Abumalloh, Rabab Ali; Nilashi, Meh bakhsh; Halabi, Osama; Ali, Raian
A icle
Does me a e se imp o e ecommenda ions quali y and cus ome
us ? A use -cen ic e alua ion amewo k based on he cogni i e-
a ec i e-beha iou al heo y
Jou nal o Inno a ion & Knowledge (JIK)
P o ided in Coope a ion wi h:
Else ie
Sugges ed Ci a ion: Abumalloh, Rabab Ali; Nilashi, Meh bakhsh; Halabi, Osama; Ali, Raian (2024) :
Does me a e se imp o e ecommenda ions quali y and cus ome us ? A use -cen ic e alua ion
amewo k based on he cogni i e-a ec i e-beha iou al heo y, Jou nal o Inno a ion & Knowledge
(JIK), ISSN 2444-569X, Else ie , Ams e dam, Vol. 9, Iss. 4, pp. 1-16,
h ps://doi.o g/10.1016/j.jik.2024.100569
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/327471
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0/
Does me a e se imp o e ecommenda ions quali y and cus ome us ?
A use -cen ic e alua ion amewo k based on he cogni i e-a ec i e-
beha iou al heo y
Rabab Ali Abumalloh
a
, Meh bakhsh Nilashi
b,d
, Osama Halabi
a
, Raian Ali
c,
a
Depa men o Compu e Science and Enginee ing, Qa a Uni e si y, Doha, Qa a
b
UCSI G adua e Business School, UCSI Uni e si y, 56000, Che as, Kuala Lumpu , Malaysia
c
College o Science and Enginee ing, Hamad Bin Khali a Uni e si y, Qa a
d
Cen e o Business In o ma ics and Indus ial Managemen (CBIIM), UCSI G adua e Business School, UCSI Uni e si y, Kuala Lumpu , Malaysia
ARTICLE INFO
A icle His o y:
Recei ed 10 Sep embe 2023
Accep ed 5 Sep embe 2024
A ailable online 30 Oc obe 2024
ABSTRACT
Recommenda ion agen s (RAs) ha e p o en o be e ec i e decision-making ools o cus ome s, as hey can
boos us and loyal y when cus ome s shop online. They can analyse la ge amoun s o da a using machine
lea ning algo i hms and p edic i e analy ics capabili ies o p o ide highly ele an ecommenda ions o
use s. In p e ious s udies, se e al app oaches ha e been implemen ed o efine and assess he e ec i eness
o hese agen s. As a new o m o i ual eali y uni e se, me a e ses can be seen as a new enue o
imp o emen s in he pe o mance o online RAs. By exploi ing he capabili ies o he me a e se and inco po-
a ing da a abou he use ’s beha iou and p e e ences, he pe o mance o hese sys ems can be enhanced
in e ms o he accu acy, di e si y, and no el y o he gene a ed ecommenda ions. The me a e se can p o-
ide isually appealing and in e ac i e ecommenda ions, and he e a e se e al po en ial ac o s ha can
a ec he cus ome ’s expe ience. The cogni i e-a ec i e-beha iou al heo y is used o de elop he p oposed
esea ch model. This s udy in es iga es he impac o he capabili ies o he me a e se on h ee quali y ac-
o s o RAs: di e si y, accu acy, and no el y. The influence o he quali y o he ecommenda ions on a ec i e
us and he influence o a ec i e us on cus ome loyal y a e also examined. In addi ion, as his is an
eme ging echnology, pe cei ed p i acy plays a c ucial ole in main aining use s’ us and confidence.
Hence, he mode a ing influence o pe cei ed p i acy on he ela ionship be ween he quali y and a ec i e
us o RAs is examined. The mode a ing impac o p oduc knowledge on he ela ionship be ween he indi-
idual pe cep ion o us and loyal y is in es iga ed. Da a we e acqui ed om 288 Malaysian esponden s
and analysed using he PLS-SEM me hod. The findings o his s udy show ha he capabili ies o he me a-
e se ha e a o able impac s on se e al quali y ac o s o he ecommende sys em, including accu acy,
di e si y, and no el y. Fu he mo e, hese quali y ac o s impac he pe cei ed quali y o RAs, which in u n
impac s cus ome us and loyal y. Pe cei ed p i acy ac s as a mode a o on he ela ionship be ween he
quali y o ecommenda ions and he indi idual’s pe cep ion o us .
© 2024 The Au ho (s). Published by Else ie España, S.L.U. on behal o Jou nal o Inno a ion & Knowledge.
This is an open access a icle unde he CC BY-NC-ND license
(h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/)
Keywo ds:
Me a e se
Recommenda ion agen s
Accu acy
No el y
Di e si y
Pe cei ed p i acy
Indus ial and inno a ion
JEL classifica ion:
C12: Hypo hesis Tes ing: Gene al
M15: IT Managemen
O39: Technological Change: O he
In oduc ion
An au oma ed ecommenda ion sys em, also called a ecommen-
da ion agen (RA), analyses he da a associa ed wi h a use and sug-
ges s p oduc s, se ices, and con en based on he use ’s pe sonal
p e e ences (Baghe i a d e al., 2017;Ga cía-S
anchez e al., 2020;
Gha ahighehi e al., 2021;Nilashi e al., 2020). The e a e many
applica ions o hese agen s in domains such as e-comme ce (Khan
e al., 2021), media (He ce-Zelaya e al., 2020;Nilashi e al., 2023),
a el (Renji h e al., 2020), news (Ka imi e al., 2018), and en e ain-
men (Ai en & Ag awal, 2023). RAs a e also playing an inc easingly
c ucial pa in indi iduals’decision-making p ocedu es (Scholz e al.,
2017;Wang e al., 2022), especially in he con ex o online e ail and
e-comme ce. An RA can o e aluable assis ance o cus ome s in he
highly compe i i e en i onmen o e-comme ce (Guo e al., 2014),
whe e he e is a as a ay o op ions ha o en o e loads he use ’s
capabili ies (Liu e al., 2022). In his case, an RA can assis he use in
Co esponding au ho .
E-mail add esses: [email p o ec ed] (R.A. Abumalloh),
[email p o ec ed] (M. Nilashi), [email p o ec ed] (O. Halabi),
[email p o ec ed] (R. Ali).
h ps://doi.o g/10.1016/j.jik.2024.100569
2444-569X/© 2024 The Au ho (s). Published by Else ie España, S.L.U. on behal o Jou nal o Inno a ion & Knowledge. This is an open access a icle unde he CC BY-NC-ND license
(h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/)
Jou nal o Inno a ion & Knowledge 9 (2024) 100569
Jou nal o Inno a ion
&Knowledge
h ps://www.jou nals.else ie .com/jou nal-o -inno a ion-and-knowledge
finding p oduc s ha will mee hei specific needs and p e e ences
mo e apidly and easily.
These agen s can analyse as amoun s o da a wi h he help o
machine lea ning algo i hms and p edic i e analy ic capabili ies, and
can p o ide highly a ge ed and ele an ecommenda ions ha can
inc ease cus ome loyal y and sa is ac ion. An RA wo ks by analysing
use da a (e.g., b owse his o y eco ds, sea ch que ies, and pas pu -
chases) in o de o build a p ofile o he use . To gene a e he lis o
ecommenda ions o he use , he RA uses a ious algo i hms such
as collabo a i e fil e ing (CF) (Kuo & Li, 2023), con en -based fil e ing
(CBF) (A oudi e al., 2021), and hyb id app oaches (Biswas & Liu,
2022). In a CBF algo i hm, he sugges ed i ems a e simila o i ems
he use has p e iously pu chased, while a CF algo i hm ecommends
i ems ha a e common among simila use s (A oudi e al., 2021). The
undamen al idea behind CBF is o sugges new p oduc s based on
hei a ibu es and he his o ical p e e ences o he cus ome (Kuo &
Cheng, 2022). Hyb id ecommenda ion app oaches combine bo h CF
and CBF o p o ide mo e p ecise and pe sonalised ecommenda ions
(Tewa i, 2020). The use o RAs has inc eased cus ome engagemen ,
e en ion, and e enue o many well-known online shopping pla -
o ms, such as Amazon.com, Alibaba.com, and Shein.com. Howe e ,
he e ficiency o an RA is de e mined by he quali y o he da a and
he algo i hms used o gene a e ecommenda ions. Use s who a e
p o ided wi h inaccu a e o i ele an ecommenda ions may lose
us in he agen and he company as a whole.
The p ima y objec i e o an RA is o p o ide cus omised selec ions
o use s based on hei p e e ences, in e es s, and pas beha iou ,
whe e se e al in e ela ed ac o s will impac on indi iduals’pe cep-
ions o he quali y o he gene a ed sugges ions. The accu acy o an
RA is e y impo an , as his has a di ec impac on he quali y o ec-
ommenda ions p o ided o use s (Zanon e al., 2022). Inaccu a e ec-
ommenda ions will nega i ely impac he use s’ us in hese
sys ems (Nilashi e al., 2016), which can lead o a dec ease in an indi-
idual’s engagemen and usage beha io . P e ious s udies ha e
shown ha he accu acy o ecommenda ions can ha e a significan
impac on hei o e all pe o mance. Schola s ha e wo ked on o he
a ious aspec s o he design o RAs in o de o imp o e use expe i-
ences such as di e si y and no el y. Di e si y ep esen s he sys em’s
capaci y o o e a a ie y o p oduc s, wi h he aim o b oadening he
consume s’p e e ences (Zanon e al., 2022). Ano he impo an ac-
o is he no el y o he ecommenda ions: an i em ha exac ly
ma ches he needs o he cus ome is ha dly a sui able choice i he
cus ome is al eady e y amilia wi h i (Kunkel & Ziegle , 2023). A
success ul RA should be able o nudge use s ou o hei "com o
zone", allowing hem o enjoy no el, di e se, ye s ill pleasan expe i-
ences (G a ino e al., 2019).
The accu acy o ecommenda ions is a ec ed by he quali y o
da a. Da a collec ion me hods, p ep ocessing echniques, and in eg a-
ion p ocesses all a ec he quali y o he da a used in RAs, meaning
ha i is essen ial o ensu e ha he da a used a e p ecise, ele an ,
and up- o-da e. P e ious s udies in he li e a u e ha e epo ed ha
he spa si y p oblem (a lack o in o ma ion abou use s’p e e ences,
in e es s, and beha iou ) has a majo e ec on he accu acy o ecom-
menda ions (Heida i e al., 2022;Joo abloo e al., 2021). Mos RAs
su e om his p oblem, as only a small pe cen age o i ems a e
a ed by use s and he e a e limi ed numbe s o in e ac ions be ween
use s and i ems. Fu he mo e, biased ecommenda ions ha a ou
popula i ems o e less popula ones can esul om a spa se da ase ,
po en ially leading o a loss o di e si y in ecommenda ions (Geng
e al., 2023). RAs may no be able o accu a ely de ec he use s’needs
and p e e ences, and add essing his issue in he con ex o ecom-
menda ion sys ems is an impo an a ea o esea ch.
