综合LDA与特征维度的丽江古城意象感知分析

doi:10.18306/dlkxjz.2020.04.008

摘要/Abstract

摘要：

论文通过建立基于LDA(Latent Dirichlet Allocation)模型和包含两级特征维度的旅游地意象感知研究框架,将LDA主题模型用于旅游微博文本分析,以特征维度半定量刻画旅游地意象感知特征,减少LDA主题凝练的主观性,帮助研究者在特定维度框架约束下准确、客观地提取旅游地意象特征。丽江古城案例证明,一级特征维度可以完整勾勒出丽江古城意象感知的基本框架,包括以聚落形态、音乐意境、标志人物、休闲空间和纳西美食为核心的5组空间与景观元素,深度旅游者、城市居民、年轻人、女孩子4类人群的特殊感知体验,及旅游者与环境要素的不同互动特征;二级特征维度进一步精细解读丽江古城的意象感知特征,表现为丽江古城慢活性、夜生活和浪漫之都的文化意象、旅游者对地方文化与现代风情融合的凝视与体验等。结合特征维度的LDA模型,准确构建了意象基本框架,成功刻画了丽江古城的形象及精细特征,并能进一步解析意象的形成机制,为旅游地意象感知研究提供了新视角,有助于深度解读意象形成的地方意义,厘清认知、情感和行为意象间的关系。

关键词: LDA模型, 特征维度, 共现关系, 意象特征, 新浪微博, 丽江古城

Abstract:

To explore whether the Latent Dirichlet Allocation (LDA) topic model is suitable for analyzing the short text of tourism microblogs and whether the result can be consistent with the research results based on interviews and other data, this study established a destination image perception framework including four first-level dimensions and 10 second-level dimensions. Then the meaning of the topics is defined based on the dominant dimensions of each topic, which can reduce the subjectivity of researchers and help them to use the LDA model to extract the destination image perception quantitatively and objectively. The case study of the Old Town of Lijiang shows that in the first-level dimensions, the basic framework of image perception can be fully outlined through the five groups of core spatial and landscape elements, including the human settlements, music culture, character, leisure space and Naxi cuisine, and the special perception of the deep tourists, urban residents, young people and girls, and the characteristics of interaction of human and environmental elements. In the second-level dimensions, more detailed perception of destination image can be vividly presented from three aspects: the slow living in the Old Town of Lijiang, the culture of nightlife and romance, and tourists' perception of the fusion of local culture and modernity. This study proves the feasibility and advantage of this method—it shows that LDA is suitable for short text analysis of social media such as Weibo. Topic analysis based on dominant semantic dimensions successfully portrays the image perception of the Old Town of Lijiang and further analyzes the mechanism of image formation, and provides a new perspective for destination image perception, which has three values. It helps to accurately establish the basic framework of destination image perception; quantitatively extract the core dimensions of image perception; and deeply interpret the local meaning of destination image and clarify the relationship between cognitive, affective, and behavioral images.

Key words: LDA model, dominant semantic dimensions, co-occurrence relationship, image characteristic, Sina Weibo, the Old Town of Lijiang

梁晨晨, 李仁杰. 综合LDA与特征维度的丽江古城意象感知分析[J]. 地理科学进展, 2020, 39(4): 614-626.

LIANG Chenchen, LI Renjie. Tourism destination image perception analysis based on the Latent Dirichlet Allocation model and dominant semantic dimensions: A case of the Old Town of Lijiang[J]. PROGRESS IN GEOGRAPHY, 2020, 39(4): 614-626.

