Analyzing the Iron Age sites in Bijar city using archaeological prediction modeling and logistic regression: appraisal, processing, and performance of the model.

One of the predominant statistical methods in archaeological predictive modelling is logistic regression. The probabilistic model created by this method is appropriate for the purposes of this paper. When the dependent variable can be studied as a binary value, the logistic regression method is applicable. In this statistical approach, the binary value is referred to as the dependent variable with the value for the presence/absence of the archaeological site. The results in the probability value between 1 and 0. Environmental factors are defined as independent variables, and logistic regression calculates the relationship between the independent variables and the dependent variable, as well as the probable value of the dependent variable at all points on the map. With such an approach to statistical concepts and the use of logistic regression models with data from archeological field studies, this article analyzed and interpreted the results in the area of Bijar city in Kurdistan province. A total of 71 Iron Age sites as input for predictive modeling led to the presentation of an optimal proposal to prepare the model for such approaches in archeology. The results not only help save time and money and increase the accuracy of archaeological investigations for future projects, but also show a predictive accuracy of 90.4% by indicating areas of high probability and reducing the scope of investigation. It has also been shown that the independent variables of rivers and settlements had the greatest impact on the model's output and on the formation of areas in the landscape.