Search Articles & Publications

Showing 262 articles found for "Negatif"

TOPSIS-BASED SYSTEM FOR THE SELECTION OF TRAINING PARTICIPANT CANDIDATES AT THE ASAHAN MANPOWER OFFICE

Maha Putra, Guntur, Wan Mariatul Kifti, Putri Amanda Nurhayati
Abstract: Abstract: Job training is one of the government’s efforts to improve the quality of human resources so that they possess competencies that meet labor market demands. The process of selecting training participants at the… e Department of Manpower of Asahan Regency is still carried out manually, which can lead to subjectivity and inefficiency in determining the most eligible candidates. This study aims to develop a decision support system using the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method to assist the selection process objectively and systematically. The study applies four evaluation criteria, namely education level, age, work experience, and interview, with a dataset consisting of 31 training candidates. The system is developed as a web-based application using PHP programming language and MySQL database. The TOPSIS method is applied through decision matrix normalization, weighting, determination of positive and negative ideal solutions, and preference value calculation to produce a ranking of candidates. The results show that the proposed system can provide objective recommendations for selecting training participants, improve the efficiency of the selection process, and support decision makers in producing more accurate and reliable decisions. Keywords: decision support system; selection; training; TOPSIS.   Abstrak: Pelatihan tenaga kerja merupakan salah satu upaya pemerintah dalam meningkatkan kualitas sumber daya manusia agar memiliki kompetensi yang sesuai dengan kebutuhan dunia kerja. Proses pemilihan calon peserta pelatihan di Dinas Tenaga Kerja Kabupaten Asahan selama ini masih dilakukan secara manual sehingga berpotensi menimbulkan subjektivitas dan kurang efektif dalam menentukan peserta yang paling layak. Penelitian ini bertujuan untuk membangun sistem pendukung keputusan menggunakan metode Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) untuk membantu proses seleksi peserta pelatihan secara objektif dan sistematis. Penelitian ini menggunakan empat kriteria penilaian yaitu pendidikan, usia, pengalaman kerja, dan wawancara dengan jumlah data sebanyak 31 calon peserta pelatihan. Sistem dikembangkan berbasis web menggunakan bahasa pemrograman PHP dan database MySQL. Metode TOPSIS digunakan untuk melakukan normalisasi matriks keputusan, pembobotan, penentuan solusi ideal positif dan negatif, serta perhitungan nilai preferensi untuk menghasilkan perankingan peserta pelatihan. Hasil penelitian menunjukkan bahwa sistem yang dibangun mampu memberikan rekomendasi peserta pelatihan secara objektif, meningkatkan efisiensi proses seleksi, serta membantu pihak dinas dalam pengambilan keputusan yang lebih akurat. Kata kunci: pelatihan; seleksi; sistem pendukung keputusan; TOPSIS.

