Electricity has now become something that is very much needed in daily life, every household has used PLN electricity lighting sources. However, there are also some regions in Indonesia that have not been able to enjoy good and even electric lighting. The data source used from the Indonesian Statistics Agency website is the Percentage of Household Data by Province and Treatment, 2013-2014. This study aims to classify households that have a source of electricity for PLN using the datamining algorithm with K-Medoid. The data is processed into 2 clusters, namely high level clusters (C1) and low level clusters (C2). Where the results of this study concluded from 33 provinces in Indonesia that the cluster level of low waste sorting behavior (C1) obtained 11 provinces namely Aceh, Kep. Bangka Belitung, DKI Jakarta, West Java, Central Java, DI Yogyakarta, East Java, Banten, Bali, West Nusa Tenggara, and North Sulawesi and 23 other provinces are included in the low-level cluster (C2).Keywords: Electricity, Datamining, K-Medoid, Clustering, Household Sources

Internet media has become one of the means of product promotion that has very good prospects today. The research aims to recommend the right social media for online businesses. The data collection method was conducted by interview and questionnaire to 300 respondents who conducted online business in Pematangsiantar city by random sampling. Based on the results of interviews and questionnaires obtained assessment criteria namely security (C1), application features (C2), community (C3), ease of access (C4) and response speed (C5). The alternatives used in this research are Facebook (A1), Instagram (A2), Line (A3) and WhatsApp (A4). The settlement method applied is a decision support system with the PROMETHEE II algorithm. The results of the algorithm show that the right alternative for doing online business is Facebook (A1) with a net flow of 0.25 and followed by WhatsApp (A4) with a net flow of ¬0.1. The results of the study are expected to provide recommendations in conducting online business.

Backpropagation is one of the methods contained in a neural network that is able to train dynamic networks using mathematical knowledge based on architectural models that have been developed in detail and systematically. Backpropagation itself is able to accommodate a lot of information that serves as a useful experience. However, the Backpropagation Algorithm tends to be slow to achieve convergence in obtaining optimum accuracy and requires large training data and the optimization used is less efficient. The purpose of this research is to optimize the learning rate on backpropagation neural networks. Source of data obtained from CV. Bona Tani Hatonduhan. There are 3 network architecture models used in this study, namely 2-51, 2-6-1, and 2-7-1 with learning rates of 0.1, 0.2, and 0.3. the results of trials carried out with MATLAB software produced the best architectural model, namely the 2-7-1 model with a learning rate of 0.3 with an accuracy of 83%. Based on this background, it is hoped that the results of the research can help in the process by optimizing the learning rate of the backpropagation Neural Network on the selection of the best architecture.