[1]刘金梅,舒远仲,张尚田,等.融合巴氏系数的加权 Slope One 算法[J].计算机技术与发展,2020,30(11):74-79.[doi:10. 3969 / j. issn. 1673-629X. 2020. 11. 014]
　LIU Jin-mei,SHU Yuan-zhong,ZHANG Shang-tian,et al.A Weighted Slope One Algorithm Based on Bhattacharyya Coefficient[J].,2020,30(11):74-79.[doi:10. 3969 / j. issn. 1673-629X. 2020. 11. 014]
点击复制

融合巴氏系数的加权 Slope One 算法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
30
期数:
2020年11期
页码:
74-79
栏目:
智能、算法、系统工程
出版日期:
2020-11-10

文章信息/Info

Title:
A Weighted Slope One Algorithm Based on Bhattacharyya Coefficient
文章编号:
1673-629X(2020)11-0074-06
作者:
刘金梅; 舒远仲; 张尚田; 唐小敏; 刘文祥
南昌航空大学 信息工程学院,江西 南昌 330063
Author(s):
LIU Jin-mei; SHU Yuan-zhong; ZHANG Shang-tian; TANG Xiao-min; LIU Wen-xiang
School of Information Engineering,Nanchang Hangkong University,Nanchang 330063,China
关键词:
巴氏系数; 加权 slope one 算法; 相似度; BCWSOA; 评分预测
Keywords:
Bhattacharyya coefficient; weighted slope one algorithm; similarity; BCWSOA; rating prediction
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2020. 11. 014
摘要:
针对用户共同评分很少甚至没有时, 传统相似度计算性能较差的问题,? 以及传统加权 slope one 算法在进行预测评分时未考虑项目相似度对预测结果的影响,? 提出一种融合巴 氏系数的加权 slope one 算法(BCWSOA)。 该算法主要针对上述两个问题进行改进,一是利用巴氏系数对用户相似度进行改进。 首先用皮尔逊相关系数计算用户局部相似度, Jaccard 相似性计算用户全局相似度,巴氏系数分析用户相关性,然后将巴氏系数作为权重因子优化用户局部相似度,最后使用参数α组合优化用户局部相似度和用户全局相似度,从而获得融合相似度。 参数α用来凸显不同相似度在融合相似度中的权重;二是利用巴氏系数改进预测评分,考虑项目相似度对预测结果的影响,计算项目相似度并将其作为权重改进加权slope one 算法预测评分公式。 通过在 Movie Lens100k 数据集上的实验表明,相比于其他算法,提出的 BCWSOA 算法准确度有所提高。
Abstract:
For the problem of poor performance of traditional similarity calculation when the user common rating is few or not,and the traditional weighted slope one algorithm does not take into account the effect of item similarity on the prediction results when making the prediction rating,a weighted slope one algorithm (BCWSOA) based on Bhattacharyya coefficient is proposed. The algorithm is mainly aimed at the above two problems. One is to use Bhattacharyya coefficient to improve the user similarity. Firstly,the local similarity is calculated by? ? ? the Pearson correlation coefficient,the global similarity is calculated by the Jaccard similarity and the user correlation is analyzed by the Bhattacharyya coefficient which is used as a weighting factor to optimize the user’s local similarity. Finally, the parameter α is used to optimize the user’s local similarity and the user’s global similarity,thus the fusion similarity is obtained. The parameter α is used to highlight the different similarities in the fusion similarity weights. The other one is to improve the prediction score by using the Bhattacharyya coefficient. Based on the impact of project similarity on prediction results,the similarity of items is calculated by using the Bhattacharyya coefficient,which is used as a weight to improve the weighted slope one algorithm prediction score formula. Experiments on MovieLens100k dataset show that compared with other algorithms,the accuracy of proposed BCWSOA is improved.

相似文献/References:

[1]刘金梅,舒远仲,张尚田.基于评分填充和时间的加权 Slope One 算法[J].计算机技术与发展,2021,31(01):35.[doi:10. 3969 / j. issn. 1673-629X. 2021. 01. 007]
　LIU Jin-mei,SHU Yuan-zhong,ZHANG Shang-tian.A Weighted Slope One Algorithm Based on Rating Filling and Time[J].,2021,31(11):35.[doi:10. 3969 / j. issn. 1673-629X. 2021. 01. 007]

更新日期/Last Update: 2020-11-10