Keyword extraction for academic papers by deep learning method

With the development of technology recently, news, social media, banking, education, etc. The amount of data is increasing rapidly in each area. With the rapid development of technology and the increase in the number of academicians, the amount of data in the academic field increases rapidly. The excessive increase in the number of data makes it difficult to reach correct information. Due to the excessive increase in data, especially in textual data, keywords are needed that can summarize the text without losing any purpose in the document. Appropriate keywords in the texts can serve as a highly concise summary of a document and can help us easily organize the documents and get them based on their content. However, most of the documents do not have assigned keywords. On the other hand, manually assigning high-quality keywords is expensive, time-consuming and error-prone. For this reason, most algorithms and systems have been proposed to help people perform automatic keyword extraction. There are studies on keyword extraction in the literature. Machine learning algorithms are generally used in the studies. In the extraction of classic keywords, the most frequently used words in a text are given and the keywords of the article are given. In this study, the keyword suggestion is made by looking at the most frequently used words in the text and by looking at how many words have passed with the other words. In addition, according to the most frequently used words, with the corpus we have, the keyword extraction process was carried out without finding the word groups in the article.