Abstract: The vast amounts of data generated from daily transactions within a community hold significant value when utilized effectively. However, the challenge lies in processing and utilizing this data as it continues to accumulate without appropriate tools. In this paper, we collected social domain data from the Al Khoums Municipality, representing citizen data for a period of 10 years. We selected, cleaned, and processed the data into a CSV file containing general statistical data about the municipality's population, categorized by the department in which they reside, known as "Mahalla." We employed machine learning methods and algorithms, along with Python, to extract information from this data. This enables decision-makers and stakeholders to use this data effectively. In this study, we utilized forecasting and aggregation methods to extract knowledge from social data, predicting the poverty rate and the growth and decline rate of the community's population. Our results provide insights into the current patterns of life within the municipality. In this paper, we used clustering and aggregation methods to extract knowledge from social data to predict the poverty rate and the rate of increase and decrease of the municipality's population. We obtained results that we interpreted and understood the current pattern of life within the municipality.

Keywords: Municipality, Dataset, Regression, Clustering, Increase, Decrease, Poverty, Machine Learning

Pages: 622-633