i
Ve nacula bounda ies o his o ic neighbo hoods
in he ci y o Lisbon
Mónica So ía Roncancio Blanco
ii
Ve nacula Bounda ies o His o ic Neighbo hoods
in he Ci y o Lisbon
Disse a ion supe ised by
P o . D . Ma co Oc  io T indade Painho, PhD
NOVA In o ma ion Managemen School
Lisbon, Po ugal
Co-supe ised by:
Vicen e De Aze edo Tang
NOVA In o ma ion Managemen School
Lisbon, Po ugal
P o . D . S en Cas eleyn, PhD
Uni e si a Jaume I
Cas ell, Spain
Feb ua y, 2024
iii
Decla a ion o O iginali y
I decla e ha he wo k desc ibed in his documen is my own and no om
someone else. All he assis ance I ha e ecei ed om o he people is duly
acknowledged and all he sou ces (published o no published) a e
e e enced.
This wo k has no been p e iously e alua ed o submi ed o NOVA
In o ma ion Managemen School o elsewhe e.
Lisbon, Feb ua y 26, 2024
Mónica So ía Roncancio Blanco
[ he signed o iginal has been a chi ed by he NOVA IMS se ices]
i
Acknowledgmen s
Fi s , I ex end my deepes g a i ude o P o . D . Ma co Painho o his
suppo and guidance h oughou his academic jou ney. He has been an
inc edible p o esso and human being.
I am also hank ul o D . S en Cas eleyn and Vicen e De Aze edo Tang o
hei insigh s and sugges ions as my hesis supe iso s. Thei pa ience and
guidance ha e been in aluable. Addi ionally, My belo ed iend Nyi Nyi
Nyan Lin dese es special men ion o his in aluable suppo o e hese 6
mon hs, o e ing always accu a e ecommenda ions o imp o e my p ojec .
This p og am has a o ded me he oppo uni y o ul ill my d eam o
s udying ab oad, allowing me o be pa o h ee Eu opean Uni e si ies
and o es ablish academic, p o essional, and pe sonal connec ions.
Finally, my hea el hanks go o my amily and iends o hei daily
suppo om a a . This new academic and pe sonal achie emen is
undoub edly o hem.
This hesis is one o he ou pu s o a p ojec unded by he Fundação pa a
a Ciência e a Tecnologia (FCT), named "Ci yMe," wi h he P ojec
Re e ence (EXPL/GESURB/1429/2021).
Ve nacula bounda ies o his o ic neighbo hoods
in he ci y o Lisbon
Abs ac
This esea ch in es iga es he delinea ion o e nacula bounda ies in
Lisbon's his o ic neighbo hoods: Al ama, Mou a ia, and Bai o Al o, he
s udy explo es he in eg a ion o pe cei ed bounda ies de i ed om
esiden s' ske ches and geo- agged da a om online ac i i ies. U ilizing a
wo-phase me hodology, he esea ch i s ex ac s pe cei ed bounda ies
h ough web-based su eys ca ego ized by esiden s' leng h o s ay,
dis inguishing be ween sho - e m (less han 10 yea s) and long- e m
esiden s (mo e han 10 yea s). This app oach allows o an o e lay
analysis, iden i ying Co e and Domain egions based on consensus
h esholds. Secondly, he s udy e ie es geo- agged bounda ies using A-
DBSCAN and alpha-shape algo i hms o analyze online ac i i y, o e ing a
compa a i e analysis wi h he pe cei ed bounda ies.
The indings e eal a nuanced unde s anding o how esiden s and online
use s concep ualize neighbo hood spaces, highligh ing disc epancies and
con e gences be ween pe cei ed and digi al mappings. By calcula ing he
In e sec ion O e Union (IOU) and F-sco es, he esea ch quan i a i ely
assesses he o e lap be ween di e en da a sou ces, iden i ying he mos
accu a e delinea ions ha e lec he his o ic neighbo hoods' spa ial
eali y. This s udy con ibu es o u ban planning and policymaking by
p o iding insigh s in o esiden s' spa ial pe cep ions, emphasizing he
impo ance o conside ing bo h li ed expe iences and digi al oo p in s in
he mapping o u ban a eas. The esea ch unde sco es he po en ial o
combining adi ional su ey me hods wi h inno a i e geo-spa ial
echnologies o enhance he p ecision and ele ance o u ban geog aphic
s udies.
Sus ainable De elopmen Goals (SGD):

i
Keywo ds
Ve nacula Bounda ies
Pe cei ed Bounda ies
Use Gene a ed Con en
Clus e ing Analysis
Geog aphic In o ma ion Science
Spa ial analysis
His o ic neighbo hoods
ii
Ac onyms
A-DBSCAN – App oxima e Densi y-Based Spa ial Clus e ing o
Applica ions wi h Noise
GIS – Geog aphic In o ma ion Sys ems
IOU – he In e sec ion O e Union
NBHD –Neighbo hood
UGC – Use Gene a ed Con en
iii
Index o he Tex
Acknowledgmen s .................................................................................... i
Abs ac .......................................................................................................
Keywo ds ................................................................................................... i
Ac onyms .................................................................................................. ii
Index o he Tex ................................................................................... iii
Index o Tables ........................................................................................... x
Index o Figu es ....................................................................................... xi
1. In oduc ion ...................................................................................... 1
1.1 Aim and Resea ch Ques ions ......................................................... 2
1.2 P ojec Objec i es ......................................................................... 2
1.3 App oach o Achie ing he P ojec Objec i es ......................... 2
1.4 Thesis O ganiza ion ........................................................................ 3
2.Li e a u e Re iew .................................................................................. 4
2.1 S udying Neighbo hoods, Ve nacula bounda ies, and Place.
.................................................................................................................... 4
2.1.1 Neighbo hoods ............................................................................... 4
2.1.2 Ve nacula Bounda ies .................................................................. 5
2.1.3 Places .............................................................................................. 6
2.2 Da a Sou ces o Pe cei ed Bounda ies in Neighbo hood
Mapping .................................................................................................... 7
2.2.1 Online Web-Based Su ey ............................................................. 7
2.2.2 Use -Gene a ed Con en (UGC) .................................................... 8
2. 3 Me hods o Bounda ies Ex ac ion ........................................... 8
2.3.1 Pe cep ion-based neighbo hood delinea ion ................................. 8
2.3.2 Bounda ies Based on Online Ac i i y ........................................... 9
2.4 The S udy Si e ................................................................................... 9
2.4.1 Lisboa ............................................................................................. 9
2.4.2 Al ama .......................................................................................... 10
2.4.3 Mou a ia ...................................................................................... 11
2.4.4 Bai o Al o .................................................................................... 12
3. Me hodology .................................................................................... 13
3.1 Online Web-Based Su ey ............................................................ 14
3.1. 1 Da a collec ion & P ocessing ...................................................... 14
3.1. 2 Pe cei ed Bounda ies Ex ac ion .......................................... 19
ix
3.2 Use Gene a ed Con en ............................................................... 22
3.2.1 Da a collec ion ............................................................................. 22
3.2.2 Geo ags selec ion & Ex ac ion. .................................................. 24
3.2.3 Spa ial Clus e ing ........................................................................ 27
3.3 Compa a i e Analysis o Pe cei ed and Geo- agged
Bounda ies ............................................................................................. 29
4. Resul s & Discussion ..................................................................... 30
4.1 Pe cei ed Bounda ies Ag eemen Rep esen a ion ................ 30
4.1.1 Al ama .......................................................................................... 30
4.1.2 Mou a ia ...................................................................................... 31
4.1.3 Bai o Al o .................................................................................... 32
4.2 Geo- agged ac i i y de i ed Bounda ies. ................................. 34
4.2.1 Al ama .......................................................................................... 34
4.2.2 Mou a ia ...................................................................................... 35
4.2.3 Bai o Al o .................................................................................... 35
4.3 Compa a i e Analysis o pe cei ed and Geo- agged
Bounda ies ............................................................................................. 37
4.3.1 Al ama .......................................................................................... 37
4.3.2 Mou a ia ...................................................................................... 41
4.3.3 Bai o Al o .................................................................................... 45
5. Final Discussion ............................................................................. 49
5.1 Answe ing Resea ch Ques ions ............................................... 49
5.2 Limi a ions ...................................................................................... 51
5.3 Fu u e Scope ................................................................................... 52
6. Conclusions ..................................................................................... 53
7. Bibliog aphic Re e ences ............................................................. 55
Appendix ................................................................................................... 58
4
Chap e 2
2.Li e a u e Re iew
2.1 S udying Neighbo hoods, Ve nacula bounda ies, and Place.
2.1.1 Neighbo hoods
Neighbo hoods, wi hin a sociological con ex , a e acknowledged as
esiden ial zones cha ac e ized by signi ican social and cul u al
composi ions. They a e places whe e esiden s sha e simila
cha ac e is ics, d i en by a collec i e s uggle o u ban space and
esou ces. This concep ualiza ion o neighbo hoods ep esen s one o he
ini ial a emp s o de ine hese a eas. I was p oposed by u ban sociologis s
Pa k and Bu gess in 1925. Thei seminal wo k, “The Ci y,” laid he
ounda ion o ea ly 20 h-cen u y u ban planning s udies, ma king a
signi ican momen in his ield, (Pa k e al.,1925).
In he ollowing yea s, con ibu ions like hose om u ban planne Ke in
Lynch would highligh i e key elemen s de ining he ci y's image, d awn
om esea ch on men al mapping and collec i e pe cep ions. Among hese
elemen s, "dis ic s" and "edges" a e closely ied o he concep o
neighbo hood s uc u e. Edges ep esen ac ual o pe cei ed bounda ies,
delinea ed as linea ea u es dema ca ing he pe iphe ies o a eas known
as dis ic s (Lynch, 1964). These dis ic s a y in size om medium o la ge
and showcase a di e si y o cha ac e is ics, including ypes o buildings,
esiden demog aphics, opog aphy, ac i i ies, his o ical signi icance, and
le els o upkeep.
