Please wait a minute...
Advanced Search
数据分析与知识发现  2018, Vol. 2 Issue (9): 22-30     https://doi.org/10.11925/infotech.2096-3467.2018.0015
  研究论文 本期目录 | 过刊浏览 | 高级检索 |
基于F-BiGRU情感分析的产品选择方法*
余本功1,2(), 张培行1, 许庆堂1
1合肥工业大学管理学院 合肥 230009
2过程优化与智能决策教育部重点实验室 合肥 230009
Selecting Products Based on F-BiGRU Sentiment Analysis
Yu Bengong1,2(), Zhang Peihang1, Xu Qingtang1
1School of Management, Hefei University of Technology, Hefei 230009, China
2Key Laboratory of Process Optimization and Intelligent Decision-making, Ministry of Education, Hefei 230009, China
全文: PDF (643 KB)   HTML ( 3
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】为提高产品选择效率, 帮助消费者更好地制定购物决策, 本文在门限递归单元的基础上, 提出一种特征强化双向门限递归单元模型(Feature Bidirectional Gated Recurrent Unit, F-BiGRU)。【方法】首先, 获取相关产品的在线评论信息; 然后对在线评论按照产品属性进行分割; 使用正向情感评论和负向情感评论对F-BiGRU模型进行训练; 最后使用F-BiGRU模型对产品各属性的评论进行情感量化, 得到产品各属性的情感满意程度, 并使用TOPSIS法对候选产品进行排序。【结果】选取汽车口碑文本评论数据进行实证, 对比相关情感分析方法, F-BiGRU方法提高了情感分析的准确度, 更适应在线评论短文本的特点。【局限】深度学习模型需要大规模的数据集, 本文方法在一些小数据集上的表现可能不佳。【结论】基于F-BiGRU情感分析的产品选择方法提高了情感分析的准确度, 能更高效快捷地帮助消费者进行产品选择。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
余本功
张培行
许庆堂
关键词 产品选择在线评论情感分析深度学习门限递归单元    
Abstract

[Objective] This paper proposes a product selection method based on the Feature Bidirectional Gated Recurrent Unit model (F-BiGRU), aiming to improve the efficiency of customers’ product selection and help them make better shopping decisions. [Methods] First, we retrieved online reviews for related products. Then, we categorized these online reviews in accordance with the product attributes. Third, we trained the F-BiGRU model using positive and negative reviews. Fourth, we quantified the sentiment of reviews on different attributes with the F-BiGRU model. Finally, we got the degrees of satisfaction on product attributes, and sorted the products using TOPSIS method. [Results] We retrieved the review texts on cars to conduct an empirical analysis. We found that the F-BiGRU method improved the accuracy of sentiment analysis, and is more appropriate for the short text reviews than traditional methods. [Limitations] The proposed deep learning model requires large dataset, which limits its performance with smaller datasets. [Conclusions] The product selection method based on F-BiGRU helps consumers choose needed products more efficiently.

Key wordsProduct Selection    Online Review    Sentiment Analysis    Deep Learning    Gated Recurrent Unit
收稿日期: 2018-01-04      出版日期: 2018-10-25
ZTFLH:  分类号: C931.6  
基金资助:*本文系国家自然科学基金项目“基于制造大数据的产品研发知识集成与服务机制研究”(项目编号: 71671057)的研究成果之一
引用本文:   
余本功, 张培行, 许庆堂. 基于F-BiGRU情感分析的产品选择方法*[J]. 数据分析与知识发现, 2018, 2(9): 22-30.
Yu Bengong,Zhang Peihang,Xu Qingtang. Selecting Products Based on F-BiGRU Sentiment Analysis. Data Analysis and Knowledge Discovery, 2018, 2(9): 22-30.