SENTIMENT ANALYSIS USING MACHINE LEARNING FOR DIGITAL SERVICE DEVELOPMENT

Balqis, Rugaiyah, Jahda Rusti Putri, Mira Afrina, Ibrahim, Ali, Fathoni, Fathoni
Abstract: Abstract: The rapid growth of e-commerce mobile applications has generated large volumes of user reviews, making manual sentiment analysis increasingly impractical. This study aims to compare the effectiveness of three machine… achine learning algorithms Support Vector Machine (SVM), Random Forest, and Naive Bayes for automated sentiment classification of Indonesian-language mobile application reviews. A dataset of 3,000 user reviews from the RupaRupa application on the Google Play Store was collected and preprocessed through normalization, tokenization, stopword removal, and stemming. TF-IDF vectorization was applied for feature extraction, while the Synthetic Minority Over-sampling Technique (SMOTE) was used to address class imbalance across three sentiment categories: positive, negative, and neutral. The results show that SVM achieved the highest accuracy of 90.02%, while Random Forest obtained the best F1-score of 88.08% when sufficient training data were available. Naive Bayes demonstrated relatively stable performance across varying training data sizes. Furthermore, TF-IDF keyword analysis revealed that negative reviews were primarily associated with delivery issues, technical problems, and pricing concerns. These findings demonstrate the effectiveness of machine learning approaches for sentiment classification and provide practical insights for improving mobile application services.   Keywords: sentiment analysis; machine learning; SMOTE; TF-IDF; text classification   Abstrak: Pertumbuhan pesat aplikasi mobile e-commerce telah menghasilkan volume ulasan pengguna yang sangat besar, sehingga analisis sentimen secara manual menjadi semakin tidak praktis. Penelitian ini bertujuan untuk membandingkan efektivitas tiga algoritma machine learning Support Vector Machine (SVM), Random Forest, dan Naive Bayes dalam melakukan klasifikasi sentimen otomatis terhadap ulasan aplikasi mobile berbahasa Indonesia. Dataset yang digunakan terdiri dari 3.000 ulasan pengguna aplikasi RupaRupa yang dikumpulkan dari Google Play Store. Data kemudian diproses melalui tahapan preprocessing yang meliputi normalisasi, tokenisasi, penghapusan stopword, dan stemming. Ekstraksi fitur dilakukan menggunakan metode Term Frequency–Inverse Document Frequency (TF-IDF), sedangkan ketidakseimbangan kelas ditangani menggunakan Synthetic Minority Over-sampling Technique (SMOTE) pada tiga kategori sentimen, yaitu positif, negatif, dan netral. Hasil penelitian menunjukkan bahwa SVM mencapai tingkat akurasi tertinggi sebesar 90,02%, sementara Random Forest memperoleh nilai F1-score terbaik sebesar 88,08% ketika tersedia data pelatihan yang memadai. Naive Bayes menunjukkan performa yang relatif stabil pada berbagai ukuran data pelatihan. Selain itu, analisis kata kunci berbasis TF-IDF mengungkapkan bahwa ulasan negatif terutama berkaitan dengan masalah pengiriman, kendala teknis aplikasi, dan isu harga. Temuan ini menunjukkan bahwa pendekatan machine learning efektif untuk klasifikasi sentimen serta memberikan wawasan yang bermanfaat dalam meningkatkan kualitas layanan aplikasi mobile.   Kata Kunci: analisis sentimen; pembelajaran mesin; SMOTE; TF-IDF; klasifikasi teks.  

SENTIMENT ANALYSIS OF CUSTOMER REVIEWS ON E-COMMERCE APPLICATIONS: LAZADA, TOKOPEDIA, AND BLIBLI

Ihza, Andika, Arifin, Muhammad, Setiawan, Arif
Abstract: Abstract: The rapid growth of e-commerce in Indonesia has increased consumer interactions with digital platforms, particularly Lazada, Tokopedia, and Blibli, resulting in a large volume of customer reviews that reflect consumer… onsumer experiences and perceptions but have not been optimally utilized in business decision-making. The main issue addressed in this study is how to process customer review data to generate meaningful information regarding consumer opinions. This research aims to apply web scraping techniques to collect customer review data and conduct sentiment analysis to identify trends in consumer opinions across the three e-commerce platforms. The dataset consists of 3,000 customer reviews, with 1,000 reviews collected from each platform, covering aspects such as shopping experience, service quality, delivery process, and customer satisfaction. The research methodology includes data collection through web scraping, text preprocessing for data cleaning and normalization, sentiment analysis using machine learning approaches, and visualization of sentiment results. The findings indicate differences in the distribution of positive, negative, and neutral sentiments across platforms, reflecting variations in consumer experiences and service strategies. These results demonstrate that sentiment analysis based on customer reviews can serve as strategic input to improve service quality, business performance, and marketing strategies in Indonesia’s e-commerce sector.   Keywords: customer reviews; digital services; e-commerce; sentiment analysis; web scarping Abstrak: Pertumbuhan pesat e-commerce di Indonesia meningkatkan interaksi konsumen dengan platform digital, khususnya Lazada, Tokopedia, dan Blibli, yang menghasilkan ulasan pelanggan dalam jumlah besar sebagai cerminan pengalaman dan persepsi konsumen, namun belum dimanfaatkan secara optimal dalam pengambilan keputusan bisnis. Permasalahan utama penelitian ini adalah bagaimana mengolah data ulasan tersebut agar dapat memberikan informasi yang bermakna mengenai opini konsumen. Penelitian ini bertujuan menerapkan web scraping untuk mengumpulkan data ulasan pelanggan serta melakukan analisis sentimen guna mengidentifikasi tren opini konsumen pada ketiga platform e-commerce tersebut. Data yang digunakan berjumlah 3.000 ulasan pelanggan, dengan masing-masing platform diwakili oleh 1.000 ulasan yang mencakup pengalaman berbelanja, kualitas layanan, proses pengiriman, dan tingkat kepuasan pelanggan. Metode penelitian meliputi pengambilan data menggunakan web scraping, pra-pemrosesan teks untuk pembersihan dan normalisasi data, analisis sentimen dengan pendekatan pembelajaran mesin, serta visualisasi hasil sentimen. Hasil penelitian menunjukkan adanya perbedaan distribusi sentimen positif, negatif, dan netral pada setiap platform, yang mencerminkan variasi pengalaman konsumen dan strategi layanan. Temuan ini menunjukkan bahwa analisis sentimen berbasis ulasan pelanggan dapat menjadi masukan strategis untuk meningkatkan kualitas layanan, kinerja bisnis, dan strategi pemasaran e-commerce di Indonesia.   Kata kunci: customer reviews; digital services;e-commerce;sentiment analysis;web scarping

