基于地理标记照片的个性化景点推荐方法

doi:10.12082/dqxxkx.2021.200608

地球信息科学学报 ›› 2021, Vol. 23 ›› Issue (8): 1391-1400.doi: 10.12082/dqxxkx.2021.200608

• 地球信息科学理论与方法 • 上一篇下一篇

基于地理标记照片的个性化景点推荐方法

叶凡¹^,²(), 孙玉¹^,²^,^*(), 陈崇成¹^,², 于大宇³

1.福州大学空间数据挖掘与信息共享教育部重点实验室,福州 350108
2.福州大学地理空间信息技术国家地方联合工程研究中心,福州 350108
3.武汉大学遥感信息工程学院,武汉 430079

收稿日期:2020-10-15 修回日期:2020-12-25 出版日期:2021-08-25 发布日期:2021-10-25
通讯作者: 孙玉
作者简介:叶凡（1995— ）,女,江西上饶人,硕士,主要从事时空数据挖掘、智能推荐。E-mail: 15770938521@163.com
基金资助:
旗山学者奖励支持计划(XRC-19001);国家重点研发计划项目(2017YFB0504202)

A Personalized Attraction Recommendation Method based on Geotagged Photos

YE Fan¹^,²(), SUN Yu¹^,²^,^*(), CHEN Chongcheng¹^,², YU Dayu³

1. Key Lab of Spatial Data Mining and Information Sharing of Ministry of Education, Fuzhou University, Fuzhou 350108, China
2. National Engineering Research Center of Geospatial Information Technology, Fuzhou University, Fuzhou 350108, China
3. School of Remote Sensing Information Engineering, Wuhan University, Wuhan 430079, China

Received:2020-10-15 Revised:2020-12-25 Online:2021-08-25 Published:2021-10-25
Contact: SUN Yu
Supported by:
Qishan Scholar Award Support Scheme(XRC-19001);National Key Research and Development Project(2017YFB0504202)

摘要/Abstract

摘要：

研究如何根据已有的海量旅游信息及数据,为游客个性化推荐旅游景点具有重要意义。本文利用从Flickr网站获取的2013—2018年香港特别行政区范围内的地理标记照片来识别旅游景点,并根据游客游览顺序重建旅游轨迹。在此基础上,针对现有方法尚未考虑游客偏好在旅行过程中会发生动态变化的问题,提出一种基于隐含狄利克雷分布模型（Latent Dirichlet Allocation, LDA）和用户长短期偏好的个性化景点推荐方法（A Recommendation Method Based on LDA and User's Long and Short-Term Preference, L-ULSP）。该方法利用LDA主题模型获取景点特征信息,挖掘景点间的相关性,再利用注意力机制和长短期记忆网络分别学习用户的长期偏好和短期偏好,最后结合长短期偏好捕捉用户偏好的动态变化。实验结果表明, L-ULSP方法所推荐的景点在命中率和平均倒数排名2个指标上均优于现有其他方法,证明了本文所提方法可以从景点序列中有效学习游客偏好,并为游客推荐下一个景点。此外,本文通过对比实验,进一步验证了同时考虑用户的长短期偏好能够更好地学习用户的偏好变化。

Abstract:

Personalized recommendation of tourist attractions for visitors is useful based on the vast amount of tourism information and data. In this paper, we use Flickr's geotagged photos from 2013 to 2018 in Hong Kong to identify tourist hot spots and reconstruct the tourism trajectory according to the tourist visiting order. On this basis, we propose a personalized recommendation method based on Latent Dirichlet Allocation (LDA) model and User's Long-term and Short-term Preference (L-ULSP) to address the problem that existing methods do not take into account the dynamic changes in visitor preferences during the travel process. In this method, the LDA model is used to obtain the feature information of attractions, and the correlation between attractions is explored. Then, attention mechanism is used to focus on the important information in the long-term sequence to capture the long-term preference of tourists, and LSTM is used to model the short-term sequence information to learn the short-term preference of tourists. Finally, the long-term and short-term preferences are weighted to obtain the final preferences of tourists to capture the dynamic changes of user preferences. The algorithm has the following advantages: (1) By mining the topic feature information of Geotagged photo text, the description information of attractions is added, which can capture users' travel preference more accurately; (2) The algorithm considers both the long-term and short-term preferences of users, and can learn the dynamic changes of users' preferences in the process of travel while modeling the sequence information of attractions. The experimental results show that: (1) The attractions recommended by the L-ULSP method outperform other existing methods in both Hit Rate and Mean Reciprocal Rank, two common evaluation metrics for recommendation algorithms, proving that the proposed method can effectively learn visitor preferences from a sequence of attractions and recommend the next attraction to visitors. It is demonstrated that the method can achieve good recommendation results in travel recommendation scenarios; (2) The comparison experiments between the model using long-term preference as the user's final preference and the model combining user's long-term and short-term preferences as the final preference further validate that considering both the user's long-term and short-term preferences can better learn the user's preference changes and thus improve the accuracy of recommendations; (3) This paper further compares the calculation efficiency of L-ULSP with different deep learning recommendation models based on RNN, and counts the running time of each model. The results show that this method is better than most methods in efficiency.

