Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Studia Ekonomiczne. Zeszyty Naukowe Uniwersytetu Ekonomicznego w Katowicach ISSN 2083-8611 Nr 234 · 2015
Krzysztof Węcel
Uniwersytet Ekonomiczny w Poznaniu Katedra Informatyki Ekonomicznej [email protected]
LINKED GEODATA FOR PROFILING OF TELCO USERS
Summary: There is a growing interest in location-based profiling of users de-fined as com-bining geo-data with anonymous on-line profiles. The profile of an entity usually consists of concepts accompanied by a weight specifying a relative importance of the given concept for making an analysed entity distinct. The proposed profiling method of telco users is a two-step approach. First, profiles of mobile tower stations (BTS) are created based on crowdsourced geographical information. Second, they are used to generalise the behaviour of a calling user, which is determined from Call Detail Records (CRD). The linked data cloud is considered as an additional knowledge source in the user modelling process. Keywords: linked data, user profiling, linked geodata, call detail record, mo-bile user, telco, cdr, bts, lgd, osm. Introduction
There is a growing interest in location-based profiling defined as combining geo-data with anonymous on-line profiles. New methods for capturing informa-tion on where are the users and how their position changes over time are con-stantly developed. This information is becoming increasingly more valuable for a growing number of location-based or location-aware services. Some research-ers try to estimate value of mobile data information utilising proximity-based advertising valuation (Baccelli, Bolot, 2011). Tourists have been identified as the most rewarding target group.
Call Detail Record (CDR) is the most widely used source of mobile loca-tion data in academic research (Song et al., 2010). Presented location is not very precise as it is “rounded” to the co-ordinates of the nearest base transceiver sta-tion (BTS). Various granularities of location impact the value of location infor-
Krzysztof Węcel
200
mation (Baccelli, Bolot, 2011). CDR has been identified to be sufficient for drawing conclusions at the area level (Qu, Zhang, 2013), hence it is also suffi-cient for our purposes.
Linked data cloud is very often considered as additional knowledge source in user modelling process. Concepts defined in various ontologies can be used to characterise entities. Profile of the entity consists then of the concept accompa-nied by a weight specifying relative importance of the given concept for making analyse entity distinct.
In this paper we focus on cell tower granularity of location information and annotate it with geographical ontology derived from OpenStreetMap. The goal of the method is to provide profiles of the users based on profiles of the BTS sta-tions. Therefore, the profiling process has been split into several steps. First, the information about BTS location and its neighbourhood has to be retrieved and analysed. Then, based on this information a summary of BTS profiles is pre-pared. We propose an improvement in profiling process by leveraging TF-IDF ranking to address the issue of uneven distribution of categories describing mo-bile tower locations (skewness). In the last step we can characterise users that log-in into specific BTS stations.
Section 2 presents related research. Section 3 explains our general approach to profiling of entities based on geographical context. Section 4 introduces a method for characterisation of BTS stations, from data collection, through analy-sis, to data aggregation. Section 5 provides a method for profiling of telco users. 1. Related research
In the literature, there are various approaches to profiling of mobile users. Some authors base purely on telco data, i.e. data available for mobile operators. Majority of methods leverage the social media where users manifest their opin-ions, feelings, reveal location etc. The most sophisticated approaches add mining for generalisation of patterns and classification of users. Having access to ano-nymised call data, we base our method on this data.
Most methods base on social data like Twitter, Foursquare, Flickr, or Insta-gram as data is relatively easily retrievable (API available). These services are then widely used and generate large volumes of data. One of the challenges is how to model the use of such social (and mobile) applications by various users. It is essential to understand the semantics of messages posted by them. Two trends are observed here: extraction of meaning and location.
Linked geodata for profiling of telco users
201
In many studies in order to enrich and disambiguate information gathered from user, semantic technologies are considered. They are particularly useful for providing context, and geographical context is one of the most important. Abel et al. introduced a user modelling framework that utilises semantic background knowledge and use it for point of interest (POI) specification (Abel et al., 2012). Two knowledge sources of linked data are considered: GeoNames and DBpedia. They demonstrate that user modelling quality improves when LOD-based back-ground knowledge is considered. DBpedia is unfortunately too coarse-grained for our purposes.