链接本文:  
https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2018.0015      或      https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/Y2018/V2/I9/22
  GRU模型结构
  基于F-BiGRU情感分析的产品选择方法
  F-BiGRU模型结构图
方法 准确率 召回率 F1值
NB 91.18% 92.21% 91.69%
SVM 93.60% 87.93% 90.68%
NN 92.39% 93.27% 92.83%
RF 91.52% 86.62% 89.00%
CNN 96.91% 94.87% 95.88%
GRU 96.17% 97.23% 96.70%
F-BiGRU 98.05% 97.18% 97.61%
  各模型分类评价对比
车型 均价(万) 口碑评论文本数量(条)
瑞虎7 13 3 755
思域 14 2 171
荣威i6 12 1 665
明锐 15 1 390
轩逸 13 1 387
朗逸 14 3 127
博越 13 4 439
荣威RX5 15 6 374
  候选车型价格及评论数量
车型 空间 动力 操控 油耗 舒适性 外观 内饰
瑞虎7 0.82664692 0.79215884 0.85923731 0.61679804 0.7480377 0.93323213 0.66287327
思域 0.751369 0.77653182 0.57529676 0.60946149 0.40987739 0.80474067 0.33736464
荣威i6 0.87899387 0.78861713 0.81631488 0.75512266 0.49496683 0.92696506 0.57103211
明锐 0.82699454 0.32145673 0.77134079 0.65305305 0.43916675 0.79899246 0.53765839
轩逸 0.80659056 0.37046233 0.54579508 0.69502336 0.67606455 0.70648897 0.49541172
朗逸 0.81036848 0.3259283 0.7747519 0.71296167 0.43799475 0.83163321 0.2792967
博越 0.6879878 0.74954957 0.82120782 0.33669931 0.75716293 0.84331733 0.79239517
荣威RX5 0.90069151 0.71247077 0.78634995 0.60119843 0.56585765 0.90103316 0.67857999
  候选车型各属性满意程度
  汽车产品属性满意程度图
空间 动力 操控 油耗 舒适性 外观 内饰
0.063517 0.156486 0.183341 0.09252 0.303589 0.166057 0.034488
  属性权重
车型 d* d 0 c 排序
瑞虎7 0.0146127 0.1470008481 0.9095824745 1
博越 0.04468999 0.1377778628 0.7550802033 2
荣威RX5 0.06288051 0.0997304454 0.6133070570 3
荣威i6 0.08037187 0.1075123034 0.5722264991 4
轩逸 0.09925965 0.0883214247 0.4708440076 1
思域 0.12167597 0.0776130979 0.3894498509 2
朗逸 0.12489039 0.0594943325 0.3226641031 3
明锐 0.1252614 0.0551273065 0.3056028650 4
  候选车型TOPSIS排序
[1] Archak N, Ghose A, Ipeirotis P G.Show me the Money!: Deriving the Pricing Power of Product Features by Mining Consumer Reviews[C]//Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Jose, California, USA. ACM, 2007: 56-65.
[2] Wang H, Guo K, Zhao Y.A TAM-Based Model of Posting Positive and Negative Reviews on Online Reputation Systems[J]. International Journal of Digital Content Technology & Its Applications, 2012, 6(22): 484-493.
[3] 殷国鹏. 消费者认为怎样的在线评论更有用? 社会性因素的影响效应[J]. 管理世界, 2012 (12): 115-124.
[3] (Yin Guopeng.Which Online Reviews are More Useful in Consumer Idea? Influence Effect of Social Factors[J]. Management World, 2012(12): 115-124.)
[4] 龚艳萍, 梁树霖. 在线评论对新技术产品消费者采用意愿的影响研究——基于ELM视角[J]. 软科学, 2014, 28(2): 96-99, 105.
[4] (Gong Yanping, Liang Shulin.Study on the Influence of Online Reviews on Consumers’ Willingness to Adopt the New Technological Products——Based on ELM View[J]. Soft Science, 2014, 28(2): 96-99, 105)
[5] 杜学美, 丁璟妤, 谢志鸿, 等. 在线评论对消费者购买意愿的影响研究[J]. 管理评论, 2016, 28(3): 173-183.
[5] (Du Xuemei, Ding Jingyu, Xie Zhihong, et al.An Empirical Study on the Impact of Online Reviews on Consumers’ Purchasing Intention[J]. Management Review, 2016, 28(3): 173-183.)
[6] 梁霞, 姜艳萍, 高梦. 基于在线评论的产品选择方法[J]. 东北大学学报: 自然科学版, 2017, 38(1): 143-147.
[6] (Liang Xia, Jiang Yanping, Gao Meng.Product Selection Methods Based on Online Reviews[J]. Journal of Northeastern University: Natural Science, 2017, 38(1): 143-147.)
[7] 周梅华, 李佩镅, 牟宇鹏. 在线评论对消费者购买意愿的影响——心理距离的中介作用[J]. 软科学, 2015, 29(1): 101-104.
[7] (Zhou Meihua, Li Peimei, Mou Yupeng.Effects of Online Reviews on Purchase Intention of Consumers——The Mediation of Psychological Distance[J]. Soft Science, 2015, 29(1): 101-104.)
