Claim Missing Document
Check
Articles

Klasterisasi Data Penduduk Untuk Penerimaan Bantuan Pangan Non Tunai (BPNT) Menggunakan K-Means (Studi Kasus : Desa Tanimulya Bandung Barat) Djuniardi Suhardinata; Ade Kania Ningsih; Fatan Kasyidi
IJESPG (International Journal of Engineering, Economic, Social Politic and Government) Vol. 1 No. 3 (2023)
Publisher : IJESPG (International Journal of Engineering, Economic, Social Politic and Government)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26638/ijespg.v1i3.55

Abstract

Dalam perkembangan yang terjadi di masyarakat desa tanimulya yang pesat terdapat pula pembagian beberapa kelompok masyarakat yang di golongkan dalam hal perekonomiannya untuk golongan yang masuk dalam kalangan menengah keatas dan menengah munkin dapat bertahan dan terus berkembang, sedangkan untuk yang masuk dalam golongan kebawah, memerlukan bantuan dari orang lain, bantuan yang di butuhkan sudah di berikan, namun pembagian dari bantuan pangan non tunai (BPNT) ini, pemberiannya belum merata dan efektif,mengapa terjadi seperti itu karena data penduduk yang digunakan belum cukup mendukung untuk pembagian yang merata dan masih manual, oleh karena itu diperlukan sebuah sistem pengklusteran yang dimana nanti disesuaikan dengan aturan atau kategori yang sudah ada untuk mengelompokan masyrakat agar nantinya penyalurannya bantuan dapat lebih maksimal lagi, klusterisasi yang merupakan sebuah proses untuk mengelompokan data kedalam beberapa kluster dimanadi data dalam kluster memiliki kemiripan dengan sesama nya, data penduduk yang sudah di olah nantinya diharapkan dapat membantu dalam program bantuan pangan non tunai (BPNT) yang dijalankan saat ini.
Application of the Modified Apriori Algorithm to Determine Sales Patterns of Capacitor Products Erna Sesarliana*; Fajri Rakhmat Umbara; Fatan Kasyidi
Riwayat: Educational Journal of History and Humanities Vol 6, No 3 (2023): Social, Political, and Economic History
Publisher : Universitas Syiah Kuala

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24815/jr.v6i3.33718

Abstract

The sales pattern becomes a benchmark for increasing product sales in a company. Sales patterns are one way that can be used to determine sales strategies by looking at how often an item is purchased simultaneously. The large number of daily transactions makes it difficult for companies to determine sales strategies. Data mining analyses extensive data to find relationships between data and can produce valuable information. Patterns of sales or consumer transactions look for relationships between one product and another in one transaction using the Association Rule method. The algorithm used is the Modified Apriori Algorithm. The data used is transaction data on capacitor products. The data used is 15513 transactions with the variables LotNo and Material Code. Processed with the Python programming language and Flask as the user interface, the minimum support used is 0.01, and the minimum confidence is 0.5, resulting in rules with the lowest reliability of 50% and the highest reliability of 100%. Based on the results of a comparison of the performance time of the Modified Apriori Algorithm and the Classic Apriori Algorithm in processing 15513 transaction data with the given conditions, namely minimum support = 0.02 and also minimum confidence = 0.5 with the time obtained by the Modified Apriori Algorithm for 5 minutes 5 seconds and the Classic Apriori Algorithm for 6 minutes 6 seconds
OPTIMASI RESPONSIVITAS WEB DENGAN PENDEKATAN MOBILE-FIRST DESIGN Insan Kamil Nurhikmat; Asep Id Hadiana; Fatan Kasyidi
Jurnal Informatika Teknologi dan Sains (Jinteks) Vol 6 No 3 (2024): EDISI 21
Publisher : Program Studi Informatika Universitas Teknologi Sumbawa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51401/jinteks.v6i3.4314

