This paper is the analysis of the main body, the object and their relationship concerning research of traditional Chinese medicine. The result proved that the main body of research of traditional Chinese medicine is r...This paper is the analysis of the main body, the object and their relationship concerning research of traditional Chinese medicine. The result proved that the main body of research of traditional Chinese medicine is researchers who possessed the fundamentals of traditional Chinese medicine. It is the objects of research of traditional Chinese medicine that origin of a class of traditional Chinese medical literature consisting of herbals which focus on descriptions of individual drugs; the relationship of the technological process of preparing Chinese medicine, as by parching, roasting, baking, steaming, soaking, and simmering, etc.;and technology of separating, extracting and purifying for composition; traditional Chinese medicine dynamics of one of the 12 two-hour periods into which the day was formerly divided before the introduction of western chronology; compatibility of medicines in active compositions; patent, effectiveness for a given period of time and constantly changing of traditional Chinese medicine.The relationship between the main body and the objects was that the former depends on the latter ones and the main body has subjective initiative as fully as possible.展开更多
以专利引证网络为载体,从知识基因稳定性、遗传性以及变异性等基本特征出发,提出一种基于subject-action-object三元组的知识基因提取方法.应用连接度算法分析专利引证关系,挖掘引证专利和被引专利之间继承和发展的知识流,建立知识进化...以专利引证网络为载体,从知识基因稳定性、遗传性以及变异性等基本特征出发,提出一种基于subject-action-object三元组的知识基因提取方法.应用连接度算法分析专利引证关系,挖掘引证专利和被引专利之间继承和发展的知识流,建立知识进化轨迹;利用文本语法分析技术,从专利权利要求书中提取subject-action-object三元组;基于语义词库WordNet进行语义加工,计算语义相似度,合并同义的subject-action-object三元组,绘制知识基因图谱.从美国专利数据库中采集了5 073项1975—1999年授权的数据挖掘领域的相关专利,分析了专利的地区分布情况和年度分布情况.从NBER(National Bureau of Economic Research)的专利数据集中查询得到专利引证关系,利用网络分析软件Pajek构建专利引证网络,作为实验数据样本,对所提出的知识基因提取方法进行验证.实验结果表明:所提取的subject-action-object三元组具备了知识基因稳定性、遗传性和变异性等特征,可以作为知识基因的一种表现形式.展开更多
针对基于关系边界框提取的谓词特征粒度相对较粗的问题,提出区域敏感的场景图生成(region-sensitive scene graph generation,RS-SGG)方法。谓词特征提取模块将关系边界框分为4个区域,基于自注意力机制抑制关系边界框中与关系分类无关...针对基于关系边界框提取的谓词特征粒度相对较粗的问题,提出区域敏感的场景图生成(region-sensitive scene graph generation,RS-SGG)方法。谓词特征提取模块将关系边界框分为4个区域,基于自注意力机制抑制关系边界框中与关系分类无关的背景区域。关系特征解码器在进行关系预测时不仅考虑了物体对的视觉特征和语义特征,也考虑了物体对的位置特征。在视觉基因组(visual genome,VG)数据集上分别计算了RS-SGG方法针对场景图生成、场景图分类和谓词分类3个子任务的图约束召回率和无图约束召回率,并与主流的场景图生成方法进行了比较。实验结果表明,RS-SGG的图约束召回率和无图约束召回率均优于主流方法。此外,可视化实验结果也进一步证明了所提出方法的有效性。展开更多
文摘This paper is the analysis of the main body, the object and their relationship concerning research of traditional Chinese medicine. The result proved that the main body of research of traditional Chinese medicine is researchers who possessed the fundamentals of traditional Chinese medicine. It is the objects of research of traditional Chinese medicine that origin of a class of traditional Chinese medical literature consisting of herbals which focus on descriptions of individual drugs; the relationship of the technological process of preparing Chinese medicine, as by parching, roasting, baking, steaming, soaking, and simmering, etc.;and technology of separating, extracting and purifying for composition; traditional Chinese medicine dynamics of one of the 12 two-hour periods into which the day was formerly divided before the introduction of western chronology; compatibility of medicines in active compositions; patent, effectiveness for a given period of time and constantly changing of traditional Chinese medicine.The relationship between the main body and the objects was that the former depends on the latter ones and the main body has subjective initiative as fully as possible.
文摘以专利引证网络为载体,从知识基因稳定性、遗传性以及变异性等基本特征出发,提出一种基于subject-action-object三元组的知识基因提取方法.应用连接度算法分析专利引证关系,挖掘引证专利和被引专利之间继承和发展的知识流,建立知识进化轨迹;利用文本语法分析技术,从专利权利要求书中提取subject-action-object三元组;基于语义词库WordNet进行语义加工,计算语义相似度,合并同义的subject-action-object三元组,绘制知识基因图谱.从美国专利数据库中采集了5 073项1975—1999年授权的数据挖掘领域的相关专利,分析了专利的地区分布情况和年度分布情况.从NBER(National Bureau of Economic Research)的专利数据集中查询得到专利引证关系,利用网络分析软件Pajek构建专利引证网络,作为实验数据样本,对所提出的知识基因提取方法进行验证.实验结果表明:所提取的subject-action-object三元组具备了知识基因稳定性、遗传性和变异性等特征,可以作为知识基因的一种表现形式.
文摘针对基于关系边界框提取的谓词特征粒度相对较粗的问题,提出区域敏感的场景图生成(region-sensitive scene graph generation,RS-SGG)方法。谓词特征提取模块将关系边界框分为4个区域,基于自注意力机制抑制关系边界框中与关系分类无关的背景区域。关系特征解码器在进行关系预测时不仅考虑了物体对的视觉特征和语义特征,也考虑了物体对的位置特征。在视觉基因组(visual genome,VG)数据集上分别计算了RS-SGG方法针对场景图生成、场景图分类和谓词分类3个子任务的图约束召回率和无图约束召回率,并与主流的场景图生成方法进行了比较。实验结果表明,RS-SGG的图约束召回率和无图约束召回率均优于主流方法。此外,可视化实验结果也进一步证明了所提出方法的有效性。
