Pengaruh Tokenisasi Kata N-Grams Spam SMS Menggunakan Support Vector Machine

Sugianto, Castaka Agus; Apandi, Tri Herdiawan

doi:10.31227/osf.io/vjc7k

Cited by 3 publications

(3 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Pada tahap Preprocessing adalah proses perubahan bentuk data yang terstruktur sembarang menjadi data yang terstruktur sesuai kebutuhan untuk proses dalam text mining (Takdirillah, 2020). Tahapan Preprocessing yang digunakan antara lain: Case Folding yaitu tahap mengubah semua huruf dalam dokumen menjadi huruf kecil dan karakter non-huruf akan dihilangkan, Tokenizing yang merupakan tahap pemotongan string input berdasarkan tiap kata yang menyusunnya (Afuan, 2018) dan secara garis besar memecah sekumpulan karakter dalam suatu teks ke dalam satuan kata, Filtering dengan mengambil kata-kata penting dari hasil tahap tokenizing yang dapat dilakukan dengan menghilangkan stoplist atau stopword (Sugianto & Apandi, 2018), dan Stemming yang mentransformasi suatu kata menjadi kata dasarnya (root word) dengan menggunakan aturan-aturan tertentu (Senthikumar, Maheswari, 2019).…”

Section: Gambar 1 Alur Metode Penelitianunclassified

Implementasi Metode K-Nearest Neighbor (K-NN) untuk Analisis Sentimen Kepuasan Pengguna Aplikasi Teknologi Finansial FLIP

Rahayu

Bororing

et al. 2022

Edumatic

View full text Add to dashboard Cite

The phenomenon of technological development can transform systems in various sectors to provide efficiency and convenience at a lower cost, including the financial sector. Flip is a financial service application that makes it easy to transfer money between banks without administrative fees. By the end of 2021, the Flip will have a 4.9 rating on the Google Play Store. The purpose of this study was to analyze user sentiment towards the Flip app to see if flip user ratings were as positive as the ratings received. This study uses a set of text mining processes on the user rating data of the Flip app on the Google Play Store, using the classification algorithm K-Nearest Neighbor with TF-IDF weighting. The results show that 77.67% of the test data are correctly classified as positive evaluation classes, with high accuracy and recall rates of 82.67% and 86.92%, respectively. In addition, from the results of applying the Flip user rating data classification method, the comparison between training data and test data is 80%:20%, and the classification accuracy using the K-Nearest Neighbor algorithm is 76.68%. User reviews of the Flip app have shown positive results, as well as the ratings obtained in the Google Play Store and the K-Nearest Neighbor algorithm, TF-IDF weighting process used to analyze user sentiment towards the Flip app.

show abstract

Section: Gambar 1 Alur Metode Penelitianunclassified

Implementasi Metode K-Nearest Neighbor (K-NN) untuk Analisis Sentimen Kepuasan Pengguna Aplikasi Teknologi Finansial FLIP

Rahayu

Bororing

et al. 2022

Edumatic

View full text Add to dashboard Cite

show abstract

“…Text mining is a text processing to extract and find information from a text or unstructured data that is useful through recognizing and exploring a pattern in a text. (Somantri, Wiyono and Dairoh, 2016) Thus text mining needs to be done in classifying a text in order to get the information needed for the data classification process [16]- [18]. Before classifying the data, we need a preprocessing which is part of text mining.…”

Section: Text Processing In Text Miningmentioning

confidence: 99%

Indonesian Language Hoax News Classification Basedn on Naïve Bayes

Sudrajat¹,

Wulandari²,

Syafwan³

2022

JAIS

View full text Add to dashboard Cite

Hoax news in Indonesia causes various problems, therefore it is necessary to classify whether a news is in the hoax category or is valid. Naive Bayes is an algorithm that can perform classification but has a weakness, namely the selection of attributes that can affect accuracy so that it needs to be optimized by giving weights to attributes using the TF-IDF method. Classification using Naive Bayes and using TF-IDF as attribute weighting on a dataset of 600 data resulted in 82% accuracy, 84% precision, and 89% recall. The suggestion put forward is that it is better to use a larger number of datasets in order to produce higher accuracy.

show abstract

“…Several studies have been conducted in handling spam with various feature selection methods and classification algorithms. For instance, the research carried out by [7] integrated the Principal Component Analysis (PCA) feature selection method and the Correlated Naïve Bayes Classifier (CNBC) algorithm, while [8] used the ranking method, and [9] applied the Genetic Algorithm as feature selection to improve the accuracy of Naïve Bayes results. Furthermore, [10] compared three classification algorithms, namely Naïve Bayes (NB), Support Vector Machine (SVM), and Artificial Neural Network (ANN), as well as four feature selection algorithms, including information gain, chi-square, forward selection, and backward elimination for analysis of movie review sentiment.…”

Section: Introductionmentioning

confidence: 99%

Comparison of Filter and Wrapper Based Feature Selection Methods on Spam Comment Classification

Anggraeni¹,

Mustofa²,

Priyanta³

2021

Indonesian J. Comput. Cybern. Syst.

View full text Add to dashboard Cite

The continuous growth of the internet has led to the use of social media for various purposes increase. For instance, some irresponsible parties take advantage of the comment feature on social media platforms to harm others by providing spam comments on the shared object. Furthermore, variation of comments creates many features to be processed, thereby negatively impacting the performance of a classification algorithm. Therefore, this study aims to solve the problem associated with spam comments by comparing filter and wrapper based feature selection using text classification techniques. Data collected from training and test data of 4944 and 100 comments showed that the best accuracy, precision, recall, and f-measure of MNB are 96%, 100%, 92%, and 95.8%. The best accuracy is achieved using feature selection by combining Chi-Square and Sequential Forward Selection methods with a subset of 500 features. Furthermore, the accuracy increase in the MNB and SVM classifications are 8% and 4%. This research concludes that the combination of feature selection improves the classification performance of Indonesian language spam comments.

show abstract

Pengaruh Tokenisasi Kata N-Grams Spam SMS Menggunakan Support Vector Machine

Cited by 3 publications

References 3 publications

Implementasi Metode K-Nearest Neighbor (K-NN) untuk Analisis Sentimen Kepuasan Pengguna Aplikasi Teknologi Finansial FLIP

Implementasi Metode K-Nearest Neighbor (K-NN) untuk Analisis Sentimen Kepuasan Pengguna Aplikasi Teknologi Finansial FLIP

Indonesian Language Hoax News Classification Basedn on Naïve Bayes

Comparison of Filter and Wrapper Based Feature Selection Methods on Spam Comment Classification

Contact Info

Product

Resources

About