<![CDATA[Water is a one of the natural resources which is very important and necessary for the activity and survival of living things, as humans, animals and plants. River is one of the source from various alternative sources of water available for processing. But nowadays as growth of population grows, industrial growth, economic development and rising standard of living cause degradation of quality of water itself. Pollution of river occurs when in the water there are various substances or conditions that can reduce water quality standards that have been determined, so it can't be used for certain needs. Therefore, there is an effort to maintain the quality, quantity and continuity of river water by monitoring and measuring the quality of water. Previously, river water quality and measurement was measured using manual methods such as Water Pollution Index (IP), Water Quality Index (WQI) and STORET with high time and cost constraints. So that another method is needed to speed up the calculation process effectively and efficiently using Learning Vector Quantization (LVQ) method which can classify data into 4 water quality class of river based on 7 input parameters. The LVQ implementation process for river water classification begins with the dataset division, data training, data testing and classification that will result in a class of good, mild, moderate and heavy contaminated classes. The best average accuracy result is 81,13% using alpha 0,1, decrement alpha 0,4, comparison of training data and testing data 100: 35 from 135 total dataset, maximum epoch 10 and minimum epoch 0,001.]]>
Copyrights © 2017
