Claim Missing Document
Check
Articles

Found 3 Documents
Search

Sistem Pakar Diagnosis Penyakit Tanaman Karet dengan Metode Fuzzy Mamdani Berbasis Web Hendrawan Hendrawan; Abdul Harris; Errissya Rasywir; Yovi Pratama
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 4, No 4 (2020): Oktober 2020
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v4i4.2521

Abstract

Rubber plants can be attacked by various diseases originating from fungi, pests, animals and even cancer cells. A method that is able to diagnose rubber disease is needed so that it is hoped that it can help farmers detect symptoms early so that the productivity of rubber plantations can increase. This study developed an analysis of the results of the diagnosis of rubber plant disease using the Mamdany Fuzzy method. The choice of this method departs from the fuzzy mamdany research which states that the fuzzy mamdany method is able to resemble the workings of the human brain intuitively. With the implementation of the Expert System for Diagnosis of Disease in Rubber Plants with the Fuzzy Mamdani Algorithm, the work of diagnosing rubber plant diseases can be done more automatically. With 33 sympthon parameter data for rubber plant disease symptoms and 14 classes of rubber disease diagnosis tested using the Mamdany Fuzzy algorithm, the results obtained an accuracy of 81.74%, a value of 5-cross validation of 80.93% and a value of 10-cross validation of 82.30%. This shows that the application of the fuzzy mamdani algorithm produces good accuracy in diagnosing rubber plants.
Optimizing Attack Detection for High Dimensionality and Imbalanced Data with SMOTE, Chi-Square and Random Forest Classifier Kurniabudi Kurniabudi; Abdul Harris; Veronica Veronica; Elvi Yanti
The IJICS (International Journal of Informatics and Computer Science) Vol 6, No 1 (2022): March 2022
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v6i1.3890

Abstract

The rapid growth of the network generates a very large and varied amount of traffic which has an impact on data and information security. This study resolves two common problems in attack detection, namely high dimensionality and high-class imbalance of the network traffic. This study used the ISCX CICIDS-2017 dataset. This study used the ISCX CICIDS-2017 dataset.  The CICIDS-2017 dataset is imbalance that contains very diverse types of traffic including normal traffic and several types of attacks (multi-class). This study proposes a combination of the Chi-Square feature selection technique with the Tree-Based Classifier Random Forest. In the experiment first the Chi-Square Correlation Based feature selection technique was applied to the imbalance dataset. The selected features are then validated using several Random Forest algorithms. The test was also performed comparisons with other classification algorithms such as Naïve Bayes, Bayes Network, J48, REPTree, and Adaboost. This study also examines the implementation of SMOTE to overcome the problem of high calass imbalance. The test results also show that the proposed ensemble method has a very good performance from the Accuracy, TPR, FPR, Precision, F-Measure, and ROC values
Optimalisasi Seleksi Fitur Untuk Deteksi Serangan Pada IoT Menggunakan Classifier Subset Evaluator Kurniabudi Kurniabudi; Abdul Harris; Elvira Rosanda
JURIKOM (Jurnal Riset Komputer) Vol 9, No 4 (2022): Agustus 2022
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v9i4.4618

Abstract

The Internet of Things (IoT) enables a wide variety of intelligent devices to connect and interact. The rapid development of technology and protocols as well as the growth of networks, makes IoT a security risk. The increasing number of interconnected intelligent electronic equipment has an impact on the complexity of the network and the increase in the volume of network traffic resulting in high-dimensional data. The feature selection technique has been proven to reduce very large (high-dimensional) network traffic data in the Intrusion Detection System (IDS). The feature selection technique is also faced with the problem of imbalanced data. In real network traffic data tends to be imbalanced, where attack traffic is less than normal data. IoT as a complex network produces a large number of features. However, not all features are relevant for identifying normal traffic and attacks. The right feature selection technique is needed to produce optimal features. In this study, a wrapper-based feature selection technique is proposed using a subset evaluator classifier with the J48 algorithm. The dataset used is CICIDS-2017 MachineLearningCSV version. Of the 78 features analyzed using the proposed method, 15 features were generated as optimal features. Optimal features are used for anomaly detection using the Random Forest algorithm. The experimental results show that attack detection with optimal features produces an average accuracy of 99.87% on training and testing data.