Garuda - Garba Rujukan Digital

Application of Data Mining For Clustering Car Sales Using The K-Means Clustering Algorithm Hutasoit, Michael Nico; Fa’rifah, Riska Yanu; Andreswari, Rachmadita
IJISTECH (International Journal of Information System and Technology) Vol 7, No 2 (2023): The August edition
Publisher : Sekolah Tinggi Ilmu Komputer (STIKOM) Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/ijistech.v7i2.307

In the digital era, data is at the core of business continuity. The need for fast, precise and accurate information is needed. Cars are one of the tertiary needs. This means of transportation is a relatively fast development and innovation business. Car sales in Indonesia recorded a reasonably high number in 2014 - 2018, namely 4.157.580 units sold. The highest sales were MPV car types being the most popular type of car, and there are many types of cars in Indonesia, including Sedans, SUV, 7 Seater SUV, and City Car types, and the enthusiasts need to play more. Hence, it is exciting to study. The variety of car brands with competing prices makes it difficult for consumers to choose the right car to buy according to their needs. This can be solved by applying data mining to cluster car sales using the k-means clustering algorithm. The goal is to know the characteristics of the car from each attribute. The k-Means algorithm is used for cluster formation based on five attributes: CC, Tank Capacity, GVW (Kg), Seater, and Door. The elbow and silhouette score methods determine the optimal number of clusters (k). The result is 4 clusters, cluster 0 (High-Performance Heavy Car), cluster 1 (Small Family Car), cluster 2 (High-Performance Small Car), and cluster 3 (Medium Performance Car). The 4 Cluster results are based on the evaluation/validation of the Elbow Method and Silhouette.

Sentiment analysis on public opinion of electric vehicles usage in Indonesia using support vector machine algorithms Poedjimartojo, Naufal Avilandi; Pramesti, Dita; Fa’rifah, Riska Yanu
Jurnal Teknika Vol 19, No 2 (2023): AVAILABLE ONLINE IN NOVEMBER 2023
Publisher : Faculty of Engineering, Universitas Sultan Ageng Tirtayasa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36055/tjst.v19i2.21967

Technological developments in the automotive industry have experienced significant progress in recent years. Currently, many electric vehicles are being produced as an environmentally friendly alternative to vehicles. The use of electric vehicles has become an intense topic of conversation in society, giving rise to various responses and opinions on Twitter. This research aims to analyze Indonesian people's sentiment regarding using electric vehicles through data collected from Twitter. Sentiment analysis is carried out using a machine-learning approach. The best method for pattern recognition problems is a Support Vector Machine (SVM) to sort each comment into positive or negative sentiments. Meanwhile, SVM classification performance was measured using the Confusion Matrix method. In this research, the Synthetic Minority Over-Sampling Technique (SMOTE) method and the Random Undersampling (RUS) method were used to overcome data imbalance. After the model creation and performance evaluation process, the best model produced was the baseline Support Vector Machine with a data sharing ratio of 70:30 without applying imbalance handling techniques. This model achieved an accuracy of 94.8%, a precision value of 95.5%, a recall value of 99.1%, and an F-1 Score value of 97.2%.

Application of Data Mining For Clustering Car Sales Using The K-Means Clustering Algorithm Hutasoit, Michael Nico; Fa’rifah, Riska Yanu; Andreswari, Rachmadita
IJISTECH (International Journal of Information System and Technology) Vol 7, No 2 (2023): The August edition
Publisher : Sekolah Tinggi Ilmu Komputer (STIKOM) Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/ijistech.v7i2.307

In the digital era, data is at the core of business continuity. The need for fast, precise and accurate information is needed. Cars are one of the tertiary needs. This means of transportation is a relatively fast development and innovation business. Car sales in Indonesia recorded a reasonably high number in 2014 - 2018, namely 4.157.580 units sold. The highest sales were MPV car types being the most popular type of car, and there are many types of cars in Indonesia, including Sedans, SUV, 7 Seater SUV, and City Car types, and the enthusiasts need to play more. Hence, it is exciting to study. The variety of car brands with competing prices makes it difficult for consumers to choose the right car to buy according to their needs. This can be solved by applying data mining to cluster car sales using the k-means clustering algorithm. The goal is to know the characteristics of the car from each attribute. The k-Means algorithm is used for cluster formation based on five attributes: CC, Tank Capacity, GVW (Kg), Seater, and Door. The elbow and silhouette score methods determine the optimal number of clusters (k). The result is 4 clusters, cluster 0 (High-Performance Heavy Car), cluster 1 (Small Family Car), cluster 2 (High-Performance Small Car), and cluster 3 (Medium Performance Car). The 4 Cluster results are based on the evaluation/validation of the Elbow Method and Silhouette.

