is calcula ed as he a e age o he local
clus e ing coe icien s o all nodes in he ne wo k. The local clus e ing coe icien is he
p obabili y, when andomly picking wo neighbo s o a node,
𝑘𝑖, ha he e is an edge be ween
hem 𝑚𝑖 (Laba u & Bos , 2020) hen a e aging i by he numbe o nodes N. The ACC is use ul
o unde s anding he gene al clus e ing endency ac oss all nodes in he ne wo k. A highe
ACC indica es a g ea e likelihood o s ong communi y s uc u es wi hin he na a i e. Dekke
e al. (2019) p o ide a compa a i e analysis o clus e ing coe icien s in social ne wo ks om
no els and a ious well-known ne wo ks such as you ube and Flick , demons a ing simila
a ia ion pa e ns ac oss di e en ne wo k ypes. The used o mula was collec ed om
Massey (2016).
𝐶 = 1
𝑁 ∑ 2𝑚𝑖
𝑘𝑖 (𝑘𝑖−1)
𝑁
𝑖=1 (𝟗)
The measu es we e calcula ed indi idually o each no el using Py hon wi h he Ne wo kX
package
23
and hen a e aged by gen e.
Mo eo e , a P incipal Componen Analysis (PCA) was conduc ed wi h he me ics. PCA is a
mul i a ia e echnique ha ex ac s he mos impo an in o ma ion om he da a by
23
h ps://ne wo kx.o g
15
educing i s dimensionali y hough he c ea ion o new a iables ha a e linea combina ions
o he o iginal a iables. This p ocess simpli ies he da ase and allows o an analysis o i s
s uc u e (Abdi & Williams, 2010). The goal is o unde s and he key me ics ha dis inguish
gen es while also ha e a isual ep esen a ions o hei g ouping.
22
Analyzing he esul s, i 's clea ha al hough each gen e gene ally g a i a es oge he , each
gen e has no clea and de ined g oup wi hin he PCA esul s, indica ing di e si y in na a i e
s uc u es wi hin e e y gen e unde s udy. This also means ha while he gen es na u ally
dis inguish hemsel es by g ouping oge he no els o he same ca ego y, hey do no clea ly
sepa a e om hose o di e en gen es, indica ing di e ences bu no comple e sepa a ion.
In ac , Mys e y p esen s i sel as a b idge ha ing i s cloud o no els loca ed be ween Fan asy
and Romance, as hese las wo a e mo e sepa a ed. This could imply ha he Mys e y gen e,
despi e he hema ic di e ences, employs simila na a i e s a egies in e ms o cha ac e
ne wo ks as he o he wo gen es.
Fan asy no els mos ly p esen nega i e alues o “Cha ac e In e connec edness and
Simila i y” and ha e mo e p ominen posi i e alues han mode a e nega i es o “Ne wo k
Size and Segmen a ion.” This ansla es in o di e se s uc u al ea u es bu a endency o
achie e ne wo ks ha a e mo e in e connec ed wi h ewe main cha ac e s in e ac ing wi h
a mo e di e se se o cha ac e s. Rega ding ne wo k size i s balanced, while some ocus on
la ge , mo e segmen ed ne wo ks wi h many main cha ac e s, o he s lean owa ds smalle ,
dense ne wo ks, e idenced by he ola ili y in sco es o “Ne wo k Size and Segmen a ion.”
This sugges s ha his gen e is mo e consensually ocused on he c ea ion o cha ac e wo k
a he han ela ionships.
Mys e y no els g a i a e owa ds he cen e o he axis, wi h mos no els showcasing
mode a e esul s on bo h PCs, b idging bo h Romance and Fan asy. This indica es a balance
in he a en ion owa ds cha ac e wo k as well as ela ionships. The cen al posi ioning on
he PCA plo sugges s ha Mys e y no els main ain a mode a e ne wo k size and densi y, wi h
a balanced numbe o main cha ac e s and in e media y connec ing cha ac e s. This balance
is c ucial o he gen e, as i o en in ol es in ica e plo s ha equi e a ne wo k s uc u e ha
suppo s bo h he de elopmen o indi idual cha ac e s and he ela ionships be ween hem.
The Romance no els p esen he highes alues o “Cha ac e In e connec edness and
Simila i y”. This means ha Romance no els ocus on c ea ing ne wo ks wi h mul iple cen al
cha ac e s, highe asso a i i y, and disconnec ed cha ac e s wi h a high a e age pa h leng h,
indica ing an emphasis on cha ac e ela ionships and in e ac ions. Each Romance no el
seems o ha e i s unique alue o “Ne wo k Size and Segmen a ion”, anging he whole scale
indica ing a wide a iabili y in ne wo k size and densi y. This a iabili y, along wi h he
p esence o bo h highly connec ed and isola ed cha ac e s, e lec s he complexi y and
di e si y o ela ionships ypically explo ed in Romance na a i es. This s uc u e suppo s he
gen e's explo a ion o a ious subplo s, p o iding a ich apes y o cha ac e in e ac ions.
