Imam Cholissodin

Improve Interval Optimization of FLR using Auto-speed Acceleration Algorithm

Inflation is a benchmark of a countrys economic development. Inflation is very influential on various things, so forecasting inflation to know on upcoming inflation will impact positively. There are various methods used to perform forecasting, one of which is the fuzzy time series forecasting with maximum results. Fuzzy logical relationships (FLR) model is a very good in doing forecasting. However, there are some parameters that the value needs to be optimised. Interval is a parameter which is highly influence toward forecasting result. The utilizing optimisation with hybrid automatic clustering and particle swarm optimisation (ACPSO). Automatic clustering can do interval formation with just the right amount. While the PSO can optimise the value of each interval and it is providing maximum results. This study proposes the improvement in find the solution using auto-speed acceleration algorithm. Auto-speed acceleration algorithm can find a global solution which is hard to reach by the PSO and time of computation is faster. The results of the acquired solutions can provide the right interval so that the value of the FLR can perform forecasting with maximum results.

Integration Method of Local-global SVR and Parallel Time Variant PSO in Water Level Forecasting for Flood Early Warning System

Flood is one type of natural disaster that can’t be predicted, one of the main causes of flooding is the continuous rain (natural events). In terms of meteorology, the cause of flood is come from high rainfall and the high tide of the sea, resulting in increased the water level. Rainfall and water level analysis in each period, still not able to solve the existing problems. Therefore in this study, the proposed integration method of Parallel Time Variant PSO (PTVPSO) and Local-Global Support Vector Regression (SVR) is used to forecast water level. Implementation in this study combine SVR as regression method for forecast the water level, Local-Global concept take the role for the minimization for the computing time, while PTVPSO used in the SVR to obtain maximum performance and higher accurate result by optimize the parameters of SVR. Hopefully this system will be able to solve the existing problems for flood detection due to erratic weather.

Optimasi Naïve Bayes Classifier Dengan Menggunakan Particle Swarm Optimization Pada Data Iris

Abstrak Klasifikasi adalah proses identifikasi obyek kedalam sebuah kelas, kelompok, atau kategori berdasarkan karakteristik yang telah ditentukan sebelumnya. Secara singkat, klasifikasi merupakan pengelompokan obyek berdasarkan kelompoknya yang biasanya disebut dengan kelas ( class ). Tak hanya klasifikasi, proses pengelompokkan obyek juga dapat dilakukan dengan menggunakan teknik clustering yang merupakan pengelompokan obyek berdasarkan kemiripan antar obyek. Salah satu metode klasifikasi yang sering digunakan adalah Naive Bayes Classifier . Menurut beberapa penelitian, Naive Bayes Classifier memiliki beberapa kelebihan yaitu, cepat dalam proses perhitungan, algoritma yang sederhana dan akurasi yang tinggi. Namun probabilitas pada Naive Bayes Classifier tidak bisa mengukur seberapa besar tingkat keakuratan sebuah prediksi, hasil akurasi metode ini juga masih kurang jika dibandingkan dengan metode C4.5, selain itu metode naive bayes juga memiliki kelemahan pada seleksi atribut. Untuk menyelesaikan permasalahan tersebut, algoritma particle swarm optimization (PSO) dapat digunakan untuk melakukan pembobotan atribut untuk meningkatkan akurasi naive bayes classifier . Kata kunci : Naive Bayes Classifier, Particle Swarm Optimization, klasifikasi, pembobotan atribut. Abstract Classification is the process of identifying objects into a class, group or category based on the predetermined characteristics. In other words, classification is a process to group objects based on their class. Grouping objects can be done not only by classification but also by clustering, which is grouping objects according to the similarity between objects. One of the most frequently used methods for classification is Naive Bayes Classifier. According to some researchers, Naive Bayes methods has its strength which is a simple and fast algorithm that can acquire a high accuracy . However, the probability of Naive Bayes methods cannot measure the level of accuracy of a prediction, the accuracy of the results of this method is still less than the C4.5 method, and Naive Bayes method has a deficiency on the selection of attributes. To solve this problem, Particle Swarm Optimization Algorithm (PSO) can be used to give weight to attributes to improve the accuracy of Naive Bayes Classifier. Keywords : Naive Bayes Classifier, Particle Swarm Optimization, classification, attribute weighting.