Cu en ly, ele ance in he depic ion o hese esiden ial a eas lies in he
ac ha neighbo hoods boos social li e, access o public se ices, and
su eillance, connec ing and exchanging esou ces wi h o he
neighbo hoods (Bae & Mon ello, 2018). On he o he hand, he
neighbo hood con o ma ion is ied o he esiden s' in e ac ions, which a e
essen ial o main aining hei ib a ion and di e si y h ough wo k,
housing, and ec ea ional ac i i ies. Acco ding o Coul on s udy,
neighbo hoods a e a collec i e and geog aphical cons uc ion, al hough
hese dimensions a e oo ed in social and psychological condi ions, hey a e
bound o geog aphic space and as a esul , ob aining meaning ul spa ial

5
ep esen a ions o neighbo hoods is necessa y o he enac men o local
policies (Coul on e al., 2013).
2.1.2 Ve nacula Bounda ies
Due o he exis ence o unclea bo de lines, e nacula egions a e
dis inguished by s ong cul u al linkages ha p oduce spa ial expansions
ha a e no sha p, such as ci y cen e o his o ic neighbo hoods. Some
w i e s, such as W. Zelinsky (Zelinsky, 1980), in es iga ed how people
pe cei e space and how place awa eness o egional consciousness c ea es
an a achmen sense based on popula cul u e. One o his mos no able
p ojec s was he iden i ica ion o cul u al egions in No h Ame ica by
analyzing he equency o local place names and businesses.
Schola s ha e ound p ac ical implica ions in he analysis o e nacula
egions because o icial adminis a i e uni s o census blocks do no
necessa ily ep esen neighbo hood o egion ex ension as spa ial and
social en i ies (Hollens ein & Pu es, 2010). Consequen ly, adminis a i e
a e ac s a e o e lain in a a ie y o ac o s associa ed o collec i e o
indi idual people beha io s be ween social, economic, and en i onmen al
ea u es such as housing sys ems, land use, and accessibili y and whose
esiden ’s pe cep ion s a e emains in cons an change hus e nacula
s udies going ocus on he ac o s' ex ension, con ibu ing o policymaking.
Fo E ans (E ans & Wa e s, 2007), we li e daily in e nacula egions ha
we accen ua e wi h geog aphical e ms ha a e no o mally ep esen ed.
Fo example, "High c ime a eas". The e ms ac as a ool o unde s and he
ne wo ks o sociolinguis ic communi ies wi h sha ed unde s andings;
while o Coul on, he e nacula egions scale is ano he ea u e o he
s udy whose dwelle s pe cep ion can be smalle o bigge acco ding o hei
own expe iences (Coul on e al., 2013).
Inco po a ing e nacula egions in o esea ch in oduces signi ican
challenges due o he subjec i e na u e o da a de i ed om indi idual and
communi y pe cep ions, leading o a ied de ini ions and in e p e a ions.
These challenges include di icul ies in da a collec ion, whe e he di e si y
in pe sonal desc ip ions o su oundings can esul in inconsis encies when
mapped agains o icial geog aphic bounda ies, Addi ionally, he eliabili y
o subjec i e pe cep ions aises ques ions, as pe sonal biases and
expe iences may hea ily in luence he delinea ion o e nacula a eas,
(Deng, 2016). To add ess hese issues, inno a i e me hodologies a e
equi ed o ha monizing e nacula bounda ies wi h o icial geog aphic
da a, necessi a ing he de elopmen o in o ma ion sys ems ha p io i ize
lay use pe spec i es o e adi ional adminis a i e geog aphy. Such
sys ems employ echniques like c owdsou ced mapping and pa icipa o y
6
GIS o in eg a e e nacula a eas in o s anda d da ase s, b idging he gap
be ween subjec i e pe cep ions and objec i e da a.
2.1.3 Places
The con ibu ion in he de ini ion o "Place" is wide due o ields such as
geog aphy, psychology, and an h opology among o he s being in ol ed.
The unique iden i y and signi icance seem o be ema kable ea u es o
place de ini ion, acco ding o he geog aphe Edwa d Relph in his book
"Place and Placelessness" he claims his esea ch me hod is "a
phenomenology o place" ha elies on he in e p e a i e s udy o human
expe ience. Fo him, a place is a geog aphical a ea wi h a dis inc iden i y
and signi icance o people. Human expe iences, memo ies, and cul u al
impo ance a e all p esen in places. They os e a eeling o iden i ica ion
and a achmen o people and communi ies.
Belonging a place shapes people's pe sonali ies c ea ing deep connec ions
wi h he sense o sel , places in luence how communi ies o m, how people
ela e o each o he , and how social no ms a e es ablished; on he o he
hand, cul u al and his o ical meaning a e con en in hese places
displaying s o ies, cul u al he i age, social memo ies and iding wi h
en i onmen al ela ionships.
As an example o he a o emen ioned. Two au ho s He nandez's esea ch
on place iden i y and Anac a e al.' s udy on spa ial ep esen a ion h ough
ou e ins uc ions con ibu e o ou unde s anding o how indi iduals
connec wi h and de ine hei neighbo hoods. He nandez's wo k del es in o
he emo ional bonds people o m wi h hei en i onmen , measu ing he
in ensi y o place a achmen and place iden i y h ough
ques ionnai es,(He nández e al., 2007). These ques ionnai es assess
a ious en i onmen al scales and neighbo hood, ci y, and island o
measu e he ' ype o bond' ha indi iduals ha e wi h each. This app oach
unde sco es he signi icance o place in shaping pe sonal iden i y and
social dynamics. On he o he hand, Anac a app oach he concep o
neighbo hood om a spa ial pe spec i e, explo ing how people cogni i ely
map a eas wi h indis inc bounda ies when gi ing ou e di ec ions,
(Anac a e al., 2017). They ocus on 'Neighbo hood' as a ca ego y wi hin
ske ch mapping, highligh ing he social use o sha ed acili ies and he
homogenei y o esiden ial o s uc u al cha ac e is ics as de ining
elemen s. By combining he indings om bo h s udies, we gain a
comp ehensi e iew o how place iden i y and spa ial cogni ion in e play
in neighbo hoods' concep ualiza ion and li ed expe ience, enhancing ou
insigh s in o u ban social geog aphy and he psychology o space.
7
2.2 Da a Sou ces o Pe cei ed Bounda ies in Neighbo hood
Mapping
In he ield o neighbo hood mapping, i is essen ial o add ess he concep
o cogni i e egions in o mal a eas cha ac e ized by unclea bounda ies,
membe ship s a us, ex ension, loca ion, and shape, which a e in luenced
by he dis ibu ion o communi ies o he pe cep ions o esiden s. These
egions' spa ial p ope ies a e also shaped by cul u al ac o s, adding o
hei complexi y. A comp ehensi e unde s anding o socio-demog aphics
and en i onmen al cha ac e is ics is i al o achie ing accu a e spa ial
delinea ion. This p ocess in ol es p ecisely de ining he bounda ies and
ea u es o hese a eas, conside ing hei in o mal na u e and he impac
o cul u al and social elemen s.
Cogni i e egions p esen signi ican challenges in geog aphic in o ma ion
science, mainly due o hei agueness, he di icul y in ep esen ing hem
accu a ely in GIS ools, hei a ying scales, and hei lack o homogenei y
o consis ency. To cap u e he di e se pe cep ions and expe iences o
esiden s, se e al da a sou ces a e used, including su eys and in e iews
ha e i y he in e iewees' de ined cha ac e iza ion, pa icipa o y
mapping exe cises, analysis o social media, and examina ion o his o ical
eco ds. Addi ionally, p eexis ing adminis a i e (no ably census) da a and
o he agency-de ined zones se e as p oxies o neighbo hood delinea ions.
This p ojec employs wo o hese da a sou ces, which will be explained in
de ail below.
2.2.1 Online Web-Based Su ey
Su eys ha e been a adi ional me hod o comp ehending cogni i e
bounda ies. They in ol e di ec ly que ying local inhabi an s abou hei
pe cep ions o neighbo hood limi s. In ad ancing he mapping o di use
a eas, su eys ha e been u ilized no only in delinea ing neighbo hood
a eas bu also in add essing issues a ec ing neighbo hoods. Fo ins ance,
he s udy by E ans & Wa e s employed a su ey o assess he high le els
o c ime in LA ci y, iden i ying a eas wi h a high isk o c ime. This
app oach allowed esea che s o ga he nuanced insigh s in o how
esiden s men ally map hei su oundings, (E ans & Wa e s, 2007). The
s udy by Wes e hol e al., used a web mapping-based su ey ha le
esponden s map and a e a eas in he ci y ela ed o poin s o in e es
ex ac ed om Google Places, showing how indi iduals' sense o place
in luences hei pe cep ion o neighbo hood bounda ies, (Wes e hol e al.,
2022).
8
2.2.2 Use -Gene a ed Con en (UGC)
UGC is ano he ich da a sou ce o s udying human beha io , sen imen ,
and pe cep ion o he u ban en i onmen . Social media analysis p o ides a
mode n, da a- ich sou ce o explo ing pe cei ed bounda ies. The s udy by
Hollens ein & Pu es demons a es how use -gene a ed con en on
pla o ms like Flick can o e insigh s in o how people pe cei e and
in e ac wi h u ban spaces. This analysis allows o he ex ac ion o
cogni i e bounda ies based on he agg ega ion o indi idual expe iences
and exp essions cap u ed h ough social media, (Hollens ein & Pu es,
2010).
Mo eo e , he wo k o Huang emphasizes he po en ial o combining "big
da a," ep esen ed by geo ags and images om Ins ag am and Twi e ,
wi h "small da a," such as ques ionnai es and ske ch maps. Thei s udy
unde lines he impo ance o analyzing social media con en o map he
cogni i e landscapes o ci ies, e ealing how esiden s and isi o s pe cei e
hem. Findings sugges ha social media analy ics can p o ide a eliable
measu e o pe cei ed ci y images, wi h pla o ms like Ins ag am
explaining public pe cep ion associa ed wi h ou is a ac ions and
landma ks, while Twi e is linked o he place’s ele ance o e e yday li e
enues o communi ies. These esul s ha e p ac ical implica ions,
especially in u ban planning, o e ing aluable in o ma ion o
policymake s and ci izens alike, (Huang e al., 2021).
