تعداد نشریات | 50 |
تعداد شمارهها | 2,079 |
تعداد مقالات | 19,525 |
تعداد مشاهده مقاله | 22,889,279 |
تعداد دریافت فایل اصل مقاله | 21,125,201 |
Providing a Model for Preprocessing the Organizational Data in Order to Predict Insurance Business Processes | ||
Journal of New Researches in Mathematics | ||
مقالات آماده انتشار، پذیرفته شده، انتشار آنلاین از تاریخ 18 تیر 1401 | ||
نوع مقاله: research paper | ||
شناسه دیجیتال (DOI): 10.30495/jnrm.2022.62870.2139 | ||
نویسندگان | ||
Mehrdad Fadaei PellehShahi1؛ Sohrab Kordrostami ![]() | ||
1PhD candidate of Applied Mathematics, Department of Mathematics, Lahijan Branch, Islamic Azad University, Lahijan, Iran | ||
2Department of Mathematics, Lahijan Branch, Islamic Azad University, Lahijan, Iran | ||
3Computer and Information Technology Department, Ahrar Institute of Technology and Higher Education, Rasht, Iran | ||
4Department of Mathematics, Lahijan Branch, Islamic Azad University, Lahijan, Iran | ||
چکیده | ||
In this paper , a new data preprocessing method for predicting business processes is presented , using recursive neural networks , Markov chains and recursive deep learning . The aim of this study is to obtain high quality data and extract the information of the most important variables involved in the disability process of the Social Security Organization (S S O ) . For this purpose, the proposed method includes reducing the number of features and normalizing the data compared to the initial features . The method is implemented for real data of the Social Security Organization and is applied in the form of predictive method . T he results show that the proposed method increases the amount of memory usage , but the amount of CPU usage time becomes significantly lower than the methods compared . In addition, the presented method signifi cantly increases the accuracy and efficiency . | ||
کلیدواژهها | ||
Preprocessing؛ Prediction؛ Social Insurance Business؛ Recurrent Neural Network؛ Data Mining | ||
آمار تعداد مشاهده مقاله: 151 |