Segmentasi Kendaraan Menggunakan Improve Blob Analysis (BA) Pada Video Lalu Lintas

Abstrak Penggunaan citra digital untuk keperluan penelitian sudah banyak dilakukan, salah satunya yaitu segmentasi. Segmentasi berfungsi untuk mendeteksi objek - objek yang terdapat pada citra, sehingga hasil segmentasi sangat penting untuk proses selanjutnya. Pada penelitian ini diusulkan teknik optimasi hasil background subtraction menggunakan kombinasi frame difference (FD) atau difference image dengan filter SDGD dan running average (RA) atau background updating dengan filter SDGD untuk diterapkan pada blob analysis. Alasan utama menggunakan penggabungan kedua metode tersebut adalah karena seringnya terdapat piksel objek yang tidak mampu dideteksi sehingga akan mengurangi tingkat optimasi pengenalan objek. Hasil pengujian akurasi dari 10 data uji yang masing – masing terdiri dari 30 frame menunjukkan bahwa aplikasi ini memiliki nilai akurasi tertinggi yakni 90% untuk pengujian threshold dan 100% untuk pengujian ukuran structure element. Sehingga dapat disimpulkan bahwa aplikasi ini mampu melakukan segmentasi kendaraan dengan baik. Kata kunci: filter SDGD, blob analysis, video lalu lintas, background subtraction. Abstract The use of digital images for the purposes of research has been often applied, one of them is segmentation. Segmentation is used to detect objects contained in the image, so the segmentation result is very important for further processing. In this study, the results of the optimization technique proposed background subtraction using a combination of frame difference (FD) or a difference image with filter SDGD and running average (RA) or background updating with SDGD filter to be applied blob analysis. The main reason to use the merger of these two methods is that often there are pixels that are not able to detect objects that will reduce the level of optimization object recognition. The results of accuracy testing using 10 data testing for each data consisting of 30 frames shows that the system proposed in this paper has best accuracy of 90% for testing the threshold and 100% for testing the size of structure element. So it can be concluded that this system capable to segmentation the vehicle properly. Keywords: filter SDGD, blob analysis, traffic video, background subtraction

Classification of campus e-complaint documents using Directed Acyclic Graph Multi-class SVM based on analytic hierarchy process

E-Complaint documents provide information that can be used to measure or evaluate the services that given by campus to its students, lecturers, staff, and public. Using text classification, the documents can be classified based on its importance and urgency. This classification will be useful for campus to make the services better. Classifying the documents can also make the complaints follow-up from campus become faster than before. This paper discussed Directed Acyclic Graph Support Vector Machine (DAGSVM) method based on Analytic Hierarchy Process (AHP) to classify E-Complaint documents into four classes based on the importance and urgecy. Highest accuracy that is obtained from this research is 82,61% with Sequential Training SVM parameters are λ = 0.5, constant of γ = 0.01, Maxiter = 10, and ε = 0.00001, training data 70%, using stemming, and Gaussian RBF kernel without using AHP weight.

Integrating Line Tracking and 2D Haar Wavelet response for palmprint identification

The progress of palmprint identification has become more advanced. Efforts to produce this kind of biometric recognition have been focused on the accuracy and speed to recognize. The critical part of recognition is features representations. It must be distinctive yet produce sparse data distribution and later could be used to differentiate intended classes. This paper proposes Line Tracking that is utilized in order to detect line points to be integrated with 2D Haar Wavelet (2D HW) as the descriptors. Recognition is evaluated by matching the feature vectors to consider who own the palmprint. The provided class is constrained into only registered users. Achieved average matching score is 97.27% using 10-fold cross-validation.