174
Web o Science Resea ch Assis an :
Func ional analysis and usage
ecommenda ions
Ca los Lopezosa
Uni e si a de Ba celona, Spain
h ps://o cid.o g/0000-0001-8619-2194
Elisenda Aguile a-Co a
Uni e si a Pompeu Fab a, Spain
h ps://o cid.o g/0000-0003-0923-9192
Lluís Codina
Uni e si a Pompeu Fab a, Spain
h ps://o cid.o g/0000-0001-7020-1631
Juan-José Bo é-Ve icad
Uni e si a de Ba celona, Spain
h ps://o cid.o g/0000-0001-9815-6190
Lopezosa, C., Aguile a-Co a, E., Codina, L., & Bo é-Ve icad, J. J. (2025). Web o
Science Resea ch Assis an : Func ional analysis and usage ecommenda ions. In J.
Gualla , M. Vállez, & A. Ven u a-Cisquella (Coo ds). Digi al communica ion. T ends and
good p ac ices (pp. 174-189). Ediciones P o esionales de la In o mación. h ps://doi.
o g/10.3145/cu icom.13.eng
175
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
Abs ac
This chap e p o ides a comp ehensi e unc ional analysis o he Web o Science Resea ch
Assis an , a gene a i e AI ool in eg a ed in o Cla i a e’s academic da abase. Designed o
enhance esea ch wo k lows, he ool suppo s asks such as li e a u e e iews, opic explo a ion,
expe iden i ica ion, and jou nal selec ion. Th ough a de ailed examina ion o i s in e ace and
unc ionali ies, including hema ic summa iza ion, co-ci a ion mapping, and end isualiza ion,
his chap e highligh s he ool’s po en ial o s eamline scien i ic disco e y. Se e al p ac ical
use cases a e p esen ed, demons a ing i s capabili ies in op imizing sea ches, iden i ying
seminal pape s, and sugges ing ele an publica ion enues. Despi e i s inno a i e ea u es,
he ool’s use mus be guided by c i ical hinking, anspa ency, and e hical conside a ions.
Resea che s a e encou aged o iew he assis an no as a subs i u e, bu as a complemen
o schola ly inqui y. The chap e concludes by emphasizing he alue o e lec i e, laye ed
in e ac ion wi h AI o esponsibly in eg a e hese echnologies in o academic p ac ice.
Keywo ds
Gene a i e a i icial in elligence; Academic sea ch ools; Web o Science; Web o Science Re-
sea ch Assis an ; Li e a u e e iew au oma ion; Resea ch e hics and anspa ency.
1. In oduc ion
Web o Science is one o he mos impo an academic da abases in he wo ld, ecognized
o i s mul idisciplina y na u e and he quali y o i s selec ion c i e ia. I is a da abase ha
co e s e e y hing om he expe imen al sciences o he humani ies, including he social
sciences. Web o Science co e s documen s published om 1900 o he p esen day in nea ly
23,000 scien i ic jou nals, o alling a ound 180 million a icles, as well as abou 150,000 books
(Cla i a e, 2025a).
A he same ime as da abases like Web o Science ha e g own in impo ance, so oo ha e
gene a i e a i icial in elligence ools, which ha e apidly expanded and now pe mea e all
a eas o socie y — especially he scien i ic and academic sphe es. This chap e in oduces
Web o Science Resea ch Assis an , he a i icial in elligence in eg a ed in o he Web o
Science da abase.
The ollowing sec ions desc ibe he ool’s in e ace and analyze i s unc ionali ies. As will be
shown, Web o Science Resea ch Assis an has he po en ial o ans o m access o, explo-
a ion o , analysis o , and syn hesis o scien i ic in o ma ion. Howe e , i s use mus always be
accompanied by c i ical hinking and an e hical pe spec i e, which includes anspa ency.
2. Wha Is Web o Science AI Resea ch Assis an and how does
I wo k?
Web o Science AI Resea ch Assis an is a ool de eloped by Cla i a e ha uses gene a i e AI
models ained on he Web o Science Co e Collec ion da abase o suppo esea che s and
s uden s seeking o ad ance hei academic wo k (Cla i a e, 2025b).
