Optimal Greedy Diversity for Recommendation

Optimal Greedy Diversity for Recommendation
复制标题

DOI:
--
复制
发表时间:
2015-07
期刊:
--
影响因子:
--
通讯作者:
Azin Ashkan;B. Kveton;S. Berkovsky;Zheng Wen
Azin Ashkan;B. Kveton;S. Berkovsky;Zheng Wen
中科院分区:
其他
文献类型:
--
作者:
Azin Ashkan;B. Kveton;S. Berkovsky;Zheng Wen

文献摘要

被引文献

相似文献

多样化的需求体现在各种推荐用例中。在这项工作中,我们提出了一种新颖的方法来多样化推荐项目列表,这使得项目的效用最大化,并受到其多样性的增加。从技术角度看,该问题可以看作是一个模函数在子模函数的多面体上的极大化问题,该问题可以用贪心法进行最优求解。我们在线下分析中评估了我们的方法,其中包括许多基线和指标,以及两个在线用户研究。在所有的实验中,我们的方法优于基线方法。
The need for diversification manifests in various recommendation use cases. In this work, we propose a novel approach to diversifying a list of recommended items, which maximizes the utility of the items subject to the increase in their diversity. From a technical perspective, the problem can be viewed as maximization of a modular function on the polytope of a submodular function, which can be solved optimally by a greedy method. We evaluate our approach in an offline analysis, which incorporates a number of baselines and metrics, and in two online user studies. In all the experiments, our method outperforms the baseline methods.