基于分布式学习自动机和用户反馈的网页排序算法研究
|
摘要
为了改善传统PageRank算法存在的不足,例如平分链接权重、主题漂移和忽略用户兴趣,提出一种基于分布式学习自动机和用户反馈的网页排序算法。利用页面内容的相似性、网页之间的超链接和用户遍历的路径,根据分布式学习自动机来确定网页间的超链接权重。考虑到用户反馈包含大量的价值信息,选择用户的转载、回复以及有效点击特征作为用户的行为特征,获得用户反馈因子。根据网页间的超链接权重和用户反馈因子计算每个网页的排名。仿真实验表明,与传统的PageRank算法和WPR算法相比,该算法在一定程度上提高了信息检索的精准度和用户满意度。
To improve the shortcomings of traditional PageRank algorithm, such as equalizing link weight, topic drift and ignoring user interest, we proposed PageRank algorithm based on distributed learning automata and user feedback. The algorithm used the similarity of page content, hyperlinks between pages and user traversal paths to determine the weight of hyperlinks between pages based on distributed learning automata. Considering that user feedback contained a lot of value information, users' reprint, reply and effective click characteristics were selected as user behavior characteristics to obtain user feedback factors. We calculated the ranking of each web page according to the weight of hyperlinks and user feedback factors. The simulation results show that compared with traditional PageRank algorithm and WPR algorithm, the algorithm improves the accuracy of information retrieval and user satisfaction to a certain extent.
To improve the shortcomings of traditional PageRank algorithm, such as equalizing link weight, topic drift and ignoring user interest, we proposed PageRank algorithm based on distributed learning automata and user feedback. The algorithm used the similarity of page content, hyperlinks between pages and user traversal paths to determine the weight of hyperlinks between pages based on distributed learning automata. Considering that user feedback contained a lot of value information, users' reprint, reply and effective click characteristics were selected as user behavior characteristics to obtain user feedback factors. We calculated the ranking of each web page according to the weight of hyperlinks and user feedback factors. The simulation results show that compared with traditional PageRank algorithm and WPR algorithm, the algorithm improves the accuracy of information retrieval and user satisfaction to a certain extent.
引文
[1] Brin S, Page L. The Anatomy of a Large-scale HyperTextual Web Search Engine[J]. Computer Networks and ISDN System, 1998,30(1-7): 107-117.
[2] Makkar A, Kumar N. User behavior analysis-based Smart Energy Management for Webpage Ranking: Learning Automata-based Solution[J]. Sustainable Computing: Informatics and Systems 2018, 20(12): 174-191.
[3] Anari Z, Meybodi M R, Anari B. Web page ranking based on fuzzy and learning automata[C]//International Conference on Management of Emergent Digital Ecosystems. ACM, 2009:162-166.
[4] 金秋, 裴斐, 林馥. 学习自动机结合节点功率自适应调整的WSN目标覆盖方案[J]. 计算机应用研究, 2017, 34(1):177-180.
[5] 陶彤彤. 学习自动机及其在随机定位应用中的研究[D]. 上海:上海交通大学, 2014.
[6] 荆羽纯, 葛昊, 江文,等. 一种基于学习自动机的推荐算法改进[J]. 计算机应用研究, 2016,33(1):32-34.
[7] Torkestani J A. An adaptive learning automata-based ranking function discovery algorithm[J]. Journal of Intelligent Information Systems, 2012, 39(2):441-459.
[8] Mehr S M, Taran M, Hashemi A B, et al. Determining web pages similarity using distributed learning automata and graph partitioning[C]//International Symposium on Artificial Intelligence and Signal Processing. IEEE, 2011:129-134.
[9] 陈超, 周静, 万懿. 基于DLA的调度算法解决定向传感器网络连接目标覆盖问题[J]. 计算机应用研究, 2015, 32(10):3108-3113.
[10] 董洋溢, 李伟华, 于会. 基于混合余弦相似度的中文文本层次关系挖掘[J]. 计算机应用研究, 2017, 34(5):1406-1409.
[11] 王冲, 纪仙慧. 基于用户兴趣与主题相关的PageRank算法改进研究[J]. 计算机科学, 2016, 43(3):275-278.
[12] 王冲, 王凯. 基于用户反馈特征聚合的网页排序算法[J]. 计算机工程与设计, 2017, 38(8):2020-2024.
[2] Makkar A, Kumar N. User behavior analysis-based Smart Energy Management for Webpage Ranking: Learning Automata-based Solution[J]. Sustainable Computing: Informatics and Systems 2018, 20(12): 174-191.
[3] Anari Z, Meybodi M R, Anari B. Web page ranking based on fuzzy and learning automata[C]//International Conference on Management of Emergent Digital Ecosystems. ACM, 2009:162-166.
[4] 金秋, 裴斐, 林馥. 学习自动机结合节点功率自适应调整的WSN目标覆盖方案[J]. 计算机应用研究, 2017, 34(1):177-180.
[5] 陶彤彤. 学习自动机及其在随机定位应用中的研究[D]. 上海:上海交通大学, 2014.
[6] 荆羽纯, 葛昊, 江文,等. 一种基于学习自动机的推荐算法改进[J]. 计算机应用研究, 2016,33(1):32-34.
[7] Torkestani J A. An adaptive learning automata-based ranking function discovery algorithm[J]. Journal of Intelligent Information Systems, 2012, 39(2):441-459.
[8] Mehr S M, Taran M, Hashemi A B, et al. Determining web pages similarity using distributed learning automata and graph partitioning[C]//International Symposium on Artificial Intelligence and Signal Processing. IEEE, 2011:129-134.
[9] 陈超, 周静, 万懿. 基于DLA的调度算法解决定向传感器网络连接目标覆盖问题[J]. 计算机应用研究, 2015, 32(10):3108-3113.
[10] 董洋溢, 李伟华, 于会. 基于混合余弦相似度的中文文本层次关系挖掘[J]. 计算机应用研究, 2017, 34(5):1406-1409.
[11] 王冲, 纪仙慧. 基于用户兴趣与主题相关的PageRank算法改进研究[J]. 计算机科学, 2016, 43(3):275-278.
[12] 王冲, 王凯. 基于用户反馈特征聚合的网页排序算法[J]. 计算机工程与设计, 2017, 38(8):2020-2024.