Please wait a minute...
Data Analysis and Knowledge Discovery  2022, Vol. 6 Issue (8): 61-74    DOI: 10.11925/infotech.2096-3467.2021.1153
Current Issue | Archive | Adv Search |
Evaluating Trust of Accommodation Sharing with Feature Grouping and Combination
Lv Wanying,Zhao Jie(),Huang Liushen,Dong Zhenning,Liang Zhouyang
School of Management, Guangdong University of Technology, Guangzhou 510520, China
Download: PDF (2116 KB)   HTML ( 4
Export: BibTeX | EndNote (RIS)      
Abstract  

[Objective] This paper tries to improve trust evaluation with feature grouping and combination. The former provides replaceable features while the latter effectively reduces the feature dimensions. [Methods] First, we used the Markov Blanket to analyze the relationship among features based on their abilities to group similar features. Then, we searched within and among groups to combine features with the RVNS methods. [Results] In the case of missing features, the proposed model could effectively provide substitutes for the missing features and yielded stable trust evaluation results. The dimension of features was reduced to 1.7%, and the average accuracy of trust evaluation was above 92%. [Limitations] More research is needed to more effectively extract knowledge from dataset with missing values. [Conclusions] The proposed model could effectively address the missing data in trust evaluation.

Key wordsAccommodation Sharing      Trust Evaluation      Feature Grouping      Feature Combination     
Received: 12 October 2021      Published: 23 September 2022
ZTFLH:  F713  
Fund:National Natural Science Foundation of China(71871069);Guangdong Research Base of Social Sciences
Corresponding Authors: Zhao Jie,ORCID:0000-0003-3315-4447     E-mail: zhaojie@gdut.edu.cn

Cite this article:

Lv Wanying, Zhao Jie, Huang Liushen, Dong Zhenning, Liang Zhouyang. Evaluating Trust of Accommodation Sharing with Feature Grouping and Combination. Data Analysis and Knowledge Discovery, 2022, 6(8): 61-74.

URL:

https://manu44.magtech.com.cn/Jwk_infotech_wk3/EN/10.11925/infotech.2096-3467.2021.1153     OR     https://manu44.magtech.com.cn/Jwk_infotech_wk3/EN/Y2022/V6/I8/61

分析视角 文献来源 特征及数据缺失情况分析
房客视角 [17-20] 考虑房客评论以及房客自身感知特征,常以调查问卷形式获取数据,数据一般较主观,难以量化和获取。
平台视角 [20-21] 考虑第三方角度,特征基于房客使用感知,常以调查问卷形式获取数据,此类信息依赖房客主观感受,没有统一标准,缺失问题较为严重。
交互视角 [20-24] 考虑两方交互过程,常通过调查问卷和数据爬取相结合获取数据,此类信息存在动态变化过程,需要多方数据同时获取,缺失问题严重。
房源视角 [25] 考虑房东展示的房源信息,常以数据爬取形式获取数据,数据完整性依赖于房东披露的程度。
房东视角 [22,26-30] 考虑房东个人信息或以披露个人信息换取的标识,最有可能触及隐私,数据缺失问题严重。
融合多视角 [20,22] 考虑视角丰富,融合主观数据与客观数据,需要获取多方信息,数据缺失问题是常见问题。
Summary of Landlord Perceived Trust Features
信任评估方法 特点
机器学习 卷积神经网络CNN[17]
信任评估模型
(TSTrustRank)[32]
决策树[33]
非线性分析方法,常结合平台真实数据进行分析。
结构方程 AMOS[19-20]
路径分析[27]