[8] 唐晓波, 朱娟, 杨丰华. 基于情感本体和KNN算法的在线评论情感分类研究[J]. 情报理论与实践, 2016, 39(6): 110-114.
[8] (Tang Xiaobo, Zhu Juan, Yang Fenghua.Research on Sentiment Classification of Online Reviews Based on Sentiment Ontology and KNN Algorithm[J]. Information Studies: Theory & Application, 2016, 39(6): 110-114.)
[9] Bai X.Predicting Consumer Sentiments from Online Text[J]. Decision Support Systems, 2011, 50(4): 732-742.
doi: 10.1016/j.dss.2010.08.024
[10] 李实, 叶强, 李一军, 等. 中文网络客户评论的产品特征挖掘方法研究[J]. 管理科学学报, 2009, 12(2): 142-152.
[10] (Li Shi, Ye Qiang, Li Yijun, et al.Mining Features of Products from Chinese Customer Online Reviews[J]. Journal of Management Sciences in China, 2009, 12(2): 142-152.)
[11] 郑丽娟, 王洪伟. 基于情感本体的在线评论情感极性及强度分析: 以手机为例[J]. 管理工程学报, 2017, 31(2): 47-54.
[11] (Zheng Lijuan, Wang Hongwei.Sentimental Polarity and Strength of Online Cellphone Reviews Based on Sentiment Ontology[J]. Journal of Industrial Engineering and Engineering Management, 2017, 31(2): 47-54.)
[12] 孙晓, 彭晓琪, 胡敏, 等. 基于多维扩展特征与深度学习的微博短文本情感分析[J]. 电子与信息学报, 2017, 39(9): 2048-2055.
[12] (Sun Xiao, Peng Xiaoqi, Hu Min, et al.Extended Multi-modality Features and Deep Learning Based Microblog Short Text Sentiment Analysis[J]. Journal of Electronics & Information Technology, 2017, 39(9): 2048-2055.)
[13] 李阳辉, 谢明, 易阳. 基于降噪自动编码器及其改进模型的微博情感分析[J]. 计算机应用研究, 2017, 34(2): 373-377.
[13] (Li Yanghui, Xie Ming, Yi Yang.Sentiment Analysis of Micro-blogging Based on DAE and Its Improved Model[J]. Application Research of Computers, 2017, 34(2): 373-377.)
[14] Kim Y.Convolutional Neural Networks for Sentence Classification[OL]. arXiv Preprint, arXiv: 1408. 5882.
[15] 刘小明, 张英, 郑秋生. 基于卷积神经网络模型的互联网短文本情感分类[J]. 计算机与现代化, 2017(4): 73-77.
[15] (Liu Xiaoming, Zhang Ying, Zheng Qiusheng.Sentiment Classification of Short Texts on Internet Based on Convolutional Neural Networks Model[J]. Computer and Modernization, 2017(4): 73-77.)
[16] Graves A, Schmidhuber J.Framewise Phoneme Classification with Bidirectional LSTM and Other Neural Network Architectures[J]. Neural Networks: The Official Journal of the International Neural Network Society, 2005, 18(5): 602-610.
doi: 10.1016/j.neunet.2005.06.042
[17] Chung J, Gulcehre C, Cho K, et al.Gated Feedback Recurrent Neural Networks [OL]. arXiv Preprint, arXiv: 1502. 02367.
[18] 王树恒, 吐尔根·依布拉音, 卡哈尔江·阿比的热西提, 等. 基于BLSTM的维吾尔语文本情感分析[J]. 计算机工程与设计, 2017, 38(10): 2879-2886.
[18] (Wang Shuheng, Turgun Ibrahim, Kahaerjiang Abiderexiti, et al.Sentiment Classification of Uyghur Text Based on BLSTM[J]. Computer Engineering and Design, 2017, 38(10): 2879-2886.)
[19] Zhu X, Sobhani P, Guo H.Long Short-Term Memory over Tree Structures[C]//Proceedings of International Conference on Machine Learning. 2015.
[20] 李雪莲, 段鸿, 许牧, 等. 基于门循环单元神经网络的中文分词法[J]. 厦门大学学报: 自然科学版, 2017, 56(2): 237-243.
[20] (Li Xuelian, Duan Hong, Xu Mu, et al.A Gated Recurrent Unit Neural Network for Chinese Word Segmentation[J]. Journal of Xiamen University: Natural Science, 2017, 56(2): 237-243.)