图/表 7

图1

表1

图2

表2

表3

图3

图4

参考文献 35

[1]	白凯, 赵安周 . 城市意象与旅游目的地意象研究中的趋同与分野[J]. 地理科学进展, 2011,30(10):1312-1320. doi: 10.11820/dlkxjz.2011.10.014
	[ Bai Kai, Zhao Anzhou . Studies on convergence and divergence of city image and destination image. Progress in Geography, 2011,30(10):1312-1320. ] doi: 10.11820/dlkxjz.2011.10.014
[2]	Çakmak E, Isaac R K . What destination marketers can learn from their visitors' blogs: An image analysis of Bethlehem, Palestine[J]. Journal of Destination Marketing & Management, 2012,1(1-2):124-133.
[3]	乌铁红, 张捷, 李文杰 . 地域文化差异对旅游者的旅游地意象感知影响: 以内蒙古自治区草原旅游地为例[J]. 旅游学刊, 2010,25(6):66-72.
	[ Wu Tiehong, Zhang Jie, Li Wenjie . The impact of regional cultural differences on tourists' perception of tourism destination image: A case study of tourists' perception of prairie image in Inner Mongolia Autonomous Region. Tourism Tribune, 2010,25(6):66-72. ]
[4]	滕茜, 杨勇, 布倩楠 , 等. 基于网络文本的景区感知及互动研究: 以上海为例[J]. 旅游学刊, 2015,30(2):33-41. doi: 10.3969/j.issn-1002-5006.2015.02.004
	[ Teng Qian, Yang Yong, Bu Qiannan , et al. Research on the perception and interaction of tourist attractions in Shanghai based on web texts. Tourism Tribune, 2015,30(2):33-41. ] doi: 10.3969/j.issn-1002-5006.2015.02.004
[5]	Arsal I, Woosnam K M, Baldwin E D , et al. Residents as travel destination information providers: An online community perspective[J]. Journal of Travel Research, 2010,49(4):400-413. doi: 10.1177/0047287509346856
[6]	郭风华, 王琨, 张建立 , 等. 成都“五朵金花”乡村旅游地形象认知: 基于博客游记文本的分析[J]. 旅游学刊, 2015,30(4):84-94. doi: 10.3969/j.issn.1002-5006.2015.04.008
	[ Guo Fenghua, Wang Kun, Zhang Jianli , et al. Rural tourism destination image of the case in "Five Golden Flowers", Chengdu: Based on the text mining of blogs. Tourism Tribune, 2015,30(4):84-94. ] doi: 10.3969/j.issn.1002-5006.2015.04.008
[7]	谢永俊, 彭霞, 黄舟 , 等. 基于微博数据的北京市热点区域意象感知[J]. 地理科学进展, 2017,36(9):1099-1110. doi: 10.18306/dlkxjz.2017.09.006
	[ Xie Yongjun, Peng Xia, Huang Zhou , et al. Image perception of Beijing's regional hotspots based on microblog data. Progress in Geography, 2017,36(9):1099-1110. ] doi: 10.18306/dlkxjz.2017.09.006
[8]	Lu Weilin, Stepchenkova S . User-generated content as a research mode in tourism and hospitality applications: Topics, methods, and software[J]. Journal of Hospitality Marketing & Management, 2015,24(2):119-154.
[9]	Alazzawi A N, Abdelmoty A I, Jones C B . What can I do there? Towards the automatic discovery of place-related services and activities[J]. International Journal of Geographical Information Science, 2012,26(2):345-364. doi: 10.1080/13658816.2011.595954
[10]	王媛, 许鑫, 冯学钢 , 等. 基于文本挖掘的古镇旅游形象感知研究: 以朱家角为例[J]. 旅游科学, 2013,27(5):86-95.
	[ Wang Yuan, Xu Xin, Feng Xuegang , et al. Research on tourists' percieved image of ancient town using web text mining methods: A case study of Zhujiajiao. Tourism Science, 2013,27(5):86-95. ]
[11]	孟奕爽, 邓森文, 徐佳 . 旅游目的地投射意象与游客感知意象差异分析: 以平江石牛寨景区为例[J]. 中南林业科技大学学报(社会科学版), 2018,12(3):67-73.
	[ Meng Yishuang, Deng Senwen, Xu Jia . Study on the differences between projection image and tourist perception image of tourist destination: Case of Shiniuzhai scenic spot, Pingjiang. Journal of Central South University of Forestry & Technology (Social Sciences), 2018,12(3):67-73. ]
[12]	Büschken J, Allenby G M . Sentence-based text analysis for customer reviews[J]. Marketing Science, 2016,35(6):1-23. doi: 10.1287/mksc.2015.0974
[14]	王永明, 王美霞, 李瑞 , 等. 基于网络文本内容分析的凤凰古城旅游地意象感知研究[J]. 地理与地理信息科学, 2015,31(1):64-67, 79.
	[ Wang Yongming, Wang Meixia, Li Rui , et al. Destination image perception of Fenghuang Ancient Town based on content analysis of travelers' web text. Geography and Geo-information Science, 2015,31(1):64-67, 79. ]
[15]	Weerdenburg D, Scheider S, Adams B , et al. Where to go and what to do: Extracting leisure activity potentials from web data on urban space[J]. Computers, Environment and Urban Systems, 2019,73:143-156.
[16]	Gao Baojun, Hu Nan, Bose I . Follow the herd or be myself? An analysis of consistency in behavior of reviewers and helpfulness of their reviews[J]. Decision Support Systems, 2016,95:1-11. doi: 10.1016/j.dss.2016.11.005