In conclusion, he PCA analysis e eals ha while each gen e exhibi s dis inc s uc u al
cha ac e is ics, being Romance and Fan asy he mos di e en in e ms o hei cha ac e
ne wo k s uc u es. Romance no els ocus on c ea ing in ica e, in e connec ed ne wo ks
wi h mul iple cen al cha ac e s, while Fan asy no els a e mo e a iable, anging om highly
in e connec ed ne wo ks o la ge , mo e segmen ed ones. Mys e y no els, ac ing as a b idge

23
and a comp omise o he s uc u e o Romance and Fan asy, showing balanced ne wo k
s uc u es ha closely mimic eal-li e social ne wo ks.
24
5. CONCLUSIONS AND FUTURE WORKS
Gen e p o ides an impo an ame o e e ence ha helps eade s iden i y, selec , and
in e p e ex s (Chandle , 1997). This wo k explo ed he s uc u al di e ences among gen es
in Po uguese ex s h ough hei cha ac e ne wo ks. By combining a e aged ne wo k
me ics, P incipal Componen Analysis (PCA) and he inclusion o eal-li e ne wo ks, we can
d aw comp ehensi e conclusions abou how hese gen es di e and ela e o one ano he .
The a e aged ne wo k me ics highligh some s uc u al ea u es wi hin each gen e. Fan asy
no els a e cha ac e ized by a single, highly cen al he o wi h high maximum deg ee and
mode a e le els o densi y and clus e ing, indica ing a balanced in e connec edness. Mys e y
no els ea u e mode a ely sized, dense ne wo ks ha highligh signi ican ela ionships
be ween cha ac e s and he p esence o b idging cha ac e s, main aining connec i i y.
Romance no els display he mos a iable ne wo ks, wi h mode a e densi y and high
maximum deg ee, bu also show endencies o isola ed cha ac e s, leading o disconnec ed
ne wo ks. Las ly, eal-li e ne wo ks do no p esen any cen ali y on a pa icula cha ac e no
in a ela ionship.
The P incipal Componen Analysis esul s suppo hese indings by showing no clea
clus e ing o gen es, sugges ing ha s uc u al di e ences in cha ac e ne wo ks a e no
s ongly ied o gen e. The dispe se in he PCA esul s wi hin each gen e, indica es di e si y in
na a i e s uc u es. While he gen es na u ally dis inguish hemsel es by g ouping oge he
no els o he same ca ego y, hey do no clea ly sepa a e om hose o di e en gen es,
indica ing di e ences bu no comple e sepa a ion. Fan asy no els show di e se s uc u al
ea u es bu end o achie e ne wo ks ha a e mo e in e connec ed wi h ewe main
cha ac e s in e ac ing wi h a mo e di e se se o cha ac e s. Mos Mys e y no els showcase a
mode a e balance be ween cha ac e and ela ionship building, while Romance emphasizes
ela ionship building wi hin i s na a i e dynamics. Mys e y p esen s i sel as a b idge
be ween Fan asy and Romance, sugges ing ha despi e hema ic di e ences, simila na a i e
s a egies a e employed in e ms o cha ac e ne wo ks om bo h gen es. The eal-li e
ne wo ks s and nea each o he and nea he li e a y ne wo ks sha ing common s uc u al
cha ac e is ics while also exhibi ing unique ea u es ha e lec hei dis inc con ex s, being
his g oup mos simila o Mys e y.
Ou conclusion is ha he simila i ies wi hin gen es and di e ences be ween gen es a e no
p ominen in g aph-speci ic measu es, e en hough each gen e exhibi s unique s uc u al
cha ac e is ics. Fan asy no els a e cha ac e ized by a ocus on a cen al he o and less
in e connec ed g oups, aligning wi h hei na a i e emphasis on indi idual jou neys and less
ela ionship building o e all. Mys e y no els exhibi mode a ely sized, dense ne wo ks wi h
key b idging cha ac e s and a balance be ween ela ionships and cha ac e de elopmen ,
e lec ing hei in ica e plo s and a e he mos simila o eal-li e ne wo ks. Romance no els
25
display he la ges and mos a iable ne wo ks, ha ing mo e han one cen al cha ac e and
cohesi e subg oups, indica i e o hei ocus on di e se and in ica e ela ionships.