2. 3 Me hods o Bounda ies Ex ac ion
2.3.1 Pe cep ion-based neighbo hood delinea ion
I s me hod in ol es ga he ing unique insigh s h ough men al mapping,
highligh ing he cogni i e mapping o neighbo hoods as desc ibed by
(Dal on & Hu ell, 2023). These mappings a e de ined as g aphic and
subjec i e ep esen a ions ha cap u e no jus he physical layou bu
also he emo ional and cul u al associa ions indi iduals hold wi h hei
communi ies. This app oach measu es cogni i e ep esen a ions by
sampling esiden s, collec ing socio-demog aphic ea u es, and e ie ing
men al ep esen a ions h ough su eys o in e iews. The collec ed da a
hen acili a es spa ial analysis abou co e a eas. The le el o ag eemen
among he men al maps se es as a measu e o membe ship, e ealing
collec i e pe cep ions o neighbo hood bounda ies.
9
2.3.2 Bounda ies Based on Online Ac i i y
(Deng, 2016) discussed he in e ac ions among esiden s and how hese
social p ocesses ela e o geog aphic pa e ns and neighbo hood o ma ion,
ocusing on physical ne wo ks. The me hod gains ele ance wi h he
inco po a ion o Use -Gene a ed Con en (UGC), as i cha ac e izes a
neighbo hood and e eals de ails o i s social composi ion and
en i onmen , ans o ming geog aphical da a in o neighbo hood pa e ns.
A case s udy in Spain u ilized da a om social media pla o ms like Google
Places, Fou squa e, Twi e , and Ins ag am. P e ious esea ch has shown
he complemen a i y o hese ou sou ces o in e ing aluable insigh s
ega ding he spa io empo al use o ci y spaces and people’s p e e ences,
aiming o le e age he a ailabili y o use -gene a ed da a sou ces
(Be nabeu-Bau is a e al., 2023). Addi ionally, s udies by (Gao e al.,
2017a) and (Tang e al., 2022) use use -gene a ed con en o de e mine he
delimi a ions o cogni i e egions. They pe o m clus e analysis using
algo i hms such as A-DBSCAN o iden i y poin clus e s based on he
collec ed da a, aiming o ind co e a eas ha ep esen a consensus among
a signi ican numbe o use s who ag ee ha hese a eas ep esen egional
o neighbo hood ex ensions.
2.4 The S udy Si e
2.4.1 Lisboa
Lisbon, he capi al o Po ugal, is a dynamic ci y wi h a ich his o y. I
s a ed as a Phoenician ade pos , was aken o e by he Romans and hen
he Muslims and became he depa u e po o he amous Po uguese
explo e s. I 's one o he oldes ci ies in Eu ope. Nowadays, i 's a popula
des ina ion o in e na ional s uden s, ou is s, and o eign esiden s
(Tang e al., 2021).
Each phase o Lisbon's de elopmen esul ed in a ious se lemen s, now
ecognized as he his o ic neighbo hoods in he ci y cen e . Fo his s udy,
h ee neighbo hoods, Al ama, Mou a ia, and Bai o Al o, ha e been
selec ed. These a eas, signi ican o hei his o ic impo ance and he need
o clea delinea ion, a e si ua ed wi hin di e se adminis a i e
bounda ies Figu e 1. Some o he main cha ac e is ics o each his o ic
neighbo hood a e depic ed in he nex sec ion.

10
Figu e 1. F eguesías Bounda ies and Pe cei ed His o ic Neighbou hoods loca ion map
2.4.2 Al ama
Named a e he A abic al-hamma. The hill whe e Al ama is loca ed, along
wi h he nea by cas le hill, we e among he i s se led a eas ha
con ibu ed o Lisbon's o igins. The pa o Al ama s e ches along he hill
be ween he Chu ch o São Miguel and San o Es ê ão, wi h Rua da
Reguei a a i s co e. The neighbo hood ex ends up o whe e he Escolas
Ge ais used o be, a esidence o uni e si y s uden s un il he 16 h
cen u y.
A he hill's lowe end lies he sho e o he Tejo Ri e , once a cen e o
i e comme ce and a ishing neighbo hood du ing he 13 h and 14 h
cen u ies.
Al ama has always been a ou is a ac ion due o being he oldes pa o
he ci y, Moo ish- oo ed a chi ec u e ha p ese es he ancien medie al
A ab oad mall wi h alleyways, na ow s ee s and s ai s ha make i
amous as well as i s p oximi y o he São Jo ge cas le and i s connec ion
wi h Fado, he adi ional Po uguese olk music. This neighbo hood has
an impo an social isibili y ha is widely ecognized in aspec s such as
li e a u e, pain ing, and music in he ci y o Lisbon, becoming a his o ical
he i age spo (Da Cos a, 2008).
In pas cen u ies, Al ama was home o bo h a is oc a s and wo king-class
indi iduals. The di e se backg ounds o i s esiden s os e ed a s ong
sense o communi y, oo ed in long- e m enancy and he social connec ions
11
o med among neighbo s (Da Cos a, 2008). Howe e , ecen challenges
such as o e - ou ism, he ise in lodging phenomena, and damage o
p ope ies indica e ha Al ama is unde going signi ican ans o ma ions,
g adually mo phing in o a p edominan ly ou is dis ic
(Cocola-Gan & Gago, 2021).
A p esen , Al ama is enowned as Lisbon's mos pic u esque
neighbo hood, o en seen as a jou ney back in ime. I 's a popula spo -on
Ins ag am, equen ly showcased by ou is s, ERASMUS s uden s, and
empo a y esiden s, (Bap is a e al., 2018).
2.4.3 Mou a ia
Mou a ia neighbo hood is loca ed be ween he alley o he Plaça Ma in
Moniz and he hills o he cas le and G aça neighbo hood ha was he
a ea designa ed o he de ea ed Moo s a e he occupa ion o he ci y by
D. A onso Hen iques as pa o he Ch is ian conques o e he Muslims in
1147. Due o i s b oken opog aphy he neighbo hood de eloped a layou
i egula , he ini ial occupa ion su ounded he cas le hill going down he
slope o he wa e alley ha exis ed in he pas and is cu en ly Almi an e
Reis A enue. Mou a ia is no able o i s medie al layou , a e lec ion o
A ab a chi ec u al in luences, cha ac e ized by small s ee s, alleys, and
na ow s ai cases. Addi ionally, he neighbo hood's his o ical seg ega ion,
as i was no included wi hin he Fe nandina’s wall in he 14 h cen u y,
mean ha i did no appeal o he weal hie classes o ha pe iod, (Da
Cos a, 2008).
Mou a ia's his o ical ac i i ies we e cen e ed a ound i s la ge Mosque,
which was loca ed whe e he Chu ch o Nossa Senho a do Soco o is
cu en ly loca ed. Addi ionally, in he a ea ha is now Ma im Moniz
Squa e, he e was once a smalle mosque along wi h public ba hs. This
a ea held signi ican comme cial impo ance, as Moo ish c a sman se ed
a la ge Ch is ian cus ome . Howe e , in 1496, he expulsion o Jewish and
Muslim mino i ies om Po ugal led o he ans o ma ion o Mosques and
Islamic s uc u es in o chu ches and con en s. These eligious buildings
we e p ima ily u ilized by he nobili y and cle gy, who played a pi o al ole
in inco po a ing Mou a ia in o he b oade ci yscape, (Langens, 2022).
Ea ly in he 19 h cen u y, he neighbo hood ell in o po e y and c ime,
wi h ba s and p os i u ion abounding. Howe e , his agic pe iod was he
pe ec se ing o he appea ance o he adi ional Po uguese musical
gen e called Fado, wi h which he adis a Ma ía Se e a ga e i be e
cul u al s a e o he “Mo o” neighbo hood. Un o una ely, in he 20 h
cen u y and in he name o he mode niza ion o he a ea, he s uc u e o
he neighbo hood was al e ed, collapsing se e al o he ep esen a i e
monumen s o he 17 h cen u y, (F anco, 2016). E en ually, as pa o
Po uguese decoloniza ion p ocesses and cu en immig a ion phenomena,
12
Mou a ia is conside ed he mul icul u al encla e o he ci y cen e , being
a comme cial, gas onomic, and social cen e ha allows mul ie hnic
ela ions, (Oli ei a, 2019). I is impo an o no e, in ecen yea s, he
municipal policies ha e es ablished s a egies aimed a he egene a ion o
Mou a ia. These s a egies in ol e ans o ming i s s igma ized image in o
ha o a ehabili a ed space, bo h in e ms o in as uc u e and by
p o iding access o cul u al ac i i ies. These ini ia i es a e designed o
enhance he neighbo hood's iden i y, adi ion, and di e si y (Es e ens e
al., 2019).
2.4.4 Bai o Al o
The Bai o Al o h o e be ween he end o he 15 h and 16 h cen u ies as
a esponse o he e oca ion o he manda e ha limi ed he ci y o de elop
only wi hin he medie al Fe nandina wall, which sough o make he
cen al a ea o he ci y and he po he social and comme cial axes. The
modi ica ion allowed his neighbo hood o begin he p ocess o Lisbon as a
mode n ci y whe e he popula classes would begin hei loca ion
ex e nally, mos ly on he i e bank in ish ades, which changed o e ime
on he no he nmos pa . Palaces, con en s, and chu ches led by
me chan s and bou geois amilies o ha ime we e loca ed on his hill, (Da
Cos a, 2008).
Wi h i s e olu ion, Bai o Al o was cha ac e ized by being a neighbo hood
wi h g ea e en ila ion han he ci y cen e , especially in imes o he
plague, and due o i s opog aphic condi ions, he ainwa e lowed di ec ly
in o he Tejo Ri e due o i s long and egula s ee s, which gua an eed
be e heal h condi ions, howe e . wi h na ow s ee s ha a he ime
educed he condi ions o cha ac e is ic mobili y ha e en oday pe sis s
as a neighbo hood wi h a la ge comme cial concen a ion, highligh ing
mee ing a eas and social ec ea ion o he ci y's esiden s, (No e, 2020).
In con empo a y imes, Bai o Al o has ea ned a epu a ion as a bohemian
encla e, a ibu ed o he many jou nalis s and li e a y au ho s who ha e
esided he e. Howe e , s a ing in he 1990s, he neighbo hood began o
de elop a ib an nigh li e, which has since become he ci y's p emie
nigh li e spo . This e olu ion is la gely due o he low o s uden s d awn
by a o dable en s and he con enience o local ameni ies (Bap is a e al.,
2018).
