CrowdDepict: Personalized Recommendation Content Generation Based on Heterogeneous Crowdsourced Data

doi:10.3778/j.issn.1673-9418.1912015

Abstract

Abstract:

With the rapid development of the E-commerce, the text of the products' advertisement and recommendation is extremely vital in the case that people cannot touch the real products. Traditional product descriptions which are written manually show the same product information to all users, which do not consider that different users pay attention to different attributes. Moreover, the efficiency of manually written product descriptions cannot match the growth rate of products, so how to automatically generate personalized product descriptions has become a frontier research problem. This paper mainly studies the generation of personalized product description, takes the personalized characteristics of users into account and generates the product description text corresponding to the interest for each user. Due to the lack of personalized product description data set, the CrowdDepict model is proposed. It collects relevant corpus through public data sources such as Douban and JD, and generates personalized product description with product comments, etc. The experimental results show that the personalized product description model CrowdDepict can automatically generate the personalized product description according to the user preference, the description covers the user interest and product characteristics, and the text expression is smooth.

Key words: deep learning, recommendation content generation, personalization, natural language generation, crowdsourced data

摘要：

随着网上购物逐渐发展，在无法接触到实体商品的情况下，商品描述显得尤为重要。传统人工撰写的商品描述语对所有用户展示相同的商品信息，但没有考虑到不同用户所关注的不同属性，并且人工撰写的效率无法与产品增长速度相匹配，因此如何自动生成个性化产品描述成为前沿研究问题。主要研究个性化商品描述内容生成，考虑用户的个性化特征，对每个用户生成对应其兴趣点的商品描述文本。因为个性化商品描述数据集的缺失，提出CrowdDepict方法，通过豆瓣、京东等公开数据源获取商品描述相关语料处理，利用商品评论等数据，生成商品个性化描述。实验结果表明，提出的个性化商品描述模型可根据用户偏好自动生成个性化的商品描述，内容覆盖用户兴趣与产品特点，文本表达流畅。

关键词: 深度学习, 商品描述内容生成, 个性化, 文本生成, 群智数据

ZHANG Qiuyun, GUO Bin, HAO Shaoyang, WANG Hao, YU Zhiwen, JING Yao. CrowdDepict: Personalized Recommendation Content Generation Based on Heterogeneous Crowdsourced Data[J]. Journal of Frontiers of Computer Science and Technology, 2020, 14(10): 1670-1680.

张秋韵，郭斌，郝少阳，王豪，於志文，景瑶. CrowdDepict：多源群智数据驱动的个性化商品描述生成方法[J]. 计算机科学与探索, 2020, 14(10): 1670-1680.

References

[1] Bengio Y, Ducharme R, Vincent P, et al. A neural probabilistic language model[J]. Journal of Machine Learning Research, 2003, 3: 1137-1155.
[2] Xu B, Xu Y, Liang J Q, et al. CN-DBpedia: a never-ending Chinese knowledge extraction system[C]//LNCS 10351: Proceedings of the 30th International Conference on Industrial Engineering and Other Applications of Applied Intelligent Systems, Arras, Jun 27-30, 2017. Berlin, Heidelberg: Springer, 2017: 428-438.
[3] Weinsberg U, Bhagat S, Ioannidis S, et al. BlurMe: inferring and obfuscating user gender based on ratings[C]//Proceedings of the 6th ACM Conference on Recommender Systems, Sep 21-26, 2012. New York: ACM, 2012: 195-202.
[4] Bi B, Shokouhi M, Kosinski M, et al. Inferring the demographics of search users: social data meets search queries[C]//Proceedings of the 22nd International World Wide Web Conference, Rio de Janeiro, May 13-17, 2013. New York: ACM, 2013: 131-140.
[5] Youyou W, Kosinski M, Stillwell D. Computer-based personality judgments are more accurate than those made by humans[J]. Proceedings of the National Academy of Sciences, 2015, 112(4): 1036-1040.
[6] Schwartz H A, Eichstaedt J C, Kern M L, et al. Personality, gender, and age in the language of social media: the open-vocabulary approach[J]. PloS One, 2013, 8(9): e73791.
[7] Tang J, Yang Y, Carton S, et al. Constext-aware natural language generation with recurrent neural networks[J]. arXiv:1611.09900, 2016.
[8] Zheng L, Noroozi V, Yu P S. Joint deep modeling of users and items using reviews for recommendation[C]//Proceedings of the 10th ACM International Conference on Web Search and Data Mining, Cambridge, Feb 6-10, 2017. New York: ACM, 2017: 425-434.
[9] Lipton Z C, Vikram S, McAuley J. Generative concatenative nets jointly learn to write and classify reviews[J]. arXiv: 1511.03683, 2015.
[10] Costa F, Ouyang S, Dolog P, et al. Automatic generation of natural language explanations[C]//Proceedings of the 23rd International Conference on Intelligent User Interfaces Companion, Tokyo, Mar 7-11, 2018. New York: ACM, 2018: 57.
[11] Ni J, McAuley J. Personalized review generation by expanding phrases and attending on aspect-aware representations[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Jul 15-20, 2018. Stroudsburg: ACL, 2018: 706-711.
[12] Nenkova A, Vanderwende L, McKeown K. A compositional context sensitive multi-document summarizer: exploring the factors that influence summarization[C]//Proceedings of the 29th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Seattle, Mar 14-18, 2006. New York: ACM, 2006: 573-580.
[13] Erkan G, Radev D R. Lexpagerank: prestige in multi-document text summarization[C]//Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing, Barcelona, Jun 25-26, 2004. Stroudsburg: ACL, 2004: 365-371.
[14] Svore K M, Vanderwende L, Burges C J C. Enhancing single-document summarization by combining RankNet and third-party sources[C]//Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, Prague, Jun 28-30, 2007. Stroudsburg: ACL, 2007: 448-457.
[15] Cheng J, Lapata M. Neural summarization by extracting sentences and words[J]. arXiv:1603.07252, 2016.
[16] Cao Z, Li W, Li S, et al. Attsum: joint learning of focusing and summarization with neural attention[J]. arXiv:1604. 00125, 2016.
[17] Paulus R, Xiong C, Socher R. A deep reinforced model for abstractive summarization[J]. arXiv:1705.04304, 2017.
[18] Pasunuru R, Bansal M. Multi-reward reinforced summarization with saliency and entailment[J]. arXiv:1804.06451, 2018.
[19] Rush A M, Chopra S, Weston J. A neural attention model for abstractive sentence summarization[J]. arXiv:1509.00685, 2015.
[20] Mihalcea R, Tarau P. Textrank: bringing order into text[C]// Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing, Barcelona, Jul 25-26, 2004. Stroudsburg: ACL, 2004: 404-411.
[21] Papineni K, Roukos S, Ward T, et al. BLEU: a method for automatic evaluation of machine translation[C]//Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, Philadelphia, Jul 6-12, 2002. Stroudsburg: ACL, 2002: 311-318.
[22] Lin C Y, Hovy E H. Automatic evaluation of summaries using n-gram co-occurrence statistics[C]//Proceedings of the 2003 Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics, Edmonton, May 27-Jun 1, 2003. Stroudsburg: ACL, 2003: 150-157.