基于协方差[29]
常进行线性分析,仅能提供一组特征,评估精度较低,通过调查问卷方式获得数据具有局限性。
回归分析 偏最小二乘法[21,28,34]
稳健回归[30]
协方差分析 ANCOVA[23]
其他方法 文献[18,22,24-26]
Summary of Trust Evaluation Method
特征类型 视角 特征数量 含缺失值
特征数量
含缺失值特征
最大值缺失率
文本 房东 10 7 99.76%
房客 5 0 /
房源 8 0 /
交互 112 0 /
图像 房东 42 41 99.22%
房源 4 0 /
数值 房东 21 6 99.29%
房客 65 65 38.12%
房源 153 21 50.92%
Statistics on Trust Features of Accommodation Sharing Platform
Feature Grouping Based Trust Evaluation Model
数据 数量
房东数据 ≥10 000
房源数据 ≥75 000
评论数据 ≥1 300 000
信任特征(条件属性) 420
感知信任(决策属性) 3类(2、3、5)
Data Statistics
DNN参数 SVM参数
隐藏层数量 3 核函数 高斯函数
隐藏层单元数 80
激活函数 ReLU 惩罚系数 188.5
解决器 Adam
正则化参数 0.000 1 核函数系数 0.003 9
最大迭代次数 200
Parameter Settings of Classifiers
数据集 特征说明 实验
数据集1 算法2输出的最优目标函数的特征组合 实验1:降维有效性;实验2:特征分组效用;实验3:未入选
组合特征替换性;实验4:入选组合特征替换性
数据集2 每次选择某一特征分组,从数据集1中去掉来自该组的所有
特征,形成新特征组合
实验2:特征分组效用
数据集3 每次选择某一特征分组,在数据集1中使用该组内未入选特征
替换原已入选特征,形成新特征组合
实验3:未入选组合特征替换性
数据集4 选择一组特征分组,在数据集1中每次去掉该组一个已入选
特征,形成新特征组合
实验4:入选组合特征替换性
数据集5 全部420维特征 实验1:降维有效性
Description of the Experimental Data Sets
F1-Score of Feature Grouping of Trust Prediction
特征类型 视角 信任特征 入选率
2类信任 3类信任 5类信任
图像 房东 房东头像人数 100.00% 100.00% 100.00%
房东头像人物表情种类数量 20.00% / 40.00%
房东头像亚洲人数量 40.00% / 20.00%
房东头像性别种类 20.00% / 40.00%
房东头像男人个数 20.00% 20.00% /
房东头像人种数量 20.00% 20.00% 20.00%
房东头像表情平静人数 20.00% / 20.00%
房东头像皮肤类型 / 60.00% 40.00%
房东头像女人个数 / 40.00% 40.00%
房东头像平均颜值(男用户打分) / 20.00% 20.00%
房东头像平均微笑程度 / 20.00% 20.00%
房东头像高兴人数 / 20.00% /
房源 房源图片数量最大值 100.00% 80.00% 100.00%
文本 交互 正向评论占比 80.00% 80.00% 80.00%
正向短句占比 20.00% 20.00% /
正向短句评论数 / / 20.00%
关注房源内部环境负向评论占比 / / 40.00%
数值 房客 历史交易房客最近住宿时间的最小值(年) 100.00% / 20.00%
历史交易房客射手座人数 80.00% / /
历史交易房客教育水平缺失人数 / 100.00% 80.00%
历史交易房客水瓶座人数 / 60.00% 80.00%
历史交易房客处女座人数 / / 20.00%
历史交易房客巨蟹座人数 / / 20.00%
Feature Statistics of Selected Trust Feature Combinations
信任类别 去除组别 DNN SVM
Precision Recall F1-Score Accuracy Precision Recall F1-Score Accuracy
2类 第一组 87.503%↓ 86.452%↓ 82.658%↓ 84.942%↓ 88.943%↓ 88.339%↓ 85.306%↓ 88.339%↓
第二组 94.372%↓ 96.533% 94.335%↓ 93.372% 96.478%↓ 96.539%↓ 96.457%↓ 96.541%↓
第三组 96.914% 96.949% 96.920% 96.947% 96.923%↓ 96.945%↓ 96.929%↓ 96.949%↓
第四组 96.118% 93.403%↓ 95.957% 96.477% 97.079% 97.118%↓ 97.071%↓ 97.129%
第五组 96.256% 95.728% 94.169%↓ 94.379% 97.106% 97.145% 97.110% 97.145%
特征组合 95.733% 94.605% 95.130% 91.584% 97.080% 97.129% 97.080% 97.123%
3类 第一组 86.873%↓ 88.705%↓ 87.309%↓ 88.705%↓ 86.223%↓ 88.144%↓ 86.754%↓ 88.144%↓
第二组 93.748%↓ 94.241%↓ 93.448%↓ 94.241%↓ 91.599%↓ 92.441%↓ 91.84%↓ 92.441%↓
第三组 93.747%↓ 94.328%↓ 93.942%↓ 94.328%↓ 89.47%↓ 92.46%↓ 90.93%↓ 92.46%↓
第四组 94.152% 94.686% 94.344% 94.686% 91.661% 92.509% 91.906% 92.509%
第五组 94.287% 94.609% 93.661%↓ 94.609% 91.692% 92.528% 91.932% 92.528%
特征组合 93.816% 94.377% 94.042% 94.377% 91.608% 92.470% 91.853% 92.470%
