Given the constantly changing job market, the recommendation of job information is inevitably hindered by factors such as the speed of recruitment information updates, leading to a lack of confidence in the suggestions provided. To address this challenge, social network data has been introduced to personalize job recommendations, solving the problem of insufficient updating speed of job information. By extracting user preferences from social networks, we have understood the interests and behaviors of job seekers, thereby improving job recommendations. The experimental results show that classifying work data and dynamically adjusting recommendations enhances confidence in work information and enhances the effectiveness of employment services.