Claim Missing Document
Check
Articles

Found 10 Documents
Search

Prediksi Harga Minyak Kelapa Sawit Menggunakan Linear Regression Dan Random Forest Yusuf Supriyanto; M. Ilhamsyah; Ultach Enri
Jurnal Ilmiah Wahana Pendidikan Vol 8 No 7 (2022): Jurnal Ilmiah Wahana Pendidikan
Publisher : Peneliti.net

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (262.781 KB) | DOI: 10.5281/zenodo.6559603

Abstract

Commodity export is an important activity because it can open up new market opportunities abroad. Besides being able to increase investment and foreign exchange for a country, palm oil is a plantation product that plays an important role in the Indonesian economy, palm oil is the largest foreign exchange earner. In palm oil exports, the volume tends to increase from 2016 to 2019 but when viewed from the export value of palm oil, it tends to fluctuate. Therefore it is necessary to predict the price of palm oil to help make commodity export decisions and also help palm oil investors in maximizing profits, in this study to predict the price of palm oil used data mining methods with the implementation of the Linear Regression and Random Forest algorithms using rapidminer, with data sharing scenarios training and testing is divided into three, namely 90:10, 80:20 and 70:30 to determine the performance of the algorithm. The data that will be used for research is historical data on palm oil prices taken from investing.com. From the results of the implementation of the algorithm obtained in the 90:10 data sharing scenario, the best algorithm is Random Forest with RMSE 25,106 results, in the second scenario with 80:20 data sharing the best algorithm is Linear Regression with RMSE 31,174, in the third scenario 70:30 Linear data sharing. regression has the best result with RMSE 30,227. then from the three scenarios, the Linear Regression algorithm gets the best performance
Penerapan Algoritma Support Vector Machine (SVM) untuk Klasifikasi Berita Hoax Covid-19 Isnin Apriyatin Ropikoh; Rijal Abdulhakim; Ultach Enri; Nina Sulistiyowati
Journal of Applied Informatics and Computing Vol 5 No 1 (2021): July 2021
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v5i1.3167

Abstract

Hoax merupakan informasi yang dibuat oleh orang tidak bertanggung jawab dengan tujuan membuat orang lain mempercayai sesuatu yang tidak benar. Berita hoax yang paling mudah beredar adalah hoax tentang kesehatan. Di Indonesia sendiri semenjak diberitakan masuknya virus Covid-19, berita hoax tentang hal itu terus meningkat berdasarkan data yang dirilis oleh Kominfo periode Januari-Agustus 2020. Agar terhindar dari berita hoax ialah dengan lebih teliti membaca judul berita pada situs yang terpercaya seperti Kompas. Karena itu penelitian ini akan mengembangkan dan menganalisis model klasifikasi berita hoax Covid-19 dengan menerapkan algoritma Support Vector Machine (SVM) dengan metodologi Knowledge Discovery in Databases (KDD). Studi kasus penelitian ini dibagi dalam 2 kategori yaitu berita hoax yang didapat dari situs Trunbackhoax & Hoax buster sedangkan berita bukan hoax diambil dari situs berita Kompas. Hasil penelitian menyatakan bahwa Algoritma Support Vector Machine (SVM) dengan kernel linear memiliki hasil prediksi yang bagus pada skenario 3 (80:20) karena model sanggup dalam mengklasifikasikan berita hoax dan bukan hoax Covid-19. Akurasi yang didapat pada skenario 3 juga memiliki nilai akurasi tertinggi sebesar 97,06%. Sedangkan pada kernel RBF memiliki akurasi terendah pada skenario 4 (90:10) yaitu 90.46% dan model kurang bagus dalam mengklasifikasikan berita hoax maupun bukan hoax Covid-19.
Penerapan Algoritma C4.5 dalam Klasifikasi Status Gizi Balita Hajar Izzatul Islam; Muhamad Khandava Mulyadien; Ultach Enri
Jurnal Ilmiah Wahana Pendidikan Vol 8 No 10 (2022): Jurnal Ilmiah Wahana Pendidikan
Publisher : Peneliti.net

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (225.278 KB) | DOI: 10.5281/zenodo.6791722

