In the last decade, technical advancements and faster Internet speeds have also led to an increasing number ofmobile devices and users. Thus, all contributors to society, whether young or old members, can use these mo...In the last decade, technical advancements and faster Internet speeds have also led to an increasing number ofmobile devices and users. Thus, all contributors to society, whether young or old members, can use these mobileapps. The use of these apps eases our daily lives, and all customers who need any type of service can accessit easily, comfortably, and efficiently through mobile apps. Particularly, Saudi Arabia greatly depends on digitalservices to assist people and visitors. Such mobile devices are used in organizing daily work schedules and services,particularly during two large occasions, Umrah and Hajj. However, pilgrims encounter mobile app issues such asslowness, conflict, unreliability, or user-unfriendliness. Pilgrims comment on these issues on mobile app platformsthrough reviews of their experiences with these digital services. Scholars have made several attempts to solve suchmobile issues by reporting bugs or non-functional requirements by utilizing user comments.However, solving suchissues is a great challenge, and the issues still exist. Therefore, this study aims to propose a hybrid deep learningmodel to classify and predict mobile app software issues encountered by millions of pilgrims during the Hajj andUmrah periods from the user perspective. Firstly, a dataset was constructed using user-generated comments fromrelevant mobile apps using natural language processing methods, including information extraction, the annotationprocess, and pre-processing steps, considering a multi-class classification problem. Then, several experimentswere conducted using common machine learning classifiers, Artificial Neural Networks (ANN), Long Short-TermMemory (LSTM), and Convolutional Neural Network Long Short-Term Memory (CNN-LSTM) architectures, toexamine the performance of the proposed model. Results show 96% in F1-score and accuracy, and the proposedmodel outperformed the mentioned models.展开更多
As the evolution of mobile technology, mobile devices have become an essential tool in people＇s daily life. Moreover, with the rapid growth of Internet and mobile networks, people can easily access various services p...As the evolution of mobile technology, mobile devices have become an essential tool in people＇s daily life. Moreover, with the rapid growth of Internet and mobile networks, people can easily access various services provided by mobile platforms. Many services can be executed on the mobile devices with various mobile applications launched to mobile platforms. People can choose what they like to install in their mobile devices and hence make their life more convenient, entertaining, and productive. However, there are too many mobile applications for users to choose. The goal of this research is to propose a methodology which can recommend top-N lists for mobile applications. A comment correlation matrix is proposed. Furthermore, a recommendation algorithm for mobile applications based on user comments and key attributes is built. With the proposed method, it outperforms Google play and is closer to user real feelings.展开更多
【目的】针对方面情感分类输入类别在不同领域之间差异较大,汽车用户评论文本语义信息不全,语义特征难以提取等问题,提出基于双通道输入的并行双向编码表征(bidirectional encoder representation from transformers,BERT)双向长短期记...【目的】针对方面情感分类输入类别在不同领域之间差异较大,汽车用户评论文本语义信息不全,语义特征难以提取等问题,提出基于双通道输入的并行双向编码表征(bidirectional encoder representation from transformers,BERT)双向长短期记忆多头自注意力模型的方面情感分类方法。【方法】首先采用了方面情感和方面抽取的双重标签进行标注;其次通过并行的方面抽取和方面情感分类任务通道,分别使用BERT、双向长短期记忆网络(bidirectional long and short-term memory networks,Bi-LSTM)及多头注意力机制(multihead self-attention,MHSA)提取更深层次的语义信息及近距离和远距离特征信息;最后采用条件随机场(conditional random field,CRF)分类器和Softmax分类器进行分类。【结果】在相关的汽车用户评论文本数据集和多语言混合数据集上,本研究提出的模型相较于主流的方面情感分类方法,具有同步抽取方面词和判断情感极性的能力,且有效提高了方面词抽取和方面情感分类的准确率和F_(1)值。【结论】本研究提出的模型更有利于汽车销售者分析用户评论,同时对识别用户评论文本的情感极性的研究也有一定的参考价值。展开更多
文摘In the last decade, technical advancements and faster Internet speeds have also led to an increasing number ofmobile devices and users. Thus, all contributors to society, whether young or old members, can use these mobileapps. The use of these apps eases our daily lives, and all customers who need any type of service can accessit easily, comfortably, and efficiently through mobile apps. Particularly, Saudi Arabia greatly depends on digitalservices to assist people and visitors. Such mobile devices are used in organizing daily work schedules and services,particularly during two large occasions, Umrah and Hajj. However, pilgrims encounter mobile app issues such asslowness, conflict, unreliability, or user-unfriendliness. Pilgrims comment on these issues on mobile app platformsthrough reviews of their experiences with these digital services. Scholars have made several attempts to solve suchmobile issues by reporting bugs or non-functional requirements by utilizing user comments.However, solving suchissues is a great challenge, and the issues still exist. Therefore, this study aims to propose a hybrid deep learningmodel to classify and predict mobile app software issues encountered by millions of pilgrims during the Hajj andUmrah periods from the user perspective. Firstly, a dataset was constructed using user-generated comments fromrelevant mobile apps using natural language processing methods, including information extraction, the annotationprocess, and pre-processing steps, considering a multi-class classification problem. Then, several experimentswere conducted using common machine learning classifiers, Artificial Neural Networks (ANN), Long Short-TermMemory (LSTM), and Convolutional Neural Network Long Short-Term Memory (CNN-LSTM) architectures, toexamine the performance of the proposed model. Results show 96% in F1-score and accuracy, and the proposedmodel outperformed the mentioned models.
文摘As the evolution of mobile technology, mobile devices have become an essential tool in people＇s daily life. Moreover, with the rapid growth of Internet and mobile networks, people can easily access various services provided by mobile platforms. Many services can be executed on the mobile devices with various mobile applications launched to mobile platforms. People can choose what they like to install in their mobile devices and hence make their life more convenient, entertaining, and productive. However, there are too many mobile applications for users to choose. The goal of this research is to propose a methodology which can recommend top-N lists for mobile applications. A comment correlation matrix is proposed. Furthermore, a recommendation algorithm for mobile applications based on user comments and key attributes is built. With the proposed method, it outperforms Google play and is closer to user real feelings.
文摘【目的】针对方面情感分类输入类别在不同领域之间差异较大,汽车用户评论文本语义信息不全,语义特征难以提取等问题,提出基于双通道输入的并行双向编码表征(bidirectional encoder representation from transformers,BERT)双向长短期记忆多头自注意力模型的方面情感分类方法。【方法】首先采用了方面情感和方面抽取的双重标签进行标注;其次通过并行的方面抽取和方面情感分类任务通道,分别使用BERT、双向长短期记忆网络(bidirectional long and short-term memory networks,Bi-LSTM)及多头注意力机制(multihead self-attention,MHSA)提取更深层次的语义信息及近距离和远距离特征信息;最后采用条件随机场(conditional random field,CRF)分类器和Softmax分类器进行分类。【结果】在相关的汽车用户评论文本数据集和多语言混合数据集上,本研究提出的模型相较于主流的方面情感分类方法,具有同步抽取方面词和判断情感极性的能力,且有效提高了方面词抽取和方面情感分类的准确率和F_(1)值。【结论】本研究提出的模型更有利于汽车销售者分析用户评论,同时对识别用户评论文本的情感极性的研究也有一定的参考价值。
