1 Introduction Information technology has been playing an ever-increasing role in geoscience.Sphisicated database platforms are essential for geological data storage,analysis and exchange of Big Data(Feblowitz,2013;Zh...1 Introduction Information technology has been playing an ever-increasing role in geoscience.Sphisicated database platforms are essential for geological data storage,analysis and exchange of Big Data(Feblowitz,2013;Zhang et al.,2016;Teng et al.,2016;Tian and Li,2018).The United States has built an information-sharing platform for state-owned scientific data as a national strategy.展开更多
Time is an essential reference system for recording objects,events,and processes in the field of geosciences.There are currently various time references,such as solar calendar,geological time,and regional calendar,to ...Time is an essential reference system for recording objects,events,and processes in the field of geosciences.There are currently various time references,such as solar calendar,geological time,and regional calendar,to represent the knowledge in different domains and regions,which subsequently entails a time conversion process required to interpret temporal information under different time references.However,the current time conversion method is limited by the application scope of existing time ontologies(e.g.,“Jurassic”is a period in geological ontology,but a point value in calendar ontology)and the reliance on experience in conversion processes.These issues restrict accurate and efficient calculation of temporal information across different time references.To address these issues,this paper proposes a Unified Time Framework(UTF)in the geosciences knowledge system.According to a systematic time element parsing from massive time references,the proposed UTF designs an independent time root node to get rid of irrelevant nodes when accessing different time types and to adapt to the time expression of different geoscience disciplines.Furthermore,this UTF carries out several designs:to ensure the accuracy of time expressions by designing quantitative relationship definitions;to enable time calculations across different time elements by designing unified time nodes and structures,and to link to the required external ontologies by designing adequate interfaces.By comparing the time conversion methods,the experiment proves the UTF greatly supports accurate and efficient calculation of temporal information across different time references in SPARQL queries.Moreover,it shows a higher and more stable performance of temporal information queries than the time conversion method.With the advent of the Big Data era in the geosciences,the UTF can be used more widely to discover new geosciences knowledge across different time references.展开更多
In this paper, we review recent emerging theoretical and technological advances of artificial intelligence （AI） in the big data settings. We conclude that integrating data-driven machine learning with human knowled...In this paper, we review recent emerging theoretical and technological advances of artificial intelligence （AI） in the big data settings. We conclude that integrating data-driven machine learning with human knowledge （common priors or implicit intuitions） can effectively lead to explainable, robust, and general AI, as follows： from shallow computation to deep neural reasoning; from merely data-driven model to data-driven with structured logic rules models; from task-oriented （domain-specific） intelligence （adherence to explicit instructions） to artificial general intelligence in a general context （the capability to learn from experience）. Motivated by such endeavors, the next generation of AI, namely AI 2.0, is positioned to reinvent computing itself, to transform big data into structured knowledge, and to enable better decision-making for our society.展开更多
在文化旅游产业快速发展的背景下,如何利用人工智能技术提升文旅信息处理和服务质量成为研究热点。提出了一种基于知识库的文旅大语言模型(Cultural Tourism Large Language Model, CT-LLM),采用了一种融合知识库和大语言模型的技术路线...在文化旅游产业快速发展的背景下,如何利用人工智能技术提升文旅信息处理和服务质量成为研究热点。提出了一种基于知识库的文旅大语言模型(Cultural Tourism Large Language Model, CT-LLM),采用了一种融合知识库和大语言模型的技术路线,通过构建富含文旅行业专家级内容的知识库,实现了对文旅内容的深度理解和生成。该模型采用了一种融合知识库和大语言模型的技术路线,有效提升了大语言模型在文旅场景下的应用能力。实验结果表明,基于知识库的文旅大语言模型在旅游推荐、景点介绍、文旅问答等任务中表现出优异的性能,显著提高了文旅信息服务的质量和用户体验。本研究为文旅领域的大语言模型构建提供了新的方法,对于推动文旅产业智能化发展具有重要的理论和实践价值。展开更多
基金granted by the National Science&Technology Major Projects of China(Grant No.2016ZX05033).
文摘1 Introduction Information technology has been playing an ever-increasing role in geoscience.Sphisicated database platforms are essential for geological data storage,analysis and exchange of Big Data(Feblowitz,2013;Zhang et al.,2016;Teng et al.,2016;Tian and Li,2018).The United States has built an information-sharing platform for state-owned scientific data as a national strategy.
基金funded by the National Natural Science Foundation of China(Grant Nos.42050101 and 42101467)the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDA23100101).
文摘Time is an essential reference system for recording objects,events,and processes in the field of geosciences.There are currently various time references,such as solar calendar,geological time,and regional calendar,to represent the knowledge in different domains and regions,which subsequently entails a time conversion process required to interpret temporal information under different time references.However,the current time conversion method is limited by the application scope of existing time ontologies(e.g.,“Jurassic”is a period in geological ontology,but a point value in calendar ontology)and the reliance on experience in conversion processes.These issues restrict accurate and efficient calculation of temporal information across different time references.To address these issues,this paper proposes a Unified Time Framework(UTF)in the geosciences knowledge system.According to a systematic time element parsing from massive time references,the proposed UTF designs an independent time root node to get rid of irrelevant nodes when accessing different time types and to adapt to the time expression of different geoscience disciplines.Furthermore,this UTF carries out several designs:to ensure the accuracy of time expressions by designing quantitative relationship definitions;to enable time calculations across different time elements by designing unified time nodes and structures,and to link to the required external ontologies by designing adequate interfaces.By comparing the time conversion methods,the experiment proves the UTF greatly supports accurate and efficient calculation of temporal information across different time references in SPARQL queries.Moreover,it shows a higher and more stable performance of temporal information queries than the time conversion method.With the advent of the Big Data era in the geosciences,the UTF can be used more widely to discover new geosciences knowledge across different time references.
文摘In this paper, we review recent emerging theoretical and technological advances of artificial intelligence （AI） in the big data settings. We conclude that integrating data-driven machine learning with human knowledge （common priors or implicit intuitions） can effectively lead to explainable, robust, and general AI, as follows： from shallow computation to deep neural reasoning; from merely data-driven model to data-driven with structured logic rules models; from task-oriented （domain-specific） intelligence （adherence to explicit instructions） to artificial general intelligence in a general context （the capability to learn from experience）. Motivated by such endeavors, the next generation of AI, namely AI 2.0, is positioned to reinvent computing itself, to transform big data into structured knowledge, and to enable better decision-making for our society.
文摘在文化旅游产业快速发展的背景下,如何利用人工智能技术提升文旅信息处理和服务质量成为研究热点。提出了一种基于知识库的文旅大语言模型(Cultural Tourism Large Language Model, CT-LLM),采用了一种融合知识库和大语言模型的技术路线,通过构建富含文旅行业专家级内容的知识库,实现了对文旅内容的深度理解和生成。该模型采用了一种融合知识库和大语言模型的技术路线,有效提升了大语言模型在文旅场景下的应用能力。实验结果表明,基于知识库的文旅大语言模型在旅游推荐、景点介绍、文旅问答等任务中表现出优异的性能,显著提高了文旅信息服务的质量和用户体验。本研究为文旅领域的大语言模型构建提供了新的方法,对于推动文旅产业智能化发展具有重要的理论和实践价值。