PENERAPAN ALGORITMA TF-IDF DAN NAÏVE BAYES UNTUK ANALISIS SENTIMEN BERBASIS ASPEK ULASAN APLIKASI FLIP PADA GOOGLE PLAY STORE Helmayanti, Sheva Aditya; Hamami, Faqih; Fa’rifah, Riska Yanu
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 4 No. 3 (2023): September
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) STMIK Indonesia Banda Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v4i3.415

The development of the internet has changed people's lifestyle with the existence of FinTech. One of the popular FinTech innovations is the Flip digital wallet application. In this study, aspect-based sentiment analysis was carried out on Flip user reviews using the naive bayes algorithm. The test results show high accuracy, with an average accuracy of 0.84. The naive bayes algorithm is effective in classifying user reviews based on aspects of speed, security, and cost, with accuracies of 0.80, 0.87, and 0.84, respectively. This research provides important insights for service providers to improve service performance and innovation. The labelling data generated the most sentiment 0 (no sentiment), followed by sentiment 1 (positive) and 2 (negative). Negative sentiments have a high frequency on speed and security aspects, while positive sentiments have a high frequency on cost aspects. Thus, improvements are needed to the security system and speed of the Flip application to increase user satisfaction in these aspects. The naive bayes algorithm can be a useful tool in processing review data on e-wallet applications and similar services.

ASPECT-BASED SENTIMENT ANALYSIS TERHADAP ULASAN APLIKASI FLIP MENGGUNAKAN PEMBOBOTAN TERM FREQUENCY-INVERSE DOCUMENT FREQUENCY (TF-IDF) DENGAN METODE KLASIFIKASI K-NEAREST NEIGHBORS (K-NN) Febrianti, Ferda Ayu Dwi Putri; Hamami, Faqih; Fa’rifah, Riska Yanu
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 4 No. 3 (2023): September
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) STMIK Indonesia Banda Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v4i3.429

The rapid growth of online transactions in Indonesia has increased the demand for efficient interbank transfer solutions. However, the costs associated with such transactions have become a significant obstacle. Flip, a company with a vision to become a global leader in customer satisfaction-driven services, offers a solution to this challenge. This study proposes an aspect-based sentiment analysis method using the K-Nearest Neighbors (K-NN) algorithm to analyze user sentiment on key aspects, namely speed, security, and the cost of using the Flip application. The results of this research provide valuable information that can be used as a basis to provide insights, suggestions and recommendations to businesses, so they can create better solutions and promote optimal user experience. The research results show that the K-NN model has the ability to predict user psychology well in all aspects, with a significant level of accuracy, specifically speed (73.04%), security (86, 05%) and costs (80.11%). In addition, this study also compares two model validation methods: simple data splitting method and K-Fold cross-validation. Although the simple data splitting method has a higher average accuracy, K-fold cross-validation is considered superior as it provides a more accurate and reliable estimate of the overall performance of the model. Sentiment analysis results show that Flip app users tend to give negative feedback on speed and security, while they give positive feedback on cost. Therefore, the main recommendation is that the company PT Fliptech Lentera Inspirasi Pertiwi improves the speed and security aspects to increase user satisfaction with the Flip application. Therefore, this customer-centric service will continue to prioritize user satisfaction as its primary goal.

Strategi Manajemen Kapasitas untuk Menangani Traffic Load dengan Algoritma Dijkstra pada PT XYZ Rahmah, Najma Syarifa; Hamami, Faqih; Fa’rifah, Riska Yanu
Jurnal Teknologi Informasi dan Pendidikan Vol. 18 No. 1 (2025): Jurnal Teknologi Informasi dan Pendidikan
Publisher : Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/jtip.v18i1.886

In the digital era marked by rapid data growth, network congestion occurs when network capacity cannot handle the transmitted data volume, becoming a significant challenge that impacts the Quality of Service (QoS). This study proposes an effective capacity management strategy, which includes alternative routing, load balancing, and capacity enhancement, involving the optimized Dijkstra Algorithm and a Decision Support System (DSS) to improve QoS. The Dijkstra Algorithm is effective in reducing the load on congested routes and ensuring balanced traffic distribution, which enhances the capacity and reliability of the network. The results indicate a reduction in packet loss from 142.353 to 66.340 packets (or 53,39%) and an increase in the number of packets successfully transmitted from 122.542 to 198.555 packets (or 61,98%). Specific node capacity increases have significantly bolstered data transmission success. This comprehensive strategy not only improves the reliability of data transmission and network performance consistency but also enables the network to adapt to growing demands without degrading performance, confirming the efficacy of capacity management in addressing capacity challenges and improving QoS.