13
Chap e 3
3. Me hodology
The me hodology o his s udy cen e s on explo ing he spa ial de ini ions
o h ee his o ic neighbo hoods in Lisbon – Al ama, Mou a ia, and Bai o
Al o. This explo a ion is achie ed h ough he analysis o da a ob ained
om cogni i e mapping, ex ac ed ia an online web-based su ey, and
geo agged online ac i i y om use -gene a ed con en (UGC). Bo h
da ase s a e in eg al componen s o he Ci yMe p ojec , an ini ia i e
unded by Po ugal's Fundação pa a a Ciência e a Tecnologia (FCT). The
Ci yMe p ojec is designed o deepen ou unde s anding o how ci izens
men ally map he ci y, he eby enhancing spa ial analysis and communi y
pa icipa ion in public policies and u ban planning.
A key con ibu ion o he Ci yMe p ojec is he de elopmen o a ully
unc ional public websi e (h ps://ci yme.no aims.unl.p /) ha acili a es
pa icipan da a collec ion h ough a web-based map su ey and p o ides
access o sou ces o geo agged con en om UGC. The wo k low o he
s udy is di ided in o wo main lines.
The i s phase ocuses on ex ac ing ep esen a i e bounda ies o he
h ee his o ic neighbo hoods. This p ocess in ol es collec ing pa icipan
ske ches h ough he web-based map su ey. This de elopmen al line
de ails da a collec ion and p ocessing me hods, ex ac ion echniques o
pe cei ed bounda ies, and culmina es in he ep esen a ion o he
ag eemen on pe cei ed bounda ies o Al ama, Mou a ia, and Bai o Al o.
The second phase o he wo k low add esses he e ie al o ep esen a i e
bounda ies om a ious sou ces o geo- agged online ac i i y. This pa o
he s udy ou lines he speci ics o da a collec ion, con en selec ion c i e ia,
coding p ocesses, clus e ing ex ac ion, and ul ima ely, he gene a ion o
bounding shapes o he his o ic neighbo hoods.
The inal phase o he s udy is whe e hese wo phases con e ge, enabling
a quan i a i e compa ison o he spa ial de ini ions de i ed om bo h
cogni i e mapping and geo agged online ac i i y. This concluding phase
in ol es calcula ing me ics o unde s and he ela ionship be ween he
neighbo hood ex ensions gene a ed by he wo me hodologies.
The igu e 2 illus a es he s uc u e o he s udy me hodology.
20
Figu e 9. O e lap ag eemen maps by polygon coun o sho - and long- e m esiden sin Al ama, Mou a ia, and Bai o Al o his o ic
neighbo hoods!

21
This s udy employed he Minimum-Maximum s anda diza ion me hod o
make uni o m he coun ield o o e lapping en i ies, es ablishing alues
anging om 0 o 100 (Pe cen age). Consensus h esholds we e chosen,
suppo ed by he li e a u e, which shows pe cen ages ha embody he
esiden 's delinea ion bounda ies and hei iden i y e ec s using ske ch
maps (Bae & Mon ello, 2018).Con o ming o he s udy, he a eas o g ea es
ag eemen we e >75 pe cen which we will call he Co e egion, and a eas
indica ing a leas 50 pe cen bu no mo e han 75 pe cen ag eemen , i
will call Domain egion.
The ca ego iza ion in o "Co e" and "Domain" egions was chosen based on
Meinig's con ibu ion, who he ex en o he egion occupied by Mo mon
cul u e o e ime in concen ic ci cles, ocusing on he s a e o U ah and
sou heas e n Idaho in he Uni ed S a es o Ame ica. His model, as
p esen ed in 1965, is adop ed and adap ed o he neighbo hood le el in his
s udy, (Meinig,1965). The in eg a ion o hese ca ego ies wi h consensus
h esholds is depic ed in Figu e 10.
Figu e 10. Neighbo hood Regions Scheme
A his s age, he ou come yields he s udy's g ound u h, which becomes
an in aluable p ima y sou ce encapsula ing uni ied bounda ies and
esiden s' de ailed pe cep ions o neighbo hoods. This g ound u h se es
as a benchma k o e alua ing he accu acy o maps p oduced om use -
gene a ed con en , and i plays a i al ole in disce ning he di e ences
and simila i ies be ween local pe cep ions and digi al depic ions de eloped
in he nex s age o he s udy.
22
3.2 Use Gene a ed Con en
3.2.1 Da a collec ion
The s udy inco po a es a second da ase de i ed om use -gene a ed
con en . This da ase is composed by geo agged pos s, pages,
ad e isemen s, poin s o in e es (POIs), and hei espec i e a ibu es,
including ex ual da a, all o which we e ga he ed om a a ie y o social
media pla o ms, i ’s impo an o no e ha he da a ex ac ed om hese
sou ces came om public p o iles o use s. The ex ac ion o his da a was
made possible using bo h o icial and uno icial APIs, and i was
unde aken as an in eg al pa o he Ci yMe p ojec .
The Ci yMe p ojec o e s a unique and comp ehensi e pe spec i e on he
spa ial dynamics o Lisbon by combining su eyed da a (Using in he i s
s age o his esea ch) wi h use -gene a ed con en om social media
pla o ms. This app oach p o ides a mo e holis ic unde s anding o he
a ea; in consequence, A key elemen o he Ci yMe p ojec UGC sec ion is
he Ci y Me-UGC online eposi o y, accessible a
h ps://gi hub.com/Ci yMe-p ojec /Ci yMe-UGC, enabling o pe o m
spa ial analyses ha compa e and cha ac e ize egions based on bo h
su eyed da a and he use -gene a ed con en .
In his segmen o he s udy, he ocus is on he me hodologies used o
ex ac ing ep esen a i e bounda ies om use -gene a ed con en . Da a
was ga he ed om h ee social ne wo ks Flick , Twi e , and Ins ag am
speci ically a ge ing men ions o he names o his o ic neighbo hoods
ele an o his s udy wi hin he ex ual con en . The ex ac ion p ocess
was ca ied ou h ough Py hon coding. This app oach allowed o a
comp ehensi e analysis o use in e ac ions and e e ences o hese
his o ic neighbo hoods.
The ollowing con en p o ides a b ie o e iew o he s uc u e o he h ee
sou ces and desc ibes he unique cha ac e is ics and con ibu ions o each
pla o m o he esea ch.
Flick
Flick ope a es as a digi al pla o m o pho o managemen , enowned o
i s ex ensi e collec ion o geo agged images, as no ed by (Gao e al., 2017a).
On Flick , use s ha e he capabili y o upload and dissemina e hei
pho og aphs, ei he wi hin speci ic Flick g oups o o a b oade audience.
These images a e o en accompanied by use -gene a ed ags, p o iding
desc ip i e con ex , and a e geo agged wi h p ecise coo dina es. This
23
ea u e is pa icula ly use ul o esea ch, as Flick is p edominan ly
u ilized by ou is s, which aligns wi h Lisbon's s a us as a sough -a e
a el des ina ion. The geo e e encing o pho og aphs on Flick can be
accomplished ei he au oma ically o manually, le e aging he loca ion-
based echnologies embedded in sma phones and came as.
Twi e
Twi e is ecognized as a p ominen social ne wo k pla o m, p ima ily
u ilized o sha ing opinions and news epo ing. Acco ding o (Li e al.,
2013), use -gene a ed da a om Twi e o e s aluable insigh s in o he
cha ac e is ics o a ious loca ions and he demog aphics o people who
equen hese places, whe he hey a e esiden s, wo ke s, o ou is s.
This aspec o Twi e makes i a signi ican sou ce o unde s anding
social dynamics and public pe cep ion.
Fu he mo e, as poin ed ou by (Sal a o e e al., 2021), Twi e da a can
se e as a supplemen a y sou ce o in o ma ion, complemen ing s a is ical
da a ob ained om o icial sou ces. The b ie , ye exp essi e messages on
Twi e o en e lec people's ins an eac ions and pe cep ions abou
a ious aspec s o hei en i onmen , such as neighbo hood sa e y and
c ime le els. These wee s no only p o ide indi idual pe spec i es bu can
also igge g oup eac ions, o e ing a eal- ime glimpse in o public
sen imen .
Ins ag am
The sha ed ex s on Ins ag am o e se e al bene i s, as ou lined by
(Be nabeu-Bau is a e al., 2023). These include insigh s in o he empo al
and spa ial ends o people a speci ic loca ions, he spon aneous eac ions
and pe cep ions indi iduals ha e o hei su oundings now o pos ing, and
he iden i ica ion o pa e ns ela ed o ac i i ies in u ban spaces. Such
pa e ns e eal he di e si y o p e e ences, mobili y, and iden i y o ci ies
a a ious poin s o in e es . Ini ially, he selec ion o geo e e enced
con en on Ins ag am was done manually. Howe e , his p ocess la e
e ol ed o u ilize in eg a ed posi ioning echnology, enhancing he
accu acy and e iciency o geo e e encing. This echnological shi has
allowed o a mo e p ecise and comp ehensi e unde s anding o how people
in e ac wi h and pe cei e hei u ban en i onmen s.
24
3.2.2 Geo ags selec ion & Ex ac ion.
In his esea ch, we ocused on s anda dizing he da a a ibu es ga he ed
om in e ac ions on Flick , Twi e , and Ins ag am. The key a iables
s anda dized we e he men ion o neighbo hood names wi hin he ex ual
con en and hei geog aphical loca ions. This da a was compiled o e a
decade, om 2012 o 2022, a pe iod du ing which con en ela ed o he ci y
o Lisbon was a ailable ac oss all hese pla o ms. The un il e ed da ase s
comp ised: 5,669,531 eco ds o Twi e , 50,946 o Flick , and 35,657 o
Ins ag am. The s anda diza ion p ocess was essen ial o ensu ing
consis ency and accu acy in he analysis o he da a collec ed om hese
di e se social media sou ces.
To ob ain he ele an neighbo hood desc ip ion o geo agged da a om
social pla o ms, names o neighbo hoods as keywo ds we e ex ac ed om
he ex ual a ibu es o he geo agged da a such as he "s a us",
"desc ip ion" and " ags" by il e ing ou exac men ions o he keywo ds and
associa ed ex ac ed neighbo hood keywo ds (Al ama, Mou a ia and
Bai o Al o) wi h he poin coo dina es. In case o mul iple neighbo hood
names we e men ioned in he same da a en y, Euclidean dis ances
be ween he da a poin and cen oids o men ioned neighbo hoods (Using
g ound u h al eady gene a ed as e e ence) we e calcula ed and he
nea es neighbo hood o he poin was assigned as he neighbo hood
a ibu e o he poin .