Key words: geotagged photos, attractions recommendation, tourism trajectory, long-term and short-term preferences, attention mechanism, long short-term memory, topic model, personalized recommendation

叶凡, 孙玉, 陈崇成, 于大宇. 基于地理标记照片的个性化景点推荐方法[J]. 地球信息科学学报, 2021, 23(8): 1391-1400.DOI:10.12082/dqxxkx.2021.200608

YE Fan, SUN Yu, CHEN Chongcheng, YU Dayu. A Personalized Attraction Recommendation Method based on Geotagged Photos[J]. Journal of Geo-information Science, 2021, 23(8): 1391-1400.DOI:10.12082/dqxxkx.2021.200608

图/表 13

图1

图2

表1

符号定义

符号	定义
$U = {u 1, u 2, …, u N}$ $V = {v 1, v 2, …, v M}$ $T = {t 1, t 2, …, t K}$ $V u = {v 1 u, v 2 u, …, v m u}$ $L u = {v 1 u, v 2 u, …, v l u$ } $S u = {v l - j + 1 u, v l - j + 2 u, …, v l u$ }	游客集合,其中N为游客数量景点集合,其中M为景点数量景点的主题特征集合,其中K是主题数量游客 $u$ 的景点序列,其中每个 $v i u$ 包含景点和主题特征 $(a i, t i)$ 游客 $u$ 的长期序列游客 $u$ 的短期序列, $j$ 是短期序列的长度

表1

表2

图3

图4

图5

图6

表3

图7

图8

表4

表5

参考文献 31

[1]	Vu H Q, Li G, Law R, et al. Exploring the travel behaviors of inbound tourists to Hong Kong using geotagged photos[J]. Tourism Management, 2015, 46:222-232. doi: 10.1016/j.tourman.2014.07.003
[2]	Becker M, Singer P, Lemmerich F, et al. Photowalking the city: Comparing hypotheses about urban photo trails on Flickr[J]. International Conference on Social Informatics, 2015, 9:227-244.
[3]	Chua A, Servillo, Marcheggiani E, et al. Mapping Cilento: Using geotagged social media data to characterize tourist flows in southern Italy[J]. Tourism Management, 2016, 57:295-310. doi: 10.1016/j.tourman.2016.06.013
[4]	Majid A, Chen L, Chen G C, et al. A context-aware personalized travel recommendation system based on geotagged social media data mining[J]. International Journal of Geographical Information Science, 2013, 27(4):662-684. doi: 10.1080/13658816.2012.696649
[5]	Jiang S, Qian X, Mei T, et al. Personalized travel sequence recommendation on multi-source big social media[J]. IEEE Transactions on Big Data, 2016, 2(1):43-56. doi: 10.1109/TBDATA.2016.2541160
[6]	Xu Z, Chen L, Guo H, et al. User similarity-based gender-aware travel location recommendation by mining geotagged photos[J]. International Journal of Embedded Systems, 2018, 10(5):356-365. doi: 10.1504/IJES.2018.095023
[7]	Lyu D D, Chen L, Xu Z X, et al. Weighted multi-information constrained matrix factorization for personalized travel location recommendation based on geo-tagged photos[J]. Applied Intelligence, 2020, 50(3):924-938. doi: 10.1007/s10489-019-01566-6
[8]	Zeng Y, Zhou X. Computing with spatial trajectories[M]. New York: Springer Science & Business Media, 2011.
[9]	Zheng Y, Zhang L, Xie X, et al. Mining Interesting Locations and Travel Sequences from GPS Trajectories[C]. Proceeding of the 18^th International Conference on World Wide Web, Madrid, Spain , 2009.
[10]	Liu C, Liu J, Xu S, et al. A spatiotemporal dilated convolutional generative network for point-of-interest recommendation[J]. ISPRS International Journal of Geo-Information, 2020, 9(2):113. doi: 10.3390/ijgi9020113
[11]	Jiang K, Yin H, Wang P, et al. Learning from contextual information of geo-tagged web photos to rank personalized tourism attractions[J]. Neurocomputing, 2013, 119:17-25. doi: 10.1016/j.neucom.2012.02.049
[12]	Xu Z, Chen L, Chen G. Topic based context-aware travel recommendation method exploiting geotagged photos[J]. Neurocomputing, 2015, 155:99-107. doi: 10.1016/j.neucom.2014.12.043
[13]	Zhang C Y, Wang K. POI recommendation through cross-region collaborative filtering[J]. Knowledge and Information Systems, 2016, 46(2):369-387. doi: 10.1007/s10115-015-0825-8
[14]	Santiago F M, López F A, Montejo-Ráez A, et al. GeOasis: A knowledge-based geo-referenced tourist assistant[J]. Expert Systems with Applications, 2012, 39(14):11737-11745. doi: 10.1016/j.eswa.2012.04.080
[15]	Binucci C, De Lu F, DI Giacomo E, et al. Designing the content analyzer of a travel recommender system[J]. Expert Systems with Applications, 2017, 87:199-208. doi: 10.1016/j.eswa.2017.06.028
[16]	Rendle S, Freudenthaler C, Schmidt-thieme L. Factorizing personalized markov chains for next-basket recommendation[C]. Proceedings of the 19^th International Conference on World Wide Web, Raleigh, NC, USA , 2010.
[17]	He R N McAuley J. Fusing similarity models with Markov chains for sparse sequential recommendation[EB/OL]. 2016: arXiv: 1609.09152[cs.IR]. https://arxiv.org/abs/1609.09152.
[18]	Liu S, Wang L. A self-adaptive point-of-interest recommendation algorithm based on a multi-order markov model[J]. Future Generation Computer Systems, 2018, 89:506-514. doi: 10.1016/j.future.2018.07.008
[19]	李昇智, 乔建忠, 林树宽. 一种基于用户移动行为相似性的位置预测方法[J]. 计算机科学, 2018, 45(12):288-292,307.
	[Li S Z, Qiao J Z, Lin S K. Location prediction method based on sinmmilarity of users moving behavior[J]. Computer Science, 2018, 45(12):28-292,307. ]
[20]	Monreale A, Pinelli F, Trasarti R, et al. Wherenext: A location predictor on trajectory pattern mining[C]. Proceedings of the 15^th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Paris, France , 2009.
[21]	Blei D M, Ng A Y, Jordan M I. Latent dirichlet allocation[J]. Journal of machine Learning research, 2003, 3(1):993-1022.
[22]	Williams R J, Zipser D. A learning algorithm for continually running fully recurrent neural networks[J]. Neural Computation, 1989, 1(2):270-280. doi: 10.1162/neco.1989.1.2.270
[23]	Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8):1735-1780. pmid: 9377276
[24]	Zhu Y, Li H, Liao Y, et al. What to do next: Modeling user behaviors by time-LSTM[C]. Proceeding of the 26 International Joint Conference on Artificial Intelligence, Melbourne, Australia: IJCAI, 2017.
[25]	Li J, Ren P, Chen Z, et al. Neural attentive session-based recommendation[C]. Proceeding of the 2017 ACM on Conference on Information and Knowledge Management, Singapore, 2017.
[26]	Liu Q, Zeng Y, Mokhosi R, et al. STAMP: Short-Term attention/memory priority model for session-based recommendation[C]. Proceeding of the 24^th ACM SIGKDD International Conference oSn Knowledge Discovery & Data Mining, London, UK , 2018.
[27]	Hidasi B, Karatzoglou A, Baltrunas L, et al. Session-based recommendations with recurrent neural networks[C]. In Proceedings of the International Conference on Learning Representations, San Juan, Puerto Rico, 2016.
[28]	Quadrana M, Cremonesi P, Jannach D. Sequence-aware recommender systems[J]. ACM Computing Surveys (CSUR), 2018, 51(4):1-36.
[29]	Sun Y R, Fan H C, Bakillah M, et al. Road-based travel recommendation using geo-tagged images[J]. Computers, Environment and Urban Systems, 2015, 53:110-122. doi: 10.1016/j.compenvurbsys.2013.07.006
[30]	Kisilevich S, Mansmann F, Keim D. P-DBSCAN: A density based clustering algorithm for exploration and analysis of attractive areas using collections of geo-tagged photos[C]. Proceeding of the 1st International Conference and Exhibition on Computing for Geospatial Research & Application, Washington, USA, 2010.
[31]	Bonnin G, Jannach D. Automated generation of music playlists: Survey and experiments[J]. ACM Computing Surveys (CSUR), 2014, 47(2):1-35.