These indings p o ide aluable insigh s in o he na a i e s yles o Po uguese-language
li e a u e, showing how each gen e employs dis inc ne wo k s uc u es o suppo hei
unique s o y elling app oaches. The lack o p onounced di e ences could be expec ed since,
as Thelwall & Kousha (2017) poin ou , gen e is a concep ha eade s, w i e s, and publishe s
may in e p e uniquely. Bookselle s may in en new gen es o ma ke books, lib a ies may
ca ego ize books in ce ain ways o a ac eade s, and gen e ypes and ypical s uc u es can
a y immensely in opics and in e na ionally.
In in e p e ing he indings o his s udy, i is impo an o conside he limi a ions associa ed
wi h he da a and me hodology. Fi s ly, all he in e p e a ions ela e o he pipeline and i s
exis ing limi a ions, such as he inabili y o p ocess s o ies wi h mul iple poin s o iew and
some deg ee o imp ecision in he co-occu ence de ec ion, which is no ully obus o non-
Po uguese names. Fu he mo e, one signi ican conce n is he use o a e ages o desc ibe
ne wo k me ics o each gen e. While a e ages p o ide a use ul summa y, hey can
some imes obscu e impo an a ia ions and nuances wi hin indi idual wo ks, his is a oided
by comple ing he in o ma ion wi h he s anda d de ia ion. Ano he limi a ion, his ela ed o
he da a, is he unde de elopmen o Po uguese hemes in gen es such as an asy o o he
as SiFi ha could be s udied, wi h ebook e sions ha would allow o a la ge da ase . In ac ,
all he small sample sizes could lead o esul s ha a e less eliable and ha de o gene alize
o he en i e gen e. Fu u e esea ch should aim o include a la ge and mo e di e se sample
o no els o alida e hese indings.
26
BIBLIOGRAPHICAL REFERENCES
Abdi, H., & Williams, L. J. (2010). P incipal componen analysis. WIREs Compu a ional
S a is ics, 2(4), 433–459. h ps://doi.o g/10.1002/wics.101
Albe ich, R., Mi o-Julia, J., & Rossello, F. (2002). Ma el Uni e se looks almos like a eal social
ne wo k (a Xi :cond-ma /0202174). a Xi . h ps://doi.o g/10.48550/a Xi .cond-
ma /0202174
Ama al, D. O. F., Fonseca, E. B., Lopes, L., & Viei a, R. (n.d.). Compa a i e Analysis o
Po uguese Named En i ies Recogni ion Tools. Re ie ed om
h ps:// eposi o io.puc s.b /dspace/bi s eam/10923/14041/2/Compa a i e_Analysis_
o _Po uguese_Named_En i ies_Recogni ion_Tools.pd
A danuy, M. C., & Spo lede , C. (2015). Clus e ing o No els Rep esen ed as Social Ne wo ks.
Linguis ic Issues in Language Technology, 12. h ps://doi.o g/10.33011/lil . 12i.1379
Ben o, F., Tagliabue, M., & Sandake , I. (2020). Complex Sys ems and Social Beha io : B idging
Social Ne wo ks and Beha io Analysis (pp. 67–91). h ps://doi.o g/10.1007/978-3-030-
45421-0_4
Bha acha ya, S., Sinha, S., Dey, P., Saha, A., Chowdhu y, C., & Roy, S. (2023). Chap e 5—
Online social-ne wo k sensing models. In D. Das, A. K. Kolya, A. Basu, & S. Sa ka (Eds.),
Compu a ional In elligence Applica ions o Tex and Sen imen Da a Analysis (pp. 113–
140). Academic P ess. h ps://doi.o g/10.1016/B978-0-32-390535-0.00010-0
Bo ga i, S. P., Meh a, A., B ass, D. J., & Labianca, G. (2009). Ne wo k Analysis in he Social
Sciences. Science, 323(5916), 892–895. h ps://doi.o g/10.1126/science.1165821
Bo ga i, S. P., E e e , M. G., & Johnson, J. C. (2018). Analyzing Social Ne wo ks. SAGE
Publica ions. h ps://books.google.p /books?id=XD1ADwAAQBAJ
Camb ia, E., & Whi e, B. (2014). Jumping NLP Cu es: A Re iew o Na u al Language P ocessing
Resea ch [Re iew A icle]. IEEE Compu a ional In elligence Magazine, 9(2), 48–57.
h ps://doi.o g/10.1109/MCI.2014.2307227
Caná io,T., & Dua e, C. (2024). TAGGUS. Gi Hub. Re ie ed July 3, 2024, om
h ps://gi hub.com/ icadu/ aggus
Chandle , D. (1997). An in oduc ion o gen e heo y. h p:// isual-
memo y.co.uk/daniel//Documen s/in gen e/chandle _gen e_ heo y.pd
Dekke , N., Kuhn, T., & E p, M. an. (2019). E alua ing named en i y ecogni ion ools o
ex ac ing social ne wo ks om no els. Pee J Compu e Science, 5, e189.
