The use of Artificial Neural Networks in Predicting Patterns of Contemporary Trends Towards Childbearing - Baghdad Governorate as A model

The research came to reveal contemporary patterns of childbearing in light of the emerging changes that affected life in all its details, the research data was extracted from a questionnaire distributed electronically to cover the nine districts of Baghdad Governorate, and (546) electronic forms were collected during a period of three weeks during the period from (1-21) June 2022, during the curfew period that accompanied the outbreak of the Corona virus. After three weeks, the electronic questionnaire was closed and the results were collected and unpacked in Excel and transferred to a SPSS file to be analyzed later. The study came in two sections, the first: descriptive statistics and the characteristics of the research sample, which included a theoretical introduction and description of the research community with a review of descriptive statistics related to demographic, social and cultural variables. The second: It came under the title of Statistical Analysis, in which we reviewed in detail the analysis of multi-layer neural networks and their role in prediction by creating a scale consisting of 8 items (the answer in which was determined by five choices according to the Likert five-way scale) to demonstrate its ability to classify the dependent variable (contemporary patterns of reproduction) Where it was classified into two categories according to reproductive preferences, the first category (prefer to have 4 children or less / Modern style) and the second category (prefer to have 5 children or more / the traditional style), this classification was according to age, where the cases were classified into two categories, the first category (with Ages 40 years and less) and the second category (those aged 40 years and over). The predictive model gave accurate results that can be considered very good and reliable in classifying the phenomenon under study.