A new cobalt(Ⅱ)-radical complex:[Co(im4-py)_(2)(PNB)_(2)](im4-py=2-(4'-pyridyl)-4,4,5,5-tetramethylimidazole-1-oxyl,HPNB=p-nitrobenzoic acid)has been synthesized and characterized by X-ray diffraction analysis,el...A new cobalt(Ⅱ)-radical complex:[Co(im4-py)_(2)(PNB)_(2)](im4-py=2-(4'-pyridyl)-4,4,5,5-tetramethylimidazole-1-oxyl,HPNB=p-nitrobenzoic acid)has been synthesized and characterized by X-ray diffraction analysis,elemental analysis,IR,and magnetic properties.X-ray diffraction analysis shows that the complex exists as mononuclear molecules and Co(Ⅱ)ion is four-coordinated with two radicals and two PNB-ligands.The magnetic susceptibility study indicates the complex exhibits weak ferromagnetic interactions between cobalt(Ⅱ)and im4-py radical.The magnetic property is explained by the magnetic and structure exchange mechanism.CCDC:976028.展开更多
In order to interpret pathologic mechanism of free radicals and thyroid hormone metabolism in cattle iodine and selenium deficiency, 20 heads of yellow cattle were selected from NiuJia town, Wu Chang City, Heilongjian...In order to interpret pathologic mechanism of free radicals and thyroid hormone metabolism in cattle iodine and selenium deficiency, 20 heads of yellow cattle were selected from NiuJia town, Wu Chang City, Heilongjiang Province, China, and were randomly devided into 4 groups with 5 for each. ① supplemented with 0.7 mg·kg -1 iodine(potassium iodine), ② supplemented with 0.2 mg·kg -1 selenium (sodium selenite), ③ supplemented with 0.7 mg·kg -1 iodine(potassium Iodine) plus 0.2 mg·kg -1 selenium (sodium selenite) per day for 30 days, respectively. ④control group. The whole blood glutathione peroxidase (GSH-px) and catalase (CAT) activities, free radicals (FR) concentration, erythrocyte superoxide dismutase (SOD) activity and molonaldehyde (MDA) concentration, the serum triiodothyronine (T 3)、thyroxine (T 4) and thyrotropin (TSH) were determined on the day of supplementation day-0 and day-30, respectively. It was showed that average iodine concentration in drinking water and diet were 3.82 μg·L -1 and 0.285mg·kg -1 , respectively, Diet selenium was 0.0498mg·kg -1 , Serum protein bound iodine(PBI) was 7.02 μg·100 mL, Blood selenium was 0.14 mg·L -1 , the schoolchildren′s goiter was 21.8%. It indicated that iodine and selenium were deficient in the investigated area. Whole blood GSH-px and CAT activities and serum T 3 concentration were significantly higher (P< 0.01 ), FR concentration and serum TSH were significantly lower(P<0.01) in the first three groups than that of the control, T 4 content in the first group was higher(P<0.05), T 4 was also higher (P>0.05) in the second group. and lower in the third group. The SOD and MDA in erythrocyte were not changed during the experimental period, The results also showed that GSH-px and CAT activities were increased, and FR decreased oberviously in the third group more than the other two groups, In addition, Thyroid hormone metabolism was more coincided with the physiologic status in the third group. the iodine and the selenium played an important role in the pathologic process of free radical metabolic disorder. selenium not only had the function of antioxidation by derectly scavenging free radicals, but also affected through GSH-px and CAT activities. iodine deficiency results in the Goiter, selenium deficiency aggravated iodine deficiency, Iodine and the selenium were dependent and restrained each other in the course of free radicals and thyroid hormone metabolism with a synergistic state.展开更多
中文命名实体识别(NER)任务旨在抽取非结构化文本中包含的实体并给它们分配预定义的实体类别。针对大多数中文NER方法在上下文信息缺乏时的语义学习不足问题,提出一种层次融合多元知识的NER框架——HTLR(Chinese NER method based on Hi...中文命名实体识别(NER)任务旨在抽取非结构化文本中包含的实体并给它们分配预定义的实体类别。针对大多数中文NER方法在上下文信息缺乏时的语义学习不足问题,提出一种层次融合多元知识的NER框架——HTLR(Chinese NER method based on Hierarchical Transformer fusing Lexicon and Radical),以通过分层次融合的多元知识来帮助模型学习更丰富、全面的上下文信息和语义信息。首先,通过发布的中文词汇表和词汇向量表识别语料中包含的潜在词汇并把它们向量化,同时通过优化后的位置编码建模词汇和相关字符的语义关系,以学习中文的词汇知识;其次,通过汉典网发布的基于汉字字形的编码将语料转换为相应的编码序列以代表字形信息,并提出RFECNN(Radical Feature Extraction-Convolutional Neural Network)模型来提取字形知识;最后,提出Hierarchical Transformer模型,其中由低层模块分别学习字符和词汇以及字符和字形的语义关系,并由高层模块进一步融合字符、词汇、字形等多元知识,从而帮助模型学习语义更丰富的字符表征。在Weibo、Resume、MSRA和OntoNotes4.0公开数据集进行了实验,与主流方法NFLAT(Non-Flat-LAttice Transformer for Chinese named entity recognition)的对比结果表明,所提方法的F1值在4个数据集上分别提升了9.43、0.75、1.76和6.45个百分点,达到最优水平。可见,多元语义知识、层次化融合、RFE-CNN结构和Hierarchical Transformer结构对学习丰富的语义知识及提高模型性能是有效的。展开更多
文摘A new cobalt(Ⅱ)-radical complex:[Co(im4-py)_(2)(PNB)_(2)](im4-py=2-(4'-pyridyl)-4,4,5,5-tetramethylimidazole-1-oxyl,HPNB=p-nitrobenzoic acid)has been synthesized and characterized by X-ray diffraction analysis,elemental analysis,IR,and magnetic properties.X-ray diffraction analysis shows that the complex exists as mononuclear molecules and Co(Ⅱ)ion is four-coordinated with two radicals and two PNB-ligands.The magnetic susceptibility study indicates the complex exhibits weak ferromagnetic interactions between cobalt(Ⅱ)and im4-py radical.The magnetic property is explained by the magnetic and structure exchange mechanism.CCDC:976028.
