With the objectives to acquire the fundamental dat a of the territorial resource, understand the impacts of human activit ies on the land use and cover patterns and evaluate the potential of the future exploitation, a...With the objectives to acquire the fundamental dat a of the territorial resource, understand the impacts of human activit ies on the land use and cover patterns and evaluate the potential of the future exploitation, an intensive land cover classification with an accuracy of 93% has been completed for North Ningxia by remote sen sing technique based on the adoption of a combination method composed o f texture training, maximum likelihood classification and post-processing such as re-allocation and aggregation. This classification result was incorporated with the contemporaneous socio-economic and meteorological d ata for cross-sectional regression modelling to reveal the spatial dete rminants of the land cover patterns and understand the human-environmen tal relationships. A tentative evaluation on the potential of soil exp loitation in the near future was carried out in combination with our land use and cover change detection results aiming at supplying some useful references for the central and local governments in their sustainable l and use planning.展开更多
给出了一个词共现改进的向量空间模型(Word Co-Occurrence Mode Based On VSM,WCBVSM)与模拟退火交叉覆盖算法(Cross Cover Algorithm Based On Simulated Annealing Algorithm,SACA)相结合的文本分类新模型。传统的向量空间模型(VSM)...给出了一个词共现改进的向量空间模型(Word Co-Occurrence Mode Based On VSM,WCBVSM)与模拟退火交叉覆盖算法(Cross Cover Algorithm Based On Simulated Annealing Algorithm,SACA)相结合的文本分类新模型。传统的向量空间模型(VSM)采用词条作为文档的语义载体,没有考虑文本上下文词语之间的语义隐含信息,在词共现模型的启发下,提出WCBVSM,它通过统计文本中的词共现信息,加入VSM,以获得文档隐含的语义信息。针对交叉覆盖算法中识别精度与泛化能力之间的一对矛盾,结合模拟退火算法的思想,提出了SACA,改进了传统交叉覆盖在覆盖初始点选取时的随机性,并通过增加每个覆盖所包含的样本点来减少覆盖数,从而增强了覆盖的泛化能力。实验结果表明提出的文本分类新模型在加快识别速度的基础上,提高了分类的精度。展开更多
基金The Sino-Belgian co-operation project on Northwest China funded by the Federal Office for the Scientific, Technical and Cultural Affairs (OSTC) of the Belgium Government, No.BL/10/C15
文摘With the objectives to acquire the fundamental dat a of the territorial resource, understand the impacts of human activit ies on the land use and cover patterns and evaluate the potential of the future exploitation, an intensive land cover classification with an accuracy of 93% has been completed for North Ningxia by remote sen sing technique based on the adoption of a combination method composed o f texture training, maximum likelihood classification and post-processing such as re-allocation and aggregation. This classification result was incorporated with the contemporaneous socio-economic and meteorological d ata for cross-sectional regression modelling to reveal the spatial dete rminants of the land cover patterns and understand the human-environmen tal relationships. A tentative evaluation on the potential of soil exp loitation in the near future was carried out in combination with our land use and cover change detection results aiming at supplying some useful references for the central and local governments in their sustainable l and use planning.
文摘给出了一个词共现改进的向量空间模型(Word Co-Occurrence Mode Based On VSM,WCBVSM)与模拟退火交叉覆盖算法(Cross Cover Algorithm Based On Simulated Annealing Algorithm,SACA)相结合的文本分类新模型。传统的向量空间模型(VSM)采用词条作为文档的语义载体,没有考虑文本上下文词语之间的语义隐含信息,在词共现模型的启发下,提出WCBVSM,它通过统计文本中的词共现信息,加入VSM,以获得文档隐含的语义信息。针对交叉覆盖算法中识别精度与泛化能力之间的一对矛盾,结合模拟退火算法的思想,提出了SACA,改进了传统交叉覆盖在覆盖初始点选取时的随机性,并通过增加每个覆盖所包含的样本点来减少覆盖数,从而增强了覆盖的泛化能力。实验结果表明提出的文本分类新模型在加快识别速度的基础上,提高了分类的精度。
