Using graph theory and Malatya centrality algorithm in summarizing news texts

Fast and accurate access to information is one of the most important requirements of the digital age. With the proliferation of information sources, the need for text summarization methods to enable users to easily access information is increasing. Effective summarization of sources containing intensive information, especially news texts, both saves time and increases the understandability of information. In this study, a method based on the Malatya Centrality Algorithm was developed for summarizing news texts. The algorithm analyzes the relationships between words within the framework of graph theory and uses centrality values to determine important points in the text. In the study, the performance of the algorithm was evaluated using 2000 news texts from the BBC dataset and 1000 news texts from the CNN dataset. During the summarization process, the aim was to select the most important information while preserving the semantic integrity of the texts. The success of the algorithm was measured with ROUGE-1, ROUGE-2 and ROUGE-L metrics and compared with popular methods such as TextRank, PageRank, LexRank and TF-IDF. The test results show that the Malatya Centrality Algorithm offers competitive performance in terms of information coverage. However, the presence of unnecessary words in the summaries indicates that the algorithm needs to be further developed. This study shows that the proposed method offers an effective approach for summarizing news texts and can provide a new perspective on text summarization techniques. In the future, it is aimed to test the algorithm on different data types and to increase its performance with further optimizations. In this respect, the proposed method has significant potential in providing fast access to information and creating meaningful summaries.