基于地理标记照片的个性化景点推荐方法

A Personalized Attraction Recommendation Method based on Geotagged Photos

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 31

相关文章 7

Metrics

本文评价

推荐阅读 0

[1]	刘新, 赵宁, 郭金运, 郭斌. 基于LSTM神经网络的青藏高原月降水量预测[J]. 地球信息科学学报, 2020, 22(8): 1617-1629.
[2]	毛文山, 赵红莉, 孙凤娇, 蒋云钟, 姜倩, 朱彦儒. 基于Item2Vec负采样优化的专题地图产品个性化推荐方法研究[J]. 地球信息科学学报, 2020, 22(11): 2128-2139.
[3]	范红超, 李万志, 章超权. 基于Anchor-free的交通标志检测[J]. 地球信息科学学报, 2020, 22(1): 88-99.
[4]	苏凯, 程昌秀, Nikita Murzintcev, 张婷. 主题模型在基于社交媒体的灾害分类中的应用及比较[J]. 地球信息科学学报, 2019, 21(8): 1152-1160.
[5]	文聪聪, 彭玲, 杨丽娜, 池天河. 主题模型与SVM组合的小尺度街区用地分类方法[J]. 地球信息科学学报, 2018, 20(2): 167-175.
[6]	仇培元, 陆锋, 张恒才, 余丽. 蕴含地理事件微博客消息的自动识别方法[J]. 地球信息科学学报, 2016, 18(7): 886-893.
[7]	刘扬, 付征叶, 郑逢斌. 高分辨率遥感影像目标分类与识别研究进展[J]. 地球信息科学学报, 2015, 17(9): 1080-1091.