176
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
This se ice goes beyond a adi ional academic da abase, unc ioning ins ead as an assis-
an ha can help (1) ind seminal and ele an a icles on a discipline o opic in seconds, (2)
s eamline ad anced asks such as li e a u e e iews, expe iden i ica ion, o jou nal selec ion
o publica ion, (3) explo e connec ions be ween academic concep s, au ho s, and scien i ic
a icles h ough in e ac i e isualiza ions like end maps and co-ci a ion ne wo ks, and e en
(4) pe o m scien i ic documen sea ches using na u al language and mul iple languages, hus
acili a ing access o scien i ic in o ma ion wi hou he need o ad anced sea ch ope a o s.
When a use submi s a que y, he ool e ie es he mos ele an documen s using seman ic
simila i y algo i hms and keywo d sea ches. I hen o ganizes he esul s based on ele ance
and gene a es esponses o summa ies using he con en o he selec ed a icles. Web o
Science Resea ch Assis an also allows bo h documen -based sea ches and syn hesis- ype
ques ions, ailo ing he esponse o ma o he use ’s needs.
Like o he ools such as Scopus AI (Aguile a-Co a e al., 2024a; 2024b), Elici (A oyo-Machado,
2024), Epsilon (Ren e al., 2025), Pe plexi y (To es-Salinas & A oyo-Machado, 2025), Cha GPT
(To es-Salinas e al., 2024; Bo é-Ve icad e al., 2024), Copilo (Lopezosa, 2023a; Bo é-Ve icad,
2024), o Sci e (Codina, 2024), Web o Science Resea ch Assis an educes he manual e o
equi ed o cons uc complex sea ches and acili a es he iden i ica ion o ends, esea ch
gaps, and non-ob ious connec ions be ween scien i ic wo ks.
In he ollowing sec ion, we p esen a unc ional analysis o he ool o illus a e, wi h exam-
ples, he possibili ies o e ed by Web o Science Resea ch Assis an o esea ch.
3. P ac ical use cases o Web o Science Resea ch Assis an
The Web o Science AI Resea ch Assis an o e s a a ie y o epo s and isualiza ions de-
signed o suppo academic esea ch. These ea u es allow use s o e icien ly analyze in o -
ma ion and gain isual insigh s in o ele an da a and esul s.
Fi s , when accessing he main Web o Science pla o m (Figu e 1), he use is p esen ed wi h
he p ima y in e ace, ocused on i s academic sea ch unc ion. A he op, he e is a menu ha
p o ides access o ea u es such as “Sea ch” and “Resea ch Assis an ,” along wi h op ions o
sign in o egis e . In ou case, o ac i a e he Web o Science AI se ice, we mus selec he
“Resea ch Assis an ” ab.
177
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
Figu e 1
Main Web o Science sea ch in e ace wi h ad anced op ions and esea ch assis ance.
Once we ha e accessed he “Resea ch Assis an ” esou ce, a new in e ace appea s in he
o m o an in e ac i e cha (Figu e 2). In he cen e o he sc een, he i ual assis an ini ia es
he con e sa ion wi h he message: “I am going o walk you h ough unde s anding a opic.
Wha is you opic o in e es o esea ch ques ion?,” in i ing he use o en e hei opic o
in e es o esea ch ques ion.
Figu e 2
Web o Science Resea ch Assis an in e ace, guiding he use in unde s anding a opic based on an ini ial ques ion.
178
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
On he le , he e is a sideba wi h he menu and he “Cha his o y” sec ion, whe e he cu en
con e sa ion appea s unde he i le “Unde s and a opic.” Na iga ion icons and sys em
op ions a e also isible.
In he lowe -le co ne , he e a e links o send eedback, ake a ou , o access in o ma ion
abou he esea ch assis an .
Once he “Resea ch Assis an ” in e ace has been desc ibed, we es i wi h a p omp : “Wha
is he ela ionship be ween SEO and he digi al media?,” indica ing ha he opic o explo e
is he ela ionship be ween sea ch engine op imiza ion (SEO) and digi al media.
The esul o his p omp (Figu e 3) shows he esponse gene a ed by he ool. A he op, he e
is a sec ion i led “O e iew,” whe e he main opics iden i ied in he abs ac s o he consul ed
a icles a e explained. In ou case, hese opics include sea ch engine op imiza ion (SEO) ech-
niques, con en and me ada a ea u es, and he impac o sea ch engines on digi al media. The
o e iew also highligh s he impo ance o SEO o imp o ing web isibili y, he ole o con en
in sea ch ankings, and how hese dynamics a ec jou nalism and academic esea ch.
