Text summarization with deep learning

Today, with the development of technology, applications of artificial intelligence have started to become widespread in our lives. Driverless vehicles, systems that detect voice commands, auxiliary robots that give different gestures and responses, drone etc. similar vehicles enter our lives and many more continue to enter. Artificial neural network are simple processors designed based on neurons in the brain. It has been developed based on the learning style of the human brain. Just as in our nerve cells, there are connections that provide information flow and learning between these nodes for the solution of certain complex problems. By training as in the brain, the accuracy rate is determined as a result of tests. Text summarization is to take a document as input and present it to the user in a shorter, simpler and more understandable form. The necessity in daily life is an indisputable issue. Daily applications such as magazines, newspapers and articles are areas where the benefits of summarizing can be seen at a high level. The first studies on text summarization were done in English about fifty years ago. Various statistical methods were used for the first time, and these methods remain valid because they meet the high performance and low cost targets. Since Turkish is an additive and canonical language, it has been cut out for language selection. In this thesis, text summarization, which is the branch of machine learning, which is one of the sub-branches of artificial intelligence, is discussed using deep learning and a summary of the text entered using the Keras library using the Tensorflow infrastructure is obtained. Very few data sets are currently available for Turkish text summarization. Generally, data sets are in English. In this thesis study, data sets for Turkish and English have been created from various databases and text summarization studies for related languages are included. It is thought that the thesis study will contribute significantly to the literature. Keywords: Text Summarization, Deep Learning, Artificial Neural Networks, Vector, Keras, Artificial Intelligence, Machine Learning