In Dynamic Global Vegetation Models （DGVMs）, the establishment of woody vegetation refers to flowering, fertiliza- tion, seed production, germination, and the growth of tree seedlings. It determines not only the pop...In Dynamic Global Vegetation Models （DGVMs）, the establishment of woody vegetation refers to flowering, fertiliza- tion, seed production, germination, and the growth of tree seedlings. It determines not only the population densities but also other important ecosystem structural variables. In current DGVMs, establishments of woody plant functional types （PFTs） are assumed to be either the same in the same grid cell, or largely stochastic. We investigated the uncertainties in the competition of establishment among coexisting woody PFTs from three aspects： the dependence of PFT establishments on vegetation states; background establishment; and relative establishment potentials of different PFTs. Sensitivity experi- ments showed that the dependence of establishment rate on the fractional coverage of a PFT favored the dominant PFT by increasing its share in establishment. While a small background establishment rate had little impact on equilibrium states of the ecosystem, it did change the timescale required for the establishment of alien species in pre-existing forest due to their disadvantage in seed competition during the early stage of invasion. Meanwhile, establishment purely fiom background （the scheme commonly used in current DGVMs） led to inconsistent behavior in response to the change in PFT specification （e.g., number of PFTs and their specification）. Furthermore, the results also indicated that trade-off between irtdividual growth and reproduction/colonization has significant influences on the competition of establishment. Hence, further development of es- tablishment parameterization in DGVMs is essential in reducing the uncertainties in simulations of both ecosystem structures and successions.展开更多
Quickly extraction of building information technology is an important application in urban development planning, electronic information, national defense and others. This paper takes Landsat-8 multispectral and panchr...Quickly extraction of building information technology is an important application in urban development planning, electronic information, national defense and others. This paper takes Landsat-8 multispectral and panchromatic data as data source, using the local variance method to select the optimal segmentation scale, normalized difference vegetation index (NDVI) and the normalized building index (NDBI) and panchromatic brightness value of an object oriented classification rule extraction. The high vegetation coverage area of buildings, and through the spatial relationships and distinguishing feature of collections of buildings independent buildings and villages. The results showed that Google earth high resolution image analysis and accuracy evaluation. the results of the extraction based on the overall accuracy of village extraction was 83%, the accuracy of extraction of independent buildings was 70%, according to the L8 remote sensing data, object oriented classification method can quickly and accurately extract the high vegetation coverage area of the building.展开更多
From a data mining perspective, sequence classification is to build a classifier using frequent sequential patterns. However, mining for a complete set of sequential patterns on a large dataset can be extremely time-c...From a data mining perspective, sequence classification is to build a classifier using frequent sequential patterns. However, mining for a complete set of sequential patterns on a large dataset can be extremely time-consuming and the large number of patterns discovered also makes the pattern selection and classifier building very time-consuming. The fact is that, in sequence classification, it is much more important to discover discriminative patterns than a complete pattern set. In this paper, we propose a novel hierarchical algorithm to build sequential classifiers using discriminative sequential patterns. Firstly, we mine for the sequential patterns which axe the most strongly correlated to each target class. In this step, an aggressive strategy is employed to select a small set of sequential patterns. Secondly, pattern pruning and serial coverage test are done on the mined patterns. The patterns that pass the serial test are used to build the sub-classifier at the first level of the final classifier. And thirdly, the training samples that cannot be covered are fed back to the sequential pattern mining stage with updated parameters. This process continues until predefined interestingness measure thresholds are reached, or all samples axe covered. The patterns generated in each loop form the sub-classifier at each level of the final classifier. Within this framework, the searching space can be reduced dramatically while a good classification performance is achieved. The proposed algorithm is tested in a real-world business application for debt prevention in social security area. The novel sequence classification algorithm shows the effectiveness and efficiency for predicting debt occurrences based on customer activity sequence data.展开更多
基金supported by the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDA05110103)the State Key Project for Basic Research Program of China(Grant No.2010CB951801)the National High Technology Research and Development Program of China(863 Program)(Grant No.2009AA122105)
文摘In Dynamic Global Vegetation Models （DGVMs）, the establishment of woody vegetation refers to flowering, fertiliza- tion, seed production, germination, and the growth of tree seedlings. It determines not only the population densities but also other important ecosystem structural variables. In current DGVMs, establishments of woody plant functional types （PFTs） are assumed to be either the same in the same grid cell, or largely stochastic. We investigated the uncertainties in the competition of establishment among coexisting woody PFTs from three aspects： the dependence of PFT establishments on vegetation states; background establishment; and relative establishment potentials of different PFTs. Sensitivity experi- ments showed that the dependence of establishment rate on the fractional coverage of a PFT favored the dominant PFT by increasing its share in establishment. While a small background establishment rate had little impact on equilibrium states of the ecosystem, it did change the timescale required for the establishment of alien species in pre-existing forest due to their disadvantage in seed competition during the early stage of invasion. Meanwhile, establishment purely fiom background （the scheme commonly used in current DGVMs） led to inconsistent behavior in response to the change in PFT specification （e.g., number of PFTs and their specification）. Furthermore, the results also indicated that trade-off between irtdividual growth and reproduction/colonization has significant influences on the competition of establishment. Hence, further development of es- tablishment parameterization in DGVMs is essential in reducing the uncertainties in simulations of both ecosystem structures and successions.
文摘Quickly extraction of building information technology is an important application in urban development planning, electronic information, national defense and others. This paper takes Landsat-8 multispectral and panchromatic data as data source, using the local variance method to select the optimal segmentation scale, normalized difference vegetation index (NDVI) and the normalized building index (NDBI) and panchromatic brightness value of an object oriented classification rule extraction. The high vegetation coverage area of buildings, and through the spatial relationships and distinguishing feature of collections of buildings independent buildings and villages. The results showed that Google earth high resolution image analysis and accuracy evaluation. the results of the extraction based on the overall accuracy of village extraction was 83%, the accuracy of extraction of independent buildings was 70%, according to the L8 remote sensing data, object oriented classification method can quickly and accurately extract the high vegetation coverage area of the building.
基金supported by Australian Research Council Linkage Project under Grant No. LP0775041the Early Career Researcher Grant under Grant No. 2007002448 from University of Technology, Sydney, Australia
文摘From a data mining perspective, sequence classification is to build a classifier using frequent sequential patterns. However, mining for a complete set of sequential patterns on a large dataset can be extremely time-consuming and the large number of patterns discovered also makes the pattern selection and classifier building very time-consuming. The fact is that, in sequence classification, it is much more important to discover discriminative patterns than a complete pattern set. In this paper, we propose a novel hierarchical algorithm to build sequential classifiers using discriminative sequential patterns. Firstly, we mine for the sequential patterns which axe the most strongly correlated to each target class. In this step, an aggressive strategy is employed to select a small set of sequential patterns. Secondly, pattern pruning and serial coverage test are done on the mined patterns. The patterns that pass the serial test are used to build the sub-classifier at the first level of the final classifier. And thirdly, the training samples that cannot be covered are fed back to the sequential pattern mining stage with updated parameters. This process continues until predefined interestingness measure thresholds are reached, or all samples axe covered. The patterns generated in each loop form the sub-classifier at each level of the final classifier. Within this framework, the searching space can be reduced dramatically while a good classification performance is achieved. The proposed algorithm is tested in a real-world business application for debt prevention in social security area. The novel sequence classification algorithm shows the effectiveness and efficiency for predicting debt occurrences based on customer activity sequence data.
