Određivanje relevantnosti reklame koja se prikazuje korisniku na temelju njegovog upita u tražilici

master's thesis

• Title (english): Determining advertisement relevancy based on user's search query
• Author: Barbara Lugar
• Mentor: Mirjana Domazet-Lošo (mentor)
• Committee member: Mirjana Domazet-Lošo (predsjednik povjerenstva)
• Committee member: Tomislav Jagušt (član povjerenstva)
• Committee member: Marko Horvat (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-08, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Google Ads najpopularnija je platforma za oglašavanje putem interneta. Jedan od glavnih izazova s kojima se Google Ads susreće jest dodjeljivanje ocjene relevantnosti reklami s obzirom na upit korisnika. Model transformera temeljen na mehanizmu pozornosti ima sposobnost dobrog obuhvaćanja semantike teksta te je moguće rješenje problema kako dodijeliti ocjenu relevantnosti. Ovaj rad bavi se istraživanjem primjene BERT-a, predstavnika transformer arhitekture, za dobivanje ocjene relevantnosti. Također, demonstriran je rad Google Ads algoritma koji koristi princip aukcije druge cijene pomoću jednostavne web aplikacije.

Abstract (english)

Google Ads is currently the most popular platform for online advertising. One of the main challenges it faces is how to assign the quality score to ads, based on the user query. Since the transformer architecture became popular in the natural language processing community due to its power to explain text semantics, it could be a solution to the problem of calculating the quality score. In this thesis, I researched how can BERT, a transformers representative, be used with the Google Ads algorithm to calculate the quality score. I also demonstrated how Google Ads and its second price auction work through a simple web application.

• Keywords: Google Reklame; BERT; obrada prirodnog jezika; transformer arhitektura; sloj pozornosti; semantika
• Keywords (english): Google Ads; BERT; natural language processing; transformer architecture; attention layer; semantics
• Language: croatian
• URN:NBN: urn:nbn:hr:168:389143
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 12:55:14