Empowering MSMEs with Data-Driven Insights: Mobile Sales Dashboard Application for MSMEs Azzahra, Zalina Fatima; Fa'rifah, Riska Yanu; Lathifah, Syfa Nur
Jurnal Nasional Teknologi dan Sistem Informasi Vol 11 No 1 (2025): April 2025
Publisher : Departemen Sistem Informasi, Fakultas Teknologi Informasi, Universitas Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25077/TEKNOSI.v11i01.2025.1-8

Micro, Small, and Medium Enterprises (MSMEs) in Indonesia have made a large contribution to GDP and the workforce but still face challenges in managing sales data and making data-driven decisions. Manual recording often causes operational inefficiencies, recording errors, and delays in business analysis. Based on the previous problem, this study develops a mobile-based sales dashboard application to help MSMEs analyse data in real-time and improve business strategies. The methodology used is Design Science Research (DSR) with a Rapid Application Development (RAD) approach for rapid and iterative development. This application was developed using Java and Firebase and provides sales summary features, best-selling cashiers, best-selling products, and less popular products, with time filters and graphical data visualization. Testing using Black Box Testing shows that all features run well, while the results of the User Acceptance Test (UAT) show that 90.625% of users feel that this application is easy to use and suits their needs. These results indicate that the application can improve operational efficiency and business transparency and support data-driven decision-making for MSMEs.

Analisis Klasifikasi Kualitas Udara Pada Tahun 2022 Menggunakan Metode Support Vector Machine Jauhari, M.Habib; Hamami, Faqih; Fa’rifah, Riska Yanu
eProceedings of Engineering Vol. 11 No. 4 (2024): Agustus 2024
Publisher : eProceedings of Engineering

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

Abstrak - Kondisi udara merujuk pada kondisi atmosfer di sekeliling kita, baik itu murni atauterkontaminasi. Kualitas atmosfer yang optimal tidakhanya sangat penting bagi manusia, melainkan jugamemiliki arti yang besar bagi makhluk hidup lainnyaseperti flora, air, dan lingkungan tanah. Dengan adanyakebijakan penduduk Jakarta banyak melakukan kegiatandiluar ruangan. Maka, dibutuhkan klasifikasi untukmendapatkan informasi kualitas udara pada Kota Jakarta.Jadi, salah satu cara melakukan klasifikasi untuk mengetahuiinformasi kualitas udara ialah menggunakan data mining.Data mining adalah langkah- langkah untuk mengambildata atau pola yang sebelumnya diketahui, secara tersirat,dan dianggap tidak memiliki nilai sebagai informasi ataupengetahuan berharga dari data dalam jumlah yang besar.Metode data mining klasifikasi digunakan karena dapatmengubah data parameter ISPU menjadi informasi yangmenunjukkan kualitas udara hariannya. Data kualitasudara dikumpulkan dalam penelitian ini pada BulanJanuari hingga November tahun 2022 dan data tersebutdiuji dengan algoritma Support Vector Machine untukmelakukan klasifikasi kualitas udara Kota Jakartaberdasarkan nilai akurasi, presisi, recall, dan skor F1.Hasil didapat dari algoritma SVM dengan rasio terbaikuntuk klasifikasi kualitas udara Kota Jakarta denganperbandingan 80:20 mendapatkan nilai precision 15%, nilai recall 54%, accuracy 90%. Kata kunci— Data Mining, Klasifikasi, Kualitas Udara, Support Vector Machine

Analisis Sentimen Terhadap Penggunaan Artis Korea Selatan Sebagai Brand Ambassador Pada Produk Makanan Dan Minuman Lokal Menggunakan Algoritma Support Vector Machine (Svm) Pariswara, Endar; Pramesti , Dita; Fa'rifah, Riska Yanu
eProceedings of Engineering Vol. 11 No. 4 (2024): Agustus 2024
Publisher : eProceedings of Engineering