Abstract

Posyandu in Dawuan Barat determines the nutritional status of children by looking at the growth chart in KIA and calculating the z-score manually and then matching the results to the category table and threshold, this takes a long time and is at risk of being inaccurate. The formulation of the problem in this study is how to do classification of data mining to determine the nutritional status of toddlers and how the evaluation results from the classification model. This study uses the C4.5 algorithm with the CRISP-DM methodology to classify the nutritional status of toddlers and uses a confusion matrix to determine the accuracy, precision, recall and f1-score values of the classification model and then the model is implemented into an application. The evaluation results of the 3 model scenarios in this study stated that scenario 1 produced the best performance among other models with 90% accuracy and 87% value of precision, recall and f1-score
Algoritma K-Means Untuk Pengelompokan Bantuan Langsung Tunai (BLT) Muhamad Khandava Mulyadien; Ultach Enri
Jurnal Ilmiah Wahana Pendidikan Vol 8 No 12 (2022): Jurnal Ilmiah Wahana Pendidikan
Publisher : Peneliti.net

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (395.262 KB) | DOI: 10.5281/zenodo.6944517

Abstract

The problem of poverty is a problem that often occurs in various regions in Indonesia, officials in West Dawuan Village continue to overcome these problems with various government assistance, one of which is Direct Cash Assistance (BLT). Often times in West Dawuan Village, there are obstacles in the distribution of the Direct Cash Aid funds received by parties who do not meet the criteria, there are residents who receive BLT successively and policy-making errors have a negative impact on the use of aid, especially in West Dawuan Village, Karawang. Using data mining clustering techniques, it is possible to group prospective beneficiaries according to the criteria to overcome these problems. This study uses the K-Means algorithm to classify BLT recipients in Dawuan Barat Village. The evaluation results from the resulting K-Means clustering model were then evaluated by using the SSE (Sum of Square Error) test. From the resulting K-Means clustering model, recommendations for prospective BLT recipients will be known according to the applicable criteria. With the help of the elbow method, the value of k=6 is obtained as the optimal k value and the results of the SSE cluster 6 test produce the best value among other clusters, which is 24551994078884.86. As a recommendation for BLT recipients, it can be done based on the best results from the SSE evaluation, namely cluster 6
Perbandingan Algoritma Backpropagation Neural Network dan Long Short-Term Memory dalam Memprediksi Harga Bitcoin Felix Andreas; Mikhael Mikhael; Ultach Enri
Jurnal Ilmiah Wahana Pendidikan Vol 8 No 12 (2022): Jurnal Ilmiah Wahana Pendidikan
Publisher : Peneliti.net

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (256.127 KB) | DOI: 10.5281/zenodo.7009768

Abstract

In actual practice, Bitcoin is the decentralized currency that allows two individuals to transact without third-party intervention. However, due to its high volatility, it has been such an attraction to investors to gain profit. But, that also mean that high volatility can also bring disadvantage if someone predicts the increase or decrease of the price of Bitcoin incorrectly. The technical analysis which is often used to predict Bitcoin prices has a weakness, that is specifically depends on the users of technical indicators. Therefore, it is necessary to use the Data Mining algorithm as an alternative solution to predict Bitcoin prices. In this paper, the implemented algorithms to predict Bitcoin prices are Long Short-Term Memory (LSTM) and Backpropagation Neural Network. The final results using T-Test showed there is no significant difference between LSTM and Backpropagation in predicting the data test with an average RMSE value of 661.580 and 1.812.503, respectively. However Backpropagation has the advantage to predict new data (outside of the dataset) with an average RMSE value of 629.545, while the average RMSE value of the LSTM is 2.818.248.
Prediksi Harga Kartu Grafis NVIDIA Berdasarkan Pengaruh Harga Cryptocurrency Menggunakan Support Vector Regression Mohammad Nurfaizy Pangestu; Mohamad Jajuli; Ultach Enri
Jurnal Ilmiah Wahana Pendidikan Vol 8 No 17 (2022): Jurnal Ilmiah Wahana Pendidikan
Publisher : Peneliti.net

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (284.592 KB) | DOI: 10.5281/zenodo.7076540