Instead of analysing separate check-ins, some approaches build activity-travel profile – a spatial trajectory is built from mobile phone call records only. The tricky part is in classification of trajectories, where data mining methods can be applied (Görnerup, 2012). Not only the sequences have to be derived but in order to make sense they have to be classified into typical activity-travel pat-terns. Their relative frequencies constitute an activity travel profile (Liu et al., 2014). Our method also bases on a number of BTSes visited. However, we do not consider sequences of visits as our experiments have shown that this would not produce meaningful results.
Trajectories, or sequences of visited locations, are not very useful unless they are confronted with activities of the users. Lie et al. investigated to what ex-tent the behavioural routines could reveal the activities being performed at mo-bile phone call locations (Liu et al., 2013). The real value is in annotating loca-tions with activity purposes but for this additional information is required. Although they have devised mathematical models to quantitatively characterize travel patterns, the motivating activities “were still in a less-explored stage” (Liu et al., 2013). Our method provides just another context for reasoning about pos-sible activities of the users, e.g. shopping, playing tennis.
According to (Cano et al., 2013) little has been done in modelling location entities. Therefore, they proposed to profile geographical areas by providing topical categorisation. Cano et al. used additional source – linked data to inter-link information from social stream with geographical objects (Cano et al., 2013). They have introduced LinkedPOI ontology, which uses DBpedia catego-ries to profile geographic space and proposed geo-lattice Awareness Stream model as one of the ways to represent location. The process consisted of filter-ing, enriching, structuring and interlinking microposts from Twitter, Facebook, and TripAdvisor. Our method in fact models location entities but as we do not analyse microposts we do not have to guess the correct location – it is provided in CDR.
Krzysztof Węcel
202
Qu proposed a framework and corresponding analytic methods to use User Generated Mobile Location Data (UGMLD) for Trade Area Analysis (Qu, Zhang, 2013). They have defined three key processes: “identifying the activity centre of a mobile user, profiling users based on their location history, and mod-elling users’ preference probability.” Application of the method is meant for analysis of customers’ visits to business venues. However, they rejected CDR as the data source in their research as it was too coarse. Our method is able to util-ise CDR in order to provide profiles of certain locations although we cannot provide profiles of certain venues belonging to bigger chains.
When specific locations are considered, Chen et al. also presented a method for profiling businesses at specific locations that was based on mining informa-tion from social media (Chen et al., 2014). They matched geo-tagged tweets against locations from Foursquare to build a profile of mentioned businesses.
Going back to the user, Ostuni et al. presented Cinemappy – a location-based application that computes film recommendations by exploiting contextual information related to current location of the user, leveraging information from DBpedia (Ostuni et al., 2013). Similarly, DBpedia was used by (Cano et al., 2013) who proposed a semantic travel mash-up as possible application. Ap-proach for museums was presented in (Ruotsalo et al., 2013).
One of the obstacles by user modelling is uneven distribution of categories of business, e.g. there are much more visits to a cinema than to second-hand and there are much more shops than theatres. It was first noted by Qu who explained this by social motivation, not necessarily by the differences in the number of various categories (Qu, Zhang, 2013). In their work the categories were very fine grained, for example Foursquare has a hierarchical category structure with 9 top categories and ca. 400 sub-categories. For the clarity of interpretation the catego-ries have been later collapsed to 6 groups. Our methods addresses this issue by applying specific methods from information retrieval domain. 2. Approach to geographical linked data-based profiling
This section describes user profiling based on BTS characteristics derived from the geographical linked data. We follow the idea of location-based user profiling – one of the approaches is geoprofiling, a commonly used method to approximate user characteristics based on neighbourhood demographic data.
In most approaches, there is a venue or a place given and authors are look-ing for coordinates. This is particularly important when text is analysed, espe-cially in social media, e.g. Twitter. In order to simplify disambiguation, some
Linked geodata for profiling of telco users
203
portals allow the so called check-ins where users select precise location, e.g. Foursquare, Facebook. We take the reverse process: starting from coordinates we are interested in the objects nearby, thus describing the geographical context, further referred to as location profile.
