A Survey on Prediction of Risk Related to Theft Activities in Municipal Areas using Deep Learning

Abstract

Deep learning techniques have been increasingly used technique in prediction and analysis. Analyzing the temporal patterns in the crime data and extracting relevant features from the demographic information is a big task. Machine learning involves using algorithms to learn patterns present in data and make predictions. It can be used to identify crime hotspots, predict criminal behavior, and forecast the likelihood of theft in specific areas. Deep learning, on the other hand, involves using artificial neural networks with multiple layers to model complex relationships in data. It is well-suited to large datasets and can be used to analyze images, audio, and text data in addition to numerical data. Deep learning can be used for theft crime prediction by identifying patterns in criminal behavior and helping to detect crime before it happens. Algorithms including Random Forest, Naive Bayes, XGBoost, and other models were used for prediction but all the mentioned models have drawbacks including low accuracy, low performance, etc. Overall, our study shows the potential of deep learning for crime prediction, emphasizing the value of using both demographic data and historical crime data in the modeling process and the shortcomings.