Sentiment analysis is a fine‐grained analysis task that aims to identify the sentiment polarity of a specified sentence.Existing methods in Chinese sentiment analysis tasks only consider sentiment features from a sin...Sentiment analysis is a fine‐grained analysis task that aims to identify the sentiment polarity of a specified sentence.Existing methods in Chinese sentiment analysis tasks only consider sentiment features from a single pole and scale and thus cannot fully exploit and utilise sentiment feature information,making their performance less than ideal.To resolve the problem,the authors propose a new method,GP‐FMLNet,that integrates both glyph and phonetic information and design a novel feature matrix learning process for phonetic features with which to model words that have the same pinyin information but different glyph information.Our method solves the problem of misspelling words influencing sentiment polarity prediction results.Specifically,the authors iteratively mine character,glyph,and pinyin features from the input comments sentences.Then,the authors use soft attention and matrix compound modules to model the phonetic features,which empowers their model to keep on zeroing in on the dynamic‐setting words in various positions and to dispense with the impacts of the deceptive‐setting ones.Ex-periments on six public datasets prove that the proposed model fully utilises the glyph and phonetic information and improves on the performance of existing Chinese senti-ment analysis algorithms.展开更多
字形信息与预训练模型(如:BERT)在中文命名实体识别任务上各有优势,它们可以被组合使用。现有的方法倾向于在预训练模型外部融合特征,这种做法需要额外设计字形特征网络与融合特征网络,没有充分利用预训练模型的结构工程(Architecture E...字形信息与预训练模型(如:BERT)在中文命名实体识别任务上各有优势,它们可以被组合使用。现有的方法倾向于在预训练模型外部融合特征,这种做法需要额外设计字形特征网络与融合特征网络,没有充分利用预训练模型的结构工程(Architecture Engineering)。为此,提出一种模型内融合算法,称为字形增强BERT(Glyph Enhanced BERT,GEBERT),通过字形适配器将字形信息注入BERT底层,使字形特征与字符特征在BERT内部融合。文中还展示了字体选取及GEBERT不同设置的消融研究。实验表明:GEBERT在MSRA数据集上达到平均F1值95.70,优于所有主流的模型外融合算法(Glyce,Glynn,FGN)获得最先进水平(State of the Art,SOTA)。在中文微博数据集上达到平均F1值70.68,仅次于FGN。展开更多
The contents of"Ba-Shu History of Science and Technology"take up a single volume of A General History of Ba-Shu Culture.This paper attempts to explore compilation modes of the history of local science and te...The contents of"Ba-Shu History of Science and Technology"take up a single volume of A General History of Ba-Shu Culture.This paper attempts to explore compilation modes of the history of local science and technology by analyzing the compilation of"Shu History of Science and Technology".There are three approaches available,i.e.a general history-based approach,a discipline-based approach and an integrated approach,also known as a discipline-specific approach based on general history.This paper considers the integrated approach to be the most feasible as it elaborates Ba-Shu culture as a sub-culture of Chinese culture,the connotations of Ba-Shu achievements in science and technology,the special significance of the history of Ba-Shu science and technology,as well as the latest progress in studying the history of Ba-Shu science and technology.This paper also puts forward a possible choice to better approach the compilation of the history of local science and technology,reveals areas worth improving,and indicates a proper direction to move towards.展开更多
First of all, the article identifies the highlights of the documentary The Romantic Shudao.It adopts a panoramic view of the broad Chinese cultural community and depicts the tangible Shudao(road to the Shu area) and t...First of all, the article identifies the highlights of the documentary The Romantic Shudao.It adopts a panoramic view of the broad Chinese cultural community and depicts the tangible Shudao(road to the Shu area) and the intangible "Romance" associated with it by combining the oral history and collective memory as perceived by common people.The vivid stories can easily reach a wide audience and can thus help to enhance people's confidence in the ever-prospering local Ba-Shu culture.Secondly, the article explains the ten major characteristics of the Ba-Shu culture by looking at the wonderful ecological landscape, the mystical culture and literature, and the magic spiritual aesthetics of the area in an international and modern context.It also studies the Ba-Shu cultural community and the origins, historical development, and main features of the city clusters within the Chengdu-Chongqing economic circle and analyzes the historical and practical values of building a cultural corridor connecting Chengdu and Chongqing.展开更多
