旅游路线个性化推荐算法比较分析

摘要

With the increasing of self-driving tours population,more and more people want to get the automatically generated route based on their specific needs,like special traveling days,traveling fees,special hotel expense quarterage. Ant colony algorithm and genetic algo-rithm are two classical algorithms used in 0-1 Knapsack Problem. It builds a mathematic model in this paper which can be applied to au-tomatically generated route. An evaluation method for assessing the best compatible recommending traveling route is also given,which is used to analyze and test the selected algorithms. The results from the experiment show that optimized ant colony algorithm and the hybrid genetic algorithm outperform the basic ant colony algorithm and genetic algorithm. And also found that the hybrid genetic algorithm can be used in the traveling route recommending system from synthetic property.%随着自助游群体的增加，越来越多的人希望能够在满足用户特定需求(如限定旅游天数、旅游费用、住宿标准等)的前提下，获取自动生成的可供参考的包括旅游景点、价格和住宿一体化的旅游推荐路线，并能够可视化呈现给用户。蚁群算法和遗传算法是0-1背包问题中的两种经典算法，通过建立应用于个性化旅游路线推荐问题中的数学模型，将蚁群算法和遗传算法应用于旅游路线个性化推荐中。依据文中所提出的最优路线推荐分值评价方法，对所选取的推荐算法进行了分析和测试。实验结果表明，优化后的蚁群算法和遗传算法均优于传统蚁群算法和遗传算法，并且从综合性能看，基于贪心解的混合遗传算法可有效应用于旅游路线个性化推荐中。