Klasifikasi Helpdesk Menggunakan Metode Support Vector Machine

Kusumahadi, Stefanie Hilda; Junaedi, Hartarto; Santoso, Joan

doi:10.30591/jpit.v4i1.1125

Cited by 7 publications

(15 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…After carrying out the data labeling process, the next step is term weighting process with TF-IDF. Because basically the machine can only process numbers, then each word in the document will be given a weight or frequency value, and then the result of this weighting process is a vector value, later the vector value will be entered into a vector space for later use for the testing process algorithm after this [14].…”

Section: Term Weightingmentioning

confidence: 99%

Combination of Support Vector Machine and Lexicon-Based Algorithm in Twitter Sentiment Analysis

Muhammadi¹,

Laksana²,

Arifa³

2022

khif

View full text Add to dashboard Cite

Data from the Ministry of Civil Works and Public Housing (Kementrian PUPR) in 2019 shows that around 81 million millennials do not own houses. Government Regulation Number 25 of 2020 on the Implementation of Public Housing Savings, commonly called PP 25 Tapera 2020, is one of the government's efforts to ensure that Indonesian people can afford houses. Tapera is a deposit of workers for house financing, which is refundable after the term expires. Immediately after enaction, there were many public responses regarding the ordinance. We investigate public sentiments commenting on the regulation and use Support Vector Machine (SVM) in the study since it has a good level of accuracy. It also requires labels and training data. To speed up labeling, we use the lexicon-based method. The issue in the lexicon-based lies in the dictionary component as the most significant factor. Therefore, it is possible to update the dictionary automatically by combining lexicon-based and SVM. The SVM approach can contribute to lexicon-based, and lexicon-based can help label datasets on SVM to produce good accuracy. The research begins with collecting data from Twitter, preprocessing raw and unstructured data into ready-to-use data, labeling the data with lexicon-based, weighting with TF-IDF, processing using SVM, and evaluating algorithm performance model with a confusion matrix. The results showed that the combination of lexicon-based and SVM worked well. Lexicon-based managed to label 519 tweet data. SVM managed to get an accuracy value of 81.73% with the RBF kernel function. Another test with a Sigmoid kernel attains the highest precision at 78.68%. The RBF kernel has the highest recall result with a value of 81.73%. Then, the F1-score for both the RBF kernel and Sigmoid is 79.60%.

show abstract

Section: Term Weightingmentioning

confidence: 99%

Combination of Support Vector Machine and Lexicon-Based Algorithm in Twitter Sentiment Analysis

Muhammadi¹,

Laksana²,

Arifa³

2022

khif

View full text Add to dashboard Cite

show abstract

“…Di google play store ada sebuah fitur yaitu rating dan ulasan (review), fitur itu bisa akan mempengaruhi calon pengguna dikarenakan kecenderungan pengguna sebelum mengunduhnya akan melihat kolom ulasan atau review ini sebagai tolak ukur bagus atau tidaknya suatu produk, serta opini opini yang di tuangkan di dalamnya. ulasan ini biasanya terbagi menjadi ulasan positif dan ulasan negatif tetapi ulasan dari sosial media mempengaruhi keputusan seseorang untuk membeli atau menggunakan produk tersebut [6] ditambah dengan kecenderungan pengguna ingin mengetahui pendapat serta pengalaman pengguna lain yang terkait aplikasi tersebut [7] oleh sebab itu diperlukan metode untuk menyortir serta menganalisis ulasan dengan cepat dan akurat serta mengkategorikan antara ulasan positif dengan negatif.…”

Section: Pendahuluanunclassified

Analisis Sentimen Ulasan Pengguna Aplikasi Myim3 Pada Situs Google Play Menggunakan Support Vector Machine

