Temporal Pattern-Based Approach for Personalized Next-Basket Recommendation on Fashion Products

Bhaswara, Julius Adetya Eka (2023) Temporal Pattern-Based Approach for Personalized Next-Basket Recommendation on Fashion Products. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 05111942000026-Undergraduate_Thesis.pdf] Text
05111942000026-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2025.

Download (3MB) | Request a copy


Fashion is one of the industries that continues to grow and undergo rapid changes. Today's consumers want fashion products that suit their needs and lifestyle, so there is a need for personalization in fashion products. However, choosing the right product can be a challenge for some consumers, due to the large number of options available and sometimes not according to their needs and style. Something that is more appealing to the eye will pique human curiosity. The fashion industry has grown over time because of this human inclination. The retail sector is developing with investments in the newest technologies to better their business as recommender systems have been introduced in several industries. Fashion has been around for millennia and will continue to be popular soon. Humans are more associated with fashion and style, and they have a wider range of options from which to choose. Because people are now more frequently judged based on their appearance, it has become an important component of life for modern families. Therefore, there is a need for a system that can help consumers in predicting the fashion products to be purchased and according to their needs and style. This paper presents a personalized fashion prediction method using temporal patterns. This method uses three popular types of recurrent neural network (RNN) models, namely Time Decay and Gated Recurrent Unit (GRU). The fashion data obtained from H&M Group and processed it to build a prediction model. The model was evaluated using Mean Average Precision with the highest score achieved using GRU method within summer season with the MAP score of 0.0022115738.

Item Type: Thesis (Other)
Uncontrolled Keywords: Fashion prediction, Gated Recurrent Unit, personalization, temporal pattern, Time Decay, Gated Recurrent Unit, prediksi fesyen, personalisasi, pola temporal, Time Decay
Subjects: T Technology > T Technology (General) > T57.5 Data Processing
T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User: Julius Adetya Eka Bhaswara
Date Deposited: 28 Jul 2023 16:06
Last Modified: 28 Jul 2023 16:06
URI: http://repository.its.ac.id/id/eprint/99676

Actions (login required)

View Item View Item