The le el o use engagemen wi h an RA can be imp o ed wi h
he help o eme ging digi al echnologies (Ma ga is e al., 2018). P e-
ious esea ch has sugges ed ha he use o social media da a o
enhance he pe o mance o an RA can yield imp o ed esul s
(Nilashi e al., 2023), and o achie e his, RAs can be connec ed o a -
ious online communi ies, meaning ha use s o social ne wo king
si es ha e many op ions o e iewing and a ing p oduc s pu chased
online (Vazquez e al., 2023). Tex ual eedback can be used as a com-
plemen o use a ings, in o de o enhance he quali y o he da a
and he accu acy o ecommenda ions, he eby ensu ing ha hey
a e ailo ed o he p e e ences o he use s (Nilashi e al., 2023). In
addi ion, connec ing RAs o popula social media si es may encou age
mo e people o use hem by making i easie o hem o ecommend
se ices o hei iends and ollowe s.
As an eme ging echnology, he me a e se o e s new oppo uni-
ies o he enhanced ope a ion o RAs and he deli e y o mo e co -
ec and ele an sugges ions o use s. As a esul o he in e sec ion
be ween he eal and i ual wo lds in he me a e se, unique mecha-
nisms o bo h gene a ing and pe cei ing alue ha e eme ged (Man-
cuso e al., 2023). Wi h he in eg a ion o se e al inno a ions wi hin
he me a e se (El Hedhli e al., 2023), he ecommenda ions can ake
no el and unexpec ed o ms in e ms o design, p esen a ion, and
implemen a ion. Insigh s in o use p e e ences and a i udes can be
gained h ough he me a e se’s i ual se ing, which allows use s o
in e ac in eal ime, bo h wi h each o he and wi h digi al i ems. The
capabili ies o he me a e se in e ms o da a a ailabili y and en ich-
men s can enhance he pe o mance o RAs, which acco dingly influ-
ences he use ’s o e all expe ience. Howe e , u he in es iga ions
a e essen ial in o de o unde s and he possible impac s o he capa-
bili ies o he me a e se on he ways in which use s pe cei e he di -
e en quali y ac o s o RAs.
The p omising capabili ies o e ed by he me a e se a e associa ed
wi h significan p i acy conce ns, which can ha e a significan impac
on he cus ome ’s expe iences and he adop ion o echnology. P i-
acy conce ns ha e been linked o se e al ou come a iables in he
li e a u e, including he accep ance o echnology in heal hca e (Dha-
ga a e al., 2020), us , in en ion, and a i ude owa ds sea ch
engines (Palanisamy, 2014), and cus ome loyal y owa d online
shopping (Wong e al., 2019). The me a e se p ima ily o e s cus om-
e s g anula se ices based on mul iple dimensions o in e ac ion, in
which da a a e collec ed om nume ous enues, unlike he da a
ga he ed om a single enue by con en ional in e ne apps (S. Zhang
e al., 2023b). In gene al, businesses ga he cus ome s’p i a e in o -
ma ion o execu e ansac ions as well as o be e comp ehend hei
needs, equi emen s, and choices, so ha hey can ailo hei ad e -
ising campaigns o hem (Degu is e al., 2023). P i acy conce ns
g ow wi h he implemen a ion o RAs, as he olumes o da a ga h-
e ed and he use p ofiles expand, o p o ide classifie s wi h mo e
da a o lea ning and in e ence (Slokom e al., 2021). The huge ol-
umes o pe sonally iden ifiable in o ma ion ha will be exposed i
he p i acy o a Me a applica ion is comp omised will be much
g ea e han o con en ional In e ne applica ions, hus posing a
se e e isk o consume s’p i acy (Liu e al., 2021;Wang e al., 2021).
In a su ey conduc ed by S a is a (2022a), 87 % o Ame ican ci izens
said hey would be wo ied abou hei p i acy i Facebook we e o
be success ul in building he me a e se. In addi ion, 50 % o hose su -
eyed said hey we e conce ned ha hacke s would be able o impe -
sona e o he people oo easily, and 41 % belie ed i would be oo
di ficul o p ese e hei ue iden i y in he me a e se.
Due o hese unce ain ies, he accep ance o RAs in he me a e se
will equi e a ce ain le el o amilia i y wi h and comp ehension o
he p oduc s in ol ed. Cus ome s mus be knowledgeable abou he
p oduc s hey in e ac wi h, as he me a e se’s capabili ies a e de el-
oped u he . In adi ional RAs, he le el o knowledge o he p oduc
will impac how he cus ome s pe cei e he quali y o he ecom-
menda ions (Yoon e al., 2013). Based on he abo e discussion, we
can d aw up he ollowing esea ch ques ions:
i. How do me a e se capabili ies influence a ious quali y ac o s o
ecommende agen s (RAs)?
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
2
ii. How does pe cei ed p i acy a ec he ela ionship be ween ecom-
menda ions quali y and cus ome a ec i e us ?
iii. How does p oduc knowledge influence he ela ionship be ween
cus ome a ec i e us and cus ome loyal y?
The emainde o his esea ch is s uc u ed as ollows: Sec ion 2
in oduces he Cogni i e-A ec i e-Beha iou al heo y as he heo e -
ical ounda ion o he s udy. Sec ion 3 p esen s an o e iew o me a-
e se capabili ies in e ailing. Hypo hesis de elopmen is de ailed in
Sec ion 4. The s udy me hodology is ou lined in Sec ion 5, and he
esul s a e discussed in Sec ion 6. Finally, Sec ion 7 p esen s he con-
clusion and he esea ch con ibu ions o he s udy.
Theo e ical backg ound
Cogni ion-a ec -beha iou al model
Toge he wi h he heo y o planned beha io (Ajzen, 1991), he
echnology accep ance heo y has been deployed as a heo e ical
base in se e al con ex s o app aise an indi idual’s ea ly accep ance
o in o ma ion sys ems, inno a ions, and ools (Da is e al., 1985).
Howe e , ea ly accep ance canno uly eflec he sys em’s pe o -
mance, as a long-s anding ela ionship be ween he use s and he
sys em is he ue measu e o he sys em’s pe o mance (Bha ache -
jee, 2001). Resea che s ha e he e o e conside ed a h ee-s age
sequence o con inuous in en ion, om quali y ea u es o long-
s anding commi men (F ank e al., 2014). Laza us (1991) explo ed
he impac o he app aisal o in e io and con ex ual a iables on he
a ec i e esponse, which in u n influences coping ac ions. D awing
on Laza us (1991) amewo k o app aisal, Bagozzi (1992) in es i-
ga ed he link be ween a i ude and beha iou based on he in e ces-
sion o a ec i e a iables be ween he cogni i e app aisal and he
coping esponse. Bagozzi (1992) s essed he p ominence o he
mo i a ional link be ween a i ude and beha io in o ming a use ’s
desi e o pe o m an ac ion. Pe using his line o esea ch, Eagly and
Chaiken (1993) p esen ed he a i ude o ma ion heo y, in which
a i udes a e o med solely based on he impac o h ee olds: cogni-
ion, a ec , and beha io . Expanding on his heo e ical amewo k,
use beha io is assessed using a h ee- ie ed chain model. The
knowledge ha use s gain abou he echnology h ough cogni i e
a iables will o mula e hei iews. As he consume engages wi h
he echnology, cogni i e a iables a e amed o de elop belie s. The
le el o expe ience wi h he inno a ion c ea es a eeling o emo ional
desi e ha ames he emo i e pe cep ions. This heo y highligh s
he in e en ion o emo i e pe cep ions be ween belie s and ac ions.
Use s’emo ional pe cep ions can be influenced by hei posi i e o
nega i e e alua ion o he echnology, which in u n shapes hei
a ec i e esponse (Kwon & Vog , 2010).
In his esea ch, we adop he cogni ion-a ec -beha iou model
o explo e he ac o s ha influence a cus ome ’s loyal y owa d an
RA in he me a e se ac oss h ee s uc u ed laye s: (i) he app aisal
o he quali y o he ecommenda ions (cogni i e), ep esen ed by
di e si y, no el y, and pe cei ed accu acy; (ii) he a ec i e compo-
nen , ep esen ed by us ; and (iii) he beha iou al componen , ep-
esen ed by loyal y.
Cogni i e o e load heo y
Indi iduals may eel s essed and men ally exhaus ed i hei cog-
ni i e load exceeds a pa icula limi (Pang & Ruan, 2023). In o ma-
ion o e load may esul in nega i e e ec s such as in o ma ion
anxie y, in o ma ion a igue, and ension (Islam e al., 2022). Acco d-
ing o cogni i e load heo y, humans ha e limi ed men al capaci ies,
and a cogni i e load ha is oo high will impac hei abili y o
acqui e in o ma ion abou p oduc s o se ices, leading hem o ha e
nega i e opinions owa ds a good o se ice. Indi iduals a e
desc ibed as ‘cogni i e mise s’who equen ly hesi a e o pu o h
mo e e o han is essen ial. RAs ha e p o en hei e ec i eness in
add essing he p oblem o in o ma ion o e load in adi ional online
en i onmen s.
The p oblem o in o ma ion o e load has been linked o e-com-
me ce in se e al con ex s (Fang e al., 2021) whe e he en i onmen in
which he cus ome p ocesses he in o ma ion influences hei pe cep-
ions o p oduc s and se ices. In si ua ed cogni ion heo y, a he han
se ing as an in angible men al p ocess, in o ma ion p ocessing is
hough o ake place in e nally and ac i ely based on he en i onmen
in which he cus ome is loca ed (Semin & Ga ido, 2015). Use s can
be e unde s and he alue o a p oduc o se ice when hei expe i-
ences enable hem o ela e in angible ac s o ac ual in e ac ions and
eal-wo ld e en s (Fan e al., 2020). Online shoppe s o en ha e di fi-
cul y in seeing how hings will fi in o hei own unique en i onmen s
(Jung e al., 2015), which inc eases hei men al wo kload. Hence, i he
cogni i e load is oo g ea , use s will expe ience nega i e emo ions as a
esul o he gap be ween hei desi es and he influence o hei su -
oundings, which will ha e a de imen al impac on hei abili y o
make decisions (Liu & Goodhue, 2012).