Abstract

The rapid growth of mobile device usage for internet access has significantly shifted the approach to web design, with responsive and mobile-first design becoming crucial in web development. This study aims to optimize the user experience of Puskesmas web through mobile-first design. The research questions include strategies for improving performance and efficiency of Puskesmas web, measuring web responsiveness, and optimizing user experience. The research methods include a literature review, user needs analysis, prototype design, testing, data collection and analysis, and report compilation. The expected results are expected to enhance the responsiveness of Puskesmas web on mobile devices, benefiting web developers and users in achieving a better and more optimal experience.
PENGAMANAN FILE DOKUMEN DENGAN KOMBINASI ALGORITMA ELGAMAL DAN TEKNIK KOMPRESI ALGORITMA STOUT CODES Szalfa Saadiatus Sakinah; Asep Id Hadiana; Fatan Kasyidi
Jurnal Informatika Teknologi dan Sains (Jinteks) Vol 6 No 3 (2024): EDISI 21
Publisher : Program Studi Informatika Universitas Teknologi Sumbawa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51401/jinteks.v6i3.4315

Abstract

The rapid advancement of technology brings both positive and negative impacts, one of the negative impacts is the risk of hacking important data and documents. Document security becomes very important to protect sensitive information from unauthorized access and unwanted modification. This research aims to secure document files by combining ElGamal cryptographic algorithm and Stout Codes compression technique. The ElGamal algorithm used for encryption, converting the original information into unintelligible codes, while Stout Codes is applied to reduce the size of the data to make it more efficient in storage and transmission. The results show that the combination of these two algorithms is able to scramble the contents of the document into ciphertext that is difficult to understand by unauthorized parties. Tests were conducted on 10 document files in .docx and .pdf formats, resulting in an average compression ratio of 382% and space saving of -282% for the encryption-compression process, and an average compression ratio of 659% and space saving of -559% for the compression-encryption process. In terms of entropy, the encryption-compression process shows a higher value of 6.59828, compared to the entropy value in the compression-encryption process of 3.214782. This indicates a better level of security in the encryption-compression process.
Model Deteksi Botnet Menggunakan Algoritma Decision Tree Dengan Untuk Mengidentifikasi Serangan Click Fraud Rafli Firdaus; Asep Id Hadiana; Fatan Kasyidi
Journal of Informatics and Communication Technology (JICT) Vol. 4 No. 2 (2022)
Publisher : PPM Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52661/j_ict.v4i2.122

Abstract

Malicious Software (Malware) merupakan program yang dibuat khusus untuk merugikan orang lain. Salah satunya Botnet, di mana Botnet dapat menginfeksi perangkat komputer serta membuat komputer tersebut sebagai suatu alat yang nantinya akan dikendalikan secara paksa oleh pemilik dari program Malware tersebut. Botnet sendiri dapat melakukan serangan Click Fraud untuk melakukan Fake Clicks terhadap iklan yang bersifat Pay Per Click. Botnet dengan serangan Click Fraud memiliki pola tingkah laku yang dapat diklasifikasikan dengan menggunakan Dataset CTU-13. Sehingga Flow Traffic dari Botnet yang melakukan serangan Click Fraud akan dapat terdeteksi dengan menggunakan algoritma CART dengan menggunakan teknik SMOTE untuk melakukan Oversampling dan teknik Random Undersampling untuk menangati ketidakseimbangan sebaran data untuk setiap kelasnya. Dengan menggunakan rasio Undersampling yaitu 50% dan terdapat 2 skenario untuk penggunaan teknik SMOTE, yaitu sebelum dan setelah data dibagi menjadi data latih dan data uji. Berdasarkan dari hasil penelitian yang telah dilakukan dapat disimpulkan bahwa dengan penggunaan teknik SMOTE dan Random Undersampling dalam kasus untuk pendeteksian Botnet yang melakukan serangan Click Fraud sebelum membagi dataset menjadi data latih dan data uji dapat meningkatkan akurasi ataupun kinerja dari model tersebut dengan mencapai tingkat akurasi sebesar 99.97%. Dan Nilai F-Score dari model yang menggunakan SMOTE dan Random Undersampling adalah 99.96%.
Penanganan Overfitting pada Klasifikasi Berita Hoax berbasis Neural Networks dengan Dropout dan Regularization Ridwan Ilyas; Fatan Kasyidi; Maulidina Norick Eriyadi
JURNAL TEKNIK KOMPUTER AMIK BSI Vol 10, No 2 (2024): Periode Juli 2024
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/jtk.v10i2.23121

