izpis_h1_title_alt

Aktivno pridobivanje vrednosti atributov s skupinskim priporočanjem
ID Špacapan, Blaž (Author), ID Kukar, Matjaž (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (1,22 MB)
MD5: 0A119CE328F39C0A0705CB2508DD7AD7

Abstract
Priporočilni sistemi se uporabljajo večinoma v komercialne namene. Po drugi strani pa se pri strojnem učenju pogosto srečujemo s problemi z manjkajočimi podatki, ki jih lahko naknadno pridobimo z nekimi meritvami. Tu bi lahko uporabili tehnike priporočilnih sistemov, da ugotovimo, katere podatke se nam najbolj splača pridobiti. Torej gre za aktivno pridobivanje značilk. Na podatkih najprej naučimo nek model strojnega učenja. S pomočjo tega modela izračunamo Shapleyeve vrednosti za atribute, ter te uporabimo kot ocene izdelkov za priporočilni sistem. Uporabniku priporočamo podmnožico meritev, ki imajo najvišje Shapleyeve vrednosti. Ugotovili smo, da se postopek na umetnih podatkih dobro obnese v primeru, ko v učni množici ni manjkajočih vrednosti, že nekaj teh pa malo pokvari rezultate. Razlog za to verjetno leži v načinu, kako računamo Shapleyeve vrednosti za manjkajoče podatke, in kako z njimi dela model strojnega učenja, vendar še ni v celoti jasen. Glede na rezultate na realnih podatkih smo zaključili, da je uspešnost naše metode precej odvisna od relacije med atributi in ciljno spremenljivko.

Language:Slovenian
Keywords:priporočilni sistemi, SVD, manjkajoče vrednosti, Shapleyeve vrednosti
Work type:Bachelor thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2024
PID:20.500.12556/RUL-161744 This link opens in a new window
Publication date in RUL:13.09.2024
Views:50
Downloads:8
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Active feature-value acquisition with collaborative filtering
Abstract:
Recommendation systems are mostly used for commercial purposes. On the other hand, in machine learning, we often encounter problems with missing data, which can be subsequently obtained with some measurements. Here, we could use recommendation system techniques to determine which data is most worthwhile to obtain. This is active feature acquisition. First, we train a machine learning model on the data. Using this model, we calculate the Shapley values for the attributes, and use these as product ratings for the recommendation system. We recommend a subset of measurements to the user that have the highest Shapley values. We found on artificial data that the procedure performs well in cases where there are no missing values in the training set, but even a few of these slightly degrade the results. The reason for this likely lies in the way we calculate Shapley values for missing data, and how the machine learning model works with them, but it is not yet fully clear. Based on the results on real data, we concluded that the performance of our method is highly dependent on the relationship between the attributes and the target variable.

Keywords:recommender systems, SVD, missing values, Shapley values

Similar documents

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

Back