Abstract

The growing popularity of cryptocurrencies has caused the market demand for graphics cards to reach unusual heights for their efficient cryptomining capabilities. Graphics cards are not only used for crypto mining but also video editing, video streaming, and video games, this causes an unavailability of graphics card supply due to high demand, especially for cryptomining needs and leads to unusual prices increases which makes it difficult for graphics card consumers and miners to buy graphics cards at normal price. Therefore, it is necessary to predict the price of NVIDIA graphics cards based on the influence of cryptocurrency prices. The methodology used is KDD, and the algorithm used to make predictions is SVR because its ability to overcome the overfitting problem so it can produce more accurate predictions, besides that in this study the grid search algorithm is applied to determine optimal parameters. In this study, 6 graphics cards and 2 cryptocurrencies were used which produced the 6 best prediction models which were chosen based on the RMSE value. GTX 1050 has RMSE value of 0.2028, GTX 1050 Ti has RMSE value of 0.14564, GTX 1060 has an RMSE value of 0.07629, while in the RTX 30 series, RTX 3070 has an RMSE value of 0.03178, RTX 3080 has RMSE value of 0.0388, and RTX 3090 has RMSE of 0.06259. From these results, it can be stated that RTX 30 series has better accuracy than GTX 10 series in making predictions. RBF is better than linear which only excels on the GTX 1060.
ANALISIS SENTIMEN TERHADAP PERUBAHAN RUTE KRL COMMUTER JABODETABEK MENGGUNAKAN ALGORITME SUPPORT VECTOR MACHINE (SVM) Fauzapril Duta Sanubari; Ultach Enri; Susilawati Susilawati
Jurnal Ilmiah Wahana Pendidikan Vol 9 No 15 (2023): Jurnal Ilmiah Wahana Pendidikan
Publisher : Peneliti.net

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.5281/zenodo.8206986

Abstract

KRL Commuter Jabodetabek is one of the modes of public transportation that is an alternative choice for residents of the capital city of Jakarta and its surroundings to reduce congestion in the Jabodetabek area. However, since there was a change in the Jabodetabek Commuter KRL route on May 28, 2022, there have been pro and con opinions from among the public users of the Jabodetabek Commuter KRL public transportation mode. The data used in this research is tweet data from Twitter with the keyword 'krl route changes' with a time span between May 26, 2022 and February 28, 2023. This research uses the Knowledge Discovery in Database (KDD) method. The purpose of this study is to determine public sentiment towards changes in the Jabodetabek Commuter KRL route and to determine the performance evaluation value of Support Vector Machine (SVM) in analyzing public sentiment. This research compares 3 SVM kernels namely RBF kernel, Linear Kernel, and Polynomial Kernel with 3 dataset sharing scenarios (90:10, 80:20, and 70:30) and also compares the effect of using the Synthetic Minority Oversampling Technique (SMOTE) algorithm to handle data imbalance. This research resulted in positive labels totaling 17, neutral labels totaling 184, and negative labels totaling 140. And the best accuracy value was obtained by RBF kernel and Polynomial kernel in scenario 2 (80:20) with the same value of 88.2%.
Analisis Sentimen Ulasan Film Oppenheimer Pada Situs Imdb Menggunakan Metode Naive Bayes Fery Anuar Ramadhan Putra; Frido Firman Fadilah; Ultach Enri
Majalah Ilmiah UNIKOM Vol 21 No 2 (2023): Majalah Ilmiah Unikom
Publisher : Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/miu.v21i2.11338

