Sustav preporučivanja restorana zasnovan na suradničkom filtriranju

undergraduate thesis

• Title (english): Restaurant Recommender System Based on Collaborative Filtering
• Author: Tin Oroz
• Mentor: Klemo Vladimir (mentor)
• Committee member: Klemo Vladimir (predsjednik povjerenstva)
• Committee member: Lucija Šikić (član povjerenstva)
• Committee member: Petar Afrić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-06-16, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Sustav preporučivanja je pod klasa sustava za filtriranje informacija, koji teži predvidjeti „ocjenu“ nekog predmeta od strane određenog korisnika na temelju informacija koje je prikupio o tom korisniku. Suradničko filtriranje je algoritam za preporučivanje koji se temelji na informacijama o ocjenama korisnika prema određenim proizvodima. Matrica korisnosti je skup podataka koji se sastoji od korisnik, predmeta, te ocjena korisnika nad određenim predmetima. Pomoću matrice korisnosti računamo međusobne sličnosti predmeta ili korisnika, te na temelju toga radimo preporuke. Uspješnost sustava mjeri se pomoću drugog korijena prosječne vrijednosti kvadratne pogreške, koji prikazuje razliku između pravih i predviđenih ocjena. Primjena ovakvih sustava je široka, od reklamiranja proizvoda, preporučivanja muzike, restorana i filmova, spajanja ljudi i sadržaja na društvenim mrežama i mnoge druge.

Abstract (english)

Recommendation system is a subclass of an information filtering system, which seeks to predict the “assessment” of an item by a user based on the information the system has collected about the user. Collaborative filtering is a recommendation algorithm based on information about user ratings for specific products. A utility matrix is a set of data consisting of user ratings over specific items. Using the utility matrix, we can calculate the similarities between items or users and based on that we can make recommendations. The accuracy of the system is measured using Root-Mean-Square error, which represents the difference between the real and predicted values in the utility matrix. The application of such systems is wide, from advertising products, recommending music, restaurants and movies, connecting people and content on social networks and many other.

• Keywords: Sustavi preporučivanja; suradničko filtriranje; kosinusna sličnost; Pearsonov koeficijent korelacije; matrica korisnosti; fenomen dugog repa; sustav preporučivanja zasnovan na sadržaju
• Keywords (english): Recommendation systems; collaborative filtering; cosine similarity; Pearson correlation coefficient; utility matrix; long tail phenomenon; content-based recommendation systems
• Language: croatian
• URN:NBN: urn:nbn:hr:168:467136
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 09:37:09