Our experiments have been carried out on anonymised data, where the only reasonable data for linking was location of BTS towers. There was just one type of information that was widely used and could supplement our records – geo-graphical information. There are several open data sources concerning geo-graphical information that we could use, DBpedia and OpenStreetMap being the most prominent. Taking into account the granularity of available data and the re-quirement to display results on the map we made a decision to base our method on OpenStreetMap and its triplified counterpart – LinkedGeoData (Auer, Leh-mann, Hellmann, 2009). As a crowdsourced data, it is kept relatively up to date and but without breaking fluctuations.
LinkedGeoData (LGD) provides an ontology for classification of locations. There are ca. 1200 categories grouped into ca. 45 top-level categories. Compar-ing LGD to Foursquare, the latter has ten top-level, 436 second-level, and 266 third-level categories1. Foursquare’s maps in fact use OpenStreetMap2. In our approach more general categories are advantageous as they can make interpreta-tion of generalisation results easier. Sub-categories could be more valuable as they can distinguish users better but the solution is then less stable from the sta-tistical point of view. This is a well-known trade-off of specificity vs. sensitivity (Fawcett, 2006). In order to best characterise the BTS stations, we have re-stricted our further analysis to some predefined objects (see Fig. 3).
The reasoning behind our profiling approach is presented in the following user story. A user often visits sport amenities. They are in the scope of some BTS stations. Profiles of such stations contain sport amenities with higher frequency than an average station. This is further reflected in a user profile where sport amenities gain higher weight when user trajectory is aggregated. Some visited venues are additionally annotated with a kind of sport, for example tennis3. Looking into calendars of sport events we can even reason further in which kind of sport user is interested or whom the user is supporting (whether team or indi-vidual).
1 https://developer.foursquare.com/categorytree. 2 https://foursquare.com/about/osm. 3 This depends on data availability in OpenStreetMap.
Krzysztof Węcel
204
3. Characteristics of BTS 3.1. Retrieval
In our experiments we have used BTS towers located in Poland, with ca. 8000 unique locations, stored in MySQL. At the beginning the information about BTS locations has been retrieved. Using a Python script, for each location a SPARQL query was prepared to retrieve list of objects in the neighbourhood along with their categories. As a source of data for our queries we have used LinkedGeoData which is a derivative of OpenStreetMap. Two main categories of objects are distinguished therein: nodes (just a point according to GIS termi-nology) and ways (lines or polygons). Separate queries for nodes and ways had to be prepared because the Virtuoso’s built-in distance function has different be-haviour for nodes and ways. As there were ca. 8000 locations, two kinds of ob-jects, two means for object capturing (bounding box and circle) and 3 various distances, we had to post ca. 80 thousand queries.
Below sample SPARQL query is presented:
PREFIX lgdm:<http://linkedgeodata.org/meta/> PREFIX geom:<http://geovocab.org/geometry#> PREFIX ogc:<http://www.opengis.net/ont/geosparql#> SELECT distinct ?class ?way WHERE { ?way a lgdm:Way . ?way a ?class . ?way geom:geometry [ogc:asWKT ?geo ] . filter(bif:st_within( ?geo,bif:st_point(%f,%f),%f )) }
Listing 1. Sample SPARQL query using geospatial functions
We have decided to use LGD’s endpoint instead of OSM’s API as it was possible to use Virtuoso built-in SPARQL functions for spatial queries. For the retrieval of nodes, it was possible to provide detailed query and the radius was in fact expressed in kilometres. For example, Figure 1a presents various venues lo-cated in a circle of 1 km diameter.
Retrieval of ways was more complicated. Function bif:st_within was not returning what it was expected for Way objects when the other parameter was a point. Several methods to get satisfactory results have been tested, includ-ing generation of boxes to simulate containment (see Fig. 1b and 1c). Finally, the circle overlap after toleration parameter tuning to 0.01 has been chosen as a method to query for neighbouring ways (Fig. 1d).
F
3
atiT
ct(u
(a)
(c)
Fig.
3.2
andtheiinteThe
catition(0.0user
) no
) wa
. 1.