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

Abstrak — Penggunaan brand ambassador pada suatuproduk dinilai dapat meningkatkan daya jual. Biasanyaperusahaan menggunakan artis atau public figure sebagai brandambassador. Penggunaan artis atau public figure bukanlahtanpa suatu alasan. Public figure yang memiliki banyakpenggemar atau pengikut di berbagai media sosialnya dapatdijadikan sebagai target dalam melakukan pemasaran.Maraknya fenomena Korean wave yang sedang terjadi diIndonesia membuat produk lokal saling bersaing dalammenjadikan artis asal Korea Selatan sebagai brand ambassador.Penggunaan artis Korea Selatan sebagai brand ambassadorproduk lokal menuai komentar positif maupun negatif darimasyarakat Indonesia. Komentar-komentar tersebut banyakdituangkan oleh masyarakat di berbagai media sosial,contohnya pada media sosial Twitter. Terhadap komentartersebut akan dilakukan analisis untuk mengetahui sentimenmasyarakat mengenai penggunaan artis Korea Selatan sebagaibrand ambassador produk makanan dan minuman lokal. Padapenelitian ini akan digunakan algoritma Support VectorMachine (SVM) untuk mengklasifikasi setiap data ke dalamsentimen positif dan sentimen negatif menggunakan metodeimbalance handling seperti SMOTE untuk oversampling, danRUS untuk undersampling. Setelah melakukan prosespengklasifikasian, selanjutnya diterapkan evaluasimenggunakan confusion matrix. Penelitian ini menghasilkanmodel SVM terbaik dengan penerapan metode SMOTE dengannilai accuracy tertinggi pada fold ke-10 mencapai 83,89%.Selain itu, terdapat nilai recall sekitar 80%, precision 85%, danF1-Score 82%. Kata kunci— Brand Ambassador, Twitter, Analisis Sentimen, Support Vector Machine (SVM), SMOTE, RUS

Analisis Sentimen Ulasan Aplikasi Maxim Untuk Peningkatan Layanan Menggunakan Algoritma Naïve Bayes Afdillah, Dwivi; Fa’rifah, Riska Yanu; Witarsyah , Deden
eProceedings of Engineering Vol. 11 No. 4 (2024): Agustus 2024
Publisher : eProceedings of Engineering

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

Abstrak — Transportasi online kian menarik perhatian danminat masyarakat untuk dijadikan salah satu kebutuhan yangdapat memudahkan aktivitas sehari-hari. Maxim merupakansalah satu transportasi online yang menepati urutan ketiga jasatransportasi online yang paling sering digunakan. Denganadanya hasil pemeringkatan jasa transportasi online tersebutmaka pihak Maxim dapat meningkatkan pelayananberdasarkan dari ulasan pengguna aplikasi Maxim melaluimetode analisis sentimen yang bertujuan untuk mengolahsejumlah besar data secara selektif dan efisien denganmengelompokkan ke dalam dua ulasan yaitu positif dan negatifdengan menggunakan algoritma Naïve Bayes. Tahap untukmelakukan analisis sentimen dibagi menjadi beberapa bagiandiantaranya yaitu pengumpulan data, preprocessing, modelling,dan evaluasi. Pada penelitian ini, menggunakan tiga skenariorasio pembanding training testing yaitu 60:40, 70:30, dan 80:20dengan menggunakan model Multinomial Naïve Bayesmenunjukkan bahwa rasio 70:30 menghasilkan model terbaikdengan nilai akurasi sebesar 87,22% yang dilakukan melaluiconfusion matrix dengan nilai recall sebesar 98,49%, nilaiprecision sebesar 86,62%, dan f1 score menghasilkan nilaisebesar 92,17%. Berdasarkan dari hasil visualisasi sentimendidapatkan hasil ulasan pengguna aplikasi Maxim cenderungmengarah ke sentimen positif dengan jumlah ulasan positifsebanyak 76% dan sentimen negatif sebanyak 24%. Sehingga,hasil kategori sentimen tersebut dapat dijadikan bahan evaluasikepada pihak developer sebagai bentuk peningkatan layananaplikasi Maxim. Kata kunci— Analisis Sentimen, Transportasi Online, Naïve Bayes

Title

Found 30 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 30 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 30 Documents
Search