2022

View full text Add to dashboard Cite

Technological developments are increasingly rapid, this makes it easier to communicate information and shopping transactions, one of the innovations that are being adopted is digital services, such as self-service. One of the self-services is myim3 which is a product of PT Indosat Ooredoo Hutchison as an internet network service provider company, with the increasing number of users of the application, many opinions or public sentiments are shared in the comments or reviews column, therefore it is necessary to analyze this MyIM3 application review to find out public opinion about the application. The review data is obtained from the Google Play website which is retrieved using the scraping method with the help of 3rd party libraries in python. The amount of data obtained in this study was 3484 data. Experts assist in data labeling to determine positive and negative. In the preprocessing stage, the data is cleaned to reduce the less influential attributes. In the next stage, perform the transformation process with TF-IDF. The classification process is divided into several scenarios with the algorithm used as a support vector machine with 2 kernels, linear and RBF. The best results are in the scenario (70:30) for the linear kernel with 87% accuracy and the scenario (90:10) with 87% accuracy in the RBF kernel. The classification process produces the most frequently occurring words in each sentiment class which is visualized with a word cloud. The word "good" is the most dominant in the positive review data, while the word "network" is the most dominant in the harmful review data of the MyIM3 application

show abstract

“…Text mining adalah proses semi otomatis penggalian pola (pengetahuan dan informasi yang berguna) dari sejumlah sumber data berukuran besar yang tidak terstruktur berupa teks [11] [12]. Text mining menggunakan natural language processing untuk memasukkan struktur ke dalam kumpulan teks.…”

Section: Tinjauan Pustaka 21 Text Miningunclassified

Klasifikasi Sentimen Terhadap Maxim Menggunakan Algoritma SVM Pada Media Sosial Twitter

Pranata

Budianita

Yusra

et al. 2022

JNKTI

View full text Add to dashboard Cite

Abstrak - Twitter merupakan media sosial yang penggunanya paling pesat. Seiring perkembangan waktu, setiap ojek online memiliki popularitas di masing-masing user. Contoh saja Maxim, pendatang baru yang menyediakan layanan yang berbeda dengan ojek lain. Salah satu aktivitas yang biasa dilakukan para fans atau haters kepada akun twitter ojek online yaitu memberikan komentar pada tweet, untuk mengetahui komentar para fans dan haters diperlukan klasifikasi dengan menggunakan machine learning. Pada penelitian ini, langkah pertama adalah dilakukan proses sortir dan pemberian label pada data tersebut. Hasilnya akan tercipta 3 label yaitu label data positif, netral dan label data negatif dengan jumlah 1200 data. Selanjutnya melakukan analisa preprocessing data yang meliputi case folding, cleansing data, tokenizing, filtering dan stemming. Lalu dilakukan pembobotan dengan metode TF-IDF dan diklasifikasikan dengan metode Support Vector Machine. Hasil pengujian dilakukan dengan metode Confussion Matrix, berdasarkan hasil pengujian diperoleh akurasi terbaik pada perbandingan data 90:10 sebesar 85% dengan menggunakan Kernel RBF dan Polynomial, dilanjutkan dengan kernel Sigmoid sebesar 82,5% dimana hasil klasifikasi didominasi kalimat positif.Kata kunci: analisis sentiment, klasifikasi, ojek online, support vector machine, twitter Abstract - Twitter is one of the Social Medias which has a rapid user. Over the time, every Ojek Online, has its own popularity among their users. Maxim, for instance, a newcomer which provides a different service from other online motorcycle taxies. One of the activities which is always do by the fans or haters toward twitter account of online motorcycle taxies is giving comments on tweet. To identify the comments from fans or haters is required classification by using Machine Learning. In this research, the first step was sorting process and labelling the data. As the result 3 labels would have created, which were positive data label, neutral data label, and negative data label with total of 1200 data. The next step was conducting the analysis of preprocessing data which included case folding, data cleansing, tokenizing, filtering and stemming. Then, the weighting was carried out using the TF-IDF method and classified by the Support Vector Machine method. The test results were carried out using the Confusion Matrix method, based on the test results, the best accuracy was obtained at a data comparison of 90:10 by 85% using the RBF Kernel and Polynomial, followed by the Sigmoid kernel of 82.5% where the classification results were dominated by positive sentences.Key word: classification, Ojek Online, sentiment analysis, support vector machine, twitter

show abstract

Klasifikasi Helpdesk Menggunakan Metode Support Vector Machine

Cited by 7 publications

References 3 publications

Combination of Support Vector Machine and Lexicon-Based Algorithm in Twitter Sentiment Analysis

Combination of Support Vector Machine and Lexicon-Based Algorithm in Twitter Sentiment Analysis

Analisis Sentimen Ulasan Pengguna Aplikasi Myim3 Pada Situs Google Play Menggunakan Support Vector Machine

Klasifikasi Sentimen Terhadap Maxim Menggunakan Algoritma SVM Pada Media Sosial Twitter

Contact Info

Product

Resources

About