h ps://doi.o g/10.7717/pee j-cs.189
27
dos Reis, C. A. A. (1993). His ó ia c í ica da li e a u a po uguesa. Edi o ial Ve bo.
h ps://books.google.p /books?id=G8YQAQAAMAAJ
Ea le, T. F., Pa kinson, S., & Alonso, C. P. (Eds.). (2013). A companion o Po uguese li e a u e
(Vol. 282). Boydell & B ewe L d. h ps://books.google.com/books?hl=p -
PT&l =&id=TNQTAgAAQBAJ&oi= nd&pg=PP1&dq=po uguese+li e a u e&o s= 812Ai5
dnC&sig=1cJBKY0nl0i89YP2OJn0VI1EI7Q
Fan, C., & Li, Y. (2022). Ne wo k ex ac ion and analysis o cha ac e ela ionships in Chinese
li e a y wo ks. Compu a ional In elligence and Neu oscience, 2022.
h ps://doi.o g/10.1155/2022/7295834
Fe nández-Peña, R., O alle-Pe andones, M.-A., Ma qués-Sánchez, P., O ego-Ma é, C., &
Se ano-Fuen es, N. (2022). The use o social ne wo k analysis in social suppo and ca e:
A sys ema ic scoping e iew p o ocol. Sys ema ic Re iews, 11(1), 9.
h ps://doi.o g/10.1186/s13643-021-01876-2
Fonseca, B. de P. F. e, Sampaio, R. B., Fonseca, M. V. de A., & Zicke , F. (2016). Co-au ho ship
ne wo k analysis in heal h esea ch: Me hod and po en ial use. Heal h Resea ch Policy
and Sys ems, 14(1), 34. h ps://doi.o g/10.1186/s12961-016-0104-5
Fos e , J. (n.d.). Examining mys e y, science ic ion, omance, and ho o : A look a popula
ic ion gen es. Jamie Fos e Science. Re ie ed June 19, 2024, om
h ps://www.jamie os e science.com/mys e y-science- ic ion- omance-and-ho o -
a e-all-conside ed/
Fowle , A. (1982). Kinds o li e a u e: An in oduc ion o he heo y o gen es and modes.
Camb idge, Mass. : Ha a d Uni e si y P ess.
h p://a chi e.o g/de ails/kindso li e a u 0000 owl
F eeman, L. C. (2004). The De elopmen o Social Ne wo k Analysis: A S udy in he Sociology
o Science. Empi ical P ess. h ps://books.google.p /books?id=VcxqQgAACAAJ
F onze i Colladon, A., & Naldi, M. (2019). P edic ing he pe o mance o TV se ies h ough
ex ual and ne wo k analysis: The case o Big Bang Theo y. PLOS ONE, 14(11), e0225306.
h ps://doi.o g/10.1371/jou nal.pone.0225306
Gessey-Jones, T., Connaugh on, C., Dunba , R., Kenna, R., MacCa on, P., O’Conchobhai , C.,
& Yose, J. (2020). Na a i e s uc u e o A Song o Ice and Fi e c ea es a ic ional wo ld
wi h ealis ic measu es o social complexi y. P oceedings o he Na ional Academy o
Sciences, 117(46), 28582–28588. h ps://doi.o g/10.1073/pnas.2006465117
Hansen, D. L., Shneide man, B., Smi h, M. A., & Himelboim, I. (2020). Chap e 6—Calcula ing
and isualizing ne wo k me ics. In D. L. Hansen, B. Shneide man, M. A. Smi h, & I.

28
Himelboim (Eds.), Analyzing Social Media Ne wo ks wi h NodeXL (Second Edi ion) (pp.
79–94). Mo gan Kau mann. h ps://doi.o g/10.1016/B978-0-12-817756-3.00006-6
He inge , L., Becke , M., Rege , I., Jannidis, F., & Ho ho, A. (2015). Gen e classi ica ion on
Ge man no els. In 2015 26 h In e na ional Wo kshop on Da abase and Expe Sys ems
Applica ions (DEXA) (pp. 249-253). IEEE. h ps://doi.o g/10.1109/DEXA.2015.62
Isinkaye, F. O., Folajimi, Y. O., & Ojokoh, B. A. (2015). Recommenda ion sys ems: P inciples,
me hods and e alua ion. Egyp ian In o ma ics Jou nal, 16(3), 261–273.
h ps://doi.o g/10.1016/j.eij.2015.06.005
Ja ynowski, A., & Boland, S. (2016). Social Ne wo ks Analysis in Disco e ing he Na a i e
S uc u e o Li e a y Fic ion. h ps://doi.o g/10.48550/a Xi .1608.05982
Kim, K.-R., & Kim, J. (2012). Recommende sys em design using mo ie gen e simila i y and
p e e ed gen es in Sma Phone. Mul imedia Tools and Applica ions - MTA, 61, 1–18.