In he me a e se, his challenge can be add essed by p o iding an
imme si e en i onmen whe e use s can in e ac wi h p oduc s o
se ices in a ealis ic and con ex ually ele an manne . Me a e se
ecommenda ions can help indi iduals o e come he p oblem o
in o ma ion o e load in di e en ways, as hey can p o ide a g ea e
esemblance o indi iduals’in e ac ions and in e p e a ions o
objec s in he eal wo ld. These ecommenda ions allow cus ome s
o o e lay i ual in o ma ion on o hei ac ual su oundings, hus
educing he cogni i e load equi ed o unde s and how a p oduc
will fi in o hei pe sonal con ex s.
Me a e se capabili ies in e ailing
Al hough he e is no single ag eed-upon desc ip ion o he me a-
e se, mos expe s concu ha i is a ne wo k o in e connec ed i -
ual se ings in which indi iduals can engage o mimic ac i i ies in
he eal wo ld (Pa k & Lim, 2023;Zallio & Cla kson, 2022). As he
name sugges s, he me a e se is no only a i ual enue bu a poin
a which he eal and i ual wo lds mee . The me a e se is also
made up o a collec ion o di e en pla o ms and inno a ions ha
wo k oge he o p o ide use s wi h an expe ience ha uni es bo h
he eal and he i ual ealms (Hazan e al., 2022). Th ough a combi-
na ion o di e se echnical in as uc u es, a ully de eloped me a-
e se can c ea e a pa allel en i onmen o cul u al exchange and
human engagemen (S ephens, 2021). De elopmen s in he me a-
e se ha e been acili a ed by he huge ad ancemen s in he a eas o
a ificial in elligence (AI), big da a (BD), i ual eali y (VR), and aug-
men ed eali y (AR), which a e all used o enhance he ac ual en i-
onmen and enhance he consume ’s expe ience in he me a e se
(Pa k & Lim, 2023). All o hese echnological inno a ions a e c ucial
o c ea ing and enhancing imme si e eali ies, as hey allow indi id-
uals o engage wi h elep esence, which o en a ec s he le el o
imme sion in he me a e se en i onmen (Giang Ba e a & Shah,
2023). The me a e se should p o ide indi iduals wi h a ealis ic
expe ience and enable hem o be imme sed in he i ual wo ld
(Dionisio e al., 2013).
The capabili ies o he me a e se a e expec ed o cause shi s in
he shopping expe iences o indi iduals in bo h he digi al and physi-
cal e ail sec o s by p o iding hem wi h engaging and imme si e
knowledge abou goods and se ices (Klaus & Kuppelwiese , 2023;
Se a alle e al., 2023). Since hey can b eak down he social and
physical ba ie s be ween cus ome s and b ands, he me a e se and
o he ully imme si e echnologies ha e he oppo uni y o s and
alone as inno a i e ad e ising ins umen s (Chekembaye a e al.,
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
3
2023). Luxu y businesses can use social media o communica e wi h
clien s in he me a e se and apply digi al ad e ising s a egies o
enhance hei pe o mance (Panga ka e al., 2023). VR, as a p omis-
ing ool o expe iencing he me a e se, is g owing in popula i y
among businesses (Sadamali Jayawa dena e al., 2023); o example,
Gucci has in oduced ‘Gucci Town,’an in e ac i e en i onmen
wi hin he ealm o he me a e se. The in e ac i e componen s o
Gucci Town include minigames, b owseable a gadge s, and he
Gucci s o e, in which consume s can pu chase clo hes o hei a a-
a s. In o de o c ea e highly ealis ic and enjoyable shopping expe i-
ences ha a e appealing o cus ome s, some me chan s ha e s a ed
wo king wi h game designe s, who we e ea ly pionee s in he ealm
o e ailing in he me a e se (Yoo e al., 2023). Fo ins ance, Uniqlo
collabo a ed wi h ideo game de elope s om Mojang S udios o
c ea e a collec ion o T-shi s wi h he Minec a heme, which we e
la e made a ailable in he physical wo ld as well as in online Mine-
c a ma ke places (Wa e s, 2020). The exis ence o a a a s is a c u-
cial elemen ha di e en ia es he expe ience in he me a e se om
ea lie online en i onmen s. Me a e se sys ems p o ide 3D a a a s
ha eflec he indi idual’s pe sonali y, and can be iewed by o he
indi iduals, in con as o cu en i ual wo lds whe e indi iduals
a e iden ifiable by hei ca dinali ies and pho os. Pla o ms ha e in e-
g a ed a a a s wi h a ious capabili ies o imp o e online in e ac-
ions, due o he impo ance o hese a a a s in he ealm o he
me a e se (Kim e al., 2023). Recen s udies indica e ha i ual
influence s on social media achie e a significan ly highe engage-
men a e compa ed o human influence s, and his end is expec ed
o in ensi y wi h he eme gence o he me a e se and he inc eased
p e alence o VR (El Hedhli e al., 2023;Li e al., 2023).
Resea ch model and hypo heses
Capabili ies o he me a e se and he no el y and quali y o RAs
The deg ee o which ecommenda ions a e pe cei ed as new, su -
p ising, o unexpec ed by a use is e e ed o as no el y (Ali Abumalloh
e al., 2020). P o iding a sui able le el o no el y is a di ficul ask o
RAs, as i necessi a es s iking a coun e balance be ween he in es iga-
ion o new p oduc s and he use o p e iously es ablished p e e ences,
while also main aining a specific accu acy le el. The usage o he me a-
e se can imp o e he no el y o ecommenda ions by p o iding a ich
and di e se en i onmen in which use in e ac ions and p e e ences
can ake place. In he me a e se, use s will be able o in e ac wi h
o he use s in ways ha a e no possible in he eal wo ld (Dong e al.,
2023). In addi ion, use s o he me a e se will be able o communica e
wi h each o he in a a ie y o ways, such as h ough cha ing and sha -
ing in o ma ion ega ding p oduc s, he eby opening up new a enues
o bo h e aile s and consume s (El Hedhli e al., 2023). These in e ac-
ions ha e he po en ial o gene a e a huge olume o in o ma ion
ega ding he use s’p e e ences, beha iou s, and ela ionships, he eby
enabling he me a e se o exploi use s’p e ious p e e ences. The da a
collec ed in his way can hen be employed in he da ase s o an RA o
op imize he no el y o he selec ions. Fo ins ance, an RA could make
use o in o ma ion gleaned om use in e ac ions in a digi al shopping
mall o make ecommenda ions o new and in e es ing p oduc s ha
he use has p obably ne e seen be o e. The me a e se has he capa-
bili y o eme ge as an inno a i e enue, pa icula ly in he con ex o
ecommenda ions gene a ed o use s. This inno a ion is eflec ed by
he da a, which a e collec ed in di e en o ms o enable he gene a ion
o mo e no el ecommenda ions. The no el y ac o will enable
enhanced expe ience h ough he gene a ion o unique ecommenda-
ions, which in addi ion o ma ching he use s’needswillopennew
choices o hem. Fo ins ance, p oduc ecommenda ions in he me a-
e se can conside he eal- ime in e ac i i y p o ided by he me a-
e se, and p o ide 3D isual e ec s o he p oduc s in addi ion o
ac ile sensa ions, unlike con en ional sys ems ha can only display
images o he goods o hype links o he esul s o he use (Wei e al.,
2023). O e all, exploi ing he capabili ies o he me a e se will imp o e
he no el y o ecommenda ions. F om s udies in he li e a u e ha
ha e explo ed he no el y o he ecommenda ions in con en ional e-
comme ce pla o ms, i is clea ha he deg ee o which ecommenda-
ionsa eno elisasignifican ac o de e mining he quali y o ecom-
menda ions p oduced by ecommende sys ems (Ali Abumalloh e al.,
2020). We he e o e p opose he ollowing wo hypo heses:
H1: The capabili ies o he me a e se ha e a posi i e impac on he
no el y o ecommenda ions.
H2: The no el y o ecommenda ions has a posi i e impac on he
quali y o ecommenda ions.
Capabili ies o he me a e se and he accu acy and quali y o RAs
The quali y o an RA can be examined in wo ways: h ough a sys-
em-cen ic o a use -cen ic e alua ion (C emonesi e al., 2013). In a
sys em-cen ic assessmen , esea che s a e in e es ed in he design
o and imp o emen s o he p edic i e echniques used in he RA,
which a e used o build a p oduc sugges ion lis , op imise he pe o -
mance o he RA, and hence gi e consume s a mo e ulfilling expe i-
ence (Knijnenbu g, 2012). Howe e , he p edic i e accu acy o he
ecommenda ion does no necessa ily mean a ulfilling use expe i-
ence, and i is essen ial o conside he con as be ween he ecom-
menda ions p esen ed o use s and he ac ual use s’choices, since
no all sugges ions will become use choices (Haz a i & Ricci, 2022).
In con as , he quali y o he RA is de e mined in a use -cen ic
assessmen using da a ga he ed om indi iduals ha ha e engaged
wi h he sys em ia di e en app oaches. The quali y o he ecom-
menda ions de e mined by sys em-cen ic me hodologies may p o-
duce conflic ing findings compa ed o he esul s o use -cen ic
me ics o he quali y o he sugges ions. Schola s ha e ecen ly
s a ed ha he objec i e o he RA should mo e beyond making accu-
a e p edic ions, hus indica ing he impo ance o conside ing use s’
pe cep ions o he quali y o he gene a ed ecommenda ions (Pu e
al., 2011). In his esea ch, we will ocus on use s’pe cep ions o he
accu acy o ecommenda ions, based on he ex en o which consum-
e s belie e he p esen ed ecommenda ions a e in ag eemen wi h
hei as es.
The capabili ies o he me a e se gi e ise o inno a i e
app oaches owa ds p o iding accu a e ecommenda ions, which a e
acili a ed h ough he collec ion o use s’da a om di e se channels,
he eby aiding he RA in gene a ing ecommenda ions ha accu-
a ely ma ch use s’ as es and p e e ences. The capabili ies o he
me a e se can acili a e use -cen e ed, conscious expe iences by suc-
cess ully adap ing isualisa ions ha mee use equi emen s, espe-
cially in e ms o iming, loca ion, and in o ma ion p esen a ion (Wei
e al., 2023). The capabili ies o he me a e se can be used o c ea e
accu a e au oma ed ecommenda ions based on use beha iou al
da a, such as clicks, acks, and gaze mo emen s (Lee, 2022). Fu he -
mo e, h ough an empi ical analysis, he capabili ies o he me a e se
can enable he a angemen o isual ecommenda ions in a way ha
does no obs uc he sigh o o he componen s o he sys em (Wei
e al., 2023). Use s can in e ac wi h and con ol hese ecommenda-
ions di ec ly o h ough non-playe cha ac e s. In elligen obo s, o
example, may o e guidance, engage wi h people h ough isualisa-
ion-based in e ac ion, and help use s pe cei e he me a e se wo ld.
