<![CDATA[The background of this research is information overload which is an effect of ease of information manipulation, storage and distribution. Massive amount of text documents available causes a decline in effectivity and efficiency of an individual when using information. Automatic Text Summarization can solve information overload by producing text document summaries. Purpose of this research is to create an Automatic Text Summarization algorithm and its implementation to create summaries of important information from text documents faster and can satisfy users needs of relevant and consistent summaries. The algorithm is based on sentence features scoring and Genetic Algorithm for determining sentence feature weights. Implementation consists of training phase (read text input, pre-summarization, summarization, and Genetic Algorithm to produce learned sentence feature weights) and testing phase (read text input, pre-summarization, summarization, and saving summary). The algorithm is evaluated by calculating summarization speed, precision, recall, F-measure, and subjective evaluation. The results are Automatic Text Summarization algorithm which is able to text documents by extracting important sentences which represent contents of original text documents. Conclusions of this research are Automatic Text Summarization algorithm can create extractive summaries which represent important information from a single document in Indonesian with faster summarization speed compared to manual process. Keywords: Automatic Text Summarization, Sentence Features, Genetic Algorithm, Latent Dirichlet Allocation]]>