. A
Vad quir nerese Fi
Aion.n an06).r pr
des
ays,
Vartion
Ana
Variouantneigstingg. 2
Anot. Gend Surofi
, cir
box
riounal
lys
ous tita
ghbog to2b stherene1.2uch les.
rcle
x, 5
us vFai
sis
aspativeourho obshor aseral reasy
.
e, 1k
km
venur
pece. Fhoobserws pecfinstauym
L
km d
ues s
ts oFor od. rve a c
ct anndinuran
mme
Link
diam
sele
of loexThtha
closnaly
ngs nts;try
ked
mete
ecte
ocaampe siat ther lyseare; onne
geo
er
d ba
ationpleize he clooked ise as n theds
odat
ased
n ch Fiof
cityk. s nufol
he os to
ta fo
d on
harg. 2the
y mo
umbllowoth be
for p
n ob
ract2a e ciost
ber ws: er e
e ad
profi
(b
(d)
bjec
erisshorclepac
of on end
ddre
filing
b) w
) wa
ct ty
sticowse spcke
venav
d aresse
g of
ways
ays,
ype
s cas Bpecied w
nueerare ued w
f tel
, bo
circ
and
an bTS ifie
with
es oge univwhe
co u
ox, 1
cle,
d dis
be asta
es thh ho
f githeversen b
user
1km
dis
stan
anaatiohe notel
ivenre asitiebui
rs
m
stanc
nce,
alysons numls in
n caare es (ildin
ce 0
nea
sed,tha
mben Po
ateg4.0
(0.0ng
0.01
ar P
boat her oolan
gory0 sh04) the
1 Pozn
oth haveof hnd i
y pehops
an loc
nań
quae hohoteis G
er Bs pe
nd ccati
Inte
alitaotel
els. Gda
BTSer lcineion
20
erna
ativls iIt i
ansk
S lolocaema
an
5
a-
ve in is k.
o-a-as nd
2
F
raaOhamst
app
206
(a)
Fig.
resua coandObjhena wmarsentthe
a typrofpark
6
Pol
. 2. N
Aults ompd GIject
nce way rkedted moA
ypicfileking
land
Num
Althopr
prehIS tts lithein
d aas
ost bAftercal e is gs a
d
mbe
ougrovihentypike ey a alms a nod
balar exBTpre
and
er o
gh nidednsive apar
are mono
desancxperTS lesend 15
f ho
nodd bve loand rkinmost
ode. aned rimlocanted5.3%
otels
des y tocathe
ng oore 110. On
nd oenti
mentatiod in% le
s lo
andthesatione Opor lpop
0 thn th
onlyity:ts won, n theisu
cate
d wse tn prpeneisupulhouhe
y 4 24
we cinc
he Fure-
ed w
waystworofinStrure lar usanothas a
4899conccludFig. -rela
K
with
s wo tyile. reetare
as wnds her a w9 wclud
ding3.
ated
Krzy
hin B
wereypes
ThtMaeas waycas
endway. waysdedg inOn
d.
yszto
(b)
BTS
e dis ofhereap careys. ses;d ar
Ths vsd thanputn av
of W
Gd
S sta
istinf obe is come be
Fo; onre Ahe ss. 2at itt frvera
Węc
dańs
ation
ngubjeca s
mmuetteror enly
ATMsam887t is omage,
cel
k
ns
uishcts tronunitr re
examy 3.Ms
me ap71 nuse
m bo, 22
ed hav
ng pty heprempl5 t– opplnodefuloth 2.1%
dueve prehas esenle, pthouoverlies des.l tonod
% o
e toto feread
ntedparusanr 6for
predesof o
o tebe enc
daptd byrkinnds500r tra
epa an
obje
echnme
ce bted y sh
ng is pa0 enam
are cnd wects
nicaergebetw
cerhows rearkintitisto
chaways wi
al lied tweenrtainwingepring ies
ops.
aracys. ithi
imito pn cn pg thesearareSh
teriComn B
itatiproateg
pattehe ante
rease rehops
isticmb
BTS
ionsvidgorernsaread a
s arepres ar
cs oine
S ar
s, de ry s. a, as re e-re
of ed re
F 4 4
fssok
pTd
d
w
Fig.
4. P
4.1
ficiesolusultof lkind
posTF-doc
deci
whe
. 3.
