Human Trafficking Detection System Using EfficientNet Model

Human trafficking is the world’s most prevalent and growing crime. Law enforcement agencies have been faced with a lot of challenges worldwide in this area. Over the years different works of literature have tried to sort out methods and techniques for detecting and controlling human trafficking by making use of different types of data available. In this research, an EfficientNet model was developed to help in the detection of human trafficking cases, by classifying an image into a class of chains. This helps to narrow the range of possibilities for an investigator. The object-oriented system analysis and design methodology were adopted and python was used in implementing the system. Data images and CSV files containing 97556 samples were obtained from the Kaggle repository and pre-processed by making use of the TensorFlow ImageDataGenerator function for normalization. After performing pre-processing, the top layers of the pre-trained model are replaced with the own data in fine-tuning. The data was used to train the model with a batch size of thirty-two (32), and the model was trained for 25 epochs. The model was evaluated against the accuracy, recall, and precision metrics, each of which recorded a score of 0.802. The metrics demonstrated that the model is robust in classifying hotel chains in detecting human trafficking activities.