Article

Land evaluation based on Boosting decision tree ensembles

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Traditional land evaluation methods are more liable to be restricted with human elements. It is significant to investigate more scientific and reasonable land evaluation methods for land use and planning. Because decision tree has the characteristics of high accuracy and intelligibility, and especially C5.0 uses Boosting technique to improve classification accuracy, land evaluation using C5.0 with the function to build Boosting decision tree ensembles was conducted. Moreover, C5.0 was used to evaluate the land resources of Guangdong Province, and the results using decision tree with and without Boosting were analyzed and compared. The experimental results demonstrate that C5.0 can be used to evaluate land resources accurately, and the land evaluation accuracy of decision tree without Boosting was lower than that of the decision tree ensemble with Boosting.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Based on the existing and traditional methods of assessing the overall productivity function of soil or land, Mueller et al (2010) advocated a straightforward indicator-based soil functional evaluation and classification system supplementing the WRB soil classifications. A number of methods have been developed to assess land quality, such as fuzzy evaluation (Burrongh, 1989;Groenemans et al., 1997;Kaufmann et al., 2009;Yang et al., 2008), boosting decision tree (Xue et al., 2008a), SFAM neural network (Xue et al., 2008b), and emergy analysis (Martin et al., 2006;Zhang et al., 2009). ...
Article
Full-text available
Agricultural land is a complex system that combines natural ecology and social economy, and the health of agricultural land directly influences regional economic development and national food security. Emergy analysis is an effective method for value analysis of ecosystems, assessment of the health of ecosystems, which is based on principles of systems ecology and energy. And fuzzy logic is a generalization tools to operate with vaguely defined parameters or concepts, which has been extensively applied in the quality evaluation of soil and land. The health of agricultural land as an aggregate that takes into account the quality and productivity of land as well as the soil environment. This paper presents an integrated evaluation model based on emergy analysis and fuzzy logic and using the geographic information system (GIS) to evaluate the health of agricultural land. An evaluation system is developed, which includes quality, productivity and the soil environment. A major grain-producing region in the North China Plain, namely Luancheng County, was selected for the case study. Field survey, household inquiry, sampling, and experimental analysis were conducted in the study area. The health of agricultural land expressed as the emergy index ranged from 0.471 to 0.763; as the fuzzy index, from 0.603 to 0.878; and as a composite index, from 0.284 to 0.666. Based on the composite index, the agricultural land was divided into four classes, namely in perfect health (which accounted for 7.6% of the agricultural land), in normal health (54.3%), in sub-health (32.9%), and in poor health (5.2%). Overall, agricultural land in Luancheng County is healthy, although 89.00 km2 of the farmland – mainly in Douyu, Nangao, and Liulintun – is in sub-health and 14.22 km2 in Douyu is in poor health because of poor productivity and unfavorable soil environment.
... They found that DT model was not dependent on experience and could reach a good result. Xue et al. used boosting DT ensembles to evaluate land resources in Guangdong [3] . Their results demonstrated that DT with boosting technique could get higher evaluation accuracy. ...
Conference Paper
Support vector machines (SVMs), which are based on statistical learning theory, have recently got considerable potential in data mining by their good ability of generalization. Land grade evaluation, which provides information for land planning and decision-making, is a process of evaluating land quality for a particular use. It can be referred to as multiclass classification problem. As a result of SVMs' binary nature, they can not be directly applied to land grading process. By integrating SVMs with a binary tree, this paper applied a binary tree based SVMs (BTSVMs) approach into land grade evaluation. Arable land in Heping County, Guangdong, was chosen as study area and BTSVMs model was then applied to the data. In addition to BTSVMs, the same data were classified using decision tree (DT) and artificial neural network (ANN). Compared with DT and ANN, results showed that BTSVMs had better classification accuracy. While decreasing the size of training data, the accuracy of each approach dropped down positively with BTSVMs relatively more accurate than others. In general, BTSVMs is potentially feasible in the application of land grade evaluation with its good performance.
Article
Brownfield assessment is a crucial precondition for sustainable urban renewal. In particular, brownfield assessment usually involve a large amount of data and information relevant to urban or land dynamic. However, obtaining these data and information in fine scale has been especially challenging via traditional datasets (like surveyed or statistics datasets). This paper proposes a method for integrating crowdsourced datasets and traditional datasets to measure dynamic information of land parcels and applies a presence and background data machine learning (PBL) model to assess the redevelopment suitability in mass. The work focuses on the study area of Shenzhen, a high-density urban context with an adequate amount of urban renewal cases and sufficient crowdsourced data. Results indicated that assessments based on the combination of crowdsourced datasets and traditional datasets were accurate and reliable for the residential and commercial land parcels. Furthermore, mass assessments at a fine spatial-temporal scale are viable if introduce crowdsourced datasets due to the cost of obtaining the crowdsourced data is much lower, and the spatial-temporal scale is much finer than investigated or surveyed data.