Pro
. T
Sient ute ts wlocad of
Ine t-IDFcum
TFided
ere
Ave
ofile
F-I
impto val
wouatiof ob
n orto uF is
mentF cad to
ni,j
erag
es o
IDF
ple corluesld b
ons. bjecrderuses act frean b
o use
is n
ge d
of l
F-in
aggrrecs. Fbe b
In cts mr toe Tctuaeqube ce re
num
distr
loca
nsp
gregctly For biasfac
mor all
TF-Ially
uenccalcelati
mbe
L
ribut
atio
pire
gatiproexa
sed ct, wre oleviIDFy a pcy (culaive f
r of
Link
tion
ons
ed
ion ofilampif w
we ofteiate
F wpro
(IDFated freq
f tim
ked
n of
s an
me
of e lople,we are
en the theweigoducF).
d in quen
mes
geo
f obj
nd
eth
geoocat as had
e inhane efghtict o
diffncy
s th
odat
jects
us
od
ogrationthe
d nonteren anffecing of tw
fferey (w
at t
ta fo
s am
ers
for
aphns. Rere ot ineste
n avct o
scwo
ent wwhic
the t
for p
mon
s
r lo
hicaRelare
nclued iveraof uchem
sta
waych is
term
profi
ng p
oca
l calative mudein i
age unevma atist
ys, es no
m ti
filing
rede
atio
ategve v
muched thinfouse
venkn
tics
e.g.orma
i occ
g of
efin
on p
gorivaluh mhis
ormer. n disnows: te
. boalis
cur
f tel
ned
pro
ies aues
morecor
matio
stribwn erm
ooleed t
rred
co u
30 c
ofili
assiare
e shrrecon,
butfro
m fre
ean oto 1
d in
user
cate
ing
igne mhopctioif g
tionm equ
or r1.0),
doc
rs
egor
g
ned moreps thon ingiv
n ofinf
uenc
raw, ex
cum
ries
to pe imhann then
f caformcy (
w frexpre
men
of o
placmpon libhe cuse
tegmati(TF
equeesse
nt dj
obje
ces ortanbrarcharers
orieion
F) an
encyd as
j.
ects
is nnt thries ractvisi
es wn rend
y. Ws fo
s
not hanthe
teriit s
we etrieinv
We hollow
20
sufn abe resticom
proevavers
havws:
7
f-b-e-cs
me
o-al. se
ve
2
n
wb
oPnTs
F T
208
nati
wheber
of tPitcnot TF-stric
Fig.
Tab
8
IDion
ere of A
the ch –ver
-IDFcted
. 4.
ble 1
DF ipow
|D|doc
An emo
– isry uF md lis
Per
1. S
is a wer
is cumentryost ps 4.2usefmethst o
rcen
amp
a meas
a nmeny topop234ful hod
of 3
ntage
ple
easuthey
numts c
o IDpula4. Cto
d ar0 ob
e of
pro
ure y ca
mbercontDF car cCondist
re gbje
f loc
file
of tan b
r oftaincalcateg
ncluting
givects.
catio
s of
termbett
f doningculagor
usionguisen in.
ons
f BT
m spter c
ocug thatiory –n: ssh ln th
con
TS l
K
pecichar
umehe teon i– Shsholocahe T
ntain
loca
Krzy
ificiract
entserms prhopps, atioTab
ning
ation
yszto
ity –teris
in m ti.
resep – i
preons. ble
g ve
ns c
of W
– lese a
cor
enteis 0esenSa
1. T
enue
calcu
Węc
ess fa do
rpu
ed i0.76nt i
ampThe
es (n
ulat
cel
freqocum
us, a
in F66 ain aple rey h
nod
ted w
quenmen
and
Fig.andalmresu
have
des)
with
nt tnt. I
d de
4. d forost ultse be
of g
h TF
ermt is
enom
TFr leha
s obeen
give
F-ID
ms hcal
min
-IDeast lf o
btainn bu
en c
DF
havelcula
nato
DF fpo
of thned
uilt
categ
e biated
or c
factopulhe
d frobas
gory
igged as
cont
tor ilar clocom sed
y
er ds fol
tain
in tcateatiotheon
discrllow
ns n
the egoonse ab
n the
rimiws:
num
casory , arbove re
i-
m-
se –
re ve e-
naSImqi5tT
mnbt
F 4
t1
4
neigare SomIn bmosqueinte5 mthosThi
meaneigbluethe
(a)
Fig.
4.2
tain10 B
4 Pl
ri
Thghbnod
me lboldst c
ent (erestmonuse cs is
Fiasurghbe – wh
Pol
. 5.
