Respiratory syncytial virus(RSV) is a leading cause of acute lower respiratory tract infections. Qingfei oral liquid(QFOL), a traditional Chinese medicine, is widely used in clinical treatment for RSV-induced pneumoni...Respiratory syncytial virus(RSV) is a leading cause of acute lower respiratory tract infections. Qingfei oral liquid(QFOL), a traditional Chinese medicine, is widely used in clinical treatment for RSV-induced pneumonia. The present study was designed to reveal the potential targets and mechanism of action for QFOL by exploring its influence on the host cellular network following RSV infection. We investigated the serum proteomic changes and potential biomarkers in an RSV-infected mouse pneumonia model treated with QFOL. Eighteen BALB/c mice were randomly divided into three groups: RSV pneumonia model group(M), QFOL-treated group(Q) and the control group(C). Serum proteomes were analyzed and compared using a label-free quantitative LC-MS/MS approach. A total of 172 protein groups, 1009 proteins, and 1073 unique peptides were successfully identified. 51 differentially expressed proteins(DEPs) were identified(15 DEPs when M/C and 43 DEPs when Q/M; 7 DEPs in common). Classification and interaction network showed that these proteins participated in various biological processes including immune response, blood coagulation, complement activation, and so forth. Particularly, fibrinopeptide B(FpB) and heparin cofactor Ⅱ(HCII) were evaluated as important nodes in the interaction network, which was closely involved in coagulation and inflammation. Further, the Fp B level was increased in Group M but decreased in Group Q, while the HCII level exhibited the opposite trend. These findings not only indicated FpB and HCII as potential biomarkers and targets of QFOL in the treatment of RSV pneumonia, but also suggested a regulatory role of QFOL in the RSV-induced disturbance of coagulation and inflammation-coagulation interactions.展开更多
The concept and advantage of reconfigurable technology is introduced. A kind of processor architecture of re configurable macro processor (RMP) model based on FPGA array and DSP is put forward and has been implemented...The concept and advantage of reconfigurable technology is introduced. A kind of processor architecture of re configurable macro processor (RMP) model based on FPGA array and DSP is put forward and has been implemented. Two image algorithms are developed: template-based automatic target recognition and zone labeling. One is estimating for motion direction in the infrared image background, another is line picking-up algorithm based on image zone labeling and phase grouping technique. It is a kind of 'hardware' function that can be called by the DSP in high-level algorithm. It is also a kind of hardware algorithm of the DSP. The results of experiments show the reconfigurable computing technology based on RMP is an ideal accelerating means to deal with the high-speed image processing tasks. High real time performance is obtained in our two applications on RMP.展开更多
基于序列到序列的情绪标签生成模型采用循环神经网络建模情绪相关性,是一种有效的多标签情绪分类方法。然而,现有序列生成模型仅通过隐状态隐式地学习标签相关性,难以有效捕捉细粒度情绪间的强相关性。针对这一问题,该文提出了一种基于...基于序列到序列的情绪标签生成模型采用循环神经网络建模情绪相关性,是一种有效的多标签情绪分类方法。然而,现有序列生成模型仅通过隐状态隐式地学习标签相关性,难以有效捕捉细粒度情绪间的强相关性。针对这一问题,该文提出了一种基于预训练序列生成模型的多标签情绪分类方法(Multi-Label Emotion Classification Based on Pre-trained BART,EmoBART)。EmoBART模型采用预训练生成式语言模型BART为情绪标签序列生成的网络骨架,使用相关性网络(CorNet)显式地学习情绪相关性。EmoBART模型由编码模块、解码模块和相关性网络模块组成。编码模块提取文本语义信息、解码模块采用生成式标签链构建情绪标签序列、相关性网络模块显式建模情绪相关性,并对情绪标签进行预测。在细粒度情绪数据集上的对比实验结果表明,EmoBART模型具有比已有模型更优的情绪识别性能。展开更多
层次主题模型可以挖掘文档中的隐含主题,建模主题间的层次结构关系,为数据治理、信息检索、内容分类和知识管理等应用提供技术支持.文中提出基于思维链和语义解耦的层次化主题模型.首先,建立基于思维链的层次主题生成模块,设计层次化主...层次主题模型可以挖掘文档中的隐含主题,建模主题间的层次结构关系,为数据治理、信息检索、内容分类和知识管理等应用提供技术支持.文中提出基于思维链和语义解耦的层次化主题模型.首先,建立基于思维链的层次主题生成模块,设计层次化主题生成思维链,指导大语言模型(Large Language Model,LLM)生成初步的主题层次结构.然后,引入基于LLM的主题相似判别机制,生成精炼的主题,并利用样例指导LLM实现主题合并,提升生成主题的质量.最后,建立基于传输规划和语义解耦的主题层次优化模块,将初始层次主题结构作为下游建模的主题先验,构建主题关键词、文档主题分布和主题距离,并将主题层次关系建模为最优运输问题,结合上下层主题关键词进行父子主题解耦,优化主题层次结构.在NeurIPS、ACL、20 Newsgroups等涵盖新闻与学术论文的多个标准公开数据集上的实验表明,文中模型在主题质量指标和层次化指标上均取得较优值.展开更多
