z-logo
open-access-imgOpen Access
Klasifikasi Keluhan Masyarakat pada Sosial Media Twitter terhadap Pelayanan Toko Online di Indonesia menggunakan Metode Cosine TF-IDF
Author(s) -
Iwan Syarif,
Rengga Asmara,
Nur Ulima Rusmayani
Publication year - 2020
Publication title -
bina insani ict journal/bina insani ict journal
Language(s) - English
Resource type - Journals
eISSN - 2527-9777
pISSN - 2355-3421
DOI - 10.51211/biict.v7i1.1334
Subject(s) - computer science , humanities , art
Abstrak: Berkembangnya toko online dan transaksi online di Indonesia pada saat ini diiringidengan berbagai permasalahan seperti keluhan pada pelayanan yang membahas mengenaiaplikasi, ketanggapan dan pengiriman. Dengan adanya permasalahan tersebut, perhitunganserta penilaian keluhan yang sering didapatkan oleh masing-masing toko online sangatdiperlukan. Dengan memanfaatkan tweet masyarakat yang ditujukan kepada toko online, datatweet tersebut akan diklasifikasikan ke dalam kategori pelayanan yang telah ditentukan.Pengolahan data berupa tweet membutuhkan proses preprocessing yaitu proses untukmendapatkan keyword dari data tweet yang telah didapatkan, proses preprocessing memilikitahapan seperti tokenizing, filtering dan stemming. Keyword yang telah didapatkan diolah untukmendapatkan nilai hasil klasifikasi yang didapatkan. Proses klasifikasi kategori pelayanan padapenelitian ini menggunakan metode Cosine TF-IDF dimana metode tersebut membutuhkanbobot dan dokumen pada setiap kategori. Metode yang dikembangkan telah diaplikasikan padapenelitian ini menghasilkan prosentase proses klasifikasi kategori pelayanan menggunakanmetode Cosine TF-IDF sebesar 63.1%. Kata kunci: analisis sentimen, klasifikasi, rule based classifier, cosine similarity, TF-IDF Abstract: The development of online stores and online transactions in Indonesia at this time isaccompanied by various problems such as complaints on services that discuss applications,responsiveness and delivery. With these problems, the calculation and assessment ofcomplaints that are often obtained by each online store is very necessary. By utilizingcommunity tweets aimed at online stores, the tweet data will be classified into predeterminedservice categories. Data processing in the form of tweets requires a preprocessing process,namely the process of getting keywords from the data tweets that have been obtained, thepreprocessing process has stages such as tokenizing, filtering and stemming. The keywordsthat have been obtained are processed to obtain the classification results obtained. The servicecategory classification process in this study uses the Cosine TF-IDF method where the methodrequires weights and documents in each category. The method developed has been applied inthis study to produce a percentage of the service category classification process using theCosine TF-IDF method of 63.1%. Keywords: sentiment analysis, classification, rule based classifier, cosine similarity, TF-IDF

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here