COMPARISON OF BILSTM, SVM FOR PBB-P2 TAX POLICY SENTIMENT ANALYSIS

Rofiqoh, Dayana, Subarkah, Pungkas, Isnaini, Khairunnisak Nur
Abstract: Abstract: The policy to increase the Rural and Urban Land and Building Tax (PBB-P2) in Indonesia often elicits mixed reactions from the public. Some support it because they believe it can strengthen regional fiscal capacity,… ity, while others reject it because they are concerned that it will increase the economic burden on the community. Understanding public sentiment towards this policy is important for evaluating the effectiveness of the policy and formulating appropriate communication strategies. This study aims to analyze public sentiment towards the PBB-P2 increase policy using data uploaded on Platform X (Twitter). The data were collected through crawling with the keyword “building tax,” then processed through several preprocessing stages before classifying tweets into positive and negative sentiments. Two models were used: Support Vector Machine (SVM) and Bidirectional Long Short-Term Memory (BiLSTM). Results show that SVM outperformed BiLSTM, achieving training accuracy of 99.4% and testing accuracy of 85.9%, with accuracy 0.8595, precision 0.8536, recall 0.8595, and F1-score 0.8449. Meanwhile, BiLSTM achieved training accuracy of 86.9% and testing accuracy of 82.9%, with accuracy 0.8294, precision 0.8150, recall 0.8294, and F1-score 0.8080. These findings suggest SVM is more effective in classifying public sentiment and can support better evaluation of regional tax policies.             Keywords: sentiment analysis; PBB-P2; BiLSTM; SVM; X platform     Abstrak: Kebijakan kenaikan tarif Pajak Bumi dan Bangunan Perdesaan dan Perkotaan (PBB-P2) di In-donesia sering memunculkan beragam reaksi dari masyarakat. Sebagian mendukung karena dianggap dapat memperkuat kapasitas fiskal daerah, sementara lainnya menolak karena kha-watir menambah beban ekonomi masyarakat. Pemahaman terhadap sentimen publik atas ke-bijakan tersebut penting untuk mengevaluasi efektivitas kebijakan dan merumuskan strategi komunikasi yang tepat. Penelitian ini bertujuan menganalisis sentimen masyarakat terhadap kebijakan kenaikan PBB-P2 menggunakan data unggahan di Platform X (Twitter). Data dik-umpulkan melalui proses crawling dengan kata kunci “pajak bangunan” kemudian diproses melalui beberapa tahap preprocessing sebelum diklasifikasikan menjadi sentimen positif dan negatif. Dua model digunakan dalam penelitian ini, yaitu Support Vector Machine (SVM) dan Bidirectional Long Short-Term Memory (BiLSTM). Hasil penelitian menunjukkan bahwa SVM memiliki kinerja lebih baik dibandingkan BiLSTM, dengan akurasi pelatihan 99,4% dan akurasi pengujian 85,9%. Nilai akurasi 0,8595, precision 0,8536, recall 0,8595, dan F1-score 0,8449. Sementara itu, BiLSTM memperoleh akurasi pelatihan 86,9% dan akurasi pengujian 82,9%, dengan akurasi 0,8294, precision 0,8150; recall 0,8294; dan F1-score 0,8080. Temuan ini menunjukkan bahwa SVM lebih efektif dalam mengklasifikasikan sentimen publik serta dapat mendukung evaluasi kebijakan pajak daerah dengan lebih baik.   Kata kunci: analisis sentimen; PBB-P2; BiLSTM; SVM; platform X