Figu e 3
Thema ic summa y gene a ed by he Web o Science Assis an on he ela ionship be ween SEO and digi al media,
accompanied by key a icles on op imiza ion echniques and web isibili y.
179
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
Below he summa y, a lis o undamen al a icles (“Seminal pape s”) ela ed o he opic is
p o ided. Among hem a e key wo ks such as B in and Page (1998), wi h mo e han 6,900
ci a ions, and o he s udies on Google ankings, con en cha ac e is ics, and echniques o
inc ease online isibili y (Figu es 3 and 4).
Each en y displays he a icle i le, au ho s, sou ce, publica ion da e, numbe o ci a ions, and
a bu on o iew mo e de ails. Addi ionally, below his lis , he e is a bu on o iew mo e doc-
umen s and an in e ac i e sec ion i led “Wha would you like o see nex ?,” o e ing comple-
men a y explo a ion op ions. These include analyzing he e olu ion o publica ions o e ime,
iewing ela ed concep s h ough a opic map, checking he mos ci ed au ho s, and accessing
new speci ic p omp s ha p o ide a guided and explo a o y app oach o he assis an .
Figu e 4
Second pa o he hema ic summa y gene a ed by he Web o Science Assis an on he ela ionship be ween
SEO and digi al media.
In addi ion o hese op ions, he Resea ch Assis an sugges s o he p omp s such as
“Documen s o e ime g aph o sea ch engine op imiza ion” (Figu e 5). This ea u e displays
a g aphical isualiza ion. A he op, Cla i a e’s cha bo no es ha his ype o cha allows
use s o obse e publica ion ends on a speci ic opic o e ime, o e ing his o ical con ex ,
iden i ying eme ging sub opics, and e ealing shi s in academic in e es . Jus beside he
explana ion, use s can adjus he cha pa ame e s using a d opdown menu o he numbe o
yea s (in his case, 25) and ano he o isualiza ion se ings.
In he example shown, he ba cha illus a es he e olu ion o he numbe o documen s published
on “sea ch engine op imiza ion” om 1999 o 2024. A s eady inc ease is obse ed s a ing in
2000, wi h a no able ise beginning a ound 2008. The peak in publica ions occu s in 2022 and
2023, wi h o e 125 documen s pe yea , wi h a sligh decline in 2024. This ype o analysis helps
iden i y key momen s o heigh ened esea ch in e es in SEO, possibly linked o echnological
ad ancemen s, changes in sea ch engine algo i hms, o he g ow h o digi al communica ion.
180
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
Figu e 5
G aph o publica ions on sea ch engine op imiza ion (SEO) o e ime, showing he sus ained g ow h o academic
in e es om 1999 o 2024.
Figu e 6
Topic map on SEO isualizing ela ed concep s and hei in e connec ion, based on he co-occu ence o hemes
in he analyzed documen s.
181
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
Addi ionally, we can c ea e an in e ac i e opic map. In he example p esen ed, we once again
use he opic o sea ch engine op imiza ion (Figu e 6). A he op, i is explained ha his ype
o isualiza ion allows use s o see he ela ionship be ween concep s wi hin a esea ch ield,
based on he documen s co e ing hose opics. The bubbles ep esen di e en hemes, and
hei size indica es he numbe o documen s ela ed o each one, while he a ows show how
he concep s e e ence one ano he .
In ou example, he cen al node o he map is sea ch engine op imiza ion, which is connec ed
o e ms such as da a b each, collec i e in elligence, web sc aping, knowledge in eg a ion,
open sou ce in elligence, and pe sonal da a. O he ela ed and ele an opics a e also dis-
played, such as cybe c ime, gene al da a p o ec ion egula ion, au oma ed decision making,
ai ness, social ne wo k analysis, secu i y awa eness, and in elligence analysis.
Figu e 7
P o iles o he six mos ci ed au ho s in Web o Science ela ed o he opic o sea ch engine op imiza ion (SEO).
182
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
This ea u e is pa icula ly aluable, as i o e s a o m o ep esen a ion ha acili a es he
explo a ion o sub opics, he iden i ica ion o in e disciplina y connec ions, and he disco e y
o eme ging ends in he ield o SEO.