文摘In order to interpret pathologic mechanism of free radicals and thyroid hormone metabolism in cattle iodine and selenium deficiency, 20 heads of yellow cattle were selected from NiuJia town, Wu Chang City, Heilongjiang Province, China, and were randomly devided into 4 groups with 5 for each. ① supplemented with 0.7 mg·kg -1 iodine(potassium iodine), ② supplemented with 0.2 mg·kg -1 selenium (sodium selenite), ③ supplemented with 0.7 mg·kg -1 iodine(potassium Iodine) plus 0.2 mg·kg -1 selenium (sodium selenite) per day for 30 days, respectively. ④control group. The whole blood glutathione peroxidase (GSH-px) and catalase (CAT) activities, free radicals (FR) concentration, erythrocyte superoxide dismutase (SOD) activity and molonaldehyde (MDA) concentration, the serum triiodothyronine (T 3)、thyroxine (T 4) and thyrotropin (TSH) were determined on the day of supplementation day-0 and day-30, respectively. It was showed that average iodine concentration in drinking water and diet were 3.82 μg·L -1 and 0.285mg·kg -1 , respectively, Diet selenium was 0.0498mg·kg -1 , Serum protein bound iodine(PBI) was 7.02 μg·100 mL, Blood selenium was 0.14 mg·L -1 , the schoolchildren′s goiter was 21.8%. It indicated that iodine and selenium were deficient in the investigated area. Whole blood GSH-px and CAT activities and serum T 3 concentration were significantly higher (P< 0.01 ), FR concentration and serum TSH were significantly lower(P<0.01) in the first three groups than that of the control, T 4 content in the first group was higher(P<0.05), T 4 was also higher (P>0.05) in the second group. and lower in the third group. The SOD and MDA in erythrocyte were not changed during the experimental period, The results also showed that GSH-px and CAT activities were increased, and FR decreased oberviously in the third group more than the other two groups, In addition, Thyroid hormone metabolism was more coincided with the physiologic status in the third group. the iodine and the selenium played an important role in the pathologic process of free radical metabolic disorder. selenium not only had the function of antioxidation by derectly scavenging free radicals, but also affected through GSH-px and CAT activities. iodine deficiency results in the Goiter, selenium deficiency aggravated iodine deficiency, Iodine and the selenium were dependent and restrained each other in the course of free radicals and thyroid hormone metabolism with a synergistic state.
文摘中文命名实体识别(NER)任务旨在抽取非结构化文本中包含的实体并给它们分配预定义的实体类别。针对大多数中文NER方法在上下文信息缺乏时的语义学习不足问题,提出一种层次融合多元知识的NER框架——HTLR(Chinese NER method based on Hierarchical Transformer fusing Lexicon and Radical),以通过分层次融合的多元知识来帮助模型学习更丰富、全面的上下文信息和语义信息。首先,通过发布的中文词汇表和词汇向量表识别语料中包含的潜在词汇并把它们向量化,同时通过优化后的位置编码建模词汇和相关字符的语义关系,以学习中文的词汇知识;其次,通过汉典网发布的基于汉字字形的编码将语料转换为相应的编码序列以代表字形信息,并提出RFECNN(Radical Feature Extraction-Convolutional Neural Network)模型来提取字形知识;最后,提出Hierarchical Transformer模型,其中由低层模块分别学习字符和词汇以及字符和字形的语义关系,并由高层模块进一步融合字符、词汇、字形等多元知识,从而帮助模型学习语义更丰富的字符表征。在Weibo、Resume、MSRA和OntoNotes4.0公开数据集进行了实验,与主流方法NFLAT(Non-Flat-LAttice Transformer for Chinese named entity recognition)的对比结果表明,所提方法的F1值在4个数据集上分别提升了9.43、0.75、1.76和6.45个百分点,达到最优水平。可见,多元语义知识、层次化融合、RFE-CNN结构和Hierarchical Transformer结构对学习丰富的语义知识及提高模型性能是有效的。