Abstract

Penelitian ini mengevaluasi efektivitas berbagai teknik deteksi hoaks di Indonesia menggunakan model klasifikasi teks dengan dua ukuran dataset berbeda, yaitu 250 dan 650 sampel. Hoaks di media sosial memiliki dampak signifikan pada masyarakat, sehingga deteksi yang akurat sangat penting. Penelitian ini menguji tiga algoritma machine learning—ID CNN, Bi-LSTM, dan LSTM—dengan teknik regulasi seperti original, regularization, dan dropout. Hasil penelitian menunjukkan bahwa teknik regularisasi pada ID CNN memberikan akurasi tertinggi pada dataset 250 sampel, sementara Bi-LSTM dengan teknik original mencapai akurasi tertinggi pada dataset yang sama. Dataset yang lebih besar (600 sampel) menunjukkan bahwa teknik regularisasi pada ID CNN tetap stabil, sedangkan teknik dropout memberikan hasil yang bervariasi. Analisis menggunakan confusion matrix dan grafik learning menunjukkan adanya overfitting pada model, terutama pada dataset yang lebih kecil. Temuan ini menegaskan pentingnya penerapan teknik regulasi untuk mengurangi overfitting dan meningkatkan generalisasi model dalam deteksi hoaks. Penelitian ini memberikan kontribusi pada pengembangan sistem deteksi hoaks yang lebih efektif di Indonesia.
Identifikasi Emosi Melalui Sinyal Elektroensephalogram Menggunakan Graph Convolutional Network LIONITAMA, VENA MEILINDA; DJAMAL, ESMERALDA CONTESSA; KASYIDI, FATAN
MIND (Multimedia Artificial Intelligent Networking Database) Journal Vol 9, No 1 (2024): MIND Journal
Publisher : Institut Teknologi Nasional Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/mindjournal.v9i1.42-51

Abstract

AbstrakEmosi merupakan bentuk respon manusia terhadap sesuatu. Pengenalan emosi menggunakan komputer dapat membantu para dokter untuk mengetahui emosi yang sedang dirasakan oleh seseorang berdasarkan aktivitas otak. Aktivitas otak dapat diketahui dengan cara merekam aktivitas sinyal Electroensephalogram (EEG). Sinyal EEG memiliki karakteristik yang berubah-ubah dan non stasioner sehingga membutuhkan metode yang dapat mengintegrasikan karakteristik temporal dan spasial. Pengenalan emosi menggunakan sinyal EEG berkaitan erat dengan pola konektivitas pada belahan otak manusia, karena setiap emosi akan memiliki pola konektivitas yang berbeda dalam belahan otak. Maka dari itu mempelajari pola konektivitas dalam belahan otak akan membantu dalam pengenalan emosi. Dan untuk menangani hal itu dibutuhkan metode deep learning yang dapat mengintegrasikan karakteristik temporal dan spasial dan dapat menerima masukan berupa pola konektivitas tersebut, metode yang dapat menanganinya yaitu, Graph Convolutional Network (GCN). Penelitian ini telah membuat sistem identifikasi emosi dengan tiga kelas menggunakan GCN dan menghasilkan akurasi data uji sebesar 35,52%.Kata kunci: Emosi; Deep Learning; Sinyal EEG; Spasial; Temporal; GCNAbstractEmotion is a form of human response to something. Emotion recognition using computers can help doctors to see the emotions that are being felt by a person based on brain activity. Brain activity can be known by recording electroencephalogram (EEG) signal activity. EEG signals have changing and non-stationary characteristics, requiring a method to integrate temporal and spatial characteristics. Emotion recognition using EEG signals is closely related to connectivity patterns in the human brain hemispheres because each emotion will have different connectivity patterns in the brain hemispheres. Therefore, studying the connectivity patterns in the cerebral hemispheres will help in emotion recognition. Moreover, a deep learning method is needed to integrate temporal and spatial characteristics and receive input in the form of connectivity patterns, a method that can handle Graph Convolutional Network (GCN). This research has created an emotion identification system with three classes using GCN and produced an accuracy of 35.52% of testing data.Keywords: Emotion; Deep Learning; EEG Signal; Spatial; Temporal; GCN
Klasifikasi Penyakit Jantung Tipe Kardiovaskular Menggunakan Adaptive Synthetic Sampling dan Algoritma Extreme Gradient Boosting Permana, Acep Handika; Umbara, Fajri Rakhmat; Kasyidi, Fatan
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5421