Coo dina es o poin s om ce ain pla o ms such as Ins ag am and
Twi e a e no o ganic, whe e geo agged coo dina es o he da a a e no
e ie ed om he use 's ac ual posi ioning sys em bu a he assigned
wi h a p e-popula ed coo dina e associa ed wi h he name o he place he
use is checking in. This na u e assigns edundan and o e lapping
coo dina e poin s o he geo agged da a ega dless o he use 's ac ual
loca ion. This becomes p oblema ic when calcula ing he (minimum)
dis ance band be ween neighbo ing poin s in iden i ying and c ea ing
clus e s since a signi ican amoun o poin s ha e iden ical coo dina es
which a ec s he s a is ical calcula ions esul ing nea sub-me e
dis ances o he dis ance band. These iden ical/ edundan /o e lapping
poin s we e emo ed ill only one unique poin pe coo dina e is le .
To add ess he imbalance in geo agged da a om sou ces like Twi e ,
which signi ican ly exceeds ha om o he sou ces, ou s udy implemen s
a me hod o calib a e he da a olume. Following he app oach by (Gao e
al., 2017a), which iden i ied a simila disc epancy, we i s es ablish he
(a e age/maximum) con ibu ion le el o no mal o o dina y use s. This
le el is de e mined by calcula ing he da a o pos s p oduced by he op
10% o ac i e use s, ep esen ing he 90 h pe cen ile o he use base. A
h eshold is hen se a his le el, and any use con ibu ions exceeding i
25
a e scaled down acco dingly. U ilizing his adjus ed h eshold ensu es ha
each use 's da a con ibu ion is p opo iona e and consis en wi hin he
da ase , which is hen used o u he analysis.
As a esul , we ob ained he il e ed da ase s wi h he h ee names o
his o ical neighbo hoods om e e y sou ce. Impo an o men ion ha he
Flick egis e s calling o Mou a ia did no yield esul s. Looking o an
explana ion, i was ound ha Mou a ia was men ioned, bu he loca ion
o he pos was caugh by Al ama cen oid in luence a ea. Figu e 11
illus a es he quan i ies de i ed a e pe o ming he da a mining p ocess.
Figu e 12 displays he ex ac ed da a and a ep esen a ion o he pos ings
used as he ex ac ion sou ce.
Figu e 11.Numbe o pos s ex ac ed based on he name o each neighbou hood om
e e y UGC sou ce.

26
Figu e 12.Dis ibu ion and depic ion o selec ed pos s om h ee di e en UGC sou ces ac oss he his o ic neighbo hoods o
Lisbon.
27
3.2.3 Spa ial Clus e ing
By ex ac ing a da ase o geo- e e enced social media pos ings om h ee
di e en sou ces, each ca ego ized by neighbo hood, he esea ch
concen a ed on employing a coding p ocess o iden i ying clus e s o poin s.
This goal was accomplished using he App oxima e DBSCAN (A-DBSCAN)
clus e ing algo i hm, de eloped by (A ibas-Bel e al., 2021)
.The s udy aimed o gain insigh s in o u ban a eas by analyzing he densi y
o poin s, pa icula ly hose associa ed wi h buildings wi hin hese a eas. Fo
he spa ial clus e ing in his esea ch, p e-ex ac ed geo ags we e u ilized,
acili a ing he iden i ica ion o high-densi y clus e s om each sou ce, and
de ining hei bounda ies.
A-DBSCAN algo i hm is use ul o he calcula ion and gene a ion o shapes
ha ep esen he spa ial ex en o neighbo hoods, illus a ing he clus e ing
pa e ns o social media geo ags, as highligh ed by (Tang e al., 2022). The
applica ion o his algo i hm equi es wo inpu pa ame e s. The i s one,
known as he eps pa ame e , ep esen s he maximum dis ance be ween wo
poin s necessa y o hem o be conside ed pa o he same neighbo hood o
clus e . Fi s ly, a nea es neighbo analysis was pe o med on he da ase .
This analysis in ol es calcula ing he dis ance om each poin in he da ase
o i s closes neighbo . The pu pose o his s ep is o unde s and he ypical
dis ances be ween poin s in he da ase , which is c ucial o de e mining he
app op ia e scale o clus e ing.
A e comple ing he nea es neighbo analysis, he nex s ep was o calcula e
a speci ic pe cen ile o hese dis ances as in (Tang & Painho, 2023, Tang e
al., 2022) s udies. The calcula ed pe cen ile alue is hen used as he ‘eps’
pa ame e in he clus e ing algo i hm. The choice o which pe cen ile o use
can depend on he da ase and he desi ed sensi i i y o he clus e ing
algo i hm o ou lie s o noise. Tha was he case o he cu en s udy we e
he pe cen ile alues oscilla e be ween he 90 h o he 99 h pe cen ile pe social
pla o ms.
The second pa ame e is called ’Minimum Poin s’ and e e s o he
minimum numbe o neighbo ing poin s equi ed o a poin o be conside ed
pa o a clus e . Gi en ha each da a sou ce exhibi s a ia ions in he
numbe o en ies, we op ed o use pe cen ages o choose Minimum Poin s
as in (Gao e al., 2017) esea ch. In his s udy, he pe cen ages chosen we e
3% and 5% o he o al numbe o pos ings pe da a sou ce, o model he ague
na u e o cogni i e egions.
28
Finally, he A-DBSCAN algo i hm was un i e a i ely. Table 2 shows he
combina ion o pa ame e s by sou ce and neighbo hood o gene a e a s able
and ep esen a i e delinea ion based on he dis ibu ion o geo ags. The
cons uc ion o ha polygons was hen unde aken o app oxima e he
bounding shape o he his o ic neighbo hoods, using he α-shape algo i hm
due o i s abili y o p o ide a smoo hed delinea ion,(Edelsb unne e al.,
1983). I is impo an o no e ha in addi ion o calcula ing polygons o each
pla o m i was c ea ed a shape de i ed om all sou ces o each o he
his o ic neighbo hoods unde s udy.
The p ojec aims o e eal pa e ns o consis ency and a iance in he
iden i ica ion o neighbo hood bounda ies by compa ing his p ima y da a
wi h he insigh s de i ed om use -gene a ed con en analysis, as discussed
in his phase. These indings will be u he explo ed in he subsequen phase.
Table 2. A-DBSCAN pa ame e s alues by sou ce and neighbo hood.
NBHD/Pa ame e s eps (%)
mins(%)
eps (%) mins(%) eps (%) mins(%) eps (%) mins(%)
Al ama 0.99 0.03 0.99 0.03 0.99 0.03 0.99 0.05
Mou a ia 0.95 0.03 0.90 0.05 0.95 0.05
Bai o_Al o 0.99 0.03 0.90 0.03 0.90 0.03 0.90 0.05
Flick
Twi e
Ins ag am
Combined
29
3.3 Compa a i e Analysis o Pe cei ed and Geo- agged Bounda ies
The inal s age o he s udy in eg a es he ini ial segmen s, unde aking a
quan i a i e compa ison o spa ial de ini ions de i ed om cogni i e mapping
and geo agged online ac i i ies wi hin he his o ic neighbo hoods o Al ama,
Mou a ia, and Bai o Al o. This compa ison aims o measu e he ag eemen
be ween neighbo hood ex ensions as iden i ied by each me hod. Cen al o
his analysis a e wo me ics: he In e sec ion O e Union (IOU) and he F-
sco es.
The IOU me ic quan i ies he o e lap be ween wo shapes, wi h alues
anging om 0 o 1 whe e 0 ep esen s no o e lap and 1 indica es pe ec
o e lap. Simila ly, F-sco es assess he accu acy and ele ance o his o e lap
by calcula ing p ecision and ecall, p o iding a balanced measu e o a
me hodology’s e ec i eness in accu a ely cap u ing and comp ehensi ely
co e ing neighbo hood bounda ies as was calcula ed in (Tang e al., 2022). F-
sco es also ange om 0 o 1, whe e 0 signi ies he wo s p ecision and ecall,
and 1 he bes .
Calcula ions o bo h IOU and F-sco es a e g ounded in he “g ound u h”
es ablished by pe cei ed bounda ies ob ained in he i s s age o he s udy
and he polygons gene a ed om use ac i i ies. Py hon was u ilized o hese
compu a ions, acili a ing he coding and analysis o he esul s. This
app oach p o ides a quan i a i e basis o discussing he spa ial de ini ion o
his o ic neighbo hoods, allowing o a compa ison o he insigh s gained om
bo h cogni i e mapping and geo- agged da a.
36
neighbo hood. Bo h Twi e and Ins ag am da a sugges ha use ac i i y in
Bai o Al o is cen e ed a ound pa icula ho spo s o enues. This is
suppo ed by he ac ha popula nigh li e a eas, such as Rua da Rosa, Rua
da A alaia, Rua São Ped o de Alcan a a, and adi ional es au an s such as
Tasca do Chico, a e loca ed wi hin hese ou lines, as con i med by (No e,
2020) .(The comple e loca ion o Bai o Al o's poin s o in e es can be ound
in he Appendix).
Figu e 18. Geo- agged ac i i y de i ed bounda ies o Bai o Al o.

37
4.3 Compa a i e Analysis o pe cei ed and Geo- agged Bounda ies
The hi d phase encompasses a compa a i e analysis. This sec ion ou lines
he compu a ion o me ics, including he In e sec ion O e Union (IOU),
which measu es he ex en o o e lap be ween he iden i ied shapes, as well
as he e alua ion o o e lapping a eas and F-sco es o each da ase . U ilizing
he pe cei ed bounda ies de ined as he benchma k in he s udy's ini ial
phase, hese a e compa ed agains he use -gene a ed shapes o de e mine
he IOU and F-sco es. The maps and me ics a e accompanied by
obse a ions, highligh ing he mos signi ican indings by his o ic
neighbo hood acco ding o esiden s’ g oups and egions o consensus.
4.3.1 Al ama
The compa a i e analysis o social media da a in ela ion o he pe cei ed
bounda ies o Al ama as de ined by sho - e m and long- e m esiden s
e eals dis inc pa e ns in bo h he domain and co e egions. The esul s a e
display in he igu es 19 and 20 and ables 6 and 7.
