Sustav za preporuku proizvoda za njegu kože temeljen na sadržaju

undergraduate thesis

• Title (english): Content-based recommender system for skincare products
• Author: Barbara Pašalić
• Mentor: Mario Brčić (mentor)
• Committee member: Mario Brčić (predsjednik povjerenstva)
• Committee member: Mihael Kovač (član povjerenstva)
• Committee member: Kristijan Poje (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Sustavi za preporučivanje popularni su alati koji znatno poboljšavaju iskustvo korisnika tijekom korištenja digitalnih platformi. U ovom radu implementiran je sustav za preporuku proizvoda za njegu kože temeljen na sadržaju. Sustavi za preporuku temeljeni na sadržaju preporučaju proizvode na temelju njihovih značajki i preferenca određenog korisnika. Nad podacima o sastojcima koji čine proizvode provedena je redukcija dimenzionalnosti pomoću tri različite metode: PCA, t-SNE i UMAP. Za ostvarenje sustava za preporuku korištena su dva algoritma: kosinusna sličnost i k-srednjih vrijednosti. Rezultati preporuka validirani su na temelju stvarne sličnosti zadanog i preporučenih proizvoda. Usporedbom rezultata dvaju implementiranih sustava utvrđeno je da daju iste preporuke za 4/5 zadanih proizvoda.

Abstract (english)

Recommendation systems are popular tools that significantly enhance the user experience when using digital platforms. A content-based recommendation system for skincare products has been implemented in this paper. Content-based recommendation systems recommend products based on their features and the preferences of a specific user. Dimensionality reduction was performed on ingredient data using three different methods: PCA, t-SNE and UMAP. Two algorithms were utilized to achieve the recommendation system: cosine similarity and k-means clustering. The recommendation results were validated based on the actual similarity between the given and recommended products. By comparing the results of the two implemented systems, it was determined that they provide the same recommendations for 4 out of 5 given products.

• Keywords: sličnost; kosinusna sličnost; k-srednjih vrijednosti; redukcija dimenzionalnosti; sustav za preporuku
• Keywords (english): similarity; cosine similarity; k-means; dimensionality reduction; recommender system
• Language: croatian
• URN:NBN: urn:nbn:hr:168:501690
• 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: 2023-07-13 10:26:20