This paper introduced a novel high performance algorithm and VLSI architectures for achieving bit plane coding （BPC） in word level sequential and parallel mode. The proposed BPC algorithm adopts the techniques of co...This paper introduced a novel high performance algorithm and VLSI architectures for achieving bit plane coding （BPC） in word level sequential and parallel mode. The proposed BPC algorithm adopts the techniques of coding pass prediction and parallel ＆ pipeline to reduce the number of accessing memory and to increase the ability of concurrently processing of the system, where all the coefficient bits of a code block could be coded by only one scan. A new parallel bit plane architecture （PA） was proposed to achieve word-level sequential coding. Moreover, an efficient high-speed architecture （HA） was presented to achieve multi-word parallel coding. Compared to the state of the art, the proposed PA could reduce the hardware cost more efficiently, though the throughput retains one coefficient coded per clock. While the proposed HA could perform coding for 4 coefficients belonging to a stripe column at one intra-clock cycle, so that coding for an NxN code-block could be completed in approximate N2/4 intra-clock cycles. Theoretical analysis and experimental results demonstrate that the proposed designs have high throughput rate with good performance in terms of speedup to cost, which can be good alternatives for low power applications.展开更多
Language, reading, and reading-related tasks were administered to 148 children from lower primary （Standard 1, 2, and 3） in Kenyan schools. The aim was to investigate the literacy developmental trends across the yea...Language, reading, and reading-related tasks were administered to 148 children from lower primary （Standard 1, 2, and 3） in Kenyan schools. The aim was to investigate the literacy developmental trends across the years and establish which of the two variables, word identification skills or target language oral proficiency influenced reading comprehension performance. The results indicated that word identification skills independently influenced reading comprehension performance in both standard 2 and standard 3 classes, but target language oral skills did not. Further, the analysis of invented spelling task performance revealed evidence of transfer of alphabetic coding skills of first language to English orthography. The spelling errors reflected letter-sound patterns not found in English orthography. Analysis of the miscues in the spelling task revealed that children rely on first language phonological processes to spell target language unfamiliar words as long the languages have a common alphabetic system.展开更多
事件抽取旨在从非结构化文本中检测事件类型并抽取事件要素。现有方法在处理文档级文本时仍存在局限性。这是因为文档级文本可能由多个事件组成,并且构成某一事件的事件要素通常分散在不同句子中。为应对上述挑战,提出了一种文档级事件...事件抽取旨在从非结构化文本中检测事件类型并抽取事件要素。现有方法在处理文档级文本时仍存在局限性。这是因为文档级文本可能由多个事件组成,并且构成某一事件的事件要素通常分散在不同句子中。为应对上述挑战,提出了一种文档级事件抽取反向推理模型(reverse inference model for document-level event extraction,RIDEE)。基于无触发词的设计,将文档级事件抽取转化为候选事件要素抽取和事件触发推理两个子任务,并行式抽取事件要素并检测事件类型。此外,设计了一种用于存储历史事件的事件依赖池,使得模型在处理多事件文本时可以充分利用事件之间的依赖关系。公开数据集上的实验结果表明,与现有事件抽取模型相比,RIDEE在进行文档级事件抽取时具有更优的性能。展开更多
"视觉词袋"(Bag of Visual Words,BOV)算法是一种有效的基于语义特征表达的物体识别算法。针对传统BOV模型存在的不足,综合利用SAR图像的灰度和纹理特征,提出基于感兴趣目标(Target of Interest,TOI)的"视觉词袋"..."视觉词袋"(Bag of Visual Words,BOV)算法是一种有效的基于语义特征表达的物体识别算法。针对传统BOV模型存在的不足,综合利用SAR图像的灰度和纹理特征,提出基于感兴趣目标(Target of Interest,TOI)的"视觉词袋"算法。首先,对训练图像进行TOI选取,用灰度共生矩阵模型提取TOI的纹理特征,再结合灰度特征,组成多维特征向量集,以簇内相似度最高、数据分布密度最大为准则,生成"视觉词袋"。其次,对测试图像,依据已生成的"视觉词袋",采用支持向量机(Support Vector Machine,SVM)分类器,实现SAR图像感兴趣目标的有效分类。实验结果表明,与传统的"视觉词袋"构建算法相比,该算法在分类正确率提高的同时,能够在训练图像较少的情况下达到良好的分类效果。展开更多
