izpis_h1_title_alt

Napovedovanje razmerja med prikazi in kliki oglasov s faktorizacijskimi metodami
ID DOVŽAN, ROBERT (Author), ID Šubelj, Lovro (Mentor) More about this mentor... This link opens in a new window, ID Kopič, Davorin (Comentor)

.pdfPDF - Presentation file, Download (1,66 MB)
MD5: 5B3A780D044B9B0EB4E2C74DEA01966C

Abstract
V panogi programatičnega spletnega oglaševanja, ki temelji na ekosistemu izbire oglasov v realnem času (angl. Real-Time Bidding) je pomembno napovedati kako uspešen bo prikaz oglasa uporabniku. Napovedovanje razmerja med prikazi in kliki (angl. Click-Through Rate prediction) oziroma verjetnosti klika je eden večjih izzivov v spletnem oglaševanju. V diplomski nalogi se lotimo napovedovanja verjetnosti z uporabo faktorizacijskih metod na podlagi podatkov, ki jih poznamo o oglasu, spletni strani, uporabniku ipd. Opišemo celoten proces obdelave podatkov, izbire značilk, implementacije in testiranja. Cilj naloge je v podjetju Zemanta d.o.o. izboljšati obstoječo rešitev, ki temelji na logistični regresiji. Z lokalnim testiranjem in testiranjem v produkcijskem okolju v obliki A/B testa naš cilj dosežemo in s tem prispevamo k izboljšanju storitve in večjemu finančnemu izkupičku podjetja.

Language:Slovenian
Keywords:faktorizacijske metode, oglaševanje, oglasi, napovedovanje, strojno učenje, podatkovno rudarjenje
Work type:Bachelor thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2019
PID:20.500.12556/RUL-107319 This link opens in a new window
Publication date in RUL:29.03.2019
Views:2004
Downloads:499
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Predicting the click-through rate of ads using factorization machines
Abstract:
In the field of programmatic advetising based on the ecosystem called realtime bidding, it is important to know, how successful an ad impression will be. Click-through rate prediction is one of the biggest challenges in online advertising. In this thesis we use factorization machines to predict the clickthrough rate based on data about the ad, website, user etc. We describe the process of data preparation, feature selection, implementation and testing. The goal is to improve the current solution in company Zemanta d.o.o. which is based on logistic regression. With local testing and online A/B testing we reach our goal and contribute to improving the service and financial performance of the company

Keywords:factorization machines, advertising, ads, prediction, machine learning, data mining

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back