Using Artificial Intelligence, A researcher in the Egyptian Russian University invents a technique for Summarizing Texts
President of the Egyptian Russian University, Dr. Sherif Fakhry Mohamed Abdel Nabi, announced that within the framework of motivating researchers to choose research topics related to practical problems, and to target research areas of global interest, including the use of artificial intelligence techniques, the researcher Eng. Mostafa Gamal Mohamed, Assistant Lecturer in the Department of Artificial Intelligence at the “Faculty of Artificial Intelligence” and a trainer at the Huawei Academy at the Egyptian Russian University, conducted a research study entitled “Summarizing Texts based on Artificial Intelligence Algorithms.” ERU president Pointed out that the scientific research activity at the university is supported by Dr. Mohamed Kamal El-Sayed Mustafa, Chairman of the University’s Board of Trustees.
For his part, Dr. Hisham Fathy, Dean of the Faculty of Artificial Intelligence at the Egyptian Russian University, added that the faculty scientific research activates the role of the university as a research and advisory institution that interacts with the needs of the society and the state. He pointed out that one of the faculty’s goals is to link academic study with scientific research to achieve the maximum possible benefits for the state and that the faculty pays great attention to areas that involve modern technologies, especially those evolving from “artificial intelligence”. Such tendencies aim at reducing the gap between man and machine, and at investing the capabilities provided by machines beyond the limited human capabilities in order to serve the society.
Engineer Mostafa Gamal Mohamed, Assistant Lecturer at the Department of Artificial Intelligence at the Faculty of Artificial Intelligence at the Egyptian Russian University, explained that one of the consequences of the huge progress that the world is witnessing in various fields is that we have a huge amount of information produced daily in all fields. Accordingly, it is very difficult to access the most important information such huge amount The massiveness makes accessing the most important information because such process requires exerting a great effort and time. He added that with the advancement of artificial intelligence and natural language processing in particular, the need for summarizing texts has become a necessity to overcome the problems of reaching excessive amount of knowledge so that the user can keep up with this huge amount of information and benefit from it so that the automatic summarization of texts using artificial intelligence techniques has become a must.
The researcher pointed out that this research overcame most of the problems that exist in the techniques that depend mainly on some statistical properties, which is defective as the resulting summary loses the general context and the links between the summary sentences. He explained that the invented algorithm relies on a set of characteristics through which the important sentences are selected from the original text, in addition to some other characteristics that keep the general context of the text and the gradual presentation of the main idea.
Engineer Mostafa Gamal Mohamed confirmed that the new algorithm offers a hybrid approach in the process of summarizing texts to solve some of the problems of the previous algorithms. He indicated that the proposed algorithm was evaluated on the standard data set from “CNN / Daily Mail”, and it was measured by: Recall-Oriented Understudy for Gisting Evaluation “ROUGE”, then the performance of the proposed method was compared with other methods, and the results showed that the new algorithm had the best performance in the quality of text summarization.
The Assistant Lecturer at the Department of Artificial Intelligence indicated that the proposed method in the research showed better accuracy than other algorithms like “ROUGE-1, ROUGE-2, and ROUGE-L”, and the increase in the highest accuracy was 4.4% compared to ROUGE-1, and 12.01% compared to ROUGE-2, and 9.8% compared to ROUGE-L.
Engineer Mustafa Gamal Mohamed noted that the results of the research were published in two peer-reviewed scientific papers in international journals included in the international Scopus classification under the title “Hybrid Algorithm Based on Chicken Swarm” Optimization and Genetic Algorithm for Text Summarization” and the other one was published under the title” Review of the Most Up-To-Date Optimization Algorithms for Extractive Text Summarization”.