PREDICTION OF ON-TIME GRADUATION OF UNIVERSITAS ROYAL STUDENTS USING MULTIPLE LINEAR REGRESSION METHOD

Rahmadani, Nurul, Kurniawan, Edi, Nurhasanah, Nurhasanah, Damanik, Wahdan
Abstract: Abstract: On-time graduation is an important indicator in measuring the success of higher education and reflects the effectiveness of the academic process in higher education. Royal University, especially the Information… Systems Study Program, still faces challenges in increasing the percentage of students who graduate on time. This study aims to identify factors that influence students' on-time graduation and build a prediction model using the multiple linear regression method. This method was chosen because it is able to analyze the simultaneous influence of several independent numeric variables on one dependent variable, making it suitable for studying the complex relationship between factors that influence student graduation. The independent variables analyzed in this study include GPA, parental income, and student part-time jobs with student graduation as the dependent variable. The results showed that parental income and part-time jobs had a significant positive effect on on-time graduation, while GPA had a negative effect. The model built had an R² value of 0.6153 and a standard error of 4.0653, indicating that the model was quite strong and accurate. These findings recommend Universitas Royal to strengthen the academic monitoring system and support working students, as well as design policies based on students' socio-economic conditions to increase the on-time graduation rate. Keywords: multiple linear regression; on-time graduation; students.    Abstrak: Kelulusan tepat waktu merupakan indikator penting dalam mengukur keberhasilan pendidikan tinggi serta mencerminkan efektivitas proses akademik di perguruan tinggi. Universitas Royal, khususnya Program Studi Sistem Informasi, masih menghadapi tantangan dalam meningkatkan persentase mahasiswa yang lulus tepat waktu. Penelitian ini bertujuan untuk mengidentifikasi faktor-faktor yang memengaruhi kelulusan tepat waktu mahasiswa serta membangun model prediksi menggunakan metode regresi linear berganda. Metode ini dipilih karena mampu menganalisis pengaruh simultan beberapa variabel independen numerik terhadap satu variabel dependen, sehingga sesuai untuk mengkaji hubungan kompleks antar faktor yang memengaruhi kelulusan mahasiswa. Variabel independen yang dianalisis dalam penelitian ini meliputi IPK, penghasilan orangtua, dan pekerjaan sambilan mahasiswa dengan kelulusan mahasiswa sebagai variabel dependen. Hasil penelitian menunjukkan bahwa penghasilan orangtua dan pekerjaan sambilan berpengaruh positif signifikan terhadap kelulusan tepat waktu, sedangkan IPK justru memiliki pengaruh negatif. Model yang dibangun memiliki nilai R² sebesar 0,6153 dan standar error 4,0653, menandakan model cukup kuat dan akurat. Temuan ini merekomendasikan Universitas Royal untuk memperkuat sistem monitoring akademik dan mendukung mahasiswa yang bekerja, serta merancang kebijakan berbasis kondisi sosial-ekonomi mahasiswa guna meningkatkan angka kelulusan tepat waktu. Kata kunci: kelulusan tepat waktu; mahasiswa; regresi linear berganda.