方面级别情感分析是对给定句子的不同方面进行情感极性预测。在长句子中有许多无关词会干扰情感预测的结果,且这些无关词与中心词存在一定的距离。对此,提出以下解决方案:设计上下文迭代学习网络。提出上下文注意力模块(context attenti...方面级别情感分析是对给定句子的不同方面进行情感极性预测。在长句子中有许多无关词会干扰情感预测的结果,且这些无关词与中心词存在一定的距离。对此,提出以下解决方案:设计上下文迭代学习网络。提出上下文注意力模块(context attention modules,CAM),模块采用上下文动态特征掩码(context features dynamic mask,CDM)遮掩距离中心词较远的词,上下文动态特征权重(context features dynamic weighted,CDW)减小较远词的权重。文中设计的CAM经过多层迭代,增强了方面词在上下文部分的特征提取。在公共的基准数据集上进行一系列的试验比对,试验结果证明文中提出的方法是有效的。展开更多
基金the Natural Science Foundation of Hubei Province (Grant No. 2006ABA370)Civil Research Project of State Defense (Grant No. C1120061304)+1 种基金National Natural Science Foundation of China (Grant No. 60572048)National High Technology Research and Develop-ment of China (863 Program) (Grant No. 2004AA119010-6)
文摘This paper introduced a novel high performance algorithm and VLSI architectures for achieving bit plane coding （BPC） in word level sequential and parallel mode. The proposed BPC algorithm adopts the techniques of coding pass prediction and parallel ＆ pipeline to reduce the number of accessing memory and to increase the ability of concurrently processing of the system, where all the coefficient bits of a code block could be coded by only one scan. A new parallel bit plane architecture （PA） was proposed to achieve word-level sequential coding. Moreover, an efficient high-speed architecture （HA） was presented to achieve multi-word parallel coding. Compared to the state of the art, the proposed PA could reduce the hardware cost more efficiently, though the throughput retains one coefficient coded per clock. While the proposed HA could perform coding for 4 coefficients belonging to a stripe column at one intra-clock cycle, so that coding for an NxN code-block could be completed in approximate N2/4 intra-clock cycles. Theoretical analysis and experimental results demonstrate that the proposed designs have high throughput rate with good performance in terms of speedup to cost, which can be good alternatives for low power applications.
文摘Language, reading, and reading-related tasks were administered to 148 children from lower primary （Standard 1, 2, and 3） in Kenyan schools. The aim was to investigate the literacy developmental trends across the years and establish which of the two variables, word identification skills or target language oral proficiency influenced reading comprehension performance. The results indicated that word identification skills independently influenced reading comprehension performance in both standard 2 and standard 3 classes, but target language oral skills did not. Further, the analysis of invented spelling task performance revealed evidence of transfer of alphabetic coding skills of first language to English orthography. The spelling errors reflected letter-sound patterns not found in English orthography. Analysis of the miscues in the spelling task revealed that children rely on first language phonological processes to spell target language unfamiliar words as long the languages have a common alphabetic system.
文摘事件抽取旨在从非结构化文本中检测事件类型并抽取事件要素。现有方法在处理文档级文本时仍存在局限性。这是因为文档级文本可能由多个事件组成,并且构成某一事件的事件要素通常分散在不同句子中。为应对上述挑战,提出了一种文档级事件抽取反向推理模型(reverse inference model for document-level event extraction,RIDEE)。基于无触发词的设计,将文档级事件抽取转化为候选事件要素抽取和事件触发推理两个子任务,并行式抽取事件要素并检测事件类型。此外,设计了一种用于存储历史事件的事件依赖池,使得模型在处理多事件文本时可以充分利用事件之间的依赖关系。公开数据集上的实验结果表明,与现有事件抽取模型相比,RIDEE在进行文档级事件抽取时具有更优的性能。
文摘"视觉词袋"(Bag of Visual Words,BOV)算法是一种有效的基于语义特征表达的物体识别算法。针对传统BOV模型存在的不足,综合利用SAR图像的灰度和纹理特征,提出基于感兴趣目标(Target of Interest,TOI)的"视觉词袋"算法。首先,对训练图像进行TOI选取,用灰度共生矩阵模型提取TOI的纹理特征,再结合灰度特征,组成多维特征向量集,以簇内相似度最高、数据分布密度最大为准则,生成"视觉词袋"。其次,对测试图像,依据已生成的"视觉词袋",采用支持向量机(Support Vector Machine,SVM)分类器,实现SAR图像感兴趣目标的有效分类。实验结果表明,与传统的"视觉词袋"构建算法相比,该算法在分类正确率提高的同时,能够在训练图像较少的情况下达到良好的分类效果。
文摘方面级别情感分析是对给定句子的不同方面进行情感极性预测。在长句子中有许多无关词会干扰情感预测的结果,且这些无关词与中心词存在一定的距离。对此,提出以下解决方案:设计上下文迭代学习网络。提出上下文注意力模块(context attention modules,CAM),模块采用上下文动态特征掩码(context features dynamic mask,CDM)遮掩距离中心词较远的词,上下文动态特征权重(context features dynamic weighted,CDW)减小较远词的权重。文中设计的CAM经过多层迭代,增强了方面词在上下文部分的特征提取。在公共的基准数据集上进行一系列的试验比对,试验结果证明文中提出的方法是有效的。