. U
FoningBTS
leasod c
heybourdes locad wchar(loctingumcate to ig. re.
bourhis
hole
land
Mo
User
or eg daS lo
e no
can b
y prrhoo
(poatio
we hractcatiog toents
egorsom5 pEa
rhostore Po
d
ost p
r pr
evalata focat
ote thbe c
rofiod ointns hhavterison io os anries me epresch od.
ricaolan
popu
rof
luatfor tion
hat t
consi
filesof tts), hav
ve mstic id 3bsend 8wi
extesentBT Th
al obnd (
ular
filin
tion3 m
ns a
this ider
L
inthe inc
ve jumark
for32),erve8 bth sent ts a
TS lhe mbjec(Fig
r ann
ng
n, wmonand
is il
red, a
Link
n thloc
cludust kedr a , thee thanksmajus
a vilocamoscts,
g. 5a
nota
we hnthsthe
llustany
ked
he catiodingone
d thgiv
e onhat ks. Wallertifieisuaatiost im oraa) a
ation
haves. Fen r
tratioothe
geo
Tabon g 2 e obhe cvenne w13 Weir nued –alisaon impangand
ns o
e usrom
rand
on oer se
odat
ble id 3bu
bjeccaten locwith
shightumb– suatiois aorta
ge –Gd
of B
sedm thdom
of thet of
ta fo
1 34 t
us stct (eegorcatih thopsts cber uch on oannoant – scdańs
BTS
d thehis dmly
he mf obj
for p
shothertop
e.g. ry wion.he hs incan of ccat
of totatobj
choosk (
loc
e dadatasel
methoject
profi
oulre as, 1id 3
with. W
highn lo
be catetegotop ted
bjecols,(Fig
(b)
catio
atababalect
od antype
filing
d bare 1 pl36)h th
Whenhest ocat
comegoorie
clawi
ts a, grg. 5
Gda
ons
basease wted u
nd ses ca
g of
be 4 glace, othe hn mt IDtionmparies
es classeith are reenb) a
ańsk
e wwe use
somean b
f tel
integeoge ofther high
manDF mn idareds geleares rthecod
n –are
k
with hav
ers4.
e debe us
co u
erprgrapf wor havhest
ny cmakd 41d beet hirly pranke todedpubatta
a sve q. Th
ecisised.
user
retephicorshve mt mcatekes 1 aretwighproked
op-rd asblicach
amquehe c
ons
rs
ed cal hipman
measegortopre
weener w
ofiled acranks foc trahed.
mpleeriedcalc
are
as obj, anny msureries
p of lessn loweige theccorked ollowansp.
d 1d ucula
arbi
foljectnd 1more, th are
f thes imocatghte lordinca
ws:por
0.0sersatio
itrar
llowts, a1 fure (huse eqe ranmpotions (c
ocatng tateg red
rt. F
000 s wn o
ry. A
ws: all ouel (e.gs bequankinortans, bcasetionto T
goryd –Figu
usewith of th
Any
in of tsta. id eingally ng.
ant but e id n. TF-y in– shures
ers at l
he c
othe
20
ththemation
41g th
freIt ithathe36)
-IDFn thhopss fo
conleaschar
er pe
9
he m n. ).
he e-is
an en ).
F he s, or
n-st r-
e-
2
aanv
F
bwhvis
210
acteas anumvisu
Fig.
bettwithhavvaluin Fshou
0
erisa wmbeuali
. 6.
Foter eh th
ve nue oFig.uld
ticsweiger oise p
Geo
or cemphis knormof g 7.
d fur
s ofghtef copro
ogra
comphakindmalgive
Grrthe
f usd sonnfile
aphi
mparasizd oliseden orouper im
sersum
nectes o
ical
risoze df chd thobjepingmpr
s ism oftionof us
Lin
on odiffeharthe vect g orov
s raf prons insers
nked
of uerents isvalutypf ce
ve th
atheofilnitias as
d D
userncess thues
pe inertahe r
r stles ateds a p
Data-
r prs an
hat as in n prain read
K
traiof vd wpie
-Ba
rofilnd saxes
chroficatedab
Krzy
ghtvisi
with cha
sed
les sims sh
hartsile hegoility
yszto
tforiteda g
art.
