Garuda - Garba Rujukan Digital

Text Mining for Customer Sentiment Using Naive Bayes and SMOTE Methods on TokopediaCare Twitter Rico Budiyanto; Indah Purnamasari; Dedi Dwi Saputra
IJISTECH (International Journal of Information System and Technology) Vol 6, No 1 (2022): June
Publisher : Sekolah Tinggi Ilmu Komputer (STIKOM) Tunas Bangsa

At this time, buying and selling online has become part of the lives of the Indonesian people and the world, especially during the pandemic, marketplace users are increasing and slowly replacing traditional markets. Tokopedia as one of the largest marketplaces in Indonesia has the largest users in the 3rd quarter of 2019. Customer complaints to Tokopedia services can be submitted through Social Media such as Twitter and also other media. Complaints submitted via Twitter to TokopediaCare are still manually identified by Tokopedia customer service so it takes a long time to respond to customer complaints because customer services need so much time to classify where is a complaint or not complaint tweeted. Text mining is used to process customer complaint data through text or sentences submitted by tweets using the Naïve Bayes method and the Synthetic Minority Oversampling Technique Method (SMOTE) feature for the implementation of machine learning can help identify the classification of complaints submitted via Twitter automatically. The use of the Naive Bayes method is added with the Syntethic Minority Oversampling Method feature which is considered better for generating predictions on tweets submitted by customers..

Validasi Dokumen Pengajuan Penelitian dan Pengabdian Masyarakat Universitas Nasional Menggunakan Metode Finite State Automata Lili Dwi Yulianto; Windu Gata; Frieyadie; Dedi Dwi Saputra
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 6 No 4 (2022): OCTOBER-DECEMBER 2022
Publisher : KITA Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v6i4.520

Non-Deterministic Finite Automata (NFA), a form of Finite State Automata (FSA), is one method for validating research proposal documents in R&D and community service activities at national universities. To carry out these activities, it is necessary to monitor the research proposal documents submitted by lecturers/researchers, one of which is to validate the research proposal document, ensuring that it follows the guidelines and avoids plagiarism, as well as the theme's suitability with the research proposal document's contents. Members of the research team can only participate in one type of study: stimulation or competitive research. Reviewers can provide research members with notes and plagiarism information, as well as refuse to submit research proposals. If the research proposal document is approved by the reviewer, the research member will receive a research certificate and will be able to continue their research with full support from the national government.

Text Mining for Customer Sentiment Using Naive Bayes and SMOTE Methods on TokopediaCare Twitter Rico Budiyanto; Indah Purnamasari; Dedi Dwi Saputra
IJISTECH (International Journal of Information System and Technology) Vol 6, No 1 (2022): June
Publisher : Sekolah Tinggi Ilmu Komputer (STIKOM) Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/ijistech.v6i1.221

At this time, buying and selling online has become part of the lives of the Indonesian people and the world, especially during the pandemic, marketplace users are increasing and slowly replacing traditional markets. Tokopedia as one of the largest marketplaces in Indonesia has the largest users in the 3rd quarter of 2019. Customer complaints to Tokopedia services can be submitted through Social Media such as Twitter and also other media. Complaints submitted via Twitter to TokopediaCare are still manually identified by Tokopedia customer service so it takes a long time to respond to customer complaints because customer services need so much time to classify where is a complaint or not complaint tweeted. Text mining is used to process customer complaint data through text or sentences submitted by tweets using the Naïve Bayes method and the Synthetic Minority Oversampling Technique Method (SMOTE) feature for the implementation of machine learning can help identify the classification of complaints submitted via Twitter automatically. The use of the Naive Bayes method is added with the Syntethic Minority Oversampling Method feature which is considered better for generating predictions on tweets submitted by customers..

Pengkategorian Komentar Instagram Terhadap Layanan Akademik dan Non-Akademik Universitas Terbuka Rhini Fatmasari; Alda Zevana Putri Widodo; Valianda Farradillah Hakim; Windu Gata; Dedi Dwi Saputra
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 7 No 1 (2023): JANUARY-MARCH 2023
Publisher : KITA Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v7i1.669

Instagram is one of the social media that has many users in Indonesia, where users are free to comment on whatever is going on, including being a form of online communication between campuses and their students. The number of topics and comments on an official Instagram account can be used as evaluation or learning material. The Open University is one of the campuses that has an official Instagram account with thousands of followers. In order to get an evaluation of academic and non-academic services, in this study a categorization analysis was carried out with 10,000 comment data taken from the official @univterbuka Instagram account. The data is categorized into 7 categories, then processed using 4 algorithms, namely SVM, Naïve Bayes, Random Forest and KNN. The highest accuracy in the category of teachers with the KNN method is 98.97% and the highest AUC is in the module category with the SVM method of 94.60%.