基金Science and Technology Innovation 2030‐“New Generation Artificial Intelligence”major project,Grant/Award Number:2020AAA0108703。
文摘Sentiment analysis is a fine‐grained analysis task that aims to identify the sentiment polarity of a specified sentence.Existing methods in Chinese sentiment analysis tasks only consider sentiment features from a single pole and scale and thus cannot fully exploit and utilise sentiment feature information,making their performance less than ideal.To resolve the problem,the authors propose a new method,GP‐FMLNet,that integrates both glyph and phonetic information and design a novel feature matrix learning process for phonetic features with which to model words that have the same pinyin information but different glyph information.Our method solves the problem of misspelling words influencing sentiment polarity prediction results.Specifically,the authors iteratively mine character,glyph,and pinyin features from the input comments sentences.Then,the authors use soft attention and matrix compound modules to model the phonetic features,which empowers their model to keep on zeroing in on the dynamic‐setting words in various positions and to dispense with the impacts of the deceptive‐setting ones.Ex-periments on six public datasets prove that the proposed model fully utilises the glyph and phonetic information and improves on the performance of existing Chinese senti-ment analysis algorithms.
文摘字形信息与预训练模型(如:BERT)在中文命名实体识别任务上各有优势,它们可以被组合使用。现有的方法倾向于在预训练模型外部融合特征,这种做法需要额外设计字形特征网络与融合特征网络,没有充分利用预训练模型的结构工程(Architecture Engineering)。为此,提出一种模型内融合算法,称为字形增强BERT(Glyph Enhanced BERT,GEBERT),通过字形适配器将字形信息注入BERT底层,使字形特征与字符特征在BERT内部融合。文中还展示了字体选取及GEBERT不同设置的消融研究。实验表明:GEBERT在MSRA数据集上达到平均F1值95.70,优于所有主流的模型外融合算法(Glyce,Glynn,FGN)获得最先进水平(State of the Art,SOTA)。在中文微博数据集上达到平均F1值70.68,仅次于FGN。
文摘The contents of"Ba-Shu History of Science and Technology"take up a single volume of A General History of Ba-Shu Culture.This paper attempts to explore compilation modes of the history of local science and technology by analyzing the compilation of"Shu History of Science and Technology".There are three approaches available,i.e.a general history-based approach,a discipline-based approach and an integrated approach,also known as a discipline-specific approach based on general history.This paper considers the integrated approach to be the most feasible as it elaborates Ba-Shu culture as a sub-culture of Chinese culture,the connotations of Ba-Shu achievements in science and technology,the special significance of the history of Ba-Shu science and technology,as well as the latest progress in studying the history of Ba-Shu science and technology.This paper also puts forward a possible choice to better approach the compilation of the history of local science and technology,reveals areas worth improving,and indicates a proper direction to move towards.
文摘First of all, the article identifies the highlights of the documentary The Romantic Shudao.It adopts a panoramic view of the broad Chinese cultural community and depicts the tangible Shudao(road to the Shu area) and the intangible "Romance" associated with it by combining the oral history and collective memory as perceived by common people.The vivid stories can easily reach a wide audience and can thus help to enhance people's confidence in the ever-prospering local Ba-Shu culture.Secondly, the article explains the ten major characteristics of the Ba-Shu culture by looking at the wonderful ecological landscape, the mystical culture and literature, and the magic spiritual aesthetics of the area in an international and modern context.It also studies the Ba-Shu cultural community and the origins, historical development, and main features of the city clusters within the Chengdu-Chongqing economic circle and analyzes the historical and practical values of building a cultural corridor connecting Chengdu and Chongqing.