[17]	Andrienko G, Andrienko N, Bosch H , et al. Thematic patterns in georeferenced tweets through space-time visual analytics[J]. Computing in Science & Engineering, 2013,15(3):72-82.
[18]	Ramage D, Dumais S T, Liebling D J. Characterizing microblogs with topic models [C]// Association for the Advancement of Artificial Intelligence. Proceedings of the fourth international conference on weblogs and social media. Washington D C, USA: AAAI Press, 2010: 130-137.
[19]	Cheng Xueqi, Yan Xiaohui, Lan Yanyan , et al. BTM: Topic modeling over short texts[J]. IEEE Transactions on Knowledge and Data Engineering, 2014,26(12):2928-2941. doi: 10.1109/TKDE.2014.2313872
[20]	Lansley G, Longley P A . The geography of twitter topics in London[J]. Computers, Environment and Urban Systems, 2016,58:85-96. doi: 10.1016/j.compenvurbsys.2016.04.002
[21]	Brandt T, Bendler J, Neumann D . Social media analytics and value creation in urban smart tourism ecosystems[J]. Information & Management, 2017,54(6):703-713.
[22]	Blei D M, Ng A Y, Jordan M I . Latent Dirichlet allocation[J]. Journal of Machine Learning Research, 2003,3(4-5):993-1022.
[23]	Ghosh D, Guha R . What are we 'tweeting' about obesity? Mapping tweets with topic modeling and geographic information system[J]. Cartography and Geographic Information Science, 2013,40(2):90-102. doi: 10.1080/15230406.2013.776210 pmid: 25126022
[24]	关鹏, 王曰芬 . 基于LDA主题模型和生命周期理论的科学文献主题挖掘[J]. 情报学报, 2015,34(3):286-299.
	[ Guan Peng, Wang Yuefen . Topic mining in scientific literature based on LDA topic model and life cycle theory. Journal of the China Society for Scientific and Technical Information, 2015,34(3):286-299. ]
[25]	谢昊, 江红 . 一种面向微博主题挖掘的改进LDA模型[J]. 华东师范大学学报(自然科版), 2013(6):93-101.
	[ Xie Hao, Jiang Hong . Improved LDA model for microblog topic mining. Journal of East China Normal University (Natural Sciences), 2013(6):93-101. ]
[26]	Hu Nan, Zhang Ting, Gao Baojun , et al. What do hotel customers complain about? Text analysis using structural topic model[J]. Tourism Management, 2019,72:417-426.
[27]	Guo Yue, Barnes S J, Jia Qiong . Mining meaning from online ratings and reviews: Tourist satisfaction analysis using latent dirichlet allocation[J]. Tourism Management, 2017,59:467-483.
[28]	南亭 . 丽江古城的前世今生[J]. 商业文化, 2017(34):88-91.
	[ Nan Ting . Past and present of Lijiang Ancient City. Business Culture, 2017(34):88-91. ]
[29]	王鲁民, 吕诗佳 . 建构丽江:秩序·形态·方法 [M]. 北京: 生活·读书·新知三联书店, 2013.
	[ Wang Lumin, Lv Shijia. Constructing Lijiang: Order, form, method. Beijing, China: SDX Joint Publishing Company, 2013. ]
[30]	Huang Yuqian, Li Yue, Shan Jie . Spatial-temporal event detection from geo-tagged tweets[J]. ISPRS International Journal of Geo-Information, 2018,7(4):150. doi: 10.3390/ijgi7040150.
[31]	白凯, 胡宪洋, 吕洋洋 , 等. 丽江古城慢活地方性的呈现与形成[J]. 地理学报, 2017,72(6):1104-1117.
	[ Bai Kai, Hu Xianyang, Lv Yangyang , et al. Study on the identity with placeness of slow living in Lijiang. Acta Geographica Sinica, 2017,72(6):1104-1117. ]
[32]	崔庆明, 和琳珊, 徐红罡 . 遗产旅游动机的核心-边缘结构研究: 以丽江为例[J]. 旅游学刊, 2016,31(10):84-93.
	[ Cui Qingming, He Linshan, Xu Honggang . The core-periphery structure of heritage tourism motivation in Lijiang. Tourism Tribune, 2016,31(10):84-93. ]
[33]	李大伟, 李诚固, 马佐澎 , 等. 旅游影响下的地方音乐及其对地方的建构: 以丽江古城为例[J]. 地理科学进展, 2018,37(6):833-843.
	[ Li Dawei, Li Chenggu, Ma Zuopeng , et al. Influence of tourism on local music and its construction to the place: A case study on the Lijiang Ancient Town. Progress in Geography, 2018,37(6):833-843. ]
[34]	孙九霞, 王心蕊 . 丽江大研古城文化变迁中的“虚无”与“实在”: 以酒吧发展为例[J]. 旅游学刊, 2012,27(9):73-83.
	[ Sun Jiuxia, Wang Xinrui . The "nothing" and "something" during the cultural changes in Lijiang Ancient Town: A case study of Lijiang Ancient Town's bars. Tourism Tribune, 2012,27(9):73-83. ]
[35]	保继刚, 苏晓波 . 历史城镇的旅游商业化研究[J]. 地理学报, 2004,59(3):427-436.
	[ Bao Jigang, Su Xiaobo . Study on control of tourism commercialization in historic town and village. Acta Geographica Sinica, 2004,59(3):427-436. ]