Article
Landslides are one of the most serious geological disasters in the world and happen quite frequently in the Three Gorges. Landslide prediction is a very important measure of landslide prevention and cure in the Three Gorges. Traditional methods lack in sufficiently mining the various complex information from a landslide system. They often need much manual intervention and possess poor intelligence and accuracy. An intelligent method proposed in this paper for landslide prediction based on an object-oriented method and knowledge driving is hopeful to solve the above problem. The method adopted Landsat ETM+ images, 1:50,000 geological map and 1:10,000 relief map in the Three Gorges as the data origins. It firstly produced the key factors influencing landslide development and used multi-resolution segmentation algorithm to segment the image objects based on the key landslide factors of engineering rock group, reservoir water fluctuation, slope structure and slope level. Secondly, the method chose some sample objects and adopted the decision tree algorithm C5.0 to mine the landslide forecast criteria according to the factor values of each sample object. Finally, under knowledge driving the method classified the image objects and realized landslide susceptibility analysis and intelligent prediction in the Three Gorges. The method proposed in this paper is object-oriented. Results of a real-world example show that: (1) the object-oriented method possesses much more compact knowledge representation, higher efficiency, more continuous classifying result and higher prediction accuracy compared with the pixel-oriented method; (2) it possesses the overall accuracy of 87.64% and kappa coefficient of 0.8305 and is more accurate than the other seven methods (such as the pixel-oriented methods of Parallelpiped, Minimum Distance, Maximum Likelihood, Mahalanobis Distance, K-means and Isodata and the object-oriented method of Nearest Neighbor); (3) about 46.97% landslides lie in the high susceptibility region, 24.24% landslides lie in the moderate susceptibility region, 27.27% landslides lie in the low susceptibility region and 1.52% landslides lie in the very low susceptibility region. Therefore the method can effectively realize landslide susceptibility analysis and provides a new idea for landslide intelligent and accurate prediction.
Article
Full-text available
Part Two of this article deals with problems related to land valuation. For that purpose, a site of 6.89 ares located in Švitrigailos Street XX, Vilnius, was selected as the research object and appraised using three approaches: the sales comparison approach, the MAMVA method and the mass valuation approach. Peculiarities of these methods in land valuation were reviewed and the recommendations were provided for land valuation for taxation purposes applicable in a new Model for Lithuanian Real Property Taxation System. Santruka Antroje šio straipsnio dalyje nagrinejamos problemos, susijusios su žemes sklypu vertinimu. Tuo tikslu buvo parinktas konkretus 6,89 aru ploto žemes sklypas, esantis Švitrigailos g. XX, Vilniuje, ir ivertintas trimis būdais: lyginamosios vertes, MAMVA metodais ir masinio vertinimo būdu. Apžvelgti šiu metodu taikymo ypatumai vertinant žemes sklypus, pateiktos rekomendacijos, kaip nustatyti mokestines žemes sklypu vertes ir pritaikyti siūlomam naujam Lietuvos nekilnojamojo turto apmokestinimo modeliui. First Published Online: 09 Jun 2011
Article
Full-text available
Proteins of the C-terminal binding protein (CtBP) family, CtBP1 and CtBP2 are closely related transcriptional regulators that are coded by two different gene loci in the vertebrate genomes. They perform redundant and unique functions during animal development. CtBP proteins mediate their transcriptional function through interaction with various DNA-binding repressors that contain PLDLS-like motifs and chromatin modifying enzymes, such as class I histone deacetylases (HDAC) that do not contain such motifs. The N-terminal region of CtBP1/2 forms a hydrophobic cleft and is involved in interaction with both PLDLS-containing factors and non-PLDLS factors. CtBP proteins function as dimers to mediate transcriptional repression and dimerization is modulated by specific binding to NAD/NADH. In this study, we have investigated the role of dimerization of CtBP2 in recruitment of PLDLS-motif cofactors and non-PLDLS cofactors. Our results indicate that mutations in CtBP2 that interfere with dimerization abolish CtBP2 interaction with most cellular factors, except the PLDLS-motif factor zinc-finger E-box binding homeobox (ZEB) and the non-PLDLS factor HDAC2. Unlike most PLDLS-containing CtBP-binding proteins, ZEB contains three PLDLS-like motifs and all three contribute to the interaction with the CtBP2 monomer. Despite the ability to interact with ZEB and HDAC, the CtBP2 monomer fails to mediate ZEB-dependent transcriptional repression. The lack of repression activity of the CtBP2 monomer is correlated with the competition between ZEB and HDAC for interaction with the CtBP2 monomer. These results suggest a competition between the canonical PLDLS-motif factors such as E1A and non-PLDLS factor HDAC for interaction with CtBP. They also indicate that the affinity for the CtBP monomer may be determined by the number as well as amino acid sequence compositions of the PLDLS-like motifs. Our results are consistent with a model that the CtBP2 dimer may interact with a PLDLS-containing repressor through one monomer and recruit HDAC and other chromatin modifying enzymes through the second monomer in the CtBP2 dimer.
ResearchGate has not been able to resolve any references for this publication.