The Internet of Things becomes Internet of Everything when in the process of communication machine-to-machine also intelligent forms of communication between human and machine are involved. Cities can be viewed as a m...The Internet of Things becomes Internet of Everything when in the process of communication machine-to-machine also intelligent forms of communication between human and machine are involved. Cities can be viewed as a microcosm of this interconnected system where ICT and emerging technologies can be enabling factors to transform cities in Smart Cities. Cities can take great advantage by using information intelligence to achieve important public-policy goals and, in particular, by enabling network communication channels between citizens and public administrators in order to provide information and online services in real time through platform systems rather than by means of humans, using Artificial Intelligence and Natural Language Processing techniques. This work was the first step of a wider project aimed at providing a Spell Checking Web Service API for Smart City communication platforms able to automatically select, among the large availability of open source spell checking tools, the most suitable tool based on the semantic structure of the specific textual data. The system should manage an enhanced Italian Vocabulary Database, specifically implemented to support all the tools of the system. The goal of the present work was to test, through an experimental research, the feasibility of the entire project by implementing a Spell Checking Prototype System designed to manage two selected spell checking tools. Results showed that the Spell Checking Prototype System significantly improves performances by allowing the user to select the most suitable tool for the specific semantic structure of the text. The system also enables to manage the list of exceptions, which continuously enhance the Italian Vocabulary Database. The experimentation proved scientific evidence of the validity of the project aimed at implementing a Spell Checking Web Service API in order to improve the quality of natural language data to be stored or processed in Smart City NCeSDP systems, through the use of existing spell checking tools.展开更多
针对汉语初学者在学习汉语时不可避免地会出现拼写错误的问题,提出一个汉语拼写检查模型,用以检测和纠正句子中的拼写错误。模型结合了汉字的视觉特征和语音特征,由一个检查网络和一个纠正网络构成。基于双向长短期记忆网络(bidirection...针对汉语初学者在学习汉语时不可避免地会出现拼写错误的问题,提出一个汉语拼写检查模型,用以检测和纠正句子中的拼写错误。模型结合了汉字的视觉特征和语音特征,由一个检查网络和一个纠正网络构成。基于双向长短期记忆网络(bidirectional long short-term memory network,BiLSTM)和条件随机场(conditional random field,CRF)构成的检测网络用于检测句子中的错误字;基于BERT(bidirectional encoder representations from transformer)模型的纠正网络用以结合全局上下文信息对检测到的错误字进行纠正。最后,在CLP-2014,SIGHAN-2013和SIGHAN-2015数据集上进行实验,结果表明:相比现有的方法,提出的模型在错字检测和错字纠正上的效果均得到了提升;相比利用视觉特征,汉字的语音特征能更好地提升错字的检测效果。展开更多
文摘The Internet of Things becomes Internet of Everything when in the process of communication machine-to-machine also intelligent forms of communication between human and machine are involved. Cities can be viewed as a microcosm of this interconnected system where ICT and emerging technologies can be enabling factors to transform cities in Smart Cities. Cities can take great advantage by using information intelligence to achieve important public-policy goals and, in particular, by enabling network communication channels between citizens and public administrators in order to provide information and online services in real time through platform systems rather than by means of humans, using Artificial Intelligence and Natural Language Processing techniques. This work was the first step of a wider project aimed at providing a Spell Checking Web Service API for Smart City communication platforms able to automatically select, among the large availability of open source spell checking tools, the most suitable tool based on the semantic structure of the specific textual data. The system should manage an enhanced Italian Vocabulary Database, specifically implemented to support all the tools of the system. The goal of the present work was to test, through an experimental research, the feasibility of the entire project by implementing a Spell Checking Prototype System designed to manage two selected spell checking tools. Results showed that the Spell Checking Prototype System significantly improves performances by allowing the user to select the most suitable tool for the specific semantic structure of the text. The system also enables to manage the list of exceptions, which continuously enhance the Italian Vocabulary Database. The experimentation proved scientific evidence of the validity of the project aimed at implementing a Spell Checking Web Service API in order to improve the quality of natural language data to be stored or processed in Smart City NCeSDP systems, through the use of existing spell checking tools.
文摘针对汉语初学者在学习汉语时不可避免地会出现拼写错误的问题,提出一个汉语拼写检查模型,用以检测和纠正句子中的拼写错误。模型结合了汉字的视觉特征和语音特征,由一个检查网络和一个纠正网络构成。基于双向长短期记忆网络(bidirectional long short-term memory network,BiLSTM)和条件随机场(conditional random field,CRF)构成的检测网络用于检测句子中的错误字;基于BERT(bidirectional encoder representations from transformer)模型的纠正网络用以结合全局上下文信息对检测到的错误字进行纠正。最后,在CLP-2014,SIGHAN-2013和SIGHAN-2015数据集上进行实验,结果表明:相比现有的方法,提出的模型在错字检测和错字纠正上的效果均得到了提升;相比利用视觉特征,汉字的语音特征能更好地提升错字的检测效果。