In addi ion, p e ious s udies in he con ex o classical RAs ha e
epo ed ha pe cei ed accu acy impac s he o e all quali y o he
RA (Ali Abumalloh e al., 2020;Nilashi e al., 2016). Acco dingly, we
p opose he ollowing wo hypo heses:
H3: The capabili ies o he me a e se ha e a posi i e impac on he
accu acy o ecommenda ions.
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
4
H4: The accu acy o ecommenda ions has a posi i e impac on he
quali y o ecommenda ions.
Capabili ies o he me a e se and he di e si y and quali y o RAs
To achie e he mos e ec i e comp omise in e ms o he choice
be ween accu acy and di e si y, a g owing numbe o esea che s
ha e ea ed he asks o an RA as an op imisa ion issue, wi h accu-
acy and di e si y me ics as conflic ing goals (Ka abadji e al., 2018).
Di e sifica ion is a popula opic wi hin esea ch on RAs, as i assis s
in add essing he issue o o e fi ing and enhances use expe ience.
A high le el o di e si y in RAs means ha he sys em makes a a ie y
o p oduc selec ions ha a e cus omised o he use ’s equi emen s
and choices (De Biasio e al., 2023). To achie e high-quali y ecom-
menda ions, a di e se se o sugges ions is essen ial. Howe e , di e -
sifica ion impac s use s di e en ly depending on hei pe sonali ies:
people who a e open o ying new hings end o a ou a wide
ange o ecommenda ions. Di e si ying ecommenda ions o boos
use sa is ac ion by ep esen ing he comple e ange o consume
p e e ences is one o he p ima y objec i es o s udies on he opic o
di e si y (Kuna e & Po
z l, 2017). RAs ha o e a di e se se o ec-
ommenda ions can make use s mo e sa isfied in se e al ways (Beel
e al., 2013): fi s ly, di e se ecommenda ions can in oduce use s o
new p oduc s ha hey migh no ha e disco e ed on hei own; and
secondly, a a ie y o ecommenda ions can p e en use s om ge -
ing bo ed wi h he sys em (Nilashi e al., 2016). This can inc ease
use s’ us by p o iding hem wi h mo e op ions ha a e aligned
wi h hei in e es s.
The me a e se can help an RA o sugges di e se p oduc s and
unusual expe iences ha use s migh like, as i p o ides he RA wi h
mo e in o ma ion abou wha use s p e e . This in o ma ion is ga he ed
om use in e ac ions and engagemen , and can be used o sugges
new and di e se p oduc s. Mo eo e , he me a e se indi ec ly influen-
ces he di e si y o ecommenda ions by exposing use s o a b oade
ange o expe iences, which can shape use s’p e e ences, leading o
mo e di e se p oduc sugges ions. Inco po a ing me a e se capabili ies
in o an RA can he e o e esul in a iche and mo e sa is ying use
expe ience. Di e si y in ecommenda ions can be eflec ed h ough he
capabili ies o he me a e se in he o m o di e se se s o p oduc s,
di e se in e ac ion expe iences wi h he ecommenda ions, and di e se
o ms o isualisa ion o ecommenda ions.
In addi ion, he inno a i eness o he me a e se can help o
add ess he p oblems o da a spa si y and cold s a , which ha e
been linked o he di e si y o ecommenda ions in se e al s udies
(Reshma e al., 2016). To o e come hese issues, he me a e se p o-
ides se e al o ms o da a collec ion ha can aid in imp o ing he
di e si y o ecommenda ions, including dynamic da a, c oss-expe i-
ence da a, and da a collec ed om a a a s ep esen ing use s (Wei e
al., 2023). I is he e o e likely ha including he capabili ies o he
me a e se will inc ease he a ie y o ecommenda ions in RAs.
Building on esul s in he li e a u e in ega d o he impac o di e -
si y on he quali y o an RA (Nilashi e al., 2016), we p opose he ol-
lowing hypo heses:
H5: The capabili ies o he me a e se ha e a posi i e impac on he
di e si y o ecommenda ions.
H6: The di e si y o ecommenda ions has a posi i e impac on he
quali y o ecommenda ions.
Quali y, a ec i e us , and loyal y
The majo i y o online e aile s now use RAs o help indi iduals
make pu chases and o lessen he cogni i e s ains associa ed wi h
in o ma ion o e load. Howe e , only a limi ed numbe o s udies
ha e explo ed use s’loyal y in he se ing o e-comme ce RAs (Ali
Abumalloh e al., 2020;Yoon e al., 2013). Gi en he fie ce compe i-
ion in he digi al wo ld, loyal y o con inuing in en ion has been
highligh ed as an indica o o he e ec i eness o digi al enues
(Kau e al., 2020;Tseng e al., 2018). Au ho s in he li e a u e ha e
explo ed indi iduals’in e ac ions wi h online s o es o e ime using
objec i e me ics ha can be measu ed by he pu chase p opo ion;
howe e , pu chase p opo ion me ics ela ed o loyal y ha e been
c i iqued o being only a pa ial es ima ion o consume s’commi -
men (Ande son & S ini asan, 2003), meaning ha i is impo an o
examine he elemen s ha can encou age people o emain loyal o
online shopping si es and echnologies ocusing on di e en beha -
io al and a i udinal indica o s.
We adop ed he cogni i e-a ec i e-beha iou model, e e ed o
he e as he ABC model, o in es iga e cus ome s’loyal y owa d an
RA in he me a e se. This is a basic model ha ou lines a h ee-pa
s uc u e o a i udes, de ailing how people e alua e p oduc s and
se ices h ough cogni i e, a ec i e, and beha io al componen s
(Eagly & Chaiken, 1998). In he li e a u e, hese h ee componen s
ha e been explo ed wi h a ocus on a ious app aisal, a i udinal, and
beha iou al ac o s; o ins ance, Sandyopasana and Ali (2020)
adop ed his model o explo e he ac o s o quali y (cogni ion), us
and sa is ac ion, swi ching ba ie s (a ec i e), and loyal y (beha io ),
whe eas ano he s udy by Q. Zhang e al. (2023a) explo ed he pe -
cep ion o alue (cogni ion), sa is ac ion (a ec i e), and loyal y
(beha iou ) in he con ex o mobile paymen s.
In ou model, he quali y dimension o an online RA can p o oke
an a ec i e us owa d he RA, which can lead o loyal y owa ds
he RA. A ec i e us is defined as eelings o confidence owa ds
he se ice p o ide , which a e p o oked by he ca e and conce n he
se ice p o ide ex ends (Johnson & G ayson, 2005). In his esea ch,
loyal y is conside ed a beha iou al cons uc ha is eflec ed by he
in en ion o use he sys em again o he pu chase p ocess (Abumal-
loh e al., 2020). Loyal y is also eflec ed by he in en ion o eac in a
posi i e manne owa d online me chan s. The beha iou al in en ion
owa ds longs anding engagemen wi h he sys em is induced by
consume s’ eelings o confidence (a ec i e us ). Thus, ollowing
he ABC amewo k and building on he ex an li e a u e, we p esen
he ollowing hypo heses:
H7: The quali y o ecommenda ions has a posi i e influence on he
a ec i e us in he RA.
H8: Consume s’a ec i e us has a posi i e influence on use s’loy-
al y o he RA.
P i acy, ecommenda ions quali y, and us
Al hough he me a e se o e s immense po en ial ad an ages, i is
impe a i e o acknowledge ha he issues o p i acy and secu i y
equi e a en ion (Fe nandez & Hui, 2022), and ha conce ns ha e
been consis en ly aised in ega d o i ual sys ems in he li e a u e
(Ba h e al., 2022;Jeon & Lee, 2022;La a & Singh, 2022;Schomake s
e al., 2021;Soumelidou & Tsohou, 2021). These conce ns a e mo e
p e alen in he con ex o he me a e se. In a su ey ca ied ou in
he Uni ed S a es, a significan p opo ion o esponden s (71 %)
epo ed app ehension o e hei p i acy and secu i y in he me a-
e se (S a is a, 2022b). Mo eo e , 43 % o all pa icipan s ha bou ed
majo conce ns o e he po en ial he o hei eal iden i y in he
me a e se. A u he subs an ial p opo ion (app oxima ely 41 %)
held he belie ha sa egua ding hei da a wi hin he me a e se
would pose o midable di ficul ies.
P i acy conce ns ha e been explo ed in he li e a u e wi h a ocus
on bo h hei an eceden s and consequences, including in he con ex
o he me a e se. Se e al s udies ha e indica ed ha p i acy con-
ce ns in he me a e se s em om he collec ion and sha ing o use s’
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
5
da a (Alkaeed e al., 2023;Canbay e al., 2022). The me a e se is sub-
jec o se e al p i acy isks, such as insecu e design, b oken au hen i-
ca ion, da a injec ion, phishing, unau ho ised access, da a he ,
ea esd opping, and pe sonal in o ma ion leakage (Huang e al.,
2023). P i acy conce ns ha e been explo ed in con ex s beyond he
me a e se in he li e a u e, wi h a ocus on hei significan impac
on use s’beha io ac oss a ious se ings. The au ho s o (Sheehan &
Hoy, 1999) epo ed ha p i acy conce ns led o mo e conse a i e
beha io in e ms o pe sonal in o ma ion sha ing. In ano he s udy
(Phelps e al., 2001), p i acy conce ns we e ound o impac pu chase
beha iou and he pu chase decision p ocess (ca alogue pu chasing
habi s). Anic e al. (2019) also indica ed ha p i acy conce ns
impac ed bo h he ab ica ion o pe sonal in o ma ion and willing-
ness o sha e in o ma ion. In a s udy by Bansal and Zahedi (2008),
p i acy conce ns we e ound o mode a e he ela ionship be ween
he quali y o p i acy s a emen s and us . In his s udy, people wi h
high le els o p i acy conce ns we e ound o ely on he adequacy o
he p i acy policy s a emen . In ou esea ch, we aim o explo e how
p i acy conce ns mode a e he ela ionship be ween RA quali y and
cus ome us . Hence, we hypo hesise ha pe cei ed p i acy has a
mode a ing impac on he ela ionship be ween he ecommenda ion
quali y and cus ome us :
H9: Pe cei ed p i acy has a mode a ing impac on he ela ionship
be ween he quali y o ecommenda ions and cus ome a ec i e
us .
