以消费者行为分析和离散选择的相关理论为基础,通过对用户生成内容进行特征粒度的情感分析,同时从产品的客观数据和用户生成的主观内容中提取模型特征,使用有监督的学习训练MNL模型预测产品的消费者剩余作为搜索排序的依据,并实现了手机、笔记本电脑和数码相机类的产品搜索系统。双盲实验表明,该文提出的产品搜索模型搜索效果比基准算法有显著的提高。
Abstract
Based on theories of consumer behavior analysis and discrete choice analysis, we perform the feature based sentiment analysis on user reviews with features selected from both objective product features and subjective user opinion data. Then we \ train a MNL model to predict products consumer surplus as the ranking criterion. With this method, we implemented product search engine for cell phone, laptop and digital camera. Double-blind trial on users shows that our model significantly outperforms the baselines.
Key wordsproduct search; MNL model; sentiment analysis; feature selection; user generated content
关键词
产品搜索 /
MNL模型 /
情感分析 /
特征选取 /
用户生成内容
{{custom_keyword}} /
Key words
product search /
MNL model /
sentiment analysis /
feature selection /
user generated content
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] R Kumar, A Tomkins. A characterization of online browsing behavior[C]//Proceedings of the 19th international conference on World wide web, Raleigh, North Carolina, USA, April 26-30, 2010: 561-570.
[2] G Singh, N Parikh, N Sundaresn. User behavior in zero-recall ecommerce queries[C]//Proceedings of SIGIR, Beijing, China, 2011: 75-84.
[3] K P Yee, K Swearingen, K Li, et al. Faceted Metadata for Image Search and Browsing[C]//Proceedings of CHI 2003, April 5-10, 2003: 401-408
[4] Z Nie, J R Wen, W Y Ma. Webpage understanding: beyond page-level search[C]//Proceedings of ACM SIGMOD Record, 37(4), 2008: 48-54
[5] C Scaffidi, K Bierhoff, E Chang, et al. Red Opal: product-feature scoring from reviews[C]//Proceedings of EC07, 2007: 182-191.
[6] K Zhang, R Narayanan, A Choudhary. Voice of the customers: mining online customer reviews for product feature-based ranking[C]//Proceedings of WSON2010. Boston, Ma, 2010:11-11
[7] B Li, A Ghose, P G Ipeirotis. A Demo Search Engine for Products[C]//Proceedings of WWW 2011, March 28-April 1, 2011.
[8] B Li , A Ghose, P G Ipeirotis. Towards a theory model for product search[C]//Proceedings of WWW 2011, March 28-April 1, 2011.
[9] A. Marshall. Principles of Economics[M]. London, Eighth ed. Macmillan and Co., 1926.
[10] K Lancaster. Consumer Demand: A New Approach[M]. Columbia University Press, New York, 1971.
[11] S Rosen. Hedonic prices and implicit markets: Product differentiation in pure competition[J]. Journal of Political Economy, 1974, 82(1): 34-55.
[12] D McFadden. Conditional Logit Analysis of Qualitative Choice Behavior[M]. Academic Press, New York, 1974.
[13] http://elsa.berkeley.edu/reprints/mcfadden/zarembka.pdf
[14] D McFadden, K Train. Mixed MNL Models for Discrete Response[J]. Journal of Applied Econometrics, 2000, 15(5):447-470.
[15] E Frank, M Hall, G Holmess, et al. Weka[J]. Data Mining and Knowledge Discovery Handbook, 2005, VIII, pp. 1305-1314.
[16] H Akaike. A new look at the statistical model identification[J]. IEEE Transactions on Automatic Control, 1974, 19(6):716-723.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
自然科学基金资助项目(61272343);教育部科技发展中心专项研究资助课题资助博士点基金项目(FSSP项目20120001110112)
{{custom_fund}}
PDF(2857 KB)