SENTIMENT ANALYSIS OF THE HALODOC APPLICATION USING THE SUPPORT VECTOR MACHINE (SVM) ALGORITHM

Rachmadi Putri, Fairuz Amani, Siswanti, Sri
Abstract: Abstract: The Halodoc application, as a digital healthcare service platform, has been widely used for various medical purposes, such as doctor consultations, medication purchases, and laboratory services. User interactions… ns and reviews play a crucial role in enhancing service quality. Sentiment analysis was conducted using the Support Vector Machine (SVM) method to assess user perceptions and satisfaction based on reviews obtained from the Google Play Store platform. The analysis process included data collection, text preprocessing, data transformation using TF-IDF, and training an SVM model to predict sentiment. The model achieved its highest accuracy of 88.32% in the first scenario. However, accuracy slightly decreased in the second and third scenarios, reaching 86.25% and 86.94%, respectively. The analysis results indicated that the model performed best in the first scenario, with the lowest number of prediction errors. Additionally, the model was more accurate in classifying negative and positive sentiments than neutral ones.             Keywords: halodoc application; sentiment analysis; support vector machine algorithm   Abstrak: Aplikasi Halodoc, sebagai platform layanan kesehatan digital, telah banyak digunakan untuk berbagai keperluan medis seperti konsultasi dokter, pembelian obat, dan layanan laboratorium. Interaksi pengguna dan ulasan mereka memiliki peran krusial dalam meningkatkan mutu layanan. Analisis sentimen dilakukan dengan menggunakan metode Support Vector Machine (SVM) untuk mengetahui persepsi dan kepuasan pengguna berdasarkan ulasan yang diperoleh dari Platform Google Play Store. Proses analisis mencakup pengumpulan data, pra-pemrosesan teks, transformasi data menggunakan TF-IDF, dan pelatihan model SVM untuk memprediksi sentimen. Hasil pelatihan model dengan akurasi tertinggi sebesar 88,32% pada skenario pertama. Akurasi sedikit menurun pada skenario kedua dan ketiga, masing-masing sebesar 86,25% dan 86,94%, Hasil analisa menunjukkan bahwa model memiliki performa terbaik pada skenario pertama dengan jumlah kesalahan prediksi terkecil. Selain itu, model cenderung lebih akurat dalam mengklasifikasikan sentimen negatif dan positif dibandingkan netral..   Kata kunci: algoritma support vector machine; analisis sentimen; aplikasi halodoc  

SENTIMENT ANALYSIS USING NAIVE BAYES ALGORITHM CASE STUDY ON AMAZON E-COMMERCE PRODUCT REVIEWS

Rahman, Erik, Namora, Namora, Anas, Lukman
Abstract: Analisis sentimen adalah proses mengidentifikasi dan mengklasifikasikan opini dalam teks menjadi kategori tertentu seperti positif, negatif, atau netral. Penelitian ini bertujuan untuk menganalisis sentimen ulasan produk… pada platform e-commerce menggunakan algoritma Naive Bayes. Dataset ulasan produk diambil dari Kaggle, terdiri dari ribuan ulasan dengan label sentimen. Metodologi mencakup tahap preprocessing teks, ekstraksi fitur menggunakan teknik TF-IDF, dan penerapan algoritma Naive Bayes untuk klasifikasi sentimen. Hasil penelitian menunjukkan bahwa algoritma Naive Bayes memberikan akurasi sebesar 94%, membuktikan kemampuannya dalam analisis sentimen dengan dataset teks pendek