P oduc knowledge, us , and loyal y
In he ealm o online comme ce, goods a e equen ly ca ego ised
in o wo main ypes: sea ch i ems, which a e dis inguished by he
abili y o e i y hei a ibu es be o e pu chase, and expe ience
i ems, whe e he ocus is on he expe ien ial aspec , which can only
be e alua ed a e pu chase (Yoon e al., 2013). Goods classified as
’sea chable’a e hose ha o e compa a i ely s aigh o wa d ways
o e i y and e iew hei a ibu es p io o comple ing a pu chase;
in con as , he cha ac e is ics o expe ien ial i ems canno be exam-
ined o e ified easily be o e hey a e consumed. Mo ies o en se e
as p ime examples o expe ien ial goods. Nume ous s udies ha e
ound ha consume s a e mo e likely o heed he ad ice o RAs o
expe ience i ems han o sea ch i ems, as he e alua ion o expe i-
ence i ems is mo e complex han o sea ch i ems (Agga wal & Vai-
dyana han, 2003). The use ’s expe ience and knowledge o he
p oduc a e impo an when e alua ing di e en a ibu es o ecom-
menda ions, pa icula ly o expe ien ial goods o se ices.
In hei esea ch, Xiao and Benbasa (2007) disco e ed ha use s
wi h highe p oduc expe ise ended o ha e less a o able e alua-
ions o a CF RA. Pe e a (2000) explo ed he in e ac ion e ec s
be ween di e en ypes o RAs (CBF e sus CF) and use s’knowledge
abou p oduc classes. The findings e ealed ha cus ome s wi h less
knowledge o he p oduc s had mo e posi i e a ec i e eac ions,
such as sa is ac ion and a ec i e us , owa ds CF RAs compa ed o
CBF RAs. Cus ome s’le el o knowledge abou he p oduc can also
influence hei pe cep ions o he ecommenda ion, since expe i-
enced consume s may ely less on RA ecommenda ions due o hei
ex ensi e knowledge o he p oduc . Howe e , his may no be he
case in he con ex o he me a e se, as ad anced algo i hms and a i-
ficial in elligence app oaches a e equen ly used o o e consume s
ailo ed ecommenda ions, based on in o ma ion collec ed om
use s. Hence, cus ome s a e gi en ecommenda ions ha a e mo e
accu a e and in line wi h hei p e e ences and needs when hey
ha e a be e le el o p oduc knowledge.
P e ious s udies ha e explo ed he ela ionships be ween p oduc
knowledge, quali y, sa is ac ion, us , and loyal y in adi ional e-
comme ce. Fo ins ance, p oduc knowledge was ound o ha e a
mode a ing impac on he ela ionship be ween ecommenda ion
quali y and sa is ac ion in a s udy by Yoon e al. (2013), and on he
ela ionship be ween es au an s imuli (e.g., quali y) and dine s’
emo ions in a s udy by Peng and Chen (2015). In ou s udy, we aim o
examine he impac o p oduc knowledge on he ela ionship
be ween us and loyal y. Based on he abo e discussion, we
hypo hesise ha :
H10: P oduc knowledge has a mode a ing impac on he ela ionship
be ween a ec i e us and cus ome loyal y.
Based on he abo e discussion, and he p oposed hypo heses, we
p esen he ini ial esea ch model in Fig. 1.
Me hod o he s udy
Da a collec ion
In his s udy, we a ge ed pa icipan s in Malaysia h ough se e al
modes, including Wha sApp, LinkedIn, and Facebook. A he begin-
ning o he su ey, pa icipan s we e p o ided wi h he ollowing
link: h ps://www.you ube.com/wa ch? =0O IM0L7lLo. The ques-
ionnai e includes h ee main pa s: he fi s pa consis s o wo
sc eening ques ions, he second pa con ains demog aphic da a, and
he hi d pa includes he main su ey i ems. Those who did no
mee he sc eening c i e ia we e excluded om he su ey. To ensu e
da a comple eness, all su ey ques ions we e se as equi ed o com-
ple ion. Re e ing o he sample size, we ollowed he 10 imes ule o
humb by Ba clay e al. (1995), which sugges s ha he sample size
should be a leas 10 imes he la ge o : (1) he highes numbe o
o ma i e indica o s used o measu e a single cons uc , o (2) he
highes numbe o s uc u al pa hs di ec ed a a pa icula cons uc
in he s uc u al model. This ule o humb e ec i ely means ha he
minimum sample size should be 10 imes he maximum numbe o
a owheads poin ing a a la en a iable in he PLS pa h model. Pa -
icipan s we e asked o indica e hei le el o amilia i y wi h ecom-
mende agen s and hei knowledge o using he me a e se in he
e ail indus y. Responses om pa icipan s who indica ed hey we e
no a all amilia wi h ei he o hese opics we e excluded om he
s udy. The main su ey comp ises nine sec ions designed o measu e
esea ch hypo heses. Each i em in he su ey is assessed using a 5-
poin Like scale (see Appendix A). The su ey i ems allow pa ici-
pan s o exp ess hei a i udes owa d he a iables subjec i ely.
The da a collec ion p ocess spanned app oxima ely ou mon hs,
om Janua y 2023 o Ap il 2023. We ecei ed 288 alid esponses,
which we e used o analysis. The demog aphic da a analysis is p e-
sen ed in Table 1. As shown in Table 1, he majo i y o esponden s
we e male, wi h he mos common age ange being 36−40. Mos pa -
icipan s epo ed an a e age income o $751−$1500. Addi ionally,
mos esponden s we e mode a ely amilia wi h bo h he ecom-
mende sys em and he use o he me a e se.
Empi ical esul s
We used Sma PLS o conduc analyses on bo h he s uc u al and
measu emen models o ensu e he alidi y and eliabili y o ou
esea ch model. We adop ed a s uc u ed app oach o e alua e he
quali y o he collec ed da a and he significance o he hypo heses
using PLS-SEM. This me hod is obus , as i can handle complex models
in ol ing mul iple cons uc s, indica o s, and laye s o ela ionships,
making i well-sui ed o he con ex o ou s udy (Hai J e al., 2020).
I can also handle bo h eflec i e and o ma i e measu emen models.
S uc u al Equa ion Modeling (SEM) is used o assess he ela ionships
be ween independen and dependen a iables (Hai e al., 2013). The
quan i a i e esea ch communi y ecognizes SEM as a eliable me hod
o ac o analysis and pa h analysis. To ensu e he su ey yielded
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
6
accu a e and alid esul s, we conduc ed h ee main checks on he
ou e model using he ool: Con e gen Validi y (CV), In e nal Consis-
ency (IC), and Disc iminan Validi y (DV). Fo he CV es , all su ey
indica o s exhibi ed ou e loadings abo e 0.4. Hence, ollowing he
guideline se by Hai e al. (2013), we decided o e ain all i ems o u -
he analysis. The nex s ep in e alua ing Con e gen Validi y (CV)
in ol es he a e age a iance ex ac ed (AVE). The AVE es equi es
ha he co ela ion be ween i ems wi hin he same ac o mus mee a
minimum h eshold o 0.5. All ac o s in he p oposed model sa isfied
his AVE c i e ion. To assess In e nal Consis ency (IC) o he ou e
model, bo h Composi e Reliabili y (CR) and C onbach’s Alpha (CA) es s
we e employed. Each ac o in he s udy model needed o achie e a
minimum h eshold o 0.7 o bo h es s. The analysis confi med ha
he IC o he ou e model was suppo ed (see Table 2).
Se e al es s we e conduc ed o assess he Disc iminan Validi y
(DV) o he model: he He e o ai -Mono ai Ra io o Co ela ions
Fig. 1. Ini ial esea ch model.
Table 1
Demog aphic esul s o he pa icipan s.
I em N = 288
F equency Pe cen
Gende Female 69 24.0
Male 219 76.0
Age Unde 30 9 3.1
30 −35 52 18.1
36 −40 81 28.1
41 −45 64 22.2
46 −50 63 21.9
51 and o e 19 6.6
Income 0−500$ 21 7.29
$501 - $750 94 32.64
$751−$1500 113 39.24
$1501-$2000 31 10.76
Abo e 2000 29 10.07
Fa o i e E-Comme ce Websi e Alibaba 4 1.4
Amazon 46 16.0
eBay 3 1.0
Lazada 62 21.5
Mudah.my 36 12.5
Namshi 4 1.4
Shein 4 1.4
Shopee 27 9.4
Taobao 66 22.9
O he 36 12.5
Le el o Familia i y wi h he Rec-
ommende Sys em
High Familia i y 117 40.63
Mode a e Familia i y 157 54.51
Low Familia i y 14 4.86
Le el o Familia i y wi h he
Usage o he Me a e se in he
Re ail Indus y
High Familia i y 125 43.4
Mode a e Familia i y 147 51.04
Low Familia i y 16 5.56
Table 2
Cons uc s eliabili y and alidi y.
I em Ou e
loadings
C onbach’s
Alpha
Composi e
Reliabili y
AVE
Accu acy 0.722 0.825 0.550
ACC1 0.641
ACC2 0.548
ACC3 0.856
ACC4 0.869
Cus ome Loyal y 0.820 0.893 0.735
CL1 0.872
CL2 0.824
CL3 0.875
Di e si y 0.731 0.882 0.788
DIV1 0.888
DIV2 0.888
No el y 0.785 0.903 0.823
NOV1 0.898
NOV2 0.916
Pe cei ed P i acy 0.826 0.884 0.657
PPV1 0.783
PPV2 0.807
PPV3 0.836
PPV4 0.816
P oduc Knowledge 0.719 0.840 0.637
PRK1 0.834
PRK2 0.800
PRK3 0.758
Recommenda ion Quali y 0.844 0.895 0.681
RECQ1 0.844
RECQ2 0.797
RECQ3 0.847
RECQ4 0.811
T us 0.890 0.948 0.900
TRU1 0.954
TRU2 0.944
Me a e se Capabili ies 0.722 0.844 0.643
MC1 0.810
MC2 0.832
MC3 0.763
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
7
(HTMT), he Fo nell-La cke (FL) c i e ion, and C oss Loadings (CL).
The DV es e alua es he deg ee o di e en ia ion be ween esea ch
ac o s. The HTMT c i e ion measu es he a e age co ela ions
be ween indica o s ac oss di e en cons uc s. The FL es ensu es
ha he co ela ion be ween each ac o and o he ac o s in he
model is less han he squa e oo o he AVE o ha ac o . The CL
es equi es ha he ou e loadings o indica o s o each ac o be
g ea e han hei c oss-loadings. The esul s o he DV es s a e
de ailed in Tables 3,4−5.
