Clustering E-Commerce Tweet Data Using the K-Means Method (Case Study of Blibli Indonesia's Twitter Account)


Alven Safik Ritonga Isnaini Muhandhis


The development of e-commerce is very rapid at this time, with the increasing number of e-commerce making competition in attracting customers and maintaining loyal customers. E-commerce players need to find a strategy for this, one way is advertising on social media, such as; Twitter, Facebook, Instagram, and so on. The purpose of this study was to obtain clustering of tweet data from Twitter using the K-Means method on tweet data from the Blibli Indonesia Twitter account to determine the type of tweet content that was retweeted by followers. The data used is follower tweet data which is pulled from the Twitter account @bliblidotcom. Testing the most optimum number of clusters by finding the largest Silhouette coefficient value. The results obtained that the optimal number of clusters is 10 clusters. From the results of this clustering, the tweet content that Blibli Indonesia consumers like the most is voucher content (cluster 4) and Opportunity series content (cluster 6). Voucher content and opporeno series content as a result of this clustering can be used by Blibli for promos to its consumers.


D. S. Indraloka and B. Santoso, ”Penerapan Text Mining untuk Melakukan Clustering Data Tweet Shopee Indonesia,” Jurnal Sains dan Seni ITS, vol. 6, no. 2, pp. 51-56, 2017.
[2] R. K. Putri, B. Warsito and Mustafid, “Implementasi Algoritma Modified Gustafson-Kessel untuk Clustering Tweets pada Akun Twitter Lazada Indonesia, ” Jurnal Gaussian, vol.8, no.3, pp. 285-295, 2019.
[3] D. H. Wahid and Azhari, “Peringkasan Sentimen Esktraktif di Twitter Menggunakan Hybrid TF-IDF dan Cosine Similarity,” IJCCS, vol. 10, no. 2, pp. 207-218, 2016.
[4] U. Rofiqoh, R.S. Perdana and M. A. Fauzi, “Analisis Sentimen Tingkat Kepuasan Pengguna Penyedia Layanan Telekomunikasi Seluler Indonesia pada Twitter dengan Metode Support Vector Machine dan Lexicon Based Features,” Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer, vol. 1, no. 12, pp. 1725-1732, 2017.
[5] S. A. D. Budiman, D. Safitri and D. Ispriyanti,”Perbandingan Metode K-Means dan Metode DBSCAN Pada Pengelompokan Rumah Kost Mahasiswa di Kelurahan Tembalang Semarang,” Jurnal Gaussian, vol. 5, no. 4, pp. 757-762, 2016.