THE ROLE OF FEATURE SELECTION IN ENHANCING THE ACCURACY OF AI ASSISTANT AUTO-LABELING

Julianto, Indri Tri, Kurniadi, Dede, B. Balilo Jr, Benedicto, Rohman, Fauza
Abstract: Abstract: The development of AI assistants such as Gemini and ChatGPT can significantly assist in daily human tasks. In the field of Sentiment Analysis, AI assistants can be utilized as an automated labeling alternative… to provide positive, negative, or neutral sentiments within a dataset. This research aims to enhance the performance of AI assistants in automated labeling processes by employing the Feature Selection algorithm, specifically Forward Selection. The methodology involves utilizing the Naïve Bayes and K-NN algorithms, and subsequently improving accuracy through the Feature Selection algorithm. The evaluation is conducted using K-Fold Cross Validation. Research findings indicate an improvement in the accuracy of the best model, which is ChatGPT, when using the Naïve Bayes algorithm and Shuffled Sampling technique. The initial accuracy of 79.09% increased to 87.18% after Feature Selection was applied. This demonstrates the effectiveness of Feature Selection, particularly Forward Selection, in enhancing the accuracy performance of the model.             Keywords: ai; assistant; chat gpt; feature selection; gemini.     Abstrak: Pekembangan Asisten AI seperti Gemini dan Chat GPT dapat membantu pekerjaan manusia sehari-hari. Dalam bidang Analisis Sentimen, Asisten AI dapat digunakan sebagai alternatif pelabelan otomatis untuk memberikan sentimen positif, negatif atau netral dalam suatu dataset. Penlitian ini bertujuan untuk meningkatkan performa yang dihasilkan oleh Asisten AI dalam proses pelabelan otomatis menggunakan Algortima Feature Selection yaitu Forward Selection. Metode yang digunakan adalah dengan menggunakan Algoritma Naïve Bayes dan K-NN kemudian hasil akurasi akan ditingkatkan menggunkan Algoritma Feature Selection. Evaluasi yang digunakan adalah K-Fold Cross Validation. Hasil penelitian menunjukkan peningkatan akurasi model terbaik berada pada Chat GPT dengan menggunakan Algoritma Naïve Bayes dan Teknik Shuffled Sampling, dari nilai akurasi awal sebesar 79.09%, setelah ditingkatkan menggunakan Feature Selection, maka nilai akurasinya meningkat menjadi 87.18%. Hal ini membuktikan peran Feature Selection, dimana yang digunakan adalah Forward Selection dalam meningkatkan akurasi ternyata memang efektif dalam meningkatkan performa akurasi model.   Kata kunci: ai; assisten; chat gpt; feature selection; gemini  

DESIGNING TAX ME SERVICE APPLICATIONS IN TAX CONSULTING

Joosten, Joosten, Halim, Fandi, Arianto, Jennifer, Yulinda, Cindy
Abstract: Abstract: "Tax Me" is an application engaged in tax consulting services that is run through a mobile platform. The problems that are often faced by the public, especially taxpayers, are difficulties in choosing reliable… and trusted tax consultants because many tax consultants embezzle money and slow down the tax reporting process, as well as the lack of knowledge of taxpayers about all tax regulations, a complicated and frequently changing tax system, and also there are still negative public assumptions about taxes such as corruption in taxes that have been paid by taxpayers.  The "Tax Me" application was developed using the System Development Life Cycle (SDLC) system development methodology. This application was developed as a forum for tax consulting services in collaboration with the Directorate General of Taxes (DGT) institution which is able to provide transparent and reliable tax consulting services. "Tax Me" is expected to be able to provide education to taxpayers, overcome taxpayers' concerns about the tax calculation and reporting process as well as in choosing a tax consultant who can be trusted. In addition, several features of the "Tax Me" application such as: NPWP, Tax Calculation, Tax Consultation, Report & Check Annual Tax Return, E-Billing & Pay Tax, Tax Reminder and Tax Article.             Keywords: directorate general of taxes (DJP); tax; tax me; taxpayer     Abstrak: “Tax Me” adalah aplikasi yang bergerak pada bidang pelayanan konsultasi pajak yang dijalankan melalui platform mobile. Masalah yang sering dihadapi oleh masyarakat khususnya Wajib Pajak adalah kesulitan dalam memilih konsultan pajak yang dapat diandalkan dan dipercaya karena banyak konsultan pajak yang menggelapkan uang dan memperlambat proses pelaporan pajak, serta kurangnya pengetahuan Wajib Pajak mengenai segala peraturan perpajakan, sistem perpajakan yang rumit dan sering berubah, dan juga masih ada anggapan negatif masyarakat tentang pajak seperti korupsi atas pajak yang telah dibayarkan oleh Wajib Pajak. Aplikasi “Tax Me” dikembangkan dengan menggunakan metodologi pengembangan sistem System Development Life Cycle (SDLC). Aplikasi ini dikembangkan sebagai wadah layanan konsultasi perpajakan yang bekerja sama dengan lembaga Direktorat Jenderal Pajak (DJP) yang mampu memberikan layanan konsultasi perpajakan yang transparan dan terpercaya. “Tax Me” diharapkan mampu memberikan edukasi kepada Wajib Pajak, mengatasi kekhawatiran Wajib Pajak terhadap proses perhitungan dan pelaporan pajak juga dalam memilih konsultan pajak yang dapat di percaya. Disamping itu, beberapa fitur dari aplikasi “Tax Me” seperti: NPWP, Hitung Pajak, Konsultasi Pajak, Lapor & Periksa SPT Tahunan, E-Billing & Bayar Pajak, Pengingat Pajak dan Artikel Pajak.   Kata kunci: direktorat jenderal pajak (DJP); pajak; tax me; wajib pajak  