Fo sho - e m esiden s
In he domain egion , Flick s ands ou wi h he highes In e sec ion o e
Union (IOU) sco e o 0.538 among all sou ces, indica ing a mode a e o e lap
wi h esiden s' domain bounda ies, see Table 5. Meanwhile, Twi e shows
he highes ecall alue o 0.755, cap u ing a signi ican po ion o he domain
egion bu also including a eas beyond i . When conside ing he combined
da a sou ces, he p ecision sco e peaks a 0.773, sugges ing ha he
agg ega e o all da a mos accu a ely aligns wi h he domain a eas ecognized
by esiden s.
Mo ing o he co e egion, while all sou ces cap u e aspec s o he co e ha
esiden s ag ee wi h a pe ec ecall , hey also encompass many a eas
esiden s do no ecognize as pa o i . P ecision alues a e low ac oss all
pla o ms, wi h he combined da a achie ing he highes p ecision a a ound
40%, indica ing ha nea ly hal o he a eas iden i ied by all sou ces all
wi hin he s ic e consensus a ea. The F-sco e o he combined da a in he
co e egion eaches 0.549, which is he highes ye e lec s he challenges in
pinpoin accu acy o an a ea pe cei ed na owly.
Fo long- e m esiden s
Flick again depic s he highes IOU sco e o 0.580 in he domain egion,
showing a s ong co ela ion wi h he a eas ha long- e m esiden s iden i y
wi h Al ama, see Table 6. The combined da a has an IOU sco e o 0.394,
sligh ly su passing Ins ag am, and a p ecision o 0.766, he highes among
38
all pla o ms, indica ing a mo e accu a e depic ion o consensus a eas when
all da a sou ces a e conside ed. Flick also exhibi s a high ecall o 0.912,
cap u ing mos o he domain egion as pe cei ed by esiden s. Fu he mo e,
Flick ’s p ecision F-sco e is signi ican ly high a 0.734, cap u ing and
comp ehensi ely co e ing neighbo hood bounda ies o long- e m esiden s.
In he co e egion, IOU sco es d op ma kedly ac oss all pla o ms. Despi e
including all a eas ha long- e m esiden s ag ee a e pa o Al ama, all
pla o ms also cap u e a eas ou side o he consensus. The combined da a’s
F-sco e is he mos no able a 0.612, showing ha he agg ega ion o all
pla o ms, despi e indi idual lowe p ecision, o e s a mo e balanced measu e
o accu acy be ween p ecision and ecall o he co e egion.
Based on he me ic esul s, Flick 's delinea ion o Al ama eme ges as he
mos ep esen a i e o he neighbo hood, pa icula ly when compa ed o he
spa ial ag eemen o long- e m esiden s in he domain egion. The da a
sugges s ha Flick use s, who a e likely o be pho og aphy en husias s,
concen a e hei pos ings on e y speci ic si es wi hin Al ama. The beha io
o use s on pho og aphy- ocused pla o ms like Flick ypically g a i a es
owa ds he mos isually appealing loca ions wi hin a neighbo hood
his o ical si es, iewpoin s, and places o signi ican in e es . This pa e n
aligns wi h he obse a ions made by Hollens ein and Pu es (2010), who also
iden i ied a simila end in use in e ac ions in hei s udy. In hei esea ch,
hey a emp ed o use Flick 's place names and desc ip ions o delinea e
e nacula bounda ies, as he me hods employed in ou s udy. Fu he mo e,
Al ama's epu a ion as Lisbon's mos pic u esque neighbo hood, as desc ibed
by (Bap is a e al., 2018), ein o ces he idea ha isually d i en pla o ms
like Flick would na u ally align closely wi h he a eas ha esiden s and
isi o s ind mos appealing.
The inal delinea ion map o he his o ic Al ama neighbo hood, based on
Flick geo ags and he pe cei ed bounda ies by long- e m esiden s, is
de ailed in he Appendix. As a suppo in he esul s desc ip ions his map
showcases enowned poin s o in e es , majo oad a e ies, and images ha
cap u e he pic u esque and his o ic cha ac e de ining Al ama.
39
Figu e 19. Compa a i e map o pe cei ed and Geo- agged bounda ies o sho - e m
esiden s in Al ama.
Table 6. Quan i a i e esul s be ween use gene a ed con en da ase s bounda ies and
Al ama Sho - e m esiden s g ound u h.
Residen s
Sho
e m
Domain
Flick
Twi e
IG
Combined
IOU
0,538
0,523
0,385
0,310
O e lap
Km²
0.276
0.281
0.274
0.127
Recall
0,741
0,755
0,736
0,341
P ecision
0,662
0,630
0,447
0,773
F-sco e
0,699
0,687
0,556
0,473
Co e
Flick
Twi e
IG
Combined
IOU
0,191
0,179
0,130
0,378
O e lap
Km²
79708,581
79708,581
79708,581
66878,3655
Recall
1,000
1,000
1,000
0,839
P ecision
0,191
0,179
0,130
0,408
F-sco e
0,321
0,303
0,230
0,549
40
Figu e 20. Compa a i e map o pe cei ed and Geo- agged bounda ies o Long- e m
esiden s in Al
Residen s
Long
e m
Domain
Flick
Twi e
IG
Combined
IOU
0,580
0,534
0,384
0,394
O e lap
Km²
0.256
0.253
0.248
0.126
Recall
0,912
0,901
0,882
0,448
P ecision
0,614
0,567
0,404
0,766
F-sco e
0,734
0,696
0,555
0,565
Co e
Flick
Twi e
IG
Combined
IOU
0,199
0,186
0,136
0,441
O e lap
Km²
0.0829
0.0829
0.0829
0.0756
Recall
1,000
1,000
1,000
0,912
P ecision
0,199
0,186
0,136
0,461
F-sco e
0,332
0,314
0,239
0,612
Table 7. Quan i a i e esul s be ween use gene a ed con en da ase s Bounda ies and
Al ama Sho - e m esiden s g ound u h.
41
4.3.2 Mou a ia
The compa a i e analysis o social media pla o m da a in delinea ing he
pe cei ed bounda ies o he Mou a ia neighbo hood p o ides insigh ul
me ics o bo h sho - e m and long- e m esiden s. The esul s a e
display in he igu es 21 and 22 and ables 8 and 9 .
Fo Sho -Te m Residen s
In he domain egion, Twi e demons a es a signi ican alignmen wi h
he pe cep ions o sho - e m esiden s, boas ing an In e sec ion o e
Union (IOU) sco e o 0.551 and a high ecall o 0.760.Twi e ’s da a
cap u es a b oad a ea ha esiden s associa e wi h he Mou a ia domain,
al hough i may ex end beyond he consensus bounda ies. Ins ag am, wi h
an excep ionally high ecall o 0.940, iden i ies a la ge po ion o he
domain. Howe e , i s p ecision sco e o 0.447 indica es i encompasses
a eas ha all ou side he esiden -de ined domain o a g ea e ex en han
Twi e . The combined sou ces shape, despi e a low IOU alue o 0.241,
achie es he highes p ecision a 0.740, poin ing o a mo e accu a e
ep esen a ion wi hin he consensus a ea.
Rega ding he co e egion, Twi e 's p ecision is no ably low a 0.216, ye
i nea ly encapsula es he en i e co e a ea wi h a ecall o 0.922. This
implies ha while Twi e 's da a ex ensi ely co e s he co e egion
acknowledged by esiden s, i is no as p ecise. Ins ag am's co e age is
comple e conce ning he co e as ecognized by esiden s bu also includes
conside able a eas beyond i .
Fo Long-Te m Residen s
The domain egion o long- e m esiden s again sees Twi e wi h a
subs an ial IOU o 0.512 and a ecall o 0.716, cap u ing a signi ican pa
o wha long- e m esiden s ega d as Mou a ia’s domain bu also eaching
ou in o a eas ou side o he ag eed domain. Ins ag am p esen s an
ex ensi e co e age wi h a ecall o 0.978, bu i s p ecision is only 0.475,
meaning i includes mo e a eas ou side o he esiden consensus compa ed
o Twi e . The ou pu om all da a sou ces, while p esen ing a smalle
o e lap, aligns mos accu a ely wi h he a eas ag eed by long- e m
esiden s.
In he co e egion, he me ics a e gene ally lowe . Twi e 's IOU is 0.288,
coupled wi h a ecall o 0.851, showing i encompasses mos o he co e as
iden i ied by long- e m esiden s. Howe e , he p ecision o 0.303 sugges s
some imp ecision in delinea ing speci ic bounda ies. The combined sou ces
ou pu ea u es an IOU o 0.359 and he bes F-sco e o 0.528, indica ing a

42
be e equilib ium in cap u ing he co e's ex en while main aining
bounda y accu acy.
The delinea ion o Mou a ia by Twi e is shown o be he mos
ep esen a i e o he neighbo hood, especially when compa ed wi h he
spa ial ag eemen o long- e m esiden s in he domain egion. The
selec ion o his delinea ion was based no solely on me ic esul s, which
we e simila ac oss all aspec s calcula ed o bo h esiden g oups, bu also
on he his o ical ex ension o he neighbo hood as desc ibed by his o ians.
Acco ding o hem, Mou a ia is loca ed be ween he alley o Plaça Ma in
Moniz and he hills o he cas le and G aça neighbo hood. This a ea was
designa ed o he de ea ed Moo s a e he ci y's conques by D. A onso
Hen iques in 1147, as pa o he Ch is ian conques o e he Muslims.
The e o e, he decision was made o in eg a e he pe cei ed bounda ies o
long- e m esiden s, which include he a ea be ween Cos a de Cas elo and
Calçada de San o And é s ee s, as i be e cap u es hose desc ibed
de ails and has mode a e alues in he me ics.
Ano he peculia i y o he esul s is ela ed o he ac ha neighbo hood
poin s o in e es , such as he Fado Vadio G a i i, known as a ibu e o
he Fado music gen e and i s bi hplace in Mou a ia, he São C is ó ão
chu ch, as ancien as i is amous, which acco ding o he Wo ld
Monumen s Wa ch websi e is pa o Mou a ia,(Chu ch o São C is ó ão,
2016) a e no included wi hin he pe cei ed limi s.
