Single document extractive automatic text summarization using genetic algorithms

Giant information have been available on the digital environments. For this reason it has hard to get the information what you want to achieve it, fast and most efficiently. For this reason, the need to obtain a summary of the texts has emerged. This research presents automatic text summarizations' approaches, studies, software, algorithms, and methods. Automatic text summarization is generally divided into two systems. There are extraction based summarizations and abstraction based summarizations. Firstly, extractions summarization approach involves selecting sentences of high score from the document based on word and sentence features. In short, finding the most important sentences from text is performed. Secondly, it is examined the semantic relationships between words and sentences in abstraction approaches. The most important difference from the first approach, made semantic analysis of words and sentences afterward new words are obtained. In this study, we focus on an extractive text summarization system and we purpose to increase the success about this issue. We have tried 3 different methods and made measurements. We applied genetic algorithm methods and scored the sentences in all three methods. We have tried to get the highest scores. In the first methods we made the scoring of sentences heuristic methods. Secondly, we have determined the range of points in the heuristic method. In the third method, we set the range from 1 to 100. We gave random integer points in this range and have tested the system. We have identified an increasing success in the measurements made in 3 methods.