ANALYSIS OF PUBLIC OPINION SENTIMENT REGARDING POLICE INSTITUTIONS BASED ON TWITTER USING THE SUPPORT VECTOR MACHINE (SVM) METHOD

Sirojudin, Said Ahmad, Susanti, Try, Aribangsa, Mhd Theo
Abstract: Abstract: Twitter occupies the top position of the most popular social media platform in Indonesia. Police and other related issues were the subject of much discussion. The aim of this research is to analyze public sentiment… ment towards the National Police Agency using Twitter with the support vector machine method. The research started by crawling Twitter data. The data contains a total of 6,925 entries for three keywords. Next, we move on to the preprocessing stage consisting of (cleaning, case folding, tokenization, and filtering). Next is the tf-idf feature extraction stage, finally the classification and evaluation stage. The results of manual data inspection (73:27) showed accuracy of 70.66%, precision of 70.68%, and recall of 99.76%. Testing the second data (82:18), found accuracy 86%, precision 86.21%, recall 99.71%. The results of manual data checking (82:18) showed accuracy of 70.66%, precision of 70.68%, recall of 99.76%. Testing the second data (82:18), found accuracy 86%, precision 86.21%, recall 99.71%. From the data system testing results (80:20), accuracy was 87.55%, positive precision 87.53%, negative precision 88.24%, positive recall 99.48%, and negative recall. the rate is 99.48.% – The result is 21.43%. Data testing results (60:40) showed accuracy of 86.89%, positive precision of 86.84%, negative precision of 88.46%, positive recall of 99.61%, and negative recall of 16.43%. Single test data validation system (80:20), accuracy 87.55, overall test cross validation system (k fold 5 accuracy) 86.673%. Keywords: data mining;police agencies;support vector machines   Abstrak: Twitter menduduki posisi teratas platform media sosial terpopuler di Indonesia. Polisi dan masalah terkait lainnya menjadi pokok bahasan banyak pembicaraan. Tujuan penelitian ini untuk menganalisis sentimen masyarakat terhadap Badan Kepolisian Nasional menggunakan Twitter dengan  metode support vector machine. Penelitian dimulai dengan  crawling  data Twitter. Data memuat total 6.925 entri dari tiga kata kunci. Selanjutnya beralih ke tahap preprocessing terdiri dari (pembersihan, pelipatan kasus, tokenisasi, dan pemfilteran). Selanjutnya tahap ekstraksi fitur tf-idf, terakhir tahap klasifikasi dan evaluasi. Hasil pemeriksaan data manual (73:27) menunjukkan akurasi 70,66%, presisi 70,68%, dan recall 99,76%. Menguji data kedua (82:18), menemukan akurasi 86%, presisi 86,21%, recall 99,71%. Hasil pemeriksaan data secara manual (82:18) menunjukkan akurasi 70,66%, presisi 70,68%, recall 99,76%. Menguji data kedua (82:18), menemukan akurasi 86%, presisi 86,21%, recall 99,71%. Dari hasil pengujian sistem data (80:20), akurasi 87,55%, presisi positif 87,53%, presisi negatif 88,24%, recall positif 99,48%, dan recall negatif. tarifnya adalah 99,48.% – Hasilnya 21,43%. Hasil pengujian data (60:40) menunjukkan akurasi 86,89%, presisi positif 86,84%, presisi negatif 88,46%, recall positif 99,61%, dan recall negatif 16,43%. Uji tunggal sistem validasi data (80:20), akurasi 87,55, uji keseluruhan sistem validasi silang  (akurasi k fold 5) 86,673%.   Kata Kunci: data mining;instansi kepolisian;mesin vektor pendukung