In he nex s age, he ela ionships be ween esea ch a iables
we e examined. E alua ing he esea ch model in ol es es ing he
p oposed hypo heses using pa h coe ficien analysis, which is a c u-
cial s ep. Addi ional es s o he inne model include examining coe -
ficien s o de e mina ion and e ec size. The final inne model is
p esen ed in Figs. 2 and 3, and Tables 6 and 7. To assess he model’s
pa hs, a boo s apping p ocedu e was conduc ed (Hai e al., 2013).
The esul s confi m he significance o all esea ch hypo heses wi hin
he model. Analysis o he inne model showed ha he impac o
me a e se capabili ies on he accu acy o RAs is he s onges among
he esea ch pa hs, wi h a coe ficien o 0.650. This is ollowed by he
influence o us on use loyal y, which has a coe ficien o 0.552.
The model’s p edic i e accu acy was assessed using he R-squa ed
es , which e alua es he p opo ion o a iance in he endogenous
a iable explained by he exogenous a iables (Hai e al., 2013). R-
squa ed alues ange om 0 o 1, wi h highe alues indica ing
g ea e p edic i e accu acy. In his s udy, R-squa ed alues ange
om 0.111 o 0.492. Gi en ha he esea ch ocuses on consume -
o ien ed opics, specifically on unde s anding us and loyal y
among consume s, an R-squa ed alue o 0.2 is conside ed significan
(Hai e al., 2013).
The esul s indica e ha he accu acy, no el y, and di e si y o
RAs accoun o 48.8 % o he a iance in ecommenda ion quali y.
Addi ionally, he esea ch model explains 46.7 % o he a iance in
use loyal y. Fu he mo e, he model accoun s o 49.2 % o he a i-
ance in use s’ us .
The mode a ion e ec indica es ha he p esence o a hi d a i-
able can ei he s eng hen o weaken he ela ionship be ween an
endogenous ac o and an exogenous ac o (Hai e al., 2013). This
s udy p ima ily in es iga es how pe cei ed p i acy a ec s he ela-
ionship be ween RA quali y and us . Ou goal is o de e mine he
significance o he mode a o ’s impac on his ela ionship. To
achie e his, we employed a wo-s age app oach o ope a ionalize
he in e ac ion e ec using Sma PLS (Hai e al., 2013). The analysis
esul s (Fig. 4) e ealed ha pe cei ed p i acy mode a es he ela-
ionship be ween RA quali y and consume us . Specifically, a
highe le el o pe cei ed p i acy is associa ed wi h a s onge posi-
i e ela ionship be ween ecommenda ion quali y and cus ome
us , as indica ed by a significan be a coe ficien o 0.092. Con-
e sely, he mode a ion e ec o p oduc knowledge on he ela ion-
ship be ween us and loyal y was no suppo ed.
Table 3
He e o ai -mono ai a io (HTMT).
Cons uc ACC CL TRU DIV NOV PPV PPV RECQ MC
Accu acy
Cus ome Loyal y 0.771
Cus ome T us 0.716 0.742
Di e si y 0.669 0.850 0.751
No el y 0.687 0.735 0.605 0.769
Pe cei ed P i acy 0.880 0.876 0.760 0.838 0.807
P oduc Knowledge 0.840 0.568 0.409 0.562 0.576 0.601
Recommenda ion Quali y 0.709 0.742 0.693 0.735 0.695 0.803 0.461
Me a e se Capabili ies 0.867 0.605 0.563 0.458 0.520 0.685 0.579 0.557
Table 4
Fo nell-La cke c i e ion.
Cons uc ACC CL TRU DIV NOV PPV PPV RECQ MC
Accu acy 0.742
Cus ome Loyal y 0.604 0.857
Cus ome T us 0.579 0.636 0.949
Di e si y 0.482 0.659 0.607 0.888
No el y 0.517 0.595 0.513 0.585 0.907
Pe cei ed P i acy 0.702 0.724 0.655 0.652 0.655 0.811
P oduc Knowledge 0.597 0.447 0.334 0.415 0.439 0.477 0.798
Recommenda ion Quali y 0.577 0.617 0.609 0.581 0.573 0.673 0.370 0.825
Me a e se Capabili ies 0.650 0.467 0.452 0.333 0.389 0.529 0.423 0.437 0.802
Table 5
C oss loadings esul s.
I ems ACC CL DIV MC NOV PPV PRK RECQ TRU
ACC1 0.641 0.357 0.314 0.381 0.330 0.345 0.567 0.300 0.354
ACC2 0.548 0.354 0.324 0.296 0.336 0.400 0.319 0.306 0.349
ACC3 0.869 0.526 0.411 0.584 0.429 0.665 0.477 0.537 0.508
ACC4 0.856 0.523 0.385 0.592 0.436 0.601 0.445 0.508 0.484
CL1 0.574 0.872 0.582 0.423 0.565 0.640 0.433 0.550 0.583
CL2 0.461 0.824 0.529 0.353 0.422 0.621 0.345 0.518 0.520
CL3 0.510 0.875 0.581 0.422 0.534 0.600 0.365 0.517 0.528
DIV1 0.422 0.577 0.888 0.281 0.520 0.545 0.331 0.524 0.551
DIV2 0.434 0.592 0.888 0.309 0.518 0.613 0.405 0.508 0.527
MC1 0.568 0.405 0.262 0.810 0.299 0.420 0.384 0.341 0.332
MC2 0.534 0.351 0.273 0.832 0.309 0.413 0.299 0.392 0.411
MC3 0.456 0.368 0.266 0.763 0.330 0.441 0.334 0.317 0.344
NOV1 0.448 0.498 0.484 0.379 0.898 0.529 0.368 0.464 0.371
NOV2 0.488 0.577 0.573 0.329 0.916 0.654 0.426 0.571 0.551
PPV1 0.703 0.608 0.495 0.519 0.519 0.783 0.414 0.560 0.548
PPV2 0.585 0.602 0.502 0.484 0.552 0.807 0.415 0.485 0.515
PPV3 0.501 0.625 0.583 0.372 0.575 0.836 0.361 0.601 0.561
PPV4 0.478 0.504 0.532 0.335 0.471 0.816 0.354 0.531 0.492
PRK1 0.497 0.327 0.298 0.348 0.341 0.346 0.834 0.258 0.241
PRK2 0.514 0.415 0.384 0.370 0.383 0.473 0.800 0.351 0.312
PRK3 0.406 0.310 0.296 0.284 0.318 0.293 0.758 0.261 0.234
RECQ1 0.505 0.499 0.465 0.397 0.511 0.598 0.345 0.844 0.493
RECQ2 0.402 0.511 0.417 0.319 0.405 0.528 0.236 0.797 0.399
RECQ3 0.498 0.534 0.527 0.373 0.449 0.569 0.340 0.847 0.522
RECQ4 0.487 0.496 0.498 0.349 0.515 0.528 0.290 0.811 0.573
TRU1 0.563 0.623 0.599 0.443 0.535 0.655 0.337 0.607 0.954
TRU2 0.535 0.582 0.550 0.413 0.434 0.585 0.296 0.546 0.944
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
8
Heida i, N., Mo adi, P., & Koocha i, A. (2022). An a en ion-based deep lea ning me hod
o sol ing he cold-s a and spa si y issues o ecommende sys ems. Knowledge-
Based Sys ems, 256, 109835. doi:10.1016/j.knosys.2022.109835.
He ce-Zelaya, J., Po cel, C., Be nab
e-Mo eno, J., Tejeda-Lo en e, A., &
He e a-Viedma, E. (2020). New echnique o alle ia e he cold s a p oblem in
ecommende sys ems using in o ma ion om social media and andom decision
o es s. In o ma ion Sciences, 536, 156–170. doi:10.1016/j.ins.2020.05.071.
Hong, I. B., & Cho, H. (2011). The impac o consume us on a i udinal loyal y and
pu chase in en ions in B2C e-ma ke places: In e media y us s. selle us .
In e na ional Jou nal o In o ma ion Managemen , 31(5), 469–479. doi:10.1016/j.
ijin omg .2011.02.001.
Huang, W.-y., Sch ank, H., & Dubinsky, A. J. (2004). E ec o b and name on consume s’
isk pe cep ions o online shopping. Jou nal o Consume Beha iou , 4(1), 40–50.
doi:10.1002/cb.156.
Huang, Y., Li, Y. J., & Cai, Z. (2023). Secu i y and p i acy in me a e se: A comp ehensi e
su ey. Big Da a Mining and Analy ics, 6(2), 234–247.
Hwang, S., & Kim, S. (2018). Does mIM expe ience a ec sa is ac ion wi h and loyal y
owa d O2O se ices? Compu e s in Human Beha io , 82(2018), 70–80.
doi:10.1016/j.chb.2017.12.044.
Islam, A. K. M. N., M€
an ym€
aki, M., Laa o, S., & Tu el, O. (2022). Ad e se consequences o
emo ional suppo seeking h ough social ne wo k si es in coping wi h s ess om
a global pandemic. In e na ional Jou nal o In o ma ion Managemen , 62, 102431.
doi:10.1016/j.ijin omg .2021.102431.
Jeon, H., & Lee, C. (2022). In e ne o hings echnology: Balancing p i acy conce ns
wi h con enience. Telema ics and In o ma ics, 70, 101816. doi:10.1016/j. el-
e.2022.101816.
Johnson, D., & G ayson, K. (2005). Cogni i e and a ec i e us in se ice ela ionships.
Jou nal o Business Resea ch, 58(4), 500–507.
Joo abloo, N., Jalili, M., & Ren, Y. (2021). A p obabilis ic g aph-based me hod o sol e
p ecision-di e si y dilemma in ecommende sys ems. Expe Sys ems wi h Appli-
ca ions, 184, 115485. doi:10.1016/j.eswa.2021.115485.
Jung, T., Chung, N., & Tom Dieck, M. C. (2015). The de e minan s o ecommenda ions
o use augmen ed eali y echnologies - The case o a Ko ean heme pa k. Tou ism
Managemen , 49,75–86. doi:10.1016/j. ou man.2015.02.013.
Ka abadji, N. E. I., Beldjoudi, S., Se idi, H., A idhi, S., & Dhifli, W. (2018). Imp o ing
memo y-based use collabo a i e fil e ing wi h e olu iona y mul i-objec i e op i-
miza ion. Expe Sys ems wi h Applica ions, 98, 153–165. doi:10.1016/j.
eswa.2018.01.015.
