AI-Driven URL Classification for Targeted Advertising in the Digital Era

In the digital era, effective advertising is based on delivering appropriate content to an adequate audience at the ideal moment. With a wealth of online material, advertisers face the challenge of ensuring their ads resonate with their target audience. Motivated by the successful implementation of recommendation systems on online platforms, our research seeks to utilize AI-based technologies to classify website topics based on its URLs and target advertisements relevant to that topic. We conducted a series of experiments on the training dataset, employing various predictive models to generate predictions. To enhance the accuracy of these predictions, we implemented optimization techniques such as hyperband and ensemble learning. Additionally, we created a website using Streamlit to visualize the experimental results, showcasing the capability of our research to target advertisements based on input URLs. The findings demonstrate the efficiency and accuracy of our predictions, particularly with larger datasets, surpassing the results of previous studies. Consequently, advertisers can deliver more engaging advertisements, increase profits, and improve the user experience in the Vietnamese market.