A Analisa Sentimen Terhadap Twitter IndihomeCare Menggunakan Perbandingan Algoritma Smote, Support Vector Machine, AdaBoost dan Particle Swarm Optimization Ferdian Syah; Hanif Fajrin; Abiyyu Nur Afif; Muhamad Rafi Saeputra; Dinda Mirranty; Dedi Dwi Saputra
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 7 No 1 (2023): JANUARY-MARCH 2023
Publisher : KITA Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v7i1.686

Indihome is one of the largest internet service providers in Indonesia with an increasing number of subscribers every year. Indihome subscribers until the end of March 2022 were recorded at 8.7 million, growing 7.2 percent over the same period last year. However, over time, many customers have complained to IndiHome about slow internet access, sudden increases in billing and so on. Based on the description above, it is interesting to conduct research on the use of Indihome tweets which are the result of channeling opinions and comments on Twitter social media. This study uses the Smote method, Support vector machine, Adaboost and Particle swarm optimization so that the results can be compared with the level of accuracy. The results of this research show that by using the Smote method, Support vector machine obtained values of Accuracy 80.48, Precision 85.29, Recall 73.75 and AUC 0.907. As for the Smote, Support vector machine and AdaBoost methods, the Accuracy values are 80.21, Precision 85.01, Recall 73.36 and AUC 0.861. Finally, the results of the Smote and Particle swarm optimization methods obtained Accuracy values of 76.59, Precision 76.57, Recall 80.35 and AUC 0.868. Based on the research results, the Smote method and the Support vector machine (SVM) have the largest results and are considered effective with the existing dataset

Sentiment Analysis Terhadap Perspektif Warganet Atas Tragedi Kanjuruhan Malang di Twitter Menggunakan Naïve Bayes Classifier Minardi Minardi; Ranita Lasepa; Santoso Riyadi; Syahrur Ramadhan; Dedi Dwi Saputra
Jurnal Informatika Vol 10, No 1 (2023): April 2023
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v10i1.14546

Media sosial Twitter adalah salah satu tempat bagi warganet dari seluruh dunia untuk menyampaikan perspektif mereka, sebuah insiden yang terjadi di Stadion Kanjuruhan Malang pada tanggal 01 Oktober 2022 sedang hangat diperbincangkan, sehingga memunculkan berbagai perspektif yang memicu timbulnya pro-kontra di masyarakat. Atas dasar itu untuk mengklasifikasikan perspektif positif atau negatif warganet di Twitter, maka dilakukan analisis sentimen menggunakan Naïve Bayes Classifier. Analisis sentimen dilakukan dengan mengambil tweet warganet di Twitter dengan hashtag UsutTuntasTragediKanjuruhan yang diambil 1.500 tweet untuk dijadikan dataset. Preprocessing data terdiri dari Annotation Removal, Remove Hashtag, Transformation Remove Url, Regexp, Indonesian Steaming, Indonesian Stopword Removal. Hasil analisis berjalan dengan baik dengan nilai akurasi 77,67%, kemudian nilai precision sebesar 77,19%, nilai recall sebesar 78,50%, dan nilai AUC 0.820 (good classification). The social media site Twitter is a place where Internet users around the world can exchange perspectives on current discussions. One of them is football; this sport is a hobby that is loved by all corners of the world, including the people of Malang. With their love for this sport, they call themselves Aremania, namely Arema Malang team supporters, but a dark incident occurred. The Kanjuruhan at Malang Stadium on January 10, 2022, raised different views from all Twitter user accounts, which led to an increase in tweets and became a trending topic at that time. To develop different perspectives based on what brings advantages and disadvantages to the community, a procedure was applied to classify Twitter users' positive or negative perspectives through sentiment analysis with the Nave Bayes classifier. Sentiment analysis was carried out by indexing Twitter user tweets with the hashtag "UsutTuntasTragediKanjuruhan," crawling data from 1,500 existing tweets as a dataset, after which the data to be processed is identified. (labeling) for the next step, namely stage Data preprocessing includes annotation removal, hashtag removal, URL removal, regexp, Indonesian steaming, and Indonesian stopword removal, as well as operators' smote upsampling. Making a confusion matrix that shows the final result of the analysis is going well, namely the value accuracy of 77.67%, the value precision of 77.19%, the value recall of 78.50%, and the value AUC of 0.820 (good classification).