h ps://doi.o g/10.1007/s11042-011-0728-y
Laba u , V., & Bos , X. (2020). Ex ac ion and Analysis o Fic ional Cha ac e Ne wo ks: A
Su ey. ACM Compu ing Su eys, 52(5), 1–40. h ps://doi.o g/10.1145/3344548
Lou enço, R.F. (2014) Guia do Colabo ado [ 1.01], 16. h ps://p ojec oadamas o .o g/wp-
con en /uploads/2014/09/Guia-do-Colabo ado .docx
Mamede, N., & Chalei a, P. (2004). Cha ac e Iden i ica ion in Child en S o ies. In J. L. Vicedo,
P. Ma ínez-Ba co, R. Muńoz, & M. Saiz Noeda (Eds.), Ad ances in Na u al Language
P ocessing (Vol. 3230, pp. 82–90). Sp inge Be lin Heidelbe g.
h ps://doi.o g/10.1007/978-3-540-30228-5_8
Massey, S. E. (2016). Social ne wo k analysis o he biblical Moses. Applied Ne wo k Science,
1(1), A icle 1. h ps://doi.o g/10.1007/s41109-016-0012-1
Ma a, A. S. da. (2020). Complex Ne wo ks: A Mini- e iew. B azilian Jou nal o Physics, 50(5),
658–672. h ps://doi.o g/10.1007/s13538-020-00772-9
Maulud, D. H., Zeeba ee, S. R. M., Jacksi, K., Sadeeq, M. A. M., & Sha i , K. H. (2021). S a e o
A o Seman ic Analysis o Na u al Language P ocessing. Qubahan Academic Jou nal,
1(2), A icle 2. h ps://doi.o g/10.48161/qaj. 1n2a44
Middha, K., Munzi , M., Goja, S., & Choudha y, S. (2022). En e ainmen con en
ecommenda ion sys em using machine lea ning. In e na ional Resea ch Jou nal o
Enginee ing and Technology (IRJET), 10(1), 279-284.
h ps://www.i je .ne /a chi es/V10/i1/IRJET-V10I148.pd
29
Mou chid, Y., Renous , B., Roupin, O., Văn, L., Che i i, H., & Hassouni, M. E. (2019). Mo iene :
A mo ie mul ilaye ne wo k model using isual and ex ual seman ic cues. Applied
Ne wo k Science, 4(1), A icle 1. h ps://doi.o g/10.1007/s41109-019-0226-0
Ne wo k Densi y—An o e iew. ScienceDi ec Topics. Re ie ed June 20, 2024, om
h ps://www.sciencedi ec .com/ opics/compu e -science/ne wo k-densi y
Newman, M. (2010). Ne wo ks: An In oduc ion. OUP Ox o d.
h ps://books.google.p /books?id=L FaU4XCsUoC
No man, M., Jones, C., Wa son, K., & P e idelli, R. L. (2023). Social Ne wo k Analysis as a Tool
in he Ca e and Wellbeing o Zoo Animals: A Case S udy o a Family G oup o Black
Lemu s (Eulemu macaco). Animals, 13(22), A icle 22.
h ps://doi.o g/10.3390/ani13223501
O e, E., & Rousseau, R. (2002). Social ne wo k analysis: A powe ul s a egy, also o he
in o ma ion sciences. Jou nal o In o ma ion Science. 28 (6): 441–453.
h ps://doi.o g/10.1177/016555150202800601
Ozella, L., Paolo i, D., Lichand, G., Rod íguez, J. P., Haenni, S., Phuka, J., Leal-Ne o, O. B., &
Ca u o, C. (2021). Using wea able p oximi y senso s o cha ac e ize social con ac
pa e ns in a illage o u al Malawi. EPJ Da a Science, 10(1), A icle 1.
h ps://doi.o g/10.1140/epjds/s13688-021-00302-w
Pa k, S.-B., Kim, Y.-W., Uddin, M. N., & Jo, G.-S. (2009). Cha ac e -Ne : Cha ac e Ne wo k
Analysis om Video. 2009 IEEE/WIC/ACM In e na ional Join Con e ence on Web
In elligence and In elligen Agen Technology, 1, 305–308. h ps://doi.o g/10.1109/WI-
IAT.2009.54
Pa k, G.-M., Kim, S.-H., & Cho, H.-G. (2013). S uc u al Analysis on Social Ne wo k Cons uc ed
om Cha ac e s in Li e a u e Tex s. Jou nal o Compu e s, 8.
h ps://doi.o g/10.4304/jcp.8.9.2442-2447
Rahul, Ayush, Aga wal, D., & Vijay, D. (2021). Gen e Classi ica ion using Cha ac e Ne wo ks.
