Information filtering via biased heat conduction
Information filtering via biased heat conduction
复制标题
通过偏热传导进行信息过滤
DOI：
10.1103/physreve.84.037101
复制
发表时间：
2011-09-07
影响因子：
2.4
通讯作者：
Guo, Qiang
中科院分区：
文献类型：
--
作者：
Liu, Jian-Guo;Zhou, Tao;Guo, Qiang
热传导的过程最近已经在个性化推荐中得到应用[Zhou et al.，Proc. Natl. Acad. Sci. USA 107，4511（2010）]，其具有高多样性但低准确性。通过降低小度数物体的温度，我们提出了一种改进的算法，称为偏置热传导，它可以同时提高精度和多样性。大量的实验分析表明，与标准热传导算法相比，MovieLens，Netflix和Delicious数据集的准确率分别提高了43.5%，55.4%和19.2%，并且多样性增加或基本不变。进一步的统计分析表明，该算法可以同时识别用户的主流和特殊口味，从而导致更好的性能比标准的热传导算法。该工作为高效的信息过滤提供了一种可靠的途径。
The process of heat conduction has recently found application in personalized recommendation [Zhou et al., Proc. Natl. Acad. Sci. USA 107, 4511 (2010)], which is of high diversity but low accuracy. By decreasing the temperatures of small-degree objects, we present an improved algorithm, called biased heat conduction, which could simultaneously enhance the accuracy and diversity. Extensive experimental analyses demonstrate that the accuracy on MovieLens, Netflix, and Delicious datasets could be improved by 43.5%, 55.4% and 19.2%, respectively, compared with the standard heat conduction algorithm and also the diversity is increased or approximately unchanged. Further statistical analyses suggest that the present algorithm could simultaneously identify users' mainstream and special tastes, resulting in better performance than the standard heat conduction algorithm. This work provides a creditable way for highly efficient information filtering.