The exclusion o key cul u al and his o ical landma ks om Mou a ia's
pe cei ed bounda ies p omp s a deepe e lec ion on he neighbo hood's
e nacula ambigui y and he challenges in accu a ely delinea ing i s eal
ex en . This cha ac e is ic o Mou a ia sugges s a nuanced complexi y in
unde s anding i s spa ial iden i y. I unde sco es he necessi y o
conside ing he s udy a ea's unique cha ac e is ics e en when applying
Ex ac ion me hods.
The inal delinea ion map o he his o ic Mou a ia neighbo hood, based on
Twi e geo ags and he pe cei ed bounda ies by long- e m esiden s, is
p o ided in de ail in he Appendix. As a suppo in he esul s desc ip ions
his map highligh s enowned poin s o in e es , majo oad a e ies, and
ea u es images ha cap u e he di e se and adi ional na u e de ining
Mou a ia.
43
Figu e 21.Compa a i e map o pe cei ed and Geo- agged bounda ies o Sho - e m
esiden s in Mou a ia.
Table 8. Quan i a i e esul s be ween use gene a ed con en da ase s bounda ies and
Mou a ia Sho - e m esiden s g ound u h.
Residen s
Sho e m
Domain
Twi e
IG
Combined
IOU
0,551
0,434
0,241
O e lap
Km²
0.0950
0.1176
0.0330
Recall
0,760
0,940
0,264
P ecision
0,668
0,447
0,740
F-sco e
0,711
0,605
0,389
Co e
Twi e
IG
Combined
IOU
0,212
0,127
0,285
O e lap
Km²
0.0308
0.0334
0.0173
Recall
0,922
1,000
0,519
P ecision
0,216
0,127
0,388
F-sco e
0,350
0,225
0,444
44
Residen s
Long e m
Domain
Twi e
IG
Combined
IOU
0,512
0,470
0,259
O e lap
Km²
0.0915
0.1250
0.0354
Recall
0,716
0,978
0,277
P ecision
0,643
0,475
0,795
F-sco e
0,677
0,639
0,411
Co e
Twi e
IG
Combined
IOU
0,288
0,193
0,359
O e lap
Km²
0.0432
0.0507
0.0252
Recall
0,851
1,000
0,496
P ecision
0,303
0,193
0,565
F-sco e 0,447 0,323 0,528
Table 9. Quan i a i e esul s be ween use gene a ed con en da ase s bounda ies and
Mou a ia Long- e m esiden s g ound u h.
Figu e 22.Compa a i e map o pe cei ed and Geo- agged bounda ies o Long- e m
esiden s in Mou a ia.
45
4.3.3 Bai o Al o
In his compa a i e analysis, we del e in o he spa ial pe cep ions o Bai o
Al o's Domain and Co e egions, as dis inguished by sho - e m and long-
e m esiden s, using da a om Flick , Twi e , Ins ag am, and a combined
app oach. The esul s a e display in he igu es 23 and 24 and ables 10 and
11.
Fo Sho -Te m Residen s
In he Domain egion, Flick 's encompassing cap u e o he a ea aligns wi h
sho - e m esiden s' ecogni ion, e en hough wi h a p ecision o only 0.327,
indica ing ex aneous a ea inclusion. Twi e , wi h a p ecision o 0.929 and
ecall o 0.487, o e s a selec i e ye highly accu a e po ayal, showing
cap u es less bu wi h g ea e ideli y. Ins ag am s ikes a balance wi h ecall
and p ecision a es o 0.564 and 0.568, espec i ely, indica ing mode a e
co e age and a e age p ecision. The combined da a sou ces achie e a
p ecision o 0.879, closely mi o ing he a eas sho - e m esiden s ag ee,
despi e no cap u ing he en i e y o he Domain.
Wi hin he Co e egion, Flick 's da a, while all-encompassing wi h a ecall o
1.000, su e s in p ecision a 0.174, sugges ing i spans many a eas ou side
he consensus. Twi e showcases a high IOU o 0.646, wi h ecall and
p ecision a es o 0.778 and 0.792, espec i ely, indica ing a selec i e ye
accu a e cap u e o he Co e. Ins ag am o e s a b oad, albei less p ecise,
iden i ica ion, whe eas he agg ega e da a p o ides a balanced
ep esen a ion wi h an IOU o 0.615, and ecall and p ecision a es o 0.753
and 0.770, espec i ely.
Fo Long-Te m Residen s
Flick again cap u es he en i e Domain ecognized by long- e m esiden s
bu wi h low p ecision (0.348), indica ing he inclusion o ex ensi e non-
consensual a eas. Twi e 's app oach, wi h a p ecision o 0.910, e lec s high
accu acy, e en co e ing less o he Domain a ea.
In he Co e egion, Flick 's co e age ex ends o e he en i e Co e and beyond,
simila o i s app oach in he Domain. Twi e , wi h i s mode a e ecall and
high p ecision, accu a ely cap u es he Co e. The combined da a sou ces show
he highes p ecision o long- e m esiden s in he Co e, sugges ing ha an
in eg a ed app oach o di e en da a sou ces yields he mos p ecise
delinea ion.
Acco ding o (F anco, 2016), he cons uc ion o he San Roque Chu ch in he
16 h cen u y. ma ked a ele an momen in he u ban his o y o Bai o Al o,
ans o ming i in o a ocal poin o Lisbon's nobili y. This shi led o he
cons uc ion o palaces on he hill's wes e n side, d i en by a desi e o be
52
5.3 Fu u e Scope
Fu u e esea ch could bene i om expanding he ex ac ion o ex ual
a ibu es ela ed o he speci ic economic, en i onmen al, and social
cha ac e is ics o each neighbo hood. E o s could include explo ing
di e se me hods o uni ying pe cei ed bounda ies, such as bounda y
agg ega ion o adial a e aging due o, as men ioned in he me hod
compa ison s udy by (Dal on & Hu ell, 2023). Addi ionally, expe imen ing
wi h a ious clus e ing algo i hms like spec al clus e ing o Ke nel
Densi y Es ima ion could p o e aluable. Inco po a ing insigh s om
e hnog aphic s udies, like hose conduc ed by (Bap is a e al, 2018), which
e eal he less posi i e social eali ies wi hin he h ee neighbo hoods,
would be pa icula ly insigh ul. These eali ies, including issues such as
ma ginaliza ion and social exclusion, a e e lec ed in he discou se on
social media and could add dep h o his s udy's indings. I his app oach
p o es success ul, i could be applied o o he his o ic neighbo hoods in he
ci y, each dis inguished by i s unique cha ac e is ics, he eby enhancing
ou unde s anding o u ban social dynamics.

Chap e 6
6. Conclusions
This s udy emba ked on he in ica e challenge o mapping ou he
e nacula bounda ies o Lisbon's his o ic neighbo hoods: Al ama,
Mou a ia, and Bai o Al o. By blending esiden s' insigh s wi h geo- agged
da a, alongside me hods and ools o ex ac ing pe cep ions, we' e
un eiled he ue po en ial o me ging cogni i e mapping wi h use -
gene a ed con en . This me hod has p o en e ec i e in delinea ing he
spa ial con ines o hese s o ied a eas, illumina ing he ich apes y o
his o y and communi y li e ha 's in e wo en wi h he u ban ab ic o
Lisbon.
The indings shed ligh on he nuanced ways digi al pla o ms and social
in e ac ions con ibu e o ou g asp o u ban bounda ies. Al ama's
ep esen a ion h ough Flick 's lens eme ged as no ably p ecise,
unde sco ing he impac o isual pla o ms in cap u ing he essence o
locales, hus emphasizing he signi ican ole o isual pe cep ions in
bounda y delinea ion. In Mou a ia, Twi e 's alignmen wi h long- e m
esiden s' unde s anding highligh ed he pla o m's capabili y o cap u e
he neighbo hood's essence ia social media dynamics. Meanwhile, Bai o
Al o's de ailed mapping ia Twi e e ealed he in luence o his o ical and
social unde pinnings in shaping u ban bo de s.
Add essing he esea ch ques ions, i 's insigh ul o no e how esiden s
spa ially de ine di e en his o ic neighbo hoods h ough a ious lenses, be
i he leng h o esidence, he o ien a ion o s ee ne wo ks, o he spa ial
ex en o a eas. A composi e app oach, which amalgama es mul iple da a
sou ces, a o ds a iche , mo e accu a e ep esen a ion o he co e egions
o e nacula bounda ies. Addi ionally, he indings ad oca e ha social
media analy ics can se e as a eliable index o pe cei ed ci y images, wi h
pla o ms like Ins ag am elucida ing public pe cep ions ied o ou is
a ac ions and landma ks.
The implica ions o his esea ch e e be a e h ough he ealms o u ban
planning and policymaking, championing an app oach ha mo e closely
esona es wi h he li ed expe iences and pe cep ions o ci y inhabi an s. A
deepe unde s anding o subjec i e neighbo hood delinea ions can
signi ican ly e ine s a egies in ou ism managemen and esou ce
alloca ion, especially wi hin highly ma ginalized a eas. This hesis
con ibu es o he en ichmen o e nacula geog aphy by o e ing new
54
insigh s in o he mode n u ban ab ic, shaped by a combina ion o social
in e ac ions and digi al oo p in s.
The success ul in eg a ion o spa ial and social dimensions showcased in
his s udy unde sco es he e icacy o employing di e se me hodologies o
da a ex ac ion, analysis, clus e ing, and delinea ion. This mul i ace ed
app oach no only enhances ou comp ehension o u ban spaces bu also
lays down a comp ehensi e amewo k o na iga ing he in ica e
in e play be ween physical geog aphy and social cons uc s in de ining
neighbo hood bounda ies.
55
7. Bibliog aphic Re e ences
Anac a, V. J. A., Humayun, M. I., Schwe ing, A., & K uka , J. (2017).
In es iga ing Rep esen a ions o Places wi h Unclea Spa ial Ex en
in Ske ch Maps. In A. B eg , T. Sa jakoski, R. Van Lamme en, & F.
Rip (Eds.), Socie al Geo-inno a ion (pp. 3–17). Sp inge
In e na ional Publishing. h ps://doi.o g/10.1007/978-3-319-56759-
4_1
A ibas-Bel, D., Ga cia-López, M.-À., & Viladecans-Ma sal, E. (2021).