5类 第一组 85.761%↓ 87.36%↓ 84.095%↓ 87.36%↓ 84.529%↓ 87.214%↓ 84.006%↓ 87.214%↓
第二组 91.685%↓ 92.189%↓ 91.514%↓ 92.189%↓ 91.247%↓ 92.034%↓ 91.569%↓ 92.034%↓
第三组 91.429%↓ 92.18%↓ 91.02%↓ 92.18%↓ 91.52%↓ 91.909%↓ 90.617%↓ 91.909%↓
第四组 91.946%↓ 92.702%↓ 92.214%↓ 92.702%↓ 92.173% 92.557% 92.131% 92.557%
第五组 92.22%↓ 92.751%↓ 92.386% 92.751%↓ 92.172% 92.557% 92.136% 92.557%
第六组 91.821%↓ 92.751%↓ 92.153%↓ 92.751%↓ 92.194% 92.567% 92.140% 92.567%
特征组合 92.907% 92.760% 92.302% 92.760% 92.161% 92.557% 92.122% 92.557%
Trust Prediction Effect of Removing a Set of Features and Feature Combination
Replacement Effect of Unselected Features
Mutual Substitution Effects of Selected Features
[1] Netter S, Pedersen E R G, Lüdeke-Freund F. Sharing Economy Revisited: Towards a New Framework for Understanding Sharing Models[J]. Journal of Cleaner Production, 2019, 221: 224-233.
doi: 10.1016/j.jclepro.2019.02.225
[2] 国家信息中心. 中国共享经济发展报告(2021)[R/OL]. [2021-05-01]. http://www.sic.gov.cn/News/557/10779.htm.
[2] (State Information Center. Report on the Development of China’s Sharing Economy(2021)[R/OL]. [2021-05-01]. http://www.sic.gov.cn/News/557/10779.htm.)
[3] 民宿大数据|2019年在线民宿行业发展分析报告[EB/OL]. (2019-12-03). [2021-05-01]. https://www.sohu.com/a/358073579_120164873.
[3] (Homestay Big Data | 2019 Online Homestay Industry Development Analysis Report[EB/OL]. (2019-12-03). [2021-05-01]. https://www.sohu.com/a/358073579_120164873.)
[4] 环球网. 小猪短租:我不是中国的Airbnb[EB/OL]. [2021-05-01]. https://quality.huanqiu.com/article/9CaKrnKeNuf.
[4] (Global Network. Xiaozhu Short-Term Rental: I am not China’s Airbnb[EB/OL]. [2021-05-01]. https://quality.huanqiu.com/article/9CaKrnKeNuf.)
[5] Pizam A. Peer-to-Peer Travel: Blessing or Blight?[J]. International Journal of Hospitality Management, 2014, 38: 118-119.
doi: 10.1016/j.ijhm.2014.02.013
[6] Hidden Cameras in Airbnb Rentals and Home Exchanges[EB/OL]. [2021-05-01]. https://sharetraveler.com/hidden-cameras-in-airbnb-rentals-and-home-exchanges/.
[7] Räisänen J, Ojala A, Tuovinen T. Building Trust in the Sharing Economy: Current Approaches and Future Considerations[J]. Journal of Cleaner Production, 2021, 279: 123724.
doi: 10.1016/j.jclepro.2020.123724
[8] Etzioni A. Cyber Trust[J]. Journal of Business Ethics, 2019, 156(1): 1-13.
doi: 10.1007/s10551-017-3627-y
[9] Liang S, Schuckert M, Law R, et al. Be a “Superhost”: The Importance of Badge Systems for Peer-to-Peer Rental Accommodations[J]. Tourism Management, 2017, 60: 454-465.
doi: 10.1016/j.tourman.2017.01.007
[10] 芮益芳. 小猪短租: 共享经济的关键是信任体系[J]. 商学院, 2015(10): 55-57.
[10] (Rui Yifang. Xiaozhu Short-Term Rental: The Key to the Sharing Economy is the Trust System[J]. Business Management Review, 2015(10): 55-57.)