Ka imi, M., Jannach, D., & Jugo ac, M. (2018). News ecommende sys ems −Su ey
and oads ahead. In o ma ion P ocessing & Managemen , 54(6), 1203–1227.
doi:10.1016/j.ipm.2018.04.008.
Kau , H., Pa u hi, M., Islam, J., & Hollebeek, L. D. (2020). The ole o b and communi y
iden ifica ion and ewa d on consume b and engagemen and b and loyal y in
i ual b and communi ies. Telema ics and In o ma ics, 46, 101321. doi:10.1016/j.
ele.2019.101321.
Khan, Z., Hussain, M. I., Il a , N., Kim, J., & Jeon, M. (2021). Con ex ual ecommende sys-
em o E-comme ce applica ions. Applied So Compu ing, 109, 107552.
doi:10.1016/j.asoc.2021.107552.
Kim, D. J., Fe in, D. L., & Rao, H. R. (2008). A us -based consume decision-making
model in elec onic comme ce: The ole o us , pe cei ed isk, and hei an ece-
den s. Decision Suppo Sys ems, 44(2), 544–564. doi:10.1016/j.dss.2007.07.001.
Kim, D. J., Song, Y. I., B ayno , S. B., & Rao, H. R. (2005). A mul idimensional us o ma-
ion model in B- o-C e-comme ce: a concep ual amewo k and con en analyses
o academia/p ac i ione pe spec i es. Decision Suppo Sys ems, 40(2), 143–165.
doi:10.1016/j.dss.2004.01.006.
Kim, D. Y., Lee, H. K., & Chung, K. (2023). A a a -media ed expe ience in he me a e se:
The impac o a a a ealism on use -a a a ela ionship. Jou nal o Re ailing and
Consume Se ices, 73, 103382. doi:10.1016/j.j e conse .2023.103382.
Kim, J., Jin, B., & Swinney, J. L. (2009). The ole o e ail quali y, e-sa is ac ion and e- us
in online loyal y de elopmen p ocess. Jou nal o Re ailing and Consume Se ices,
16(4), 239–247. doi:10.1016/j.j e conse .2008.11.019.
Klaus, P. P., & Kuppelwiese , V. (2023). A glimpse o he u u e e ail cus ome expe i-
ence −Guidelines o esea ch and p ac ice. Jou nal o Re ailing and Consume Se -
ices, 73, 103205. doi:10.1016/j.j e conse .2022.103205.
Knijnenbu g, B. P. (2012). Conduc ing use expe imen s in ecommende sys ems. In
P oceedings o he six h ACM con e ence on ecommende sys ems (pp. 3−4). h p://
deli e y.acm.o g.lib a y.iau.edu.sa/10.1145/2370000/2365956/p3-knijnenbu g.
pd ?ip=91.227.24.65&id=2365956&acc=ACTIVE SERVICE&-
key=4A313E3191DFF965.CD6AA1E1785C7CCE.4D4702B0C3E38B35.4D4702B0-
C3E38B35&CFID=849210383&CFTOKEN=51692793&__acm__=1515437624_.
Knijnenbu g, B. P., Willemsen, M. C., Gan ne , Z., Soncu, H., & Newell, C. (2012).
Explaining he use expe ience o ecommende sys ems. Use Modeling and Use -
Adap ed In e ac ion, 22(4-5), 441–504. doi:10.1007/s11257-011-9118-4.
Komiak, S. Y. X., & Benbasa , I. (2006). The e ec s o pe sonaliza ion and amilia i y on
us and adop ion o ecommenda ion agen s. MIS Qua e ly, 30(4), 941–960.
doi:10.2307/25148760.
Kuna e , M., & Po
z l, T. (2017). Di e si y in ecommende sys ems −A su ey. Knowl-
edge-Based Sys ems, 123, 154–162. doi:10.1016/j.knosys.2017.02.009.
Kunkel, J., & Ziegle , J. (2023). A compa a i e s udy o i em space isualiza ions o ec-
ommende sys ems. In e na ional Jou nal o Human-Compu e S udies, 172, 102987.
doi:10.1016/j.ijhcs.2022.102987.
Kuo, R. J., & Cheng, H.-R. (2022). A con en -based ecommende sys em wi h conside -
a ion o epea pu chase beha io . Applied So Compu ing, 127, 109361.
doi:10.1016/j.asoc.2022.109361.
Kuo, R. J., & Li, S.-S. (2023). Applying pa icle swa m op imiza ion algo i hm-based col-
labo a i e fil e ing ecommende sys em conside ing a ing and e iew. Applied
So Compu ing, 135, 110038. doi:10.1016/j.asoc.2023.110038.
Kwon, J., & Vog , C. A. (2010). Iden i ying he ole o cogni i e, a ec i e, and beha io al
componen s in unde s anding esiden s’a i udes owa d place ma ke ing. Jou nal
o T a el Resea ch, 49(4), 423–435.
La a, S., & Singh, D. (2022). In usion de ec ion sys em in cloud en i onmen : Li e a u e
su ey & u u e esea ch di ec ions. In e na ional Jou nal o In o ma ion Manage-
men Da a Insigh s, 2,(2) 100134. doi:10.1016/j.jjimei.2022.100134.
Laza us, R. S. (1991). Cogni ion and mo i a ion in emo ion. Ame ican Psychologis , 46
(4), 352–367. doi:10.1037/0003-066X.46.4.352.
Lee, U.-K. (2022). Tou ism using i ual eali y: Media ichness and in o ma ion sys em
successes. Sus ainabili y, 14(7), 3975.
Lenzini, G., an Hou en, Y., Huijsen, W., & Melenho s , M. (2010). Shall i us a ecom-
menda ion? Towa ds an e alua ion o he us wo hiness o ecommende si es:
5968 (pp. 121−128). LNCS. doi:10.1007/978-3-642-12082-4_16.
Li, H., Lei, Y., Zhou, Q., & Yuan, H. (2023). Can you sense wi hou being human? Com-
pa ing i ual and human influence s endo semen e ec i eness. Jou nal o Re ail-
ing and Consume Se ices, 75, 103456. doi:10.1016/j.j e conse .2023.103456.
Liang, Y., Zhang, X., Wang, H., & Liu, M. (2024). Use s’willingness o adop Me a e se
d awing on flow heo y: An empi ical s udy using PLS-SEM and FsQCA. Heliyon,
e33394. doi:10.1016/j.heliyon.2024.e33394.
Lin, H.-H., & Wang, Y.-S. (2006). An examina ion o he de e minan s o cus ome loy-
al y in mobile comme ce con ex s. In o ma ion & Managemen , 43(3), 271–282.
doi:10.1016/j.im.2005.08.001.
Liu, B. Q., & Goodhue, D. L. (2012). Two wo lds o us o po en ial e-comme ce use s
[21_Publica ion in e e eed jou nal]. Humans as Cogni i e Mise s, 23(4), 1246–
1262. doi:10.1287/is e.1120.0424.
Liu, J., Shi, C., Yang, C., Lu, Z., & Yu, P. S. (2022). A su ey on he e ogeneous in o ma ion
ne wo k based ecommende sys ems: Concep s, me hods, applica ions and
esou ces. AI Open, 3,40–57. doi:10.1016/j.aiopen.2022.03.002.
Liu, Q., Hao, Z., Peng, Y., Jiang, H., Wu, J., Peng, T., Wang, G., & Zhang, S. (2021). SecVKQ:
Secu e and e ifiable kNN que ies in senso −cloud sys ems. Jou nal o Sys ems
A chi ec u e, 120, 102300. doi:10.1016/j.sysa c.2021.102300.
Mancuso, I., Messeni Pe uzzelli, A., & Panniello, U. (2023). Digi al business model inno-
a ion in me a e se: How o app oach i ual economy oppo uni ies. In o ma ion
P ocessing & Managemen , 60,(5) 103457. doi:10.1016/j.ipm.2023.103457.
Ma ga is, D., Vassilakis, C., & Geo giadis, P. (2018). Que y pe sonaliza ion using social
ne wo k in o ma ion and collabo a i e fil e ing echniques. Fu u e Gene a ion
Compu e Sys ems, 78, 440–450. doi:10.1016/j. u u e.2017.03.015.
McNee, S.M., Riedl, J., & Kons an, J.A. (2006). Being accu a e is no enough: how accu-
acy me ics ha e hu ecommende sys ems. CHI’06 ex ended abs ac s on
Human ac o s in compu ing sys ems,
Nilashi, M., Ali Abumalloh, R., Samad, S., Minaei-Bidgoli, B., Hang Thi, H.,
Alghamdi, O. A., Yousoo Ismail, M., & Ahmadi, H. (2023). The impac o mul i-c i e-
ia a ings in social ne wo king si es on he pe o mance o online ecommenda-
ion agen s. Telema ics and In o ma ics, 76, 101919. doi:10.1016/j.
ele.2022.101919.
Nilashi, M., Asadi, S., Abumalloh, R. A., Samad, S., & Ib ahim, O. (2020). In elligen ec-
ommende sys ems in he COVID-19 ou b eak: The case o wea able heal hca e
de ices. Jou nal o So Compu ing and Decision Suppo Sys ems, 7(4), 8–12.
Nilashi, M., Jannach, D., Ib ahim, O.b., Es ahani, M. D., & Ahmadi, H. (2016). Recommen-
da ion quali y, anspa ency, and websi e quali y o us -building in ecommen-
da ion agen s. Elec onic Comme ce Resea ch and Applica ions, 19,70–84.
doi:10.1016/j.ele ap.2016.09.003.
O’Dono an, J., & Smy h, B. (2005). T us in ecommende sys ems.
Oli e , R. L. (1999). Whence consume loyal y? Jou nal o Ma ke ing, 63,33–44.
doi:10.2307/1252099.
Ozdemi , S., Zhang, S., Gup a, S., & Bebek, G. (2020). The e ec s o us and pee influ-
ence on co po a e b and—Consume ela ionships and consume loyal y. Jou nal
o Business Resea ch, 117, 791–805. doi:10.1016/j.jbus es.2020.02.027.
Palanisamy, R. (2014). The impac o p i acy conce ns on us , a i ude and in en ion
o using a sea ch engine: an empi ical analysis. In e na ional Jou nal o Elec onic
Business, 11(3), 274–296.
Pang, H., & Ruan, Y. (2023). Can in o ma ion and communica ion o e load influence
sma phone app use s’social ne wo k exhaus ion, p i acy in asion and discon-
inuance in en ion? A cogni ion-a ec -cona ion app oach. Jou nal o Re ailing and
Consume Se ices, 73, 103378. doi:10.1016/j.j e conse .2023.103378.