Abstract

Cardiovascular diseases are conditions that commonly affect the cardiovascular system, such as heart disease and stroke. According to data from the World Health Organization (WHO), 17.9 million deaths worldwide in 2019 were attributable to cardiovascular disease. Early detection is crucial, but diagnosing heart disease is complex in developing countries due to the limited availability of diagnostic tools and medical personnel. This study uses the Heart Disease Dataset from Kaggle, consisting of 15 attributes and 4238 records, to develop a heart disease classification model using XGBoost. The research stages include data imputation, data transformation using LabelEncoder, data balancing using ADASYN, data splitting (80% training data, 20% testing data), and hyperparameter tuning with Bayesian Optimization. The results show that the XGBoost model with ADASYN performs better, with a ROC-AUC of 0.971 and an accuracy of 0.916, compared to the model without ADASYN, which has a ROC-AUC of 0.698 and an accuracy of 0.841. Based on the research results, ADASYN has proven effective in improving model performance on imbalanced datasets. Additionally, Bayesian Optimization plays an important role in finding the optimal parameter combination, which can further enhance model performance. With this research, the impact is quite significant in the development of early detection methods for cardiovascular heart disease, particularly through the application of the XGBoost classification algorithm
Klasifikasi Sentimen Untuk Mengetahui Kecenderungan Politik Pengguna X Pada Calon Presiden Indonesia 2024 Menggunakan Metode IndoBert Oktariansyah, Indro Abri; Umbara, Fajri Rakhmat; Kasyidi, Fatan
Building of Informatics, Technology and Science (BITS) Vol 6 No 2 (2024): September 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i2.5435

Abstract

X has evolved into one of the most popular social media platforms in the world. In Indonesia, the use of X is quite widespread, especially in discussions about the presidential election, which is currently a hot topic. Everyone has different views on the candidates, both positive and negative. With a large amount of tweet data from users, this information can serve as a data source for processing and analysis. Various methods can be used to analyze and classify sentiment from this data, one of which is using BERT. This research conducts sentiment classification using BERT with the IndoBert model. The research aims to classify sentiments towards tweets related to the 2024 Indonesian presidential election to understand the political inclinations of X users, evaluate the performance of the IndoBert model in sentiment classification, and assess the extent to which back translation augmentation and synonym augmentation techniques can enhance the model's performance. Data was collected using crawling techniques for seven days leading up to the election and manually labeled by annotators. Synonym augmentation and back translation techniques were used to balance data in minority classes. The data was divided into 80% training data, 10% test data, and 10% validation data. The classification process was conducted using the IndoBert model that had been fine-tuned. The research results show that IndoBert with synonym augmentation achieved the highest accuracy, which was 82% in the first experiment and 81% in the second experiment. On the other hand, back translation only reached an accuracy of 78% in the first experiment and 74% in the second experiment. This indicates that synonym augmentation proved to be more effective in increasing data variation and model performance on the dataset used in this research.
PREDIKSI PENDAPATAN PADA MITRA TOKO PARFUME TRENDS MENGGUNAKAN METODE VECTOR AUTOREGRESSIVE INTEGRETED MOVING AVERAGE (VARIMA) Afifah, Hira Nur; Witanti, Wina; Kasyidi, Fatan
Technologia : Jurnal Ilmiah Vol 15, No 3 (2024): Technologia (Juli)
Publisher : Universitas Islam Kalimantan Muhammad Arsyad Al Banjari

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31602/tji.v15i3.15352

Abstract

Penelitian ini bertujuan untuk memprediksi pendapatan pada mitra Toko Parfume Trends menggunakan metode Vector Autoregressive Integrated Moving Average (VARIMA). Metode VARIMA dipilih karena kemampuannya dalam menganalisis dan meramalkan data deret waktu multivariat, serta menangkap berbagai pola dalam data, termasuk tren musiman dan hubungan antar variabel. Data yang digunakan adalah data sekunder dari Toko Parfume Trends, mencakup periode Januari 2021 hingga Juni 2024. Analisis kestasioneran data dilakukan menggunakan uji Augmented Dickey-Fuller (ADF), dan model dievaluasi berdasarkan Mean Absolute Percentage Error (MAPE). Hasil penelitian menunjukkan bahwa model VARIMA efektif dalam memprediksi pendapatan dengan nilai MAPE sebesar 0.3997. Temuan ini diharapkan dapat membantu mitra Toko Parfume Trends dalam merancang strategi bisnis yang lebih efektif dan mengoptimalkan pengelolaan risiko serta peluang pasar.