[1]何柔萤,徐 建.基于注意力卷积神经网络的工作票专家推荐方法[J].南京理工大学学报(自然科学版),2019,43(01):13.[doi:10.14177/j.cnki.32-1397n.2019.43.01.002]
 He Rouying,Xu Jian.Expert recommendation for trouble tickets usingattention-based CNN model[J].Journal of Nanjing University of Science and Technology,2019,43(01):13.[doi:10.14177/j.cnki.32-1397n.2019.43.01.002]
点击复制

基于注意力卷积神经网络的工作票专家推荐方法()
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
43卷
期数:
2019年01期
页码:
13
栏目:
出版日期:
2019-02-28

文章信息/Info

Title:
Expert recommendation for trouble tickets usingattention-based CNN model
文章编号:
1005-9830(2019)01-0013-09
作者:
何柔萤徐 建
南京理工大学 计算机科学与工程学院,江苏 南京 210094
Author(s):
He RouyingXu Jian
School of Computer Science and Engineering,Nanjing University of Scienceand Technology,Nanjing 210094,China
关键词:
专家推荐 卷积神经网络 注意力机制 系统运维
Keywords:
expert recommendation convolutional neural network attention mechanism system operation and maintenance
分类号:
TP391
DOI:
10.14177/j.cnki.32-1397n.2019.43.01.002
摘要:
为了更准确地将工作票推荐给具备解决问题能力的系统运维专家,对历史工作票数据进行研究提出基于深度学习的工作票专家推荐算法。首先根据专业熟练度水平和领域知识构建专家能力模型,然后设计卷积神经网络框架,在输入层中引入注意力来提高模型对工作票文本特征提取能力,并度量与专家模型的匹配度,实现以推荐质量为依据的专家推荐。在真实的数据集上进行了实验,结果表明与传统的基于机器学习的推荐方法相比,该方法的准确率提升了6%,引入注意力可以有效学习特征权重。
Abstract:
In order to improve the accuracy of recommending trouble tickets to experts with problem-solving ability,the expert recommendation algorithm based on deep learning are studied by learning the historical trouble ticket data. According to the expert’s professional proficiency level and domain knowledge,an expert’s ability model is constructed,and an expert recommendation framework based on convolution neural network is defined. Attention mechanism is introduced into input layer of the model to enhance the ability of describing the feature extraction of tickets. This paper measures the similarity match score between the problem description and the expert’s model to realize expert recommendation based on quality. Experimental results on real ticket datasets show that the proposed method can improve the accuracy by about 6% compared with the traditional machine learning classification recommendation methods,and can effectively learn the weight of ticket feature by introducing attention.

参考文献/References:

[1] Li T,Peng W,Perng C,et al. An integrated data-driven framework for computing system management[J]. IEEE Transactions on Systems Man & Cybernetics Part A Systems & Humans,2010,40(1):90-99.
[2]Xu J,Tang L,Li T. System situation ticket identification using SVMs ensemble[J]. Expert Systems with Applications,2016,60:130-140.
[3]Zeng C,Zhou W,Li T,et al. Knowledge guided hierarchical multi-label classification over ticket data[J]. IEEE Transactions on Network & Service Management,2017,14(2):246-260.
[4]Sun H,Srivatsa M,Tan S,et al. Analyzing expert behaviors in collaborative networks[C]//Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York,USA:ACM,2014:1486-1495.
[5]李颖,朱保平. 基于交互链路的相似用户推荐算法[J]. 南京理工大学学报,2018,42(2):183-190.
Li Ying,Zhu Baoping. Similar user recommendation algorithm based on interactive link[J]. Journal of Nanjing University of Science and Technology,2018,42(2):183-190.
[6]Shao Q,Chen Y,Tao S,et al. Efficient ticket routing by resolution sequence mining[C]//Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York,USA:ACM,2008:605-613.
[7]Xu J,He R,Zhou W,et al. Trouble ticket routing models and their applications[J]. IEEE Transactions on Network & Service Management,2018,15(2):530-543.
[8]李贤慧,余正涛,魏斯超,等. 基于Listwise的深度学习专家排序方法[J]. 模式识别与人工智能,2015,28(11):976-982.
Li Xianhui,Yu Zhengtao,Wei Sichao. Deep learning expert ranking method based on Listwise[J]. Pattern Recognition and Artificial Intelligence,2015,28(11):976-982.
[9]梁斌,刘全,徐进,等. 基于多注意力卷积神经网络的特定目标情感分析[J]. 计算机研究与发展,2017,54(8):1724-1735.
Liang Bin,Liu Quan,Xu Jin,et al. Aspect-based sentiment analysis based on multi-attention CNN[J]. Journal of Computer Research and Development,2017,54(8):1724-1735.

[10] Zhou W,Xue W,Baral R,et al. STAR:a system for ticket analysis and resolution[C]//Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York,USA:ACM,2017:2181-2190.
[11]Bahdanau D,Cho K,Bengio Y. Neural machine translation by jointly learning to align and translate[J]. Proceedings of ICLR,2015:1-15.
[12]Yin W,Schütze H,Xiang B,et al. ABCNN:attention-based convolutional neural network for modeling sentence pairs[J]. Computer Science,2015.2015,arXiv:1512.05193.
[13]Wang Lin,Cao Zhu,de Melo G,et al. Relation classification via multi-level attention cnns[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics. Stroudsburg,PA:ACL,2016:1298 1307.
[14]Momtazi S,Naumann F. Topic modeling for expert finding using latent Dirichlet allocation[J]. Wiley Interdisciplinary Reviews:Data Mining and Knowledge Discovery,2013,3(5):346-353.

相似文献/References:

[1]王 林,董 楠.基于Gabor特征与卷积神经网络的人体轮廓提取[J].南京理工大学学报(自然科学版),2018,42(01):89.[doi:10.14177/j.cnki.32-1397n.2018.42.01.013]
 Wang Lin,Dong Nan.Human silhouette identification based on Gabor featureand convolutional neural network[J].Journal of Nanjing University of Science and Technology,2018,42(01):89.[doi:10.14177/j.cnki.32-1397n.2018.42.01.013]

备注/Memo

备注/Memo:
收稿日期:2018-10-26 修回日期:2018-12-04
基金项目:国家自然科学基金(61872186,61802205); 江苏省研究生科技与实践创新计划项目(KYCX17_0403)
作者简介:何柔萤(1994-),硕士生,主要研究方向:文本数据挖掘,E-mail:bingdan_hry@163.com; 通讯作者:徐建(1979-),男,博士,教授,主要研究方向:数据挖掘,E-mail:dolphin.xu@njust.edu.cn。
引文格式:何柔萤,徐建. 基于注意力卷积神经网络的工作票专家推荐方法[J]. 南京理工大学学报,2019,43(1):12-21.
投稿网址:http://zrxuebao.njust.edu.cn
更新日期/Last Update: 2019-02-28