Abstract

Penelitian ini bertujuan untuk menganalisis keakuratan sentimen film Oppenheimer berdasarkan ulasan penonton yang ditulis melalui situs web Internet Movie Database (IMDb) menggunakan metode Naive Bayes. Ulasan penonton di situs IMDb merupakan sumber informasi yang berharga dalam memahami pendapat dan tanggapan penonton terhadap suatu film. Dalam penelitian ini, peneliti mengimplementasikan metode klasifikasi algoritma Naive Bayes untuk mengklasifikasikan ulasan sebagai sentimen positif atau negatif. Data ulasan film dari IMDb dikumpulkan dan masuk ke tahap pre-processing, kemudian fitur-fitur yang relevan diekstraksi untuk melatih model Naive Bayes. Hasil evaluasi menunjukkan bahwa metode Naive Bayes dapat mengenali sentimen dalam ulasan film Oppenheimer dengan tingkat akurasi yang signifikan. Temuan penelitian ini memberikan wawasan berharga bagi industri film dalam memahami respons penonton terhadap film ini, dan informasi sentimen yang diperoleh dapat digunakan sebagai dasar untuk pengambilan keputusan yang lebih baik dalam pengembangan film dan pemasaran. Meskipun demikian, peneliti mengakui adanya keterbatasan, terutama dalam akurasi klasifikasi pada ulasan yang menggunakan bahasa yang ambigu atau tidak jelas. Oleh karena itu, untuk penelitian ke depannya dapat melibatkan metode lain atau menggabungkan beberapa metode untuk meningkatkan akurasi dan keandalan analisis sentimen ulasan film ini. Kata Kunci: Analisis Sentimen, IMDb, Naive Bayes, Ulasan Film
IMPLEMENTASI PHT: PENGGUNAAN AGENS BIOLOGIS BURUNG HANTU PUTIH UNTUK PENGENDALIAN HAMA TIKUS DI DESA PULOMULYA KECAMATAN LEMAHABANG KABUPATEN KARAWANG Afifah, Lutfi; Nurcahyo W. Saputro; Ultach Enri
Jurnal Abditani Vol. 7 No. 1 (2024): April
Publisher : FAKULTAS PERTANIAN UNIVERSITAS ALKHAIRAAT

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31970/abditani.v7i1.285

Abstract

The rat pest Rattus argentiventer is a severe problem in Karawang, it harms by damaging rice plants. Farmers do not have enough knowledge about how to deal with the pest. The partners in this activity are the Pulomulya Village Farmer Group, Lemahabang, with approximately 30 members. This community service provides knowledge about natural enemy conservation through planting refugia plants that can attract natural enemies and become a shelter for predators and parasitoids. As a result, farmers are able to conserve natural enemies by planting refugia around rice fields, and farmers are able to make artificial barn owl nests (rubuha) to control rat pests as an effort to implement Biointensive IPM. Farmer groups and communities have learned how to make rubuha and the required tools and materials. After the installation of the rubuha, Tyto alba barn owls came naturally and nested inside. This service allows farmers to use the biological agent Tyto alba and reduce the impact of using synthetic pesticides. With the successful use of these rubuha, the local government provided funding for a program to install 15 rubuha in 2023. The more rubuha installed, the better the natural rat control in Pulomulya Village. This innovation is an environmentally friendly solution to the rat problem in the area.
Analisis Sentimen Isu Childfree Di Media Sosial Twitter Menggunakan Algoritma Support Vector Machine Lidya Nurhidayati; Yuyun Umaidah; Ultach Enri
Jurnal Ilmiah Wahana Pendidikan Vol 10 No 4 (2024): Jurnal Ilmiah Wahana Pendidikan
Publisher : Peneliti.net

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.5281/zenodo.10521284

Abstract

AbstractSentiment analysis is a process to process, convert and interpret a text and classify it in the form of positive and negative sentiments. The phenomenon of childfree in Indonesia is currently causing debate and has become a trending topic on several social media, especially Twitter. The assumption that childfree decisions are categorized as selfish decisions is certainly closely related to the patriarchal culture that exists in Indonesia. This patriarchal culture is certainly very much in line with the concept of gender construction, where the childfree decision for women is considered a form of female selfishness. Based on this, an analysis of public sentiment related to the issue of childfree on Twitter social media using the Support Vector Machine (SVM) algorithm using 4 kernels. This research uses the KDD method by going through the stages of data selection, preprocessing, transformation, data mining, and evaluation. The data used are tweets totaling 1,447 tweets. The data was then selected into 1,447 which were divided into 1178 positive label data and 226 negative label data. In the data mining stage, the data is divided into 4 scenarios, namely 90:10, 80:20, 70:30, and 60:40. The best results were found in the first scenario with the Linear kernel, resulting in 75.93% accuracy, 83.33% precision, and 68.97% recall, showing the effectiveness of the algorithm in analyzing sentiment regarding the childfree phenomenon on Twitter. Keywords: Sentiment Analysis, Childfree, Support Vector Machine