Analisis Sentimen Terhadap Telkomsel dan XL Berbasis Machine Learning Pada Data Twitter Trisiwi Indra Cahyani; Windu Gata; Dedi Dwi Saputra; Hafifah Bella Novitasari; Hernawati Hernawati
INTECOMS: Journal of Information Technology and Computer Science Vol 6 No 1 (2023): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/intecoms.v6i1.5765

Di Indonesia pengguna internet mencapai lebih dari 200 juta pengguna. Telkomsel dan XL bersaing untuk menjadi penyedia layanan internet nomor satu. Media sosial Twitter membuat pengguna lebih jujur dalam memberikan review. Umpan balik pengguna akan menjadi rekomendasi dari mulut ke mulut (WoW). Pada penelitian ini bertujuan untuk mengetahui pandangan masyarakat terhadap provider Telkomsel dan XL berdasarkan data tweet di Twitter pada bulan Juli dan Agustus 2022. Dataset dikumpulkan dari Twitter menggunakan Twitter API dengan kata kunci â€œXL Internetâ€, â€œTelkomsel Internetâ€, â€œMyXLâ€, dan â€œMyTelkomselâ€ dan diperoleh sebanyak 17.543 data. Kemudian dataset akan dilakukan case folding, tokenized, normalized, stopword removal, stemming, dan proses pembobotan TF-IDF. Model klasifikasi menggunakan Entropy Maksimum, Multinomial NaÃ¯ve Bayes, dan Complement NaÃ¯ve Bayes. Untuk menguji kemampuan menggeneralisasi, dilakukan 10-Fold Cross Validation untuk masing-masing model. Hasil menunjukkan bahwa metode ME lebih baik dari MNB dan CNB dengan nilai akurasi 84,11%, 81,53%, dan 79,95%.

Analisis Sentimen Dengan Metode Naïve Bayes, SMOTE Dan Adaboost Pada Twitter Bank BTN Kurnia; Indah Purnamasari; Dedi Dwi Saputra
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 7 No 2 (2023): APRIL-JUNE 2023
Publisher : KITA Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v7i3.707

Social media has become a means of sharing information with each other, from being a means of promotion to being used to spread opinions. Because of the ease of technology, anyone can access social media and comment on issues that are being discussed. Branding image is also an important thing, because people can interact directly through social media, be it support or criticism. So this study was conducted to analyze the sentiment on twitter towards Bank BTN as a bank that has long focused on housing loans. To analyze the sentiment, an experiment was carried out by a combination of SMOTE, Naïve Bayes and Adaboost algorithms. Before calculating the algorithm, stemming and stopwords are carried out so that the data used does not contain noise. The results showed that the combination of SMOTE, Naïve Bayes, and Adaboost showed the best modeling results with an accuracy of 87.05%, precision of 90.63%, recall of 83.00%, and AUC of 0.909.