一级维度	二级维度	维度释义	特征词样例(以丽江古城为例)
空间与景观 (SL)	外部空间(SL₁)	目的地周边旅游景区、客源地等空间单元	昆明、大理、拉市海、虎跳峡、北京
	内部空间(SL₂)	目的地内部的客栈、餐馆、酒吧等空间单元	客栈、酒店、房间、院子、阳台、酒吧、电影院
	自然景观(SL₃)	描述地文、水域、生物、气候等自然景观要素	阳光、蓝天、微风、小雨、白云
	人文景观(SL₄)	包括建筑、民俗、文化、历史等人文景观要素	四方街、情人节、鲜花饼、小倩、乐队
行为与心理 (BP)	体验行为(BP₁)	旅行中融入目的地空间,与景观交互的行为活动	骑马、喝咖啡、散步、晒太阳、唱歌、泡吧、跳舞
	旅行方式(BP₂)	游览观光的出行方式,如交通工具、步行等	搭车、行走、火车、飞机
	感知心理(BP₃)	旅游者对环境的感知与旅游体验心理活动	开心、愉快、激动、热闹、崩溃
时间与行程 (TR)	旅行时间(TR₁)	描述旅游行为的特定时间节点	旺季、暑假、周末、白天、早晨
时间与行程 (TR)	行程安排(TR₂)	描述旅游行程的节奏与旅行过程安排	两天、第一天、一站、早餐、起床、出门
其他类(OT)	其他类(OT)	不属于以上任一维度的特征词语	日子、听说、商业、距离