基金supported by the National Natural Science Foundation of China(No.81574025)the Open Project Program of Jiangsu Key Laboratory of Pediatric Respiratory Disease,Nanjing University of Chinese Medicine(No.JKLPRD201410)
文摘Respiratory syncytial virus(RSV) is a leading cause of acute lower respiratory tract infections. Qingfei oral liquid(QFOL), a traditional Chinese medicine, is widely used in clinical treatment for RSV-induced pneumonia. The present study was designed to reveal the potential targets and mechanism of action for QFOL by exploring its influence on the host cellular network following RSV infection. We investigated the serum proteomic changes and potential biomarkers in an RSV-infected mouse pneumonia model treated with QFOL. Eighteen BALB/c mice were randomly divided into three groups: RSV pneumonia model group(M), QFOL-treated group(Q) and the control group(C). Serum proteomes were analyzed and compared using a label-free quantitative LC-MS/MS approach. A total of 172 protein groups, 1009 proteins, and 1073 unique peptides were successfully identified. 51 differentially expressed proteins(DEPs) were identified(15 DEPs when M/C and 43 DEPs when Q/M; 7 DEPs in common). Classification and interaction network showed that these proteins participated in various biological processes including immune response, blood coagulation, complement activation, and so forth. Particularly, fibrinopeptide B(FpB) and heparin cofactor Ⅱ(HCII) were evaluated as important nodes in the interaction network, which was closely involved in coagulation and inflammation. Further, the Fp B level was increased in Group M but decreased in Group Q, while the HCII level exhibited the opposite trend. These findings not only indicated FpB and HCII as potential biomarkers and targets of QFOL in the treatment of RSV pneumonia, but also suggested a regulatory role of QFOL in the RSV-induced disturbance of coagulation and inflammation-coagulation interactions.
文摘The concept and advantage of reconfigurable technology is introduced. A kind of processor architecture of re configurable macro processor (RMP) model based on FPGA array and DSP is put forward and has been implemented. Two image algorithms are developed: template-based automatic target recognition and zone labeling. One is estimating for motion direction in the infrared image background, another is line picking-up algorithm based on image zone labeling and phase grouping technique. It is a kind of 'hardware' function that can be called by the DSP in high-level algorithm. It is also a kind of hardware algorithm of the DSP. The results of experiments show the reconfigurable computing technology based on RMP is an ideal accelerating means to deal with the high-speed image processing tasks. High real time performance is obtained in our two applications on RMP.
文摘基于序列到序列的情绪标签生成模型采用循环神经网络建模情绪相关性,是一种有效的多标签情绪分类方法。然而,现有序列生成模型仅通过隐状态隐式地学习标签相关性,难以有效捕捉细粒度情绪间的强相关性。针对这一问题,该文提出了一种基于预训练序列生成模型的多标签情绪分类方法(Multi-Label Emotion Classification Based on Pre-trained BART,EmoBART)。EmoBART模型采用预训练生成式语言模型BART为情绪标签序列生成的网络骨架,使用相关性网络(CorNet)显式地学习情绪相关性。EmoBART模型由编码模块、解码模块和相关性网络模块组成。编码模块提取文本语义信息、解码模块采用生成式标签链构建情绪标签序列、相关性网络模块显式建模情绪相关性,并对情绪标签进行预测。在细粒度情绪数据集上的对比实验结果表明,EmoBART模型具有比已有模型更优的情绪识别性能。
文摘层次主题模型可以挖掘文档中的隐含主题,建模主题间的层次结构关系,为数据治理、信息检索、内容分类和知识管理等应用提供技术支持.文中提出基于思维链和语义解耦的层次化主题模型.首先,建立基于思维链的层次主题生成模块,设计层次化主题生成思维链,指导大语言模型(Large Language Model,LLM)生成初步的主题层次结构.然后,引入基于LLM的主题相似判别机制,生成精炼的主题,并利用样例指导LLM实现主题合并,提升生成主题的质量.最后,建立基于传输规划和语义解耦的主题层次优化模块,将初始层次主题结构作为下游建模的主题先验,构建主题关键词、文档主题分布和主题距离,并将主题层次关系建模为最优运输问题,结合上下层主题关键词进行父子主题解耦,优化主题层次结构.在NeurIPS、ACL、20 Newsgroups等涵盖新闻与学术论文的多个标准公开数据集上的实验表明,文中模型在主题质量指标和层次化指标上均取得较优值.