Analisis Sentimen Persepsi Masyarakat Terhadap Penggunaan Aplikasi My Pertamina Pada Media Sosial Twitter Menggunakan Metode Naïve Bayes Classifier Rini Maria; Retno Umi Umayah; Syifa Mahardinny; Diki Kalana; Dedi Dwi Saputra
Jurnal Komputer Antartika Vol. 1 No. 1 (2023): Maret 2023
Publisher : Antartika Media Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Penyebaran informasi saat ini menjadi lebih mudah dan cepat. Karena semakin berkembangkan sistem informasi khususnya dalam penyebaran berita. Salah satunya media sosial Twitter yang sangat memudahkan masyarakat mencari informasi atau berinteraksi langsung dengan pengguna Twitter lainnya dan Studi kasus yang diambil oleh peneliti yaitu mengenai Persepsi Masyarakat Terhadap Penggunaan Aplikasi My Pertamina. Kasus tersebut diambil karena ramai diperbincangkan masyarakat Indonesia di media sosial Twitter dimana Perusahaan Pertamina menerapkan sistem pembayaran Non-tunai sehingga memunculkan perspektif yang memicu timbulnya Pro-Kontra di masyarakat mengenai penggunaan Aplikasi tersebut. Salah satu pemanfaatan penelitian ini adalah untuk mengetahui kecenderungan komentar atau tweets pengguna Twitter terhadap adanya Kebijakan pembelian Pertalite menggunakan Aplikasi MyPertamina dengan melakukan analisis sentimen. Oleh karena itu dibutuhkan metode untuk mengklasifikasikan komentar publik berupa analisis sentimen pada media sosial Twitter, algoritma yang digunakan dalam melakukan analisis sentimen tersebut adalah metode Naïve Bayes Classifier (NBC). Analisis sentimen dilakukan dengan mengambil 1000 tweet untuk dijadikan sebagai dataset, sentimen akan diklasifikasikan dengan memberikan label yang dibagi menjadi dua kategori yaitu complaint dan non complaint. Tahapan dalam melakukan analisis sentimen pada penelitian ini adalah preprocessing data, pengolahan data, klasifikasi, dan evaluasi. Dari analisis tersebut diperoleh hasil akurasi sebesar 82.96%, precision sebesar 81.17%, Recall sebesar 86.07%, kemudian hasil dari AUC sebesar 0.906. Berdasarkan Hasil presentase yang diperoleh tersebut sudah mencapai hasil yang maksimal untuk mengklasifikasi komentar-komentar publik. Dissemination of information is now easier and faster. Due to the growing development of information systems, especially in the dissemination of news. One of them is social media Twitter, which makes it very easy for people to find information or interact directly with other Twitter users. The case study taken by the researcher is regarding Public Perceptions of Using the My Pertamina Application. This case was taken because it was widely discussed by the Indonesian people on social media Twitter where the Pertamina Company implemented a non-cash payment system, giving rise to views that sparked pros and cons regarding the people who use the application. One of the uses of this research is to find out trends in comments or tweets by Twitter users regarding Pertalite purchasing policies using the MyPertamina application by conducting sentiment analysis. Therefore we need a method to classify public comments in the form of sentiment analysis on social media Twitter, the algorithm used in sentiment analysis is the Naïve Bayes Classifier (NBC) method. Sentiment analysis is carried out by taking 1000 tweets to be used as a dataset, sentiment will be classified by labeling it which is divided into two categories, namely complaint and non complaint. The stages in conducting sentiment analysis in this study are data preprocessing, data processing, classification, and evaluation. From this analysis, the results obtained were 82.96% accuracy, 81.17% precision, 86.07% recall, then the AUC result was 0.906. Based on the percentage results obtained, maximum results have been achieved for classifying public comments.

Implementasi Machine Learning Untuk Klasifikasi Narasi Informative dan Non-Informative Pada Media Sosial Twitter TMC Polda Metro Jaya Menggunakan Naïve Bayes Classifier Nendi Riska Fadhila; Tri Sutarti; Melly Patende; Galistia Yudha; Dedi Dwi Saputra
Jurnal Komputer Antartika Vol. 1 No. 1 (2023): Maret 2023
Publisher : Antartika Media Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Twitter adalah layanan jejaring sosial yang memungkinkan penggunanya untuk mengirim dan membaca pesan berbasis teks hingga 140 karakter. Sejak diluncurkan, twitter telah menjadi salah satu dari sepuluh situs yang paling sering dikunjungi di internet. Adapun pengklasifikasian narasi yang bersifat informative atau non-informative pada media sosial twitter TMC Polda Metro Jaya terhadap pandangan masyarakat, menggunakan metode klasifikasi alogoritma Naive Bayes Classifier. Penelitian ini dilakukan dengan mengambil 1.000 sampel data dengan keyword @TMCPoldaMetro untuk dijadikan dataset. Dengan menggunakan preprocessing data, annotation removal, remove hastag, regexp, transformation remove URL, Indonesian stemming dan Indonesia stop removal. Hasil analisis penelitian yaitu nilai Accuracy sebesar 81,7%, nilai Pression sebesar 81,30%, nilai Recall sebesar 83,03% dan AUC sebesar 0,847. Twitter is a social networking service that enables its users to send and read text-based messages of up to 140 characters. Since its launch, Twitter has become one of the ten most visited websites on the internet. As for the classification of narratives that are informative or non-informative on TMC Polda Metro Jaya's social media twitter to the views of the public, using the Naive Bayes Classifier algorithm classification method. This research was conducted by taking 1,000 data samples with the keyword @TMCPoldaMetro to be used as a dataset. By using data preprocessing, annotation removal, remove hashtags, regexp, transform remove URLs, Indonesian stemming, and Indonesian stop removal. The results of the research analysis are Accuracy of 81.7%, Pressure of 81.30%, Recall of 83.03%, and AUC of 0.847.

Title

Found 11 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 11 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 11 Documents
Search