pro
mumilarhous inhasoriey of
of W
rward BTgiveFig
ofile
uch ritie
uld hn su vas orf th
Węc
rd. TS
en Bg. 6
e of
moes bhavuch aluer ev
he ch
cel
Thloc
BTS pre
f the
ore betwve th
a w 1.0venhar
he pcatiS. Sesen
e sam
suitweehe sway0. Tn redrt.
profons
Simnts
mpl
tabln usamy, t
The duc
file s, wilarpro
e us
le iuserme mthatsam
cing
of wherrly tofile
sers
s a s. O
meat usme g nu
f a re tto les f
s
radOneasurser use
umb
usethe ocafor
dar e of re. Twit
ers ber
er iswe
atiosam
chaf theTheth tare of
s preighns,
mple
art te prerefthe
comcat
repht iswe
e us
thatroblforehigmptego
ares the casers
t caleme, wghesareorie
ed he an s.
an ms we
st ed es
F C
Lueio
pJtbtttca
Fig.
Con
Linusererating or r
profJordtionblesthatto pthe catean e
. 7.
ncl
Inkedrs. Ttorsany
reveSo
filedann is s out areprephie
egorexte
Com
lusi
n thdGeThes. Ny buenuo fa
e. Inn, 20
mour ae pr
pareerarriesensi
mpa
ion
his eoDe m
Neveusine esar wn th012odelapprove aprchys whion
ariso
ns a
paDatameth
erthnesstimwe e fu
2). Illed
proavidepproy ofhenof
on o
and
apea anhod heles re
matihav
uturIt isd asach ed boprif can anthe
L
of u
d fu
er wnd a
hasess, equiionve re ws a gs a f
forby liateategnnote me
Link
user
utur
we alsos be theirin. app
we pgenfinir proca
e migoritatietho
ked
pro
re w
ha theeene el
ng p
plieplan
neraite mrofiatioixtues. ng od t
geo
ofile
wo
avee TF
n aplaboprof
d an toativemixilingns. ure.Asspeto h
odat
es on
ork
e cF-ID
pplieoratfile
a simo use prxturg: uWh Th of
ecifhier
ta fo
n ra
ontDF-ed ited of
mpse Lrobre ouserhat he mf nofic vrarc
for p
adar
tribu-bain tmenei
ple aLateabi
overrs awe
methw Lvenuchic
profi
cha
uteasedhe pethoighb
appent listr anare e nehodLinkue
cal L
filing
arts
d d meparod cbou
proaDirtic mn un
moeed d caked(e.gLDA
g of
– re
theethorticucan
urho
ach richmodnderodeto
an adGeg. bA (h
f tel
estr
mod fularn beood
forhlet del rlyi
elleddet
alsoeoDbothhLD
co u
ricte
methfor r see us, e.
r agAllwh
ing d astermo beDatah shDA
user
ed to
hodpro
ettinsed g. f
ggrlocahere
sets a
minee ima cohop A).
rs
o tw
d foofilings
unfor m
regaatioe eat of
mie ar
mproontaand
wo u
for ing of
nivemar
atioon (ach f topixture wovedainsd am
users
exof mo
ersarke
n toLDitem
picsure weigd b
s allmen
s
ploloc
obilelly
eting
o oDA) m os. Tof ghty cl intnity
oitatcatioe tefor
g pu
obta(B
of a This
catts alonsterm
y). T
tiononselcor prurp
ain lei,cores
tegollowsidemedThe
21
n os ano oprofilose
use Ngllecsemoriewinerindiatere i
1
of nd p-l-es
er g, c-
m-es ng ng te is
Krzysztof Węcel
212
References Abel F., Hauff C., Houben G.-J., Tao K. (2012), Leveraging User Modeling on the So-
cial Web with Linked Sata [in:] Web Engineering, Springer-Verlag, Berlin-Heidelberg, pp. 378-385.
Auer S., Lehmann J., Hellmann S. (2009), LinkedGeoData: Adding a Spatial Dimension to the Web of Data, ISWC 2009, Vol. 5823, Springer, Heidelberg, pp. 731-746.
Baccelli F., Bolot J. (2011), Modeling the Economic Value of the Location Data of Mo-bile Users, INFOCOM, IEEE, pp. 1467-1475.
Blei D.M., Ng A.Y., Jordan M.I. (2012), Latent Dirichlet Allocation, “Journal of Machine Learning Research”, Vol. 3(4-5), pp. 993-1022, doi:10.1162/jmlr.2003.3.4-5.993.
