BALI TOURIST VISITS CLUSTERED VIA TRIPADVISOR REVIEWS USING K-MEANS ALGORITHM

Abstract

Bali is one of the provinces with the most popular destinations for tourists. However, there are obstacles in developing tourist destinations in the province of Bali in terms of absorbing more tourist visits. Tripadvisor, the world's largest tourism information platform. In order to improve its service to users, Tripadvisor conducts online reviews to obtain ratings based on travel experience. The purpose of this study is to determine clustering and accuracy in tourist visits to tourist destinations in the province of Bali. Clustering is done using 3 clusters using the KDD method. The first process is data selection, then data processing which consists of several stages, first deleting rows of empty data, then cleaning duplicate data and the final result is 5261 clean data then data transformation, so that data can be read by python, The next process is data mining, this process uses the K-Means clustering algorithm which produces 3 clusters with cluster 1 being medium with 1495 data, high cluster 2 with 2315 data, and low cluster 3 with 1451 data. The Davies Boldin Index is used to evaluate the K-Algorithm means clustering, the result is 0.3 where the value is very good because it is not minus and the value is close to zero.

Keywords

• data mining
• dbi
• kdd
• k-means clustering
• tripadvisor