[21] 王博立, 史晓东, 苏劲松. 一种基于循环神经网络的古文断句方法[J]. 北京大学学报: 自然科学版, 2017, 53(2): 255-261.
[21] (Wang Boli, Shi Xiaodong, Su Jinsong.A Sentence Segmentation Method for Ancient Chinese Texts Based on Recurrent Neural Network[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2017, 53(2): 255-261.)
[22] 郑雄风, 丁立新, 万润泽. 基于用户和产品Attention机制的层次BGRU模型[J]. 计算机工程与应用, 2018, 54(11): 145-152.
[22] (Zheng Xiongfeng, Ding Lixin, Wan Runze.User and Product Attention Mechanism Based Hierarchical BGRU Model[J]. Computer Engineering and Applications, 2018, 54(11): 145-152.)
[23] Bahdanau D, Cho K, Bengio Y.Neural Machine Translation by Jointly Learning to Align and Translate[OL]. arXiv Preprint, arXiv: 1409. 0473.
[24] 余本功, 张连彬. 基于CP-CNN的中文短文本分类研究[J]. 计算机应用研究, 2018, 35(4): 1001-1004.
[24] (Yu Bengong, Zhang Lianbin.Chinese Short Text Classification Based on CP-CNN[J]. Application Research of Computers, 2018, 35(4): 1001-1004.)
[1] 周泽聿,王昊,赵梓博,李跃艳,张小琴. 融合关联信息的GCN文本分类模型构建及其应用研究*[J]. 数据分析与知识发现, 2021, 5(9): 31-41.
[2] 徐月梅, 王子厚, 吴子歆. 一种基于CNN-BiLSTM多特征融合的股票走势预测模型*[J]. 数据分析与知识发现, 2021, 5(7): 126-138.
[3] 赵丹宁,牟冬梅,白森. 基于深度学习的科技文献摘要结构要素自动抽取方法研究*[J]. 数据分析与知识发现, 2021, 5(7): 70-80.
[4] 黄名选,蒋曹清,卢守东. 基于词嵌入与扩展词交集的查询扩展*[J]. 数据分析与知识发现, 2021, 5(6): 115-125.
[5] 钟佳娃,刘巍,王思丽,杨恒. 文本情感分析方法及应用综述*[J]. 数据分析与知识发现, 2021, 5(6): 1-13.
[6] 张国标,李洁. 融合多模态内容语义一致性的社交媒体虚假新闻检测*[J]. 数据分析与知识发现, 2021, 5(5): 21-29.
[7] 刘彤,刘琛,倪维健. 多层次数据增强的半监督中文情感分析方法*[J]. 数据分析与知识发现, 2021, 5(5): 51-58.
[8] 马莹雪,甘明鑫,肖克峻. 融合标签和内容信息的矩阵分解推荐方法*[J]. 数据分析与知识发现, 2021, 5(5): 71-82.
[9] 王雨竹,谢珺,陈波,续欣莹. 基于跨模态上下文感知注意力的多模态情感分析 *[J]. 数据分析与知识发现, 2021, 5(4): 49-59.
[10] 常城扬,王晓东,张胜磊. 基于深度学习方法对特定群体推特的动态政治情感极性分析*[J]. 数据分析与知识发现, 2021, 5(3): 121-131.
[11] 冯勇,刘洋,徐红艳,王嵘冰,张永刚. 融合近邻评论的GRU商品推荐模型*[J]. 数据分析与知识发现, 2021, 5(3): 78-87.
[12] 胡昊天,吉晋锋,王东波,邓三鸿. 基于深度学习的食品安全事件实体一体化呈现平台构建*[J]. 数据分析与知识发现, 2021, 5(3): 12-24.
[13] 张琪,江川,纪有书,冯敏萱,李斌,许超,刘浏. 面向多领域先秦典籍的分词词性一体化自动标注模型构建*[J]. 数据分析与知识发现, 2021, 5(3): 2-11.
[14] 吕学强,罗艺雄,李家全,游新冬. 中文专利侵权检测研究综述*[J]. 数据分析与知识发现, 2021, 5(3): 60-68.
[15] 成彬,施水才,都云程,肖诗斌. 基于融合词性的BiLSTM-CRF的期刊关键词抽取方法[J]. 数据分析与知识发现, 2021, 5(3): 101-108.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址:北京市海淀区中关村北四环西路33号 邮编:100190
电话/传真:(010)82626611-6626,82624938
E-mail:jishu@mail.las.ac.cn