当前位置>主页 > 期刊在线 > 计算机技术 >

计算机技术2018年4期

基于协同过滤推荐算法的公园个性化推荐系统的研建
张伊, 张军霞, 邹雨纯, 徐丹阳, 王丽婷
(北京林业大学 信息学院,北京 100083)

摘  要:当今,公园游玩已经是大众日常休闲娱乐的普遍选择,但由于公园种类繁多以及相关信息的冗杂和有效信息的缺乏,游客不易选择适合自己出行计划的公园。随着互联网的迅猛发展,推荐系统已经渗透到了互联网的各个方面,协同过滤技术是目前应用最为广泛和成功的技术。虽然协同过滤推荐技术取得了很大的成功,但传统的算法本身还存在一些问题,包括数据稀疏问 题、冷启动性问题,这些问题都是协同过滤必须解决的问题。基于此背景,本文对协同过滤推荐算法中的数据稀疏和冷启动问题做了深入的分析和探讨,并将改进的方案运用到一个具有推荐功能的公园个性化推荐系统中,以解决人们日常的休闲娱乐需求。


关键词:协同过滤;公园个性化推荐;数据稀疏;冷启动



中图分类号:TP391.3         文献标识码:A         文章编号:2096-4706(2018)04-0082-03


Park Personalization Recommendation System Based on the Collaborative Filtering Recommendation Algorithm
ZHANG Yi,ZHANG Junxia,ZOU Yuchun,XU Danyang,WANG Liting
(School of Information Science & Technology,Beijing Forestry University,Beijing 100083,China)

Abstract:Today,more and more people tend to play in the park at their leisure time. But the redundancy of the related information and the lack of effective information usually result in the aimlessness and restrictiveness of people's trip. With the rapid development of the internet,recommend system has been penetrated into all parts of the internet,and collaborative filtering is the most widely-used and successful technology currently. However,the traditional algorithm itself has some problems,including scalability problem and coldstart problem which are the problems need to be solved. This paper,based on this background,makes in-depth analysis and exploration in scalability problem and cold-start problem in collaborative filtering recommendation algorithm and applies the improved method to a park personalization recommendation system with a recommend function to solve people’s daily needs of leisure and entertainment.

Keywords:collaborative filtering;park personalization recommendation;data sparsity;cold-start


参考文献:

[1] 周武忠,林宝荣,周康,等. 世界公园评价指标体系初探 [J].中国名城,2012(9):34-38.

[2] HONG X C,LIN Z Y,ZHANG W,et al. Construction of a Comprehensive Soundscape Evaluation System in Forest Parks Based on Grey Statistical Theory [J]. Journal of Landscape Research,2016,8(6):25-28.

[3] 郭艳红. 推荐系统的协同过滤算法与应用研究 [D]. 大连:大连理工大学,2008.

[4] MU Z,JIE M,JING L,et al. Research on PersonalizedRecommendation Technology for Tourism Industry-A Perspective of a System Framework Design [J]. Advanced Materials Research,2011:1276-1280.

[5] Badrul Munir Sarwar. Sparsity,Scalability and Distribution in Recommender Systems [D]. Ph. Ddissertation,University of Minisota,2001:33-5813.

[6] 赵宁,王学军. 推荐系统中协同过滤技术的研究 [J]. 河北省科学院学报,2013,30(2):62-65.

[7] ScheinA.,PopescnlA.,UngarL.,and PennockD.,Methods and metrics for cold-start recommendations,Proceedings of the 25th annual international ACMSIGIR conference on Research and development in information retrieval,2002:112-225.

[8] David Goldberg,David Nichols,Brian M. Oki,et al.Using collaborative filtering to weave an information tapestry [J].
Communications of the ACM,1992,35(12):61-70.

[9] 季昀. 基于协同过滤推荐算法电影网站的构建 [D]. 哈尔滨:哈尔滨工业大学,2009.

[10] 张莉,秦桃,滕丕强. 一种改进的基于用户聚类的协同过滤算法 [J]. 情报科学,2014,32(10):24-27+32.


作者简介:

张伊(1996.08-),女,汉族,北京人,本科,研究方向:信息管理与信息系统。

张军霞(1997.01-), 女,汉族,河南人,本科,研究方向:信息管理与信息系统。

邹雨纯(1996.02-),女,汉族,四川人,本科,研究方向:信息管理与信息系统。

徐丹阳(1997.03-),女,汉族,河南人,本科,研究方向:信息管理与信息系统。

王丽婷(1997.11-),女,汉族,河北人,本科,研究方向:信息管理与信息系统。