[11] Yang S B, Lee K, Lee H N, et al. In Airbnb We Trust: Understanding Consumers’ Trust-Attachment Building Mechanisms in the Sharing Economy[J]. International Journal of Hospitality Management, 2019, 83: 198-209.
doi: 10.1016/j.ijhm.2018.10.016
[12] Alfina I, Ero J, Hidayanto A N, et al. The Impact of Cognitive Trust and E-WOM on Purchase Intention in C2C E-Commerce Site[J]. Journal of Computer Science, 2014, 10(12): 2518-2524.
doi: 10.3844/jcssp.2014.2518.2524
[13] Ert E, Fleischer A, Magen N. Trust and Reputation in the Sharing Economy: The Role of Personal Photos in Airbnb[J]. Tourism Management, 2016, 55: 62-73.
doi: 10.1016/j.tourman.2016.01.013
[14] Ranzini G, Etter M, Lutz C, et al. Privacy in the Sharing Economy[OL]. (2017-04-30). [2021-05-01]. https://ssrn.com/abstract=2960942.
[15] Ranzini G, Etter M, Vermeulen I. My Home on the Platform: Exploring the Physical Privacy Concerns of Home-Sharing Providers[J]. International Journal of Hospitality Management, 2020, 86: 102433.
doi: 10.1016/j.ijhm.2019.102433
[16] ter Huurne M, Ronteltap A, Corten R, et al. Antecedents of Trust in the Sharing Economy: A Systematic Review[J]. Journal of Consumer Behaviour, 2017, 16(6): 485-498.
doi: 10.1002/cb.1667
[17] Cheng X S, Fu S X, Sun J S, et al. An Investigation on Online Reviews in Sharing Economy Driven Hospitality Platforms: A Viewpoint of Trust[J]. Tourism Management, 2019, 71: 366-377.
doi: 10.1016/j.tourman.2018.10.020
[18] Sthapit E, Björk P. Sources of Distrust: Airbnb Guests’ Perspectives[J]. Tourism Management Perspectives, 2019, 31: 245-253.
doi: 10.1016/j.tmp.2019.05.009
[19] Mao Z X, Jones M F, Li M M, et al. Sleeping in a Stranger’s Home: A Trust Formation Model for Airbnb[J]. Journal of Hospitality and Tourism Management, 2020, 42: 67-76.
doi: 10.1016/j.jhtm.2019.11.012
[20] Ye S, Ying T Y, Zhou L Q, et al. Enhancing Customer Trust in Peer-to-Peer Accommodation: A “Soft” Strategy via Social Presence[J]. International Journal of Hospitality Management, 2019, 79: 1-10.
doi: 10.1016/j.ijhm.2018.11.017
[21] Kong Y, Wang Y, Hajli S, et al. In Sharing Economy We Trust: Examining the Effect of Social and Technical Enablers on Millennials’ Trust in Sharing Commerce[J]. Computers in Human Behavior, 2020, 108: 105993.
doi: 10.1016/j.chb.2019.04.017
[22] Xu X, Zeng S, He Y J. The Impact of Information Disclosure on Consumer Purchase Behavior on Sharing Economy Platform Airbnb[J]. International Journal of Production Economics, 2021, 231: 107846.
doi: 10.1016/j.ijpe.2020.107846
[23] Lu L, Cai R Y, King C. Building Trust Through a Personal Touch: Consumer Response to Service Failure and Recovery of Home-Sharing[J]. Journal of Business Research, 2020, 117: 99-111.
doi: 10.1016/j.jbusres.2020.05.049
[24] Cui W J, Chen P J, Zhang T T. A Qualitative Research Approach for Developing a Trust Model of P2P Accommodations for Chinese Travelers[J]. Journal of Tourism & Hospitality, 2018, 7(3): 1000359.
