20.500.12556/RUL-140250 Priporočilni sistem za izdelke vsakdanje rabe z uporabo nevronskih mrež Recommender System for Everyday Products Using Neural Networks V okviru diplomske naloge je bil ustvarjen priporočilni sistem za napovedovanje izdelkov spletne trgovine. Uporabljeni podatki prihajajo iz podatkovne zbirke Instacart, ki je dostopna na spletni strani Kaggle. Priporočilni sistemi filtrirajo veliko količino podatkov in uporabnikom prikažejo le njim zanimivo vsebino. Za gradnjo priporočilnega sistema je bila uporabljena nevronska mreža, ki v tem področju še ni dosti raziskana, vendar obeta zelo pomenljive rezultate. Nevronsko mreža je zgrajena po principu skupnega filtriranja. Testiranje rezultatov je ločeno na izdelke, ki jih uporabnik še ni kupil, in izdelke, ki jih je uporabnik kupil že v preteklosti. Za testiranje novih izdelkov je uporabljena mera HR@10. Za že kupljene pa sta uporabljeni meri priklica in točnosti. Pridobljeni rezultati so primerjani z rezultati napovedanja najbolj popularnih izdelkov. Pri napovedovanju novih izdelkov je naš model dosegel boljše rezultate in slabše pri izdelkih, ki jih je uporabnik že kupil v preteklosti. As part of the diploma thesis, we created a recommender system for predicting items in an online shop. The data used come from the Instacart database, which is accessible on the Kaggle website. Recommender systems filter a large amount of data and show users only the content that interests them. The recommender system has been built using a neural network. This approach has not been researched much in this field, but it is yielding promising results. The neural network has been built using collaborative filtering. The testing of results is divided into items that the user has never bought and items that the user already bought in the past. For evaluating the new items, the metric HR@10 was used and for evaluating the items the user bought in the past, the metrics precision and recall were used. The obtained results were compared with the results of predicting the most popular items. Our model performed better when predicting new products and worse when predicting products that the user bought in the past. priporočilni sistem nevronske mreže napoved nakupov recommender system neural network purchase prediction true false false Slovenski jezik Angleški jezik Diplomsko delo/naloga 2022-09-13 20:25:06 2022-09-13 20:25:10 2022-09-16 06:58:23 0000-00-00 00:00:00 2022 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 34958 121575939 Grzinic_Matej_-_Priporocilni_sistem_za_izdelke_vsakdanje_rabe_z_uporabo_nevronskih_mrez.pdf Grzinic_Matej_-_Priporocilni_sistem_za_izdelke_vsakdanje_rabe_z_uporabo_nevronskih_mrez.pdf 1 1D0A0B6EEE81609B721CD5AF940E0233 1c44e20d6d2f4d634a34cd7a62ec5c1659a4c8029d85fb2efce37a347fe65312 55fc6b9b-3391-11ed-92af-00155dcfd717 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=160957 Fakulteta za računalništvo in informatiko 0 0 0