Cano A.E., Dadzie A.-S., Burel G., Ciravegna F. (2013), Topica-Profiling Locations through Social Streams. Semantic Technology, Springer-Verlag, Berlin-Heidelberg, pp. 290-305.
Chen F., Joshi D., Miura Y., Ohkuma T. (2014), Social Media-based Profiling of Busi-ness Locations, Proceedings of the 3rd ACM Multimedia Workshop on Geotagging and Its Applications in Multimedia, Orlando, FL, pp. 1-6.
Fawcett T. (2006), An Introduction to ROC Analysis, “Pattern Recognition Letters”, Vol. 27(8), pp. 861-874, doi:10.1016/j.patrec.2005.10.010.
Görnerup O. (2012), Scalable Mining of Common Routes in Mobile Communication Network Traffic Data [in:] J. Kay, P. Lukowicz, H. Tokuda, P. Olivier, A. Krüger (eds.), “Pervasive Computing”, Vol. 7319, Springer-Verlag London, pp. 99-106, doi:10.1007/978-3-642-31205-2_7.
Liu F., Janssens D., Cui J., Wang Y., Wets G., Cools M. (2014), Building a Validation Measure for Activity-based Transportation Models Based on Mobile Phone Data, “Expert Systems with Applications”, Vol. 41(14), pp. 6174-6189, doi: 10.1016/ j.eswa.2014.03.054.
Liu F., Janssens D., Wets G., Cools M. (2013), Annotating Mobile Phone Location Data with Activity Purposes Using Machine Learning Algorithms, “Expert Systems with Applications”, Vol. 40(8), pp. 3299-3311. doi:10.1016/j.eswa.2012.12.100.
Ostuni V.C., Gentile G., Di Noia T., Mirizzi R., Romito D., Di Sciascio E. (2013), Mobile Movie Recommendations with Linked Data [in:] Availability, Reliability, and Security in Information Systems and HCI, Springer, Berlin-Heidelberg, pp. 400-415.
Qu Y., Zhang J. (2013), Trade Area Analysis Using User Generated Mobile Location Data, Proceedings of the 22nd International Conference on World Wide Web, Re-public and Canton of Geneva, Switzerland, International World Wide Web Confer-ences Steering Committee, pp. 1053-1064, http://dl.acm.org/citation.cfm?id= 2488388.2488480 (accessed: 30.08.2015).
Ruotsalo T., Haav K., Stoyanov A., Roche S., Fani E., Deliai R., Mäkelä E., Kauppinen T., Hyvönen E. (2013), SMARTMUSEUM: A Mobile Recommender System for the Web of Data. “Web Semantics: Science, Services and Agents on the World Wide Web”, Vol. 20(0), pp. 50-67, doi:10.1016/j.websem.2013.03.001.
Song C., Qu Z., Blumm N., Barabási A.-L. (2010), Limits of Predictability in Human Mobility, “Science”, Vol. 327(5968), pp. 1018-1021.
Linked geodata for profiling of telco users
213
POWIĄZANE GEODANE DLA PROFILOWANIA UŻYTKOWNIKÓW TELCO
Streszczenie: Obserwuje się rosnące zainteresowanie geograficznym profilowaniem użytkowników, rozumianym jako łączenie danych geograficznych z anonimowymi pro-filami użytkowników. Profil jednostki zazwyczaj składa się z pojęć geograficznych oznaczonych wagami, odzwierciedlającymi względną ważność poszczególnych pojęć dla odróżniania użytkowników. Proponowana metoda profilowania użytkowników sieci komórkowych jest dwuetapowa. W pierwszej kolejności tworzone są profile stacji prze-kaźnikowych (BTS) na podstawie społecznie dostarczonych informacji geograficznych. Następnie te profile są wykorzystywane do uogólnienia zachowania użytkownika, wyni-kającego z analizy logów jego połączeń (CDR). Chmura danych powiązanych (linked data) jest wykorzystywana jako dodatkowe źródło wiedzy w procesie modelowania użytkownika. Słowa kluczowe: dane powiązane, profilowanie użytkownika, powiązane geodane, logi połączeń, użytkownik mobilny, telco, cdr, bts, lgd, osm.