Panga ka , A., Pa el, J., & Kuma , S. K. (2023). D i e s o eWOM engagemen on social
media o luxu y consume s: Analysis, implica ions, and u u e esea ch di ec ions.
Jou nal o Re ailing and Consume Se ices, 74, 103410. doi:10.1016/j.j e con-
se .2023.103410.
Pa k, H., & Lim, R. E. (2023). Fashion and he me a e se: Cla i ying he domain and
es ablishing a esea ch agenda. Jou nal o Re ailing and Consume Se ices, 74,
103413. doi:10.1016/j.j e conse .2023.103413.
Peng, N., & Chen, A. H. (2015). Dine s’loyal y owa d luxu y es au an s: he mode a -
ing ole o p oduc knowledge. Ma ke ing In elligence & Planning, 33(2), 179–196.
Pe e a, R. E. (2000). Op imizing human-compu e in e ac ion o he elec onic com-
me ce en i onmen . Jou nal o Elec onic Comme ce Resea ch, 1(1), 23–44.
Phelps, J. E., D’Souza, G., & Nowak, G. J. (2001). An eceden s and consequences o con-
sume p i acy conce ns: An empi ical in es iga ion. Jou nal o In e ac i e Ma ke -
ing, 15(4), 2–17. doi:10.1002/di .1019.
Pu, P., Chen, L., & Hu, R. (2011). A use -cen ic e alua ion amewo k o ecommende
sys ems. In P oceedings o he fi h ACM con e ence on ecommende sys ems.
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
15
Renji h, S., S eekuma , A., & Ja ha edan, M. (2020). An ex ensi e s udy on he e olu ion
o con ex -awa e pe sonalized a el ecommende sys ems. In o ma ion P ocess-
ing & Managemen , 57,(1) 102078. doi:10.1016/j.ipm.2019.102078.
Reshma, R., Ambikesh, G., & Thilagam, P. S. (2016). Alle ia ing da a spa si y and cold
s a in ecommende sys ems using social beha iou . 2016 in e na ional con e -
ence on ecen ends in in o ma ion echnology (ICRTIT).
Sadamali Jayawa dena, N., Thaichon, P., Quach, S., Razzaq, A., & Behl, A. (2023). The pe -
suasion e ec s o i ual eali y (VR) and augmen ed eali y (AR) ideo ad e ise-
men s: A concep ual e iew. Jou nal o Business Resea ch, 160, 113739.
doi:10.1016/j.jbus es.2023.113739.
Sandyopasana,T.,&Ali ,M.G.(2020).Influence o pe cei ed quali y o mobile paymen
applica ion owa ds loyal y. Unde s anding digi al indus y (pp. 163−167). Rou ledge.
Scholz, M., Do ne , V., Sch yen, G., & Benlian, A. (2017). A configu a ion-based ecom-
mende sys em o suppo ing e-comme ce decisions. Eu opean Jou nal o Ope a-
ional Resea ch, 259(1), 205–215. doi:10.1016/j.ejo .2016.09.057.
Schomake s, E.-M., Bie mann, H., & Ziefle, M. (2021). Use s’p e e ences o sma home
au oma ion −in es iga ing aspec s o p i acy and us . Telema ics and In o ma ics,
64, 101689. doi:10.1016/j. ele.2021.101689.
Semin, G. R., & Ga ido, M. V. (2015). Socially si ua ed cogni ion. Theo y and explana ion
in social psychology (pp. 283−302). The Guil o d P ess.
Se a alle, F., Vanheems, R., & Viassone, M. (2023). Does p oduc in ol emen d i e con-
sume flow s a e in he AR en i onmen ? A s udy on beha iou al esponses. Jou nal o
Re ailing and Consume Se ices, 72, 103279. doi:10.1016/j.j e conse .2023.103279.
Sheehan, K. B., & Hoy, M. G. (1999). Flaming, complaining, abs aining: How online use s
espond o p i acy conce ns. Jou nal o Ad e ising, 28(3), 37–51.
Slokom, M., Hanjalic, A., & La son, M. (2021). Towa ds use -o ien ed p i acy o ecom-
mende sys em da a: A pe sonaliza ion-based app oach o gende ob usca ion o
use p ofiles. In o ma ion P ocessing & Managemen , 58,(6) 102722. doi:10.1016/j.
ipm.2021.102722.
Soumelidou, A., & Tsohou, A. (2021). Towa ds he c ea ion o a p ofile o he in o ma-
ion p i acy awa e use h ough a sys ema ic li e a u e e iew o in o ma ion p i-
acy awa eness. Telema ics and In o ma ics, 61, 101592. doi:10.1016/j.
ele.2021.101592.
S a is a. (2022). Po en ial secu i y conce ns acco ding o me a e se en husias s in he
Uni ed S a es as o Augus 2022. Re ie ed 2023 om h ps://www.s a is a.com/s a
is ics/1346474/us-me a e se-secu i y-conce ns-2022/
S a is a. (2022). Conce ns posed by he me a e se acco ding o adul s in he Uni ed S a es
as o Decembe 2021. Re ie ed Ma ch om h ps://www.s a is a.com/s a is ics/
1288065/uni ed-s a es-adul s-conce ns-abou - he-me a e se/
S ephens, D. (2021). The me a e se will adically change e ail. Business o ashion. .
Re ie ed June, 6 om h ps://www.businesso ashion.com/opinions/ e ail/ he-
me a e se-will- adically-change- e ail/.
Tewa i, A. S. (2020). Gene a ing i ems ecommenda ions by using con en and use -
i em based collabo a i e fil e ing. P ocedia Compu e Science, 167, 1934–1940.
doi:10.1016/j.p ocs.2020.03.215.
Tsai, Y. C., & Yeh, J. C. (2010). Pe cei ed isk o in o ma ion secu i y and p i acy in
online shopping: A s udy o en i onmen ally sus ainable p oduc s. A ican Jou nal
o Business Managemen , 4(18), 4057.
Tseng, F.-C., Pham, T. T. L., Cheng, T. C. E., & Teng, C.-I. (2018). Enhancing cus ome loy-
al y o mobile ins an messaging: Pe spec i es o ne wo k e ec and sel -de e mi-
na ion heo ies. Telema ics and In o ma ics, 35(5), 1133–1143. doi:10.1016/j.
ele.2018.01.011.
Vazquez, E. E., Pa el, C., Al id ez, S., & Siliceo, L. (2023). Images, e iews, and pu chase
in en ion on social comme ce: The ole o men al image y i idness, cogni i e and
a ec i e social p esence. Jou nal o Re ailing and Consume Se ices, 74, 103415.
doi:10.1016/j.j e conse .2023.103415.
Wang, S., Zhang, P., Wang, H., Yu, H., & Zhang, F. (2022). De ec ing shilling g oups in
online ecommende sys ems based on g aph con olu ional ne wo k. In o ma ion
P ocessing & Managemen , 59,(5) 103031. doi:10.1016/j.ipm.2022.103031.
Wang, Y., Pan, J., & Chen, Y. (2021). Fine-g ained secu e a ibu e-based enc yp ion.
Ad ances in c yp ology
−
c yp o 2021: 41s annual in e na ional c yp ology con e -
ence, c yp o 2021, i ual e en (p. 41) Augus 16−20, 2021, P oceedingsPa IV.
Wa e s, M. (2020). Why ideo games a e he nex e ail on ie . Re ie ed June 5, 2023
om h ps://www.mode n e ail.co/ e aile s/why- ideo-games-a e- he-nex -
e ail- on ie /
Wei, L., Wang, X., Wang, T., Duan, Z., Hong, Y., He, X., & Huang, H. (2023). Recommen-
da ion sys ems o he me a e se. Blockchains, 1(1), 19–33.
We sch, L. R. (2013). T us , sa is ac ion and loyal y in cus ome ela ionship managemen :
an applica ion o jus ice heo y: 4 (pp. 29−42). The Hawo h P ess, Inc..
doi:10.1300/J366 04n03_03.
Wong, W. P. M., Tan, K. L., Ida, A. K., & Lim, B. C. Y. (2019). The e ec o echnology us
on cus ome e-loyal y in online shopping and he media ing e ec o us wo hi-
ness. Jou nal o Ma ke ing Ad ances and P ac ices, 1(2), 38–51.
Xiao, & Benbasa (2007). E-comme ce p oduc ecommenda ion agen s: Use, cha ac-
e is ics, and impac . MIS Qua e ly, 31(1), 137–209. doi:10.2307/25148784.
Yoo, K., Welden, R., Hewe , K., & Haenlein, M. (2023). The me chan s o me a: A
esea ch agenda o unde s and he u u e o e ailing in he me a e se. Jou nal o
Re ailing, 99(2), 173–192. doi:10.1016/j.j e ai.2023.02.002.
Yoon, V. Y., Hos le , R. E., Guo, Z., & Guima aes, T. (2013). Assessing he mode a ing
e ec o consume p oduc knowledge and online shopping expe ience on using
ecommenda ion agen s o cus ome loyal y. Decision Suppo Sys ems, 55(4),
883–893. doi:10.1016/j.dss.2012.12.024.
Zallio, M., & Cla kson, P. J. (2022). Designing he me a e se: A s udy on inclusion, di e -
si y, equi y, accessibili y and sa e y o digi al imme si e en i onmen s. Telema ics
and In o ma ics, 75, 101909. doi:10.1016/j. ele.2022.101909.
Zanon, A. L., Rocha, L. C. D.d., & Manza o, M. G. (2022). Balancing he ade-o be ween
accu acy and di e si y in ecommende sys ems wi h pe sonalized explana ions
based on linked open da a. Knowledge-Based Sys ems, 252, 109333. doi:10.1016/j.
knosys.2022.109333.
Zhang, Q., A i fin, S. K., Richa dson, C., & Wang, Y. (2023). Influencing ac o s o cus-
ome loyal y in mobile paymen : A consump ion alue pe spec i e and he ole o
al e na i e a ac i eness. Jou nal o Re ailing and Consume Se ices, 73, 103302.
doi:10.1016/j.j e conse .2023.103302.
Zhang, S., Wang, Y., Luo, E., Liu, Q., Gu, K., & Wang, G. (2023). A aceable and e ocable
decen alized mul i-au ho i y p i acy p o ec ion scheme o social me a e se.
Jou nal o Sys ems A chi ec u e, 140, 102899. doi:10.1016/j.sysa c.2023.102899.
R.A. Abumalloh, M. Nilashi, O. Halabi e al. Jou nal o Inno a ion & Knowledge 9 (2024) 100569
16
