Yera Gil, AinhoaPerona Balda, IñigoArbelaitz Gallego, OlatzMuguerza Rivero, Javier2024-11-272024-11-27production.39946https://dx.doi.org/10.26876/ikergazte.ii.03.16https://gordailua.ueu.eus/handle/123456789/2323Lan honetan UPV/EHUko matrikulazioaren inguruko eZerbitzuko nabigazioa aztertu dugu web-meatzaritzako tekniken bidez, nabigazio-saioen sailkapena era automatikoan egiten saiatu gara.Emaitzen arabera, baieztatu dugu maila handi batean definitutako arrakasta / porrota nabigazio-portaerak detektatzeko gai garela. Esaterako, arrakasta gisa etiketatu diren klusterretan dauden saioen% 90etik gora arrakasta motakoak dira eta porroten kasuan % 90 inguru. Gainera, gainbegiratutakoikasketa bidez, bi saio-motak era automatikoan desberdintzeko gai gara % 96ko asmatze-tasa batekin.Beraz, lan hau etorkizunean eZerbitzu hau hobetu ahal izateko oinarri ona dela uste dugu.In this work we have analyzed the enrollment eService navigation of the UPV/EHU and using datamining techniques we have attempted to automatically perform navigation sessions classification. Theresults show that we are able to detect the defined success and failure navigation behaviours. Forexample, more than 90 % of the sessions of the clusters labelled as success are of success type and inthe failure case, around 90%. Besides, using supervised learning we are able to automaticallydistinguish the two nabigation types with an accuracy rate of 96 %. Thus, we think that this researchis a suitable basis to improve the eService analyzed in a near future.eGobernuaeZerbitzuakweb erabilpenaren meatzaritzanabigazio-ereduak.eGovernmenteServicesweb usage miningnavigation models.InformatikaIngeniaritzaintroduction