Ano he aluable ea u e o e ed by he Resea ch Assis an is he mos ci ed au ho s panel
(Figu e 7). In ou example, o he opic o sea ch engine op imiza ion, 84 ele an documen s
we e iden i ied, and he p o iles o he six au ho s whose wo ks ha e been mos ci ed wi hin
ha se a e displayed. Each au ho is ep esen ed in an indi idual ca d ha includes hei name,
ins i u ional a ilia ion, Web o Science esea che ID, he mos equen opics in hei publica ions
(up o 10), and a bu on o iew hei ull p o ile. This panel p o ides a clea o e iew o he
leading esea che s in he ield and allows use s o explo e hei con ibu ions in g ea e dep h.
Figu e 8
Thema ic and bibliog aphic analysis on SEO and digi al media, including a summa y o key echniques, impac on
jou nalism, and e e ences o seminal a icles.
189
Web o Science Resea ch Assis en : Func ional analysis and usage ecommenda ions
Ca los Lopezosa; Elisenda Aguile a-Co a; Lluís Codina; Juan-José Bo é-Ve icad
Digi al communica ion. T ends and good p ac ices
Bo é-Ve icad, J.-J, Fabei o, R., & Anglada, L. (2024). C eando un cha bo con Cha GPT como
sopo e a la ca alogación en biblio ecas, a chi os y cen os de documen ación. Compa ación
de modelos de lenguaje en e sión g a ui a y p emium. h ps://hdl.handle.ne /2445/213600
Bo é-Ve icad, J.-J. (2024). Copilo Mic oso y Cha GPT: Análisis de da os, dispe sión, elación
y co elaciones de a iables [ ídeo]. h ps://hdl.handle.ne /2445/213581
Cla i a e. (2025a). Resou ces o Lib a ians and Adminis a o s. h ps://cla i a e.libguides.
com/lib a ian esou ces/co e age
Cla i a e (2025b). Web o Science Resea ch Assis an . h ps://cla i a e.com/academia-go -
e nmen /scien i ic-and-academic- esea ch/ esea ch-disco e y-and- e e encing/web-o -sci-
ence/web-o -science- esea ch-assis an
Codina, L. (2024). In eligencia a i icial gene a i e pa a abajos académicos: mé odo de
análisis y compa ación de Elici , Epsilon, Pe plexi y, SciSpace y Sci e. Lluís Codina. h ps://
www.lluiscodina.com/in eligencia-a i icial-gene a i a-academia
Codina, L. (2025). In e acción c í ica escalonada con in eligencia a i icial: explo ando caminos
pa a usa la IA en la academia. Lluís Codina. h ps://www.lluiscodina.com/in e accion-c i ica-
ia
Fon -Julián, C. I., O duña-Malea, E., & Codina, L. (2024). Cha GPT Sea ch as a ool o schola ly
asks: e olu ion o de olu ion?. In onomy, 2(5). h ps://doi.o g/10.3145/in onomy.24.059
Lopezosa, C. (2023a). Bing cha : hacia una nue a o ma de en ende las búsquedas. Anua io
ThinkEPI, 17. h ps://doi.o g/10.3145/ hinkepi.2023.e17a04
Lopezosa, C. (2023b). La In eligencia a i icial gene a i a en la comunicación cien í ica: e os
y opo unidades. Re is a de In es igación e Inno ación en Ciencias de la Salud, 5(1), 1-5.
h ps://doi.o g/10.46634/ iics.211
O duña-Malea, E., & Cabezas-Cla ijo, Á. (2023). Cha GPT and he po en ial g owing o ghos
bibliog aphic e e ences. Scien ome ics, 128(9), 5351-5355. h ps://doi.o g/10.1007/s11192-
023-04804-4
Ren, Y., Mohammadi, L., & Codina, L. (2025). Aplicación de sis emas de IA en un abajo de
inal de más e uni e si a io: p esen ación de un caso. Ba celona: Uni e si a Pompeu Fab a.
Depa amen o de Comunicación. h p://hdl.handle.ne /10230/70069
To es-Salinas, D., & A oyo Machado, W. (2025). Flujos de abajo in eligen es con Cha GPT,
Pe plexi y y No ebookLM. h ps://doi.o g/10.5281/zenodo.14981646
To es-Salinas, D., Thelwall, M., & A oyo Machado, W. (2024). Cha GPT o Bibliome ics:
Po en ial applica ions and limi a ions. h ps://hdl.handle.ne /10481/92547