2021 5 h In e na ional Con e ence on In elligen Compu ing and Con ol Sys ems
(ICICCS), 216–222. h ps://doi.o g/10.1109/ICICCS51141.2021.9432303
Reddy, S. R., Nallu i, S., Kunise i, S., Ashok, S., & Venka esh, B. (2019). Con en -based mo ie
ecommenda ion sys em using gen e co ela ion. In S. C. Sa apa hy, V. Bha eja, & H. S.
Saini (Eds.), Sma In elligen Compu ing and Applica ions (Vol. 105, pp. 391-397).
Sp inge . h ps://doi.o g/10.1007/978-981-13-1927-3_42
Rocha, C., Jo ge, A., Oli ei a, M., B i o, P., Gama, J., & Pimen a, C. (2014). F om en i y
ex ac ion o ne wo k analysis: A me hod and an applica ion o a Po uguese ex ual
sou ce. h ps://www. ep.up.p /docen es/cpimen a/ ex os/pd /ACM15_Rocha2.pd
30
Rossi, R. A., & Ahmed, N. K. (2015). The Ne wo k Da a Reposi o y wi h in e ac i e g aph
analy ics and isualiza ion. In P oceedings o he AAAI Con e ence on A i icial
In elligence , 29(1). h ps://doi.o g/10.1609/aaai. 29i1.9277
Sil a, M. O., Oli ei a, G. P., & Mo o, M. M. (2023). Analyzing Cha ac e Ne wo ks in
Po uguese-language Li e a y Wo ks. Anais Do B azilian Wo kshop on Social Ne wo k
Analysis and Mining (B aSNAM), 115–126.
h ps://doi.o g/10.5753/b asnam.2023.230585
Sudhaha , S., C is ianini, N. (2013). Au oma ed analysis o na a i e con en o digi al
humani ies. In e na ional Jou nal o Ad anced Compu e Science. 3.
h ps://ci esee x.is .psu.edu/documen ? epid= ep1& ype=pd &doi=58e9de78dc9c534
8339e83 774c689bb 7d9c45a
Sugishi a, K., & Masuda, N. (2023). Social ne wo k analysis o manga: Simila i ies o eal-wo ld
social ne wo ks and ends o e decades. Applied Ne wo k Science, 8(1), A icle 1.
h ps://doi.o g/10.1007/s41109-023-00604-0
Suen, C., Kuenzel, L., & Gil, S. (2013). Ex ac ion and Analysis o Cha ac e In e ac ion Ne wo ks
F om Plays and Mo ies. Digi al Humani ies Con e ence.
h ps://www.seman icschola .o g/pape /Ex ac ion-and-Analysis-o -Cha ac e -
In e ac ion-Suen-Kuenzel/a57b 7098ea7adcc 626caba682 3b4350aa3955
Tamen, M., & Buescu, H. C. (2013). A e isiona y his o y o Po uguese li e a u e. Rou ledge.
h ps://api. aylo ancis.com/con en /books/mono/download?iden i ie Name=doi&id
en i ie Value=10.4324/9780203726877& ype=googlepd
Tan a dini, M., Ie a, F., Tajoli, L., & Picca di, C. (2019). Compa ing me hods o compa ing
ne wo ks. Scien i ic Repo s, 9(1), 17557. h ps://doi.o g/10.1038/s41598-019-53708-y
Thelwall, M., & Kousha, K. (2017). Good eads: A social ne wo k si e o book eade s. Jou nal
o he Associa ion o In o ma ion Science and Technology, 68(4), 972–983.
h ps://doi.o g/10.1002/asi.23733
Wellek, Ren., & Wa en, A. (1956). Theo y o li e a u e. Ha cou , B ace & Wo ld.
Zacha y, W. W. (1977). An In o ma ion Flow Model o Con lic and Fission in Small G oups.
Jou nal o An h opological Resea ch, 33(4), 452–473.