Building(s and) ci ies: Delinea ing u ban a eas wi h a machine
lea ning algo i hm. Jou nal o U ban Economics, 125, 103217.
h ps://doi.o g/10.1016/j.jue.2019.103217
Bae, C., & Mon ello, D. (2018). Rep esen a ions o an U ban E hnic
Neighbou hood: Residen s’ Cogni i e Bounda ies o Ko ea own, Los
Angeles. Buil En i onmen , 44, 218–240.
h ps://doi.o g/10.2148/ben .44.2.218
Bap is a, L. V., No e, J., & Jo ge, M. do R. (2018). Mobilidade, Cidade e
Tu ismo: Pis as pa a analisa as ans o mações em cu so no cen o
his ó ico de Lisboa. Sociologia: Re is a da Faculdade de Le as da
Uni e sidade do Po o, 14–32.
B own, G., Raymond, C. M., & Co co an, J. (2015). Mapping and measu ing
place a achmen . Applied Geog aphy, 57, 42–53.
h ps://doi.o g/10.1016/j.apgeog.2014.12.011
Chuguno , A., Misniko , Y., Roshchin, E., & T u ne , D. (Eds.). (2019).
Elec onic Go e nance and Open Socie y: Challenges in Eu asia: 5 h
In e na ional Con e ence, EGOSE 2018, S . Pe e sbu g, Russia,
No embe 14-16, 2018, Re ised Selec ed Pape s (Vol. 947). Sp inge
In e na ional Publishing. h ps://doi.o g/10.1007/978-3-030-13283-5
Chu ch o São C is ó ão. (n.d.). Wo ld Monumen s Fund. Re ie ed
Feb ua y 23, 2024, om h ps://www.wm .o g/p ojec /chu ch-
s%C3%A3o-c is %C3%B3 %C3%A3o
Cocola-Gan , A., & Gago, A. (2021). Ai bnb, buy- o-le in es men and
ou ism-d i en displacemen : A case s udy in Lisbon. En i onmen
and Planning A: Economy and Space, 53(7), 1671–1688.
h ps://doi.o g/10.1177/0308518X19869012
Coul on, C. J., Jennings, M. Z., & Chan, T. (2013). How Big is My
Neighbo hood? Indi idual and Con ex ual E ec s on Pe cep ions o
Neighbo hood Scale. Ame ican Jou nal o Communi y Psychology,
51(1–2), 140–150. Scopus. h ps://doi.o g/10.1007/s10464-012-9550-
6
Da Cos a, A. F. (2008). Sociedade de Bai o Dinâmicas Sociais da
Iden idade Cul u al (2 h ed.). Cel a edi o a.
Dal on, N. S., & Hu ell, M. (2023). Me hods o neighbou hood Mapping,
bounda y ag eemen . En i onmen and Planning B: U ban
Analy ics and Ci y Science, 50(2), 401–415.
h ps://doi.o g/10.1177/23998083221115195
56
Deng, Y. (2016). Challenges and complica ions in neighbo hood mapping:
F om neighbo hood concep o ope a ionaliza ion. Jou nal o
Geog aphical Sys ems, 18(3), 229–248.
h ps://doi.o g/10.1007/s10109-016-0232-z
Edelsb unne , H., Ki kpa ick, D., & Seidel, R. (1983). On he shape o a
se o poin s in he plane. IEEE T ansac ions on In o ma ion Theo y,
29(4), 551–559. h ps://doi.o g/10.1109/TIT.1983.1056714
Es e ens, A., Cocola-Gan , A., Cal o, D. M., & Ma os, F. (2019). As a es e
a cul u a nas p á icas hegemónicas e al e na i as na cidade de
Lisboa. O caso do La go do In enden e. Fo um Sociológico. Sé ie II,
35, A icle 35. h ps://doi.o g/10.4000/sociologico.7826
E ans, A. J., & Wa e s, T. (2007). Mapping e nacula geog aphy: Web-
based GIS ools o cap u ing “ uzzy” o “ ague” en i ies.
In e na ional Jou nal o Technology, Policy and Managemen , 7(2),
134. h ps://doi.o g/10.1504/IJTPM.2007.014547
F anco, A. (2016). Caminha po Lisboa 7 pe cu sos pelas his ó ias e
seg edos da cidade ( i s ). Po o Edi o a.
Gao, S., Janowicz, K., Mon ello, D. R., Hu, Y., Yang, J.-A., McKenzie, G.,
Ju, Y., Gong, L., Adams, B., & Yan, B. (2017a). A da a-syn hesis-
d i en me hod o de ec ing and ex ac ing ague cogni i e egions.
In e na ional Jou nal o Geog aphical In o ma ion Science, 31(6),
1245–1271. h ps://doi.o g/10.1080/13658816.2016.1273357
Gao, S., Janowicz, K., Mon ello, D. R., Hu, Y., Yang, J.-A., McKenzie, G.,
Ju, Y., Gong, L., Adams, B., & Yan, B. (2017b). A da a-syn hesis-
d i en me hod o de ec ing and ex ac ing ague cogni i e egions.
In e na ional Jou nal o Geog aphical In o ma ion Science, 1–27.
h ps://doi.o g/10.1080/13658816.2016.1273357
Glebo a, M. (n.d.). Town and Gown: Visualising Uni e si y
Neighbou hoods as Places wi hin he U ban En i onmen .
He nández, B., Ca men Hidalgo, M., Salaza -Laplace, M. E., & Hess, S.
(2007). Place a achmen and place iden i y in na i es and non-
na i es. Jou nal o En i onmen al Psychology, 27(4), 310–319.
Scopus. h ps://doi.o g/10.1016/j.jen p.2007.06.003
Hollens ein, L., & Pu es, R. S. (2010). Explo ing place h ough use -
gene a ed con en : Using Flick ags o desc ibe ci y co es. Jou nal
o Spa ial In o ma ion Science, 1(2010), 21–48. Scopus.
h ps://doi.o g/10.5311/JOSIS.2010.1.3
Huang, J., Ob ach -P ondzynska, H., Kam owska-Zaluska, D., Sun, Y., &
Li, L. (2021). The image o he ci y on social media: A compa a i e
s udy using “Big Da a” and “Small Da a” me hods in he T i-Ci y
Region in Poland. Landscape and U ban Planning, 206, 103977.
h ps://doi.o g/10.1016/j.landu bplan.2020.103977
Langens, J. (2022). Mou a ia c oss oads o cul u es as old as he na ion. In
10 Lisbon S o ies. casadasle as.
Li, L., Goodchild, M. F., & Xu, B. (2013). Spa ial, empo al, and
socioeconomic pa e ns in he use o Twi e and Flick . Ca og aphy
57
and Geog aphic In o ma ion Science, 40(2), 61–77.
h ps://doi.o g/10.1080/15230406.2013.777139
Lynch, K. (1964). The Image o he Ci y. MIT P ess.
Meinig, D. W. (1965). The Mo mon Cul u e Region: S a egies and
Pa e ns in he Geog aphy o he Ame ican Wes , 1847–1964.
Annals o he Associa ion o Ame ican Geog aphe s, 55(2), 191–219.
h ps://doi.o g/10.1111/j.1467-8306.1965. b00515.x
No e, J. (2020). No e, J. (2020). Pa y ou ism, pub c awls and c azy
nigh s in Bai o Al o, Lisbon: Celeb a ing cosmopoli anism in a
acialized he e o opic pleasu escape. IS Wo king Pape s, Se ie III,
No. 87: 1-13.
Oli ei a, N. (2019). A Lisboa cosmopoli a e o ascínio da di e sidade.
CIDADES, Comunidades e Te i ó ios, 115–128.
h ps://doi.o g/10.15847/ci iescommuni ies e i o ies.dec2019.039.a
04
Sal a o e, C., Bi ignandi, S., & Bianchi, A. (2021). Social Media and
Twi e Da a Quali y o New Social Indica o s. Social Indica o s
Resea ch, 156(2), 601–630. h ps://doi.o g/10.1007/s11205-020-
02296-w
Tang, V., Acedo, A., & Painho, M. (2021). Sense o place and he ci y: The
case o non-na i e esiden s in Lisbon. Jou nal o Spa ial
In o ma ion Science, 23, A icle 23.
h ps://doi.o g/10.5311/JOSIS.2021.23.165
Tang, V., & Painho, M. (2023). Explo ing he ela ionships be ween
pe cei ed neighbo hood bounda ies and s ee ne wo k o ien a ion.
T ansac ions in GIS, 27(3), 877–899.
h ps://doi.o g/10.1111/ gis.13058
Tang, V., Pu i, J., & Painho, M. (2022). Mapping a his o ic neighbou hood
h ough use -gene a ed con en : The case o Al ama, Lisbon
(Po ugal). AGILE: GIScience Se ies, 3, 1–8.
h ps://doi.o g/10.5194/agile-giss-3-63-2022
Tou ism indus y eclipses p e-pandemic le els in 2023—P oduc s Eu os a
News—Eu os a . (n.d.). Re ie ed Feb ua y 5, 2024, om
h ps://ec.eu opa.eu/eu os a /web/p oduc s-eu os a -news/w/ddn-
20240116-1
Ve non, H. (2023, Feb ua y 27). 2022, el mejo año u ís ico de la his o ia
de Po ugal. Benoi P ope ies.
h ps://www.benoi p ope ies.com/es/news/2022-ma ks-bes -e e -
yea - o -po ugal- ou ism/
Wes e hol , R., Acedo, A., & Na anjo-Zolo o , M. (2022). Explo ing sense o
place in ela ion o u ban acili ies – e idence om Lisbon. Ci ies,
127, 103750. h ps://doi.o g/10.1016/j.ci ies.2022.103750
Zelinsky, W. (1980). No h Ame ica’s Ve nacula Regions. Annals o he
Associa ion o Ame ican Geog aphe s, 70(1), 1–16.

Appendix
Figu e 25. Map o he His o ic Al ama Neighbo hood Delinea ion Based on Flick Geo ags and Pe cei ed Bounda ies by Long-Te m Residen s
59
Figu e 26.Map o he His o ic Mou a ia Neighbo hood Delinea ion Based on Twi e Geo ags and Pe cei ed Bounda ies by Long-Te m Residen s
60
Figu e 27.Map o he His o ic Bai o Al o Neighbo hood Delinea ion Based on Twi e Geo ags and Pe cei ed Bounda ies by Long-Te m.Residen s
61
Ve nacula Bounda ies o His o ic
Neighbo hoods in he Ci y o Lisbon
2024
Mónica So ía Roncancio Blanco