Šiuo metu internetiniai aukcionai valdo didelius informacijos srautus, tačiau visa turima informacija yra panaudojama nevisiškai efektyviai arba nepakanka laiko priimti reikiamus sprendimus realiu laiku, taikant turimas technologijas. Problema atsiranda todėl, kad turima informacija nėra tinkamai analizuojama aukciono pirkėjo agento, tai yra visiškai neįvertinama skirtingų vartotojų aplinka ir jų specifinė elgsena, kuri formuoja visiškai kitokius poreikius Interneto aukciono vartotojo aplinkai. Darbe yra pateikiamas aukciono pirkėjo agento modelis formuojantis aukciono dalyvio grafinę sąsaja pagal jo elgsenos modelį. Modelis suformuotas, pritaikant neuroninių tinklų (Kohoneno žemėlapius) ir agentinių sistemų technologijas. / Nowadays internet auctions manage massive data flow, however, all available data is not being used efficiently or there is not enough time to make needed decisions applying available technology. The problems occur because available data is not analysed properly, that is graphic users’ interface and their specific behaviour, which form different user‘s needs for internet auction interface, are not asessed. In order to improve internet auction efficiency and to optimizme its operation, the auction buyer agent is applied in a way that GUI satisfy user‘s (auction participant‘s) demands. This technique considers user‘s demands accourding to specific criteria such as user‘s nationality , his/her loyality to particular auction system, the statistics of successful auctions using the clustering method, which present GUI in a user friendly way. This technique would increase both auction buyer agent and auction seller agent‘s efficiency.
| Identifer | oai:union.ndltd.org:LABT_ETD/oai:elaba.lt:LT-eLABa-0001:E.02~2008~D_20140623_182143-78862 |
| Date | 23 June 2014 |
| Creators | Balčius, Evaldas |
| Contributors | Dilijonas, Darius, Vilnius University |
| Publisher | Lithuanian Academic Libraries Network (LABT), Vilnius University |
| Source Sets | Lithuanian ETD submission system |
| Language | Lithuanian |
| Detected Language | English |
| Type | Master thesis |
| Format | application/pdf |
| Source | http://vddb.library.lt/obj/LT-eLABa-0001:E.02~2008~D_20140623_182143-78862 |
| Rights | Unrestricted |
Page generated in 0.002 seconds