Recommender systems for travel industry  using visual information

Burda, Marko

Home
→
Students Research & Project Works | Роботи студентів
→
Факультет прикладних наук
→
Бакалаврська програма "Комп'ютерні науки"
→
2022
→
View Item

dc.contributor.author	Burda, Marko
dc.date.accessioned	2024-02-14T11:46:38Z
dc.date.available	2024-02-14T11:46:38Z
dc.date.issued	2022
dc.identifier.citation	Burda, Marko. Recommender systems for travel industry using visual information / Burda, Marko; Supervisor: Taras Firman; Ukrainian Catholic University, Department of Computer Sciences. – Lviv: 2022. – 34 p.	uk
dc.identifier.uri	https://er.ucu.edu.ua/handle/1/4432
dc.language.iso	en	uk
dc.title	Recommender systems for travel industry using visual information	uk
dc.type	Preprint	uk
dc.status	Публікується вперше	uk
dc.description.abstracten	When choosing the right establishment to order from, we rely on the average ratings of those places. The ratings are calculated based on numeric values left by different users who have had experience with the establishment. The problem with such ratings is that having a high score does not guarantee one’s satisfaction with a meal. It is a mere number evaluated based on other people’s experiences. However, every human being is an individual with specific tastes, and the concept of good food differs for everyone. With the unexpected strike of the COVID-19 pandemic, there was also a sharp spike in need for food delivery services. For applications like Glovo or its competitors Rocket and Bolt Food, the only way of approximating the food quality is a score averaged from other users’ ratings, which makes it, quite often, a hit-or-miss experience. Such services, unfortunately, neither offer written reviews nor any personalized recommendations based on your order history. This project is built around this topic and is aimed to train a model that learns from vi- sual information and user behavior and produces recommendations based on given information.	uk