The increasing deployment of public wireless access points (hotspots) and the rise of wireless computing devices such as tablets and mobiles have made it easier for people to access information on the Internet. Though convenient, such networks have potential security and privacy risks. However, most users are neglecting the privacy threats because currently there is no way for them to know to what extent their privacy is revealed. This paper has two goals, the first goal examine the users’ awareness of privacy leakage in public hotspots from activities such as web browsing, search engine querying, and use Social Networking, the second goal for this paper help the university decision-makers to consider the interests of public Wi-Fi users to open or close those frequently searched sites within the university’s domain and to make the ideal use of university resources. We have collected real data for 7295 users from three public Wi-Fi points distributed in the range of Minia University, as it covers the large range of the university. After analyzing the collected data and classify all visited website using machine learning model, we discover that 85% from the users use an unwanted links category, Which may cause his/her device’s information and privacy to be leaked, and 90% of the users use Social Network category, and 60% from the users use search engine category.