[25] Martin-Fuentes E, Fernandez C, Mateu C, et al. Modelling a Grading Scheme for Peer-to-Peer Accommodation: Stars for Airbnb[J]. International Journal of Hospitality Management, 2018, 69: 75-83.
doi: 10.1016/j.ijhm.2017.10.016
[26] Ert E, Fleischer A. The Evolution of Trust in Airbnb: A Case of Home Rental[J]. Annals of Tourism Research, 2019, 75: 279-287.
doi: 10.1016/j.annals.2019.01.004
[27] Su N, Mattila A S. Does Gender Bias Exist? The Impact of Gender Congruity on Consumer’s Airbnb Booking Intention and the Mediating Role of Trust[J]. International Journal of Hospitality Management, 2020, 89: 102405.
doi: 10.1016/j.ijhm.2019.102405
[28] Park S, Tussyadiah I P. How Guests Develop Trust in Hosts: An Investigation of Trust Formation in P2P Accommodation[J]. Journal of Travel Research, 2020, 59(8): 1402-1412.
doi: 10.1177/0047287519884654
[29] Tussyadiah I P, Park S. When Guests Trust Hosts for Their Words: Host Description and Trust in Sharing Economy[J]. Tourism Management, 2018, 67: 261-272.
doi: 10.1016/j.tourman.2018.02.002
[30] Zhang L, Yan Q, Zhang L. A Text Analytics Framework for Understanding the Relationships Among Host Self-Description, Trust Perception and Purchase Behavior on Airbnb[J]. Decision Support Systems, 2020, 133: 113288.
doi: 10.1016/j.dss.2020.113288
[31] Thuy N N, Wongthanavasu S. An Efficient Stripped Cover-Based Accelerator for Reduction of Attributes in Incomplete Decision Tables[J]. Expert Systems with Applications, 2020, 143: 113076.
doi: 10.1016/j.eswa.2019.113076
[32] Chen X, Yuan Y Y, Orgun M A, et al. A Topic-Sensitive Trust Evaluation Approach for Users in Online Communities[J]. Knowledge-Based Systems, 2020, 194: 105546.
doi: 10.1016/j.knosys.2020.105546
[33] Andrade A A, Cardoso M G M S, Lopes V V. Logical Rules Characterization of Online Consumer Trust[J]. International Transactions in Operational Research, 2021, 28(3): 1393-1416.
doi: 10.1111/itor.12630
[34] Agag G, Eid R. Examining the Antecedents and Consequences of Trust in the Context of Peer-to-Peer Accommodation[J]. International Journal of Hospitality Management, 2019, 81: 180-192.
doi: 10.1016/j.ijhm.2019.04.021
[35] Shen H T, Zhu Y H, Zheng W, et al. Half-Quadratic Minimization for Unsupervised Feature Selection on Incomplete Data[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(7): 3122-3135.
doi: 10.1109/TNNLS.2020.3009632
[36] 梁家铭, 赵洁, 郑鹏, 等. 特征选择下融合图像和文本分析的在线短租平台信任计算框架[J]. 数据分析与知识发现, 2021, 5(2): 129-140.
[36] (Liang Jiaming, Zhao Jie, Zheng Peng, et al. Framework for Computing Trust in Online Short-Rent Platform Using Feature Selection of Images and Texts[J]. Data Analysis and Knowledge Discovery, 2021, 5(2): 129-140.)
[37] Zhang L, Yan Q, Zhang L. A Computational Framework for Understanding Antecedents of Guests’ Perceived Trust Towards Hosts on Airbnb[J]. Decision Support Systems, 2018, 115: 105-116.
doi: 10.1016/j.dss.2018.10.002
[38] García-Torres M, Gómez-Vela F, Melián-Batista B, et al. High-Dimensional Feature Selection via Feature Grouping: A Variable Neighborhood Search Approach[J]. Information Sciences, 2016, 326: 102-118.
doi: 10.1016/j.ins.2015.07.041
[39] Yu L, Liu H. Efficient Feature Selection via Analysis of Relevance and Redundancy[J]. Journal of Machine Learning Research, 2004, 5: 1205-1224.
[40] Dash M, Liu H. Feature Selection for Classification[J]. Intelligent Data Analysis, 1997, 1(1): 131-156.
doi: 10.3233/IDA-1997-1302
[41] Hall M A. Correlation Based Feature Selection Algorithm for Machine Learning[D]. Hamilton: University of Waikato, 1999.
[1] Xu Yuemei, Wang Zihou, Wu Zixin. Predicting Stock Trends with CNN-BiLSTM Based Multi-Feature Integration Model[J]. 数据分析与知识发现, 2021, 5(7): 126-138.
  Copyright © 2016 Data Analysis and Knowledge Discovery   Tel/Fax:(010)82626611-6626,82624938   E-mail:jishu@mail.las.ac.cn