主题6			主题12			主题17			主题22			主题24
概率	词语	维度	概率	词语	维度	概率	词语	维度	概率	词语	维度	概率	词语	维度
0.222	客栈	SL₂	0.168	旅行	BP₁	0.145	偷笑	BP₃	0.078	朋友	SL₄	0.069	玉龙雪山	SL₁
0.047	老板	SL₄	0.068	心情	BP₃	0.076	浮云	SL₃	0.028	唱歌	BP₁	0.048	睡觉	BP₁
0.042	月亮	SL₃	0.059	风景	SL₃	0.071	音乐	SL₄	0.027	爱情	SL₄	0.044	纳西	SL₄
0.033	离开	BP₁	0.047	一米阳光	SL₄	0.065	天气	SL₃	0.026	日子	SL₄	0.038	火锅	SL₄
0.032	舒服	BP₃	0.034	小伙伴	SL₄	0.044	夜晚	TR₁	0.023	忘记	BP₃	0.033	真心	BP₃
0.026	惬意	BP₃	0.024	江湖	OT	0.042	来到	OT	0.021	啤酒	SL₄	0.031	酸奶	SL₄
0.022	院子	SL₂	0.022	早晨	TR₁	0.036	夏天播放	SL₄	0.021	一场	OT	0.022	好吃	BP₃
0.022	房间	SL₂	0.021	哈气	OT	0.035	酒吧	SL₂	0.017	不用	OT	0.021	美食	SL₄
0.019	一家	OT	0.016	最美	BP₃	0.023	播放	BP₁	0.017	浪漫	BP₃	0.020	小吃	SL₄
0.016	晚餐	TR₁	0.014	小桥流水	SL₃	0.018	听歌	BP₁	0.016	孩子	SL₄	0.019	自然	SL₃
0.015	印象	BP₃	0.013	一座	OT	0.018	还好	BP₃	0.016	一群	OT	0.018	到达	BP₁
0.015	外面	OT	0.012	景色	SL₃	0.016	见到	OT	0.014	听说	OT	0.016	烤鱼	SL₄
0.013	入住	BP₁	0.011	木有	OT	0.013	妹子	SL₄	0.014	旅程	TR₂	0.016	海拔	SL₃
0.012	笑哈哈	BP₁	0.01	倾城	BP₃	0.011	辛苦	BP₃	0.013	静静的	BP₃	0.013	拜拜	BP₁
0.011	环境	OT	0.01	向往	BP₃	0.010	懂得	BP₃	0.013	收获	OT	0.012	雪山	SL₁

类别	主题	空间与景观	行为与心理	时间与行程	其他类	类别	主题	空间与景观	行为与心理	时间与行程	其他类
单维度主导	主题1	20.2	29.9	5.7	9.6	单维度主导	主题19	13.9	26.1	9.8	16.8
单维度主导	主题3	12.0	26.8	6.4	17.3	单维度主导	主题20	17.6	29.8	7.6	13.8
单维度主导	主题4	15.1	28.5	1.9	22.5	单维度主导	主题21	9.7	37.6	2.6	21.4
单维度主导	主题5	21.7	31.7	4.4	16.3	单维度主导	主题23	8.8	45.6	1.3	21.1
单维度主导	主题6	39.0	17.1	3.0	14.5	单维度主导	主题24	38.4	18.8	0.0	7.4
单维度主导	主题7	9.1	26.2	1.5	16.0	双维度主导	主题0	20.6	13.9	14.9	13.6
单维度主导	主题8	25.3	9.5	15.8	19.9	双维度主导	主题2	24.2	29.1	0.4	15.6
单维度主导	主题10	24.6	33.7	6.2	12.1	双维度主导	主题9	27.6	30.9	1.9	19.5
单维度主导	主题11	6.3	15.6	3.5	36.6	双维度主导	主题16	24.6	29.7	0.0	15.4
单维度主导	主题12	22.0	33.8	3.5	14.7	双维度主导	主题17	35.0	27.2	4.4	13.0
单维度主导	主题13	40.5	17.4	4.6	15.2	双维度主导	主题22	23.0	16.5	1.4	17.6
单维度主导	主题14	9.9	29.8	0.0	19.6	双维度主导	主题25	22.1	17.6	3.7	26.7
单维度主导	主题15	31.2	22.8	10.5	13.0	双维度主导	主题26	26.0	31.8	1.4	12.8
单维度主导	主题18	19.2	32.7	1.2	11.7