[目的/意义]针对现有图书馆数字参考咨询人机对话机器人在对话回复内容方面的局限，提出一种融合人物画像的对话生成模型，使其回复更具个性化和趣味性，以提升图书馆智能咨询服务效果。[方法/过程]利用人机对话技术，对图书馆数字参考咨询服务中的用户和问题进行自动建模，建模方式分为个性化回复风格建模和特定用户属性建模。在个性化回复风格建模上，提出一种基于对话表示和相关性回复建模方法，该方法在学习到对话相关性的同时，利用个性化文本生成个性化的回复；在用户个人属性建模方面，基于信息抽取技术生成用户的人物画像。[结果/结论]实验结果表明，所提出的个性化回复生成模型优于已有的回复生成模型，人物画像识别的F值达到了99.8%。
[Purpose/significance] In view of the limits of response generation of conversational robots in the existing in library digital reference, this paper proposes a dialogue generation model which integrates the portraits of characters, making the reply more personalized and interesting, in order to improve the effect of library intelligent reference service.[Method/process] We automatically model the specific roles and questions in digital reference service of library in two separate ways. First is to model the personalized responding style of specific role and second is to model the aspects of the role. In modeling personalized responding style, we propose an utterance representation and responding relevance-based approach to simultaneously learn the relevance of dialogue and utilize the personalized text to generate personalized responses. In modeling aspects of a specific role, we establish human profile by employing the information extraction techniques. [Result/conclusion] The experimental results show that, the personalized reply generation model proposed by us is superior to the best one, and the F score of user profiling recognition is 99.8%.
[1] 郭山.智能机器人技术在公共图书馆实时参考咨询服务中的应用[J].图书馆学研究,2017(10):58-61.
[2] 姚飞, 纪磊, 张成昱,等. 实时虚拟参考咨询服务新尝试——清华大学图书馆智能聊天机器人[J]. 现代图书情报技术, 2011, 27(4):77-81.
[3] 孙翌, 李鲍, 曲建峰. 图书馆智能化IM咨询机器人的设计与实现[J]. 现代图书情报技术, 2011(5):88-92.
[4] 李文江, 陈诗琴. AIMLBot智能机器人在实时虚拟参考咨询中的应用[J]. 数据分析与知识发现, 2012, 28(7):127-132.
[5] 沈奎林, 邵波, 赵华. 利用微信构建图书馆智能问答系统[J]. 图书馆学研究, 2015(8):75-80.
[6] VINYALS O, LE Q. A neural conversational model[EB/OL].[2015-07-22].https://arxiv.org/pdf/1506.05869.pdf.
[7] SORDONI A, GALLEY M, AULI M, et al. A neural network approach to context-sensitive generation of conversational responses[EB/OL].[2018-06-22].https://arxiv.org/pdf/1506.06714.pdf.
[8] SHANG L, LU Z, LI H. Neural responding machine for short-text conversation[EB/OL].[2018-04-27].https://arxiv.org/pdf/1503.02364.pdf.
[9] MOU L, SONG Y, YAN R, et al. Sequence to backward and forward sequences:a content-introducing approach to generative short-text conversation[C]//Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics:Technical Papers, 2016:3349-3358. 2016.
[10] XING C, WU W, WU Y, et al. Topic augmented neural response generation with a joint attention mechanism[EB/OL].[2018-09-19]. http://cn.arxiv.org/pdf/1606.08340.pdf.
[11] SERBAN I V, SORDONI A, LOWE R, et al. A hierarchical latent variable encoder-decoder model for generating dialogues[C]//Proceedings of the 39th international conference on Association for the Advancement of Artificial Intelligence. San Francisco:AAAI, 2017:3295-3301.
[12] ZHAO T, ZHAO R, ESKENAZI M. Learning discourse-level diversity for neural dialog models using conditional variational autoencoders[EB/OL].[2017-10-21].https://arxiv.org/pdf/1703.10960.pdf.
[13] KAMBHATLA N. Combining lexical, syntactic, and semantic features with maximum entropy models for extracting relations[C]//Proceedings of the ACL 2004 on Interactive poster and demonstration sessions. Stroudsburg:Association for Computational Linguistics, 2004:22.
[14] ZENG D, LIU K, LAI S, et al. Relation classification via convolutional deep neural network[C]//Proceedings of COLING 2014, the 25th international conference on computational linguistics:Technical Papers. Dublin:ACL,2014:2335-2344.
[15] LODHI H, SAUNDERS C, SHAWE-TAYLOR J, et al. Text classification using string kernels[J]. Journal of machine learning research, 2002(2):419-444.
[16] 刘克彬,李芳,刘磊,等. 基于核函数中文关系自动抽取系统的实现[J]. 计算机研究与发展,2007, 4(8), 1406-1411.
[17] BANKO M, CAFARELLA M J, SODERLAND S, et al. Open information extraction from the web[C]//Proceeding IJCAI'07 San Francisco:Morgan Kaufmann Publishers, 2007:2670-2676.
[18] 周蓝珺. 音乐领域中文实体关系抽取研究[D]. 哈尔滨:哈尔滨工业大学, 2009.
[19] 刘勇杰,姜天文,秦兵,等. 融入搜索引擎的启发式开放域三元组抽取[C]//第五届全国知识图谱与语义计算大会. 成都:CCKS, 2017.
[20] FADER A, SODERLAND S, ETZIONI O. Identifying relations for open information extraction[C]//Proceedings of the conference on empirical methods in natural language processing. Edinburgh:Association for Computational Linguistics, 2011:1535-1545.
[21] BANCHS R E. Movie-DiC:a movie dialogue corpus for research and development[C]//Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics:Short Papers-Volume 2. Jeju Island:Association for Computational Linguistics, 2012:203-207.
[22] LOWE R, POW N, SERBAN I, et al. The ubuntu dialogue corpus:a large dataset for research in unstructured multi-turn dialogue systems[EB/OL].[2018-02-04].https://arxiv.org/pdf/1506.08909.pdf.
[23] KINGMA D P, WELLING M. Auto-encoding variational bayes[EB/OL].[2018-05-01].https://arxiv.org/pdf/1312.6114.pdf.