h ps://doi.o g/10.1086/ja .33.4.3629752
31
APPENDIX A
Figu e 5 - Cha ac e Ne wo k o "A Cos a dos Mu mú ios"
Gen e: Romance
38
João
Deus
Jaime
Paixão
Amélia
Alice
Domingos
Sequei a
João
Pessoa
Nicolau
Eunice
TVN
William
Alenque
0
0
0
0
0
1
1
2
Guilhe me
Vau
0
1
0
0
0
0
0
0
Albe o
0
0
0
0
0
0
0
0
Es ela
1
1
0
0
1
0
3
0
A onso Soa es
0
0
0
0
0
0
1
0
Joel F anco
0
19
20
13
0
0
0
2
Ba adas
0
1
0
0
0
0
1
0
Di ei o
0
1
0
1
0
0
0
0
João Deus
0
0
0
0
0
0
0
0
Jaime Paixão
0
0
5
0
0
0
0
0
Amélia
0
5
0
8
0
0
0
0
Alice
0
0
8
0
0
0
0
0
Domingos
Sequei a
0
0
0
0
0
0
0
0
João Pessoa
0
0
0
0
0
0
0
0
Nicolau
0
0
0
0
0
0
0
1
Eunice TVN
0
0
0
0
0
0
1
0
Ana Paula
0
0
0
0
3
0
0
0
Humbe o
0
0
0
0
0
0
0
0
Be na dino
Reis
0
3
0
0
0
0
0
3
Sebas ião
Ca alho
0
0
0
0
0
0
0
1
Inês
Albe ga ia
0
0
0
0
0
0
0
0
No h ope
0
0
0
0
0
0
0
0
Lond es
0
0
0
0
0
0
0
0
An ónio
Monsan o
0
0
0
0
0
0
0
3
Conceição
Boni ácio
0
0
0
1
0
0
0
0
Ma lene
0
0
0
0
0
0
0
0
Jo ge
0
0
0
0
0
0
0
0
Golden
In es men s
0
0
0
0
0
0
0
0
Aníbal
0
0
0
0
0
0
0
0
An on G isak
0
0
0
0
0
0
0
0

39
Ana Paula
Humbe o
Be na din
o Reis
Sebas ião
Ca alho
Inês
Albe ga ia
No h ope
Lond es
William
Alenque
0
3
9
0
0
4
3
Guilhe me
Vau
1
0
33
0
0
13
4
Albe o
0
0
0
0
0
0
0
Es ela
0
0
0
0
0
0
0
A onso
Soa es
0
0
15
0
0
0
0
Joel F anco
0
0
20
0
10
1
0
Ba adas
0
0
1
0
0
0
0
Di ei o
0
0
1
0
0
0
0
João Deus
0
0
0
0
0
0
0
Jaime Paixão
0
0
3
0
0
0
0
Amélia
0
0
0
0
0
0
0
Alice
0
0
0
0
0
0
0
Domingos
Sequei a
3
0
0
0
0
0
0
João Pessoa
0
0
0
0
0
0
0
Nicolau
0
0
0
0
0
0
0
Eunice TVN
0
0
3
1
0
0
0
Ana Paula
0
0
0
0
0
0
0
Humbe o
0
0
0
0
0
0
0
Be na dino
Reis
0
0
0
0
1
3
0
Sebas ião
Ca alho
0
0
0
0
0
0
0
Inês
Albe ga ia
0
0
1
0
0
0
0
No h ope
0
0
3
0
0
0
6
Lond es
0
0
0
0
0
6
0
An ónio
Monsan o
0
0
12
0
1
3
1
Conceição
Boni ácio
0
0
0
0
0
0
0
Ma lene
0
0
0
0
0
0
0
Jo ge
0
0
0
0
0
0
0
Golden
In es men s
0
0
1
0
0
5
3
Aníbal
0
0
1
0
0
0
0
An on G isak
0
0
0
0
0
1
0
40
An ónio
Monsan o
Conceição
Boni ácio
Ma lene
Jo ge
Golden
In es men s
Aníbal
An on
G isak
William
Alenque
10
4
3
0
0
0
0
Guilhe me
Vau
17
2
0
0
3
0
1
Albe o
0
0
0
0
0
0
0
Es ela
1
0
0
0
0
0
0
A onso Soa es
4
0
0
0
0
0
0
Joel F anco
9
4
0
0
0
1
0
Ba adas
1
0
0
1
0
0
0
Di ei o
1
0
1
0
0
0
0
João Deus
0
0
0
0
0
0
0
Jaime Paixão
0
0
0
0
0
0
0
Amélia
0
0
0
0
0
0
0
Alice
0
1
0
0
0
0
0
Domingos
Sequei a
0
0
0
0
0
0
0
João Pessoa
0
0
0
0
0
0
0
Nicolau
0
0
0
0
0
0
0
Eunice TVN
3
0
0
0
0
0
0
Ana Paula
0
0
0
0
0
0
0
Humbe o
0
0
0
0
0
0
0
Be na dino
Reis
12
0
0
0
1
1
0
Sebas ião
