Tea commodity has a very strategic role for Indonesian economy. In 2012 the tea commodity was able to generate foreign exchange of US $ 156.74 million. Nationally, tea industry contributes a Gross Domestic Product (GDP) of around Rp. 1.2 trillion. The total tea plantation area in Indonesia entirely covering 123 938 hectares. The conception of sustainable tea production includes three aspects, economic development, social development and environmental protection. One of the steps towards sustainable tea production is survey and identification. The process continues with the selection of suitable planting material (seedlings). Gambung series, are planting material seeds that have been recommended by the Indonesia Ministry of Agriculture. The Gambung series has a potential yield of 4.000 - 5.800 kg/hectare of dried tea. The morphological similarity level of Gambung series is very high, because the elders of the clones are from the same crossing parents. Experts who are able to identify Gambung clone are very limited. This process is susceptible to errors and is very dependent on the presence of experts. If an error occurs in the process of identifying the type of clone, it will interferes with the breeding process. Errors in the selection of recommended clones will be detrimental to the process for long time period, due to the economic age of the tea plant can reach 50 years. From this issue, it is very necessary to design a system that is able to identify the planting material of Gambung series clones. The system is designed to classify Gambung and Non Gambung tea series using Convolutional Neural Network (CNN) with a high accuracy and low loss rate.

Cardiovascular disorders, including atrial fibrillation (AF) and congestive heart failure (CHF), are the major causes of mortality worldwide. The diagnosis of cardiovascular disorders is heavily reliant on electrocardiogram (ECG) signals. Therefore, extracting significant features from ECG signals is the most challenging aspect to represent each condition of the ECG signals. Earlier studies have claimed that the Hjorth descriptor is assigned as a simple feature extraction algorithm that has the capability of class separation among AF, CHF, and normal sinus rhythm (NSR) conditions. However, owing to noise interference, certain features do not represent the characteristics of the ECG signals. This study addressed this important gap by applying a discrete wavelet transform (DWT) prior to applying the Hjorth descriptor as a feature extraction method. Furthermore, the feature selection process and optimization of various classifier algorithms, including k-nearest neighbor (k-NN), support vector machine (SVM), and artificial neural network (ANN), were investigated to provide the best system performance. This study obtained accuracies of 95 %, 92 %, and 95 % for the k-NN, SVM, and ANN classifiers, respectively. The results demonstrated that the optimization of the classifier algorithm could improve the classification accuracy of AF, CHF, and NSR conditions, compared to earlier studies.