Decision tree model

About: Decision tree model is a research topic. Over the lifetime, 2256 publications have been published within this topic receiving 38142 citations.

Vegetation Classification in Hyperspectral Image with CART Decision Tree

Abstract: To improve the accuracy of vegetation classification,the influences of training sample sizes and terrain should be considered when extracting vegetation information from the hyperspectral image.Taking the Changbai Mountain as the study area,this paper built a decision tree model based on CART(Classification and Regression Tree) algorithm to classify vegetation in hyperspectral image.In order to reduce the influence of the mixed pixels,using PPI(Pixel Purity Index) to extract pure pixel as the training samples.CART decision tree was built based on these classification feature variables,such as vegetation index,texture,terrain and so on,the tree was applied on vegetation classification and the result was compared with the maximum likelihood classification.The result showed that CART decision tree method combined with spectrum,texture and terrain,had a better effect of classification.

Memory Complexity Analysis on Numerical Programs

Abstract: Memory systems become more and more complicated with so many efforts on bridging the large speed gap between processor and main memory. It is now difficult to gain high performance from a processor or a large parallel processing systems without considering the specific memory system features. Thus it becomes not enough just to use the time and space complexity to explain why different forms of one algorithm explore so different performance on one same platform. The complexity of memory systems must be incorporated into the analysis of algorithms. In 1996, Sun Jiachang first presented a new concept on memory complexity. It is believed that the complexity of an algorithm should consist of its computational complexity and memory complexity, among them, computational complexity consists of time complexity and space complexity, which are the basic characteristics of algorithm; while memory complexity is a varying characteristic, which will change with different implementations of the same algorithm and different platforms. The purpose of algorithmic optimization is to reduce the memory complexity, while the reduction of computational complexity needs new algorithmic research activity. In this paper, we try to analyze the different implementations of an algorithm and to predict the relative performance differences among them through combining the memory complexity analysis and the data movement/floating point operation ratio analysis. Further analysis with remote communication in parallel processing will be our future work.

Lower bounds on quantum query complexity for read-once decision trees with parity nodes

Abstract: We introduce a complexity measure for decision trees called the soft rank, which measures how well-balanced a given tree is. The soft rank is a somehow relaxed variant of the rank. Among all decision trees of depth d, the complete binary decision tree (the most balanced tree) has maximum soft rank d, the decision list (the most unbalanced tree) has minimum soft rank √d, and any other trees have soft rank between √d and d. We show that, for any decision tree T in some class G of decision trees which includes all read-once decision trees, the soft rank of T is a lower bound on the quantum query complexity of the Boolean function that T represents. This implies that for any Boolean function f that is represented by a decision tree in G, the deterministic query complexity of f is only quadratically larger than the quantum query complexity of f.

C4.5-Based Classification Rules Mining of High-Rise Building SFIO

Abstract: Firstly, we show that the C4.5 algorithm inherits all the advantages of the ID3 algorithm, at the same time, it overcomes the defect of ID3 algorithm such as can not deal directly with continuous attributes, tend to choose more attributes when using information gain attributes to confirm the test values; we introduced the basic thought, problem solving process, the theoretical basis and classification rules extraction methods of C4.5 algorithms. Then, application of the high-rise building as the background, we used C4.5 algorithm to establish decision tree model of structural form intelligent optimization (SFIO). We extract the classification rules which could be distinguish structural form such as high-rise building structural function, height, fortification intensity etc. It provides new ways and scientific basis for the structural form optimal selection. Theory and Practice shows that C4.5 algorithm can effective mine form optimization data from the cases. This paper provides a new approach and methodology on the full use of abundant cases to address SFIO of knowledge acquisition "bottleneck".

Handheld terminal traffic identification method and system based on machine learning

Abstract: The invention discloses a handheld terminal traffic identification method and a system based on machine learning The method comprises the following steps: 1, UA keyword matching is carried out on to-be-identified traffic, the to-be-identified traffic is directly identified as the handheld device traffic or the non handheld device traffic in the case of matching, and a second step is carried out in the case of not matching; 2, based on a C45 decision tree algorithm and traffic attributes, the information gain rate of each traffic attribute is calculated and a decision tree model is built, and the unmatched to-be-identified traffic is identified as the handheld device traffic or the non handheld device traffic through the decision tree model When the method adopts the C45 decision tree algorithm for classifying the traffic which can not be identified by the UA method, comparison on the traffic attribute values only needs to be carried out, the processing is simple relatively, the processing time is shortened obviously, and the handheld terminal identification accuracy and the non handheld terminal identification accuracy can be greatly improved