Ca alho
0
0
0
0
0
0
0
Inês
Albe ga ia
1
0
0
0
0
0
0
No h ope
3
0
0
0
5
0
1
Lond es
1
0
0
0
3
0
0
An ónio
Monsan o
0
2
0
0
0
0
1
Conceição
Boni ácio
2
0
0
0
0
0
0
Ma lene
0
0
0
0
0
0
0
Jo ge
0
0
0
0
0
0
0
Golden
In es men s
0
0
0
0
0
0
0
Aníbal
0
0
0
0
0
0
0
An on G isak
1
0
0
0
0
0
0
41
Figu e 7 - Cha ac e Ne wo k o "Limbo"
Gen e: Fan asy
42
Table 7 - Adjacency Ma ix o "Limbo"
Gen e: Fan asy
Ve dade
Lili h
Azazel
Ra ae
Ca los
Tomoe
Ma sílio
Oli ie
Ca los
Magno
Du andal
Ve dade
0
1
0
0
0
0
0
0
0
0
Lili h
1
0
12
0
23
1
0
1
1
0
Azazel
0
12
0
2
1
0
0
0
0
0
Ra ael
0
0
2
0
0
0
0
0
0
0
Ca los
0
23
1
0
0
1
0
0
0
1
Tomoe
0
1
0
0
1
0
0
0
0
0
Ma sílio
0
0
0
0
0
0
0
1
0
0
Oli ie
0
1
0
0
0
0
1
0
1
0
Ca los
Magno
0
1
0
0
0
0
0
1
0
0
Du andal
0
0
0
0
1
0
0
0
0
0
Musa
Musas
0
1
0
0
0
0
0
0
0
0
Finn
Cumhail
0
0
0
0
3
0
0
0
0
0
Ex-
Gene al
1
1
0
0
0
0
0
0
0
0
P o esso
0
1
0
0
1
0
0
0
0
0
Hen ique
Xiangu
0
1
0
0
2
0
0
0
0
0
Tio
0
1
0
0
0
0
0
0
0
0
Adolesce
n e
1
0
0
0
0
0
0
0
0
0
Ancião
0
0
0
0
0
0
0
0
0
0
Odin
0
1
0
0
0
0
0
0
0
0
Tho
0
0
0
0
0
0
0
0
0
0
Miguel
0
0
1
1
0
0
0
0
0
0
Eleos
0
5
0
0
2
0
0
0
0
0
A u
0
3
0
0
1
0
0
0
0
0
Samael
0
1
0
0
1
0
0
0
0
0
Gab iel
0
4
0
1
0
0
0
0
0
0
43
Musa
Musas
Finn
Cumhail
Ex-
Gene al
P o esso
Hen iqu
e Xiangu
Tio
Adolesce
n e
Ancião
Ve dade
0
0
1
0
0
0
1
0
Lili h
1
0
1
1
1
1
0
0
Azazel
0
0
0
0
0
0
0
0
Ra ael
0
0
0
0
0
0
0
0
Ca los
0
3
0
1
2
0
0
0
Tomoe
0
0
0
0
0
0
0
0
Ma sílio
0
0
0
0
0
0
0
0
Oli ie
0
0
0
0
0
0
0
0
Ca los
Magno
0
0
0
0
0
0
0
0
Du andal
0
0
0
0
0
0
0
0
Musa
Musas
0
0
0
0
0
0
0
0
Finn
Cumhail
0
0
0
0
0
0
0
0
Ex-Gene al
0
0
0
4
1
0
0
1
P o esso
0
0
4
0
1
2
1
1
Hen ique
Xiangu
0
0
1
1
0
0
0
0
Tio
0
0
0
2
0
0
1
1
Adolescen
e
0
0
0
1
0
1
0
1
Ancião
0
0
1
1
0
1
1
0
Odin
0
0
0
0
0
0
0
0
Tho
0
0
0
0
0
0
0
0
Miguel
0
0
0
0
0
0
0
0
Eleos
0
0
0
0
0
0
0
0
A u
0
1
0
0
0
0
0
0
Samael
0
0
0
0
0
0
0
0
Gab iel
0
0
0
0
0
0
0
0

44
Odin
Tho
Miguel
Eleos
A u
Samael
Gab iel
Ve dade
0
0
0
0
0
0
0
Lili h
1
0
0
5
3
1
4
Azazel
0
0
1
0
0
0
0
Ra ael
0
0
1
0
0
0
1
Ca los
0
0
0
2
1
1
0
Tomoe
0
0
0
0
0
0
0
Ma sílio
0
0
0
0
0
0
0
Oli ie
0
0
0
0
0
0
0
Ca los
Magno
0
0
0
0
0
0
0
Du andal
0
0
0
0
0
0
0
Musa Musas
0
0
0
0
0
0
0
Finn
Cumhail
0
0
0
0
1
0
0
Ex-Gene al
0
0
0
0
0
0
0
P o esso
0
0
0
0
0
0
0
Hen ique
Xiangu
0
0
0
0
0
0
0
Tio
0
0
0
0
0
0
0
Adolescen e
0
0
0
0
0
0
0
Ancião
0
0
0
0
0
0
0
Odin
0
1
0
0
0
0
0
Tho
1
0
0
0
0
0
0
Miguel
0
0
0
0
0
0
1
Eleos
0
0
0
0
0
0
0
A u
0
0
0
0
0
0
0
Samael
0
0
0
0
0
0
0
Gab iel
0
0
1
0
0
0
0
45