TFR: A Tourist Food Recommender System based on Collaborative Filtering

Rajabpour, Neda; Mohammadighavam, Amirmahdi; Naserasadi, Ali; Estilayee, Majid

doi:10.5120/ijca2018917695

Cited by 3 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study raises the CF capability to be applied in spatial information systems that recommend desired culinary, including location and rating by public. This research is in line with the research of [20], [21], [22]. CF is also applied to restaurant recommendation systems, where their research is focused on computing CF performance, such as MAE and accuracy.…”

Section: Discussionsupporting

confidence: 67%

SISRES: Web-Based Culinary Recommendation with Collaborative Filtering

Aini,

Muhammad,

Rustamaji

et al. 2024

Appl. Inf. Syst. Manage.

View full text Add to dashboard Cite

Due to its advantageous location on the border of Jakarta and Tangerang, Tangerang Selatan is a highly developed city. In this instance, the Tangerang Selatan Department of Culture and Tourism (Dispar Tangsel) is required to use the application to assist the general public in realizing the availability of information in the Tangerang Selatan area. Twenty to twenty-five restaurants submit applications each year to the Dispar Tangsel Tourism Business Registry (TDUP). Dispar Tangsel must choose and decide on TDUP licensing priorities from among TDUP requests in order to open a restaurant. The purpose of the research is to offer recommendations to the community on food choices. Rapid application development (RAD) was used in the system's development. Additionally, the collaborative filtering technique has been employed by the decision support system to determine the amount of criteria or weight for restaurants using the weighted sum algorithm and for restaurants using cosine-based similarity algorithms. Additionally, the system design tool made use of MySQL as a database, PHP, the Codeigniter framework, and the unified modeling language (UML). The result demonstrates that the system is capable of displaying the output in accordance with the user's expectations during black box testing, which evaluates the functionality of the system based on the specifications. Collaborative filtering in SISRES can yield a significant improvement in recommendation accuracy. By collectively analyzing user preferences and behaviors, the algorithm can provide more relevant and personalized recommendations.

show abstract

Section: Discussionsupporting

confidence: 67%

SISRES: Web-Based Culinary Recommendation with Collaborative Filtering

Aini,

Muhammad,

Rustamaji

et al. 2024

Appl. Inf. Syst. Manage.

View full text Add to dashboard Cite

show abstract

“…To achieve and create a good restaurant RS, the efficiency and accuracy of the recommendation algorithm in generating recommendations should be taken into consideration. Many research studies had been conducted to improve the efficiency and accuracy of a RS (Rajabpour et. al., 2018;Tang & Wang, 2018;Roy et.…”

Section: Introductionmentioning

confidence: 99%

The The Efficiency of Hybridised Genetic Algorithm and Ant Colony Optimisation (HGA-ACO) in a Restaurant Recommendation System

Kassim¹

2022

ASMScJ

View full text Add to dashboard Cite

A recommendation system (RS) is used to provide recommendations to users by filtering items based on given inputs. Metaheuristic algorithms such as Genetic Algorithm (GA) and Ant Colony Optimisation (ACO) are known to be used in many RS to provide optimal and good recommendations. Both algorithms are designed based on nature-inspired events, where GA is designed based on the natural evolution process while ACO is based on ants’ behaviour in their natural habit. In this paper, both GA and ACO algorithms were implemented in a restaurant RS and evaluated by using the restaurant’s attributes, which was then followed by a list of recommended restaurants as the output. With the highest score of 99.64% of accuracy, GA overtakes ACO in terms of recommendation accuracy while ACO computed 67.12% lesser runtime than GA. Considering the results acquired, a new hybrid framework known as the HGA-ACO algorithm was proposed. The proposed HGA-ACO has a recommendation accuracy of 99.57% and achieved a 31.37% runtime reduction from GA. Thus, the proposed framework was observed to have improved the output accuracy of ACO and improved the processing time in GA, thus, improving the overall efficiency of the RS.

show abstract