<p>Providing personalized recommendations for travelers has always been a key research topic in the field of cultural tourism. The existing recommendation algorithms are mainly based on tourists' historical behavior and preferences for recommendations. However, these algorithms are inadequate in recommending complex cultural tourism attractions and activities in Guangxi. In order to provide more accurate personalized travel recommendations, this study proposes an improved weighted association rule algorithm based on the Frequent Pattern (FP) growth algorithm. At the same time, this study also proposes to add anti pop recommendation algorithms in personalized attraction recommendations, so that personalized travel recommendations can better meet user needs. This study uses Wanro collector, crawler technology, and public data to collect user basic data. The data preprocessing methods include three methods: participle, stop use and keyword extraction. The results showed that the proposed scenic spot recommendation algorithm had an accuracy rate of over 95% in personalized scenic spot recommendation, while traditional recommendation algorithms had a recommendation accuracy rate of below 95%. The reverse popularity recommendation algorithm designed in the study can better enhance users' satisfaction with choosing tourist attractions. The anti pop recommendation algorithm designed in this study can effectively improve user satisfaction in selecting tourist attractions. This algorithm not only improves the personalized travel satisfaction of travelers, but also provides guidance for regional tourism industry development planning.</p>