The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLVIII-4/W3-2022
https://doi.org/10.5194/isprs-archives-XLVIII-4-W3-2022-67-2022
https://doi.org/10.5194/isprs-archives-XLVIII-4-W3-2022-67-2022
02 Dec 2022
 | 02 Dec 2022

A WEB-BASED APPLICATION FOR MAKING LOW-COST VACATION RESERVATIONS FOR TOURISTS USING THE K-NEAREST NEIGHBORS ALGORITHM

S. O. Mariwa and T. K. Tunduny

Keywords: Machine Learning, K-Nearest Neighbors, Vacation, Holiday Reservation, Home Exchange, Home Tier, Vacation Home, Scikit-learn

Abstract. An occasional recreational vacation is a necessity for many people. It provides a perfect opportunity for the body and the mind to get much-needed rest after weeks, months, or years of daunting tasks and a break from routine. It also gives people morale as they perform their usual tasks afterward. Unfortunately, many people are unable to afford a vacation not only internationally but also locally due to the high costs involved. This makes many people prefer spending their holidays with extended families, by, for instance, traveling to their rural homes as opposed to taking a vacation. To boost the tourism sector in our country that is being promoted by initiatives such as ‘Tembea Kenya’, we should encourage domestic tourism. Another challenge is the experience in hotels that some people do not like that would entirely cause them to opt to spend their holidays differently, for example, the lack of privacy in the shared accommodation facility, the limited space in hotels, the numerous restrictions, the level of cleanliness in the shared facility especially during the COVID-19 crisis, etcetera. The aim of this project was to solve the problem by coming up with a technological means of enabling people to make reservations for vacation homes with each other such that they mutually benefit from the program thus eliminating the fee for renting out the house. This solution was implemented using a web-based application and applied the K-Nearest Neighbors machine learning algorithm that was used to classify homes based on the features available.