Topic
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.
Papers published on a yearly basis
Papers
More filters
22 Oct 2019
TL;DR: This work presents a design pattern for incremental tree data processing as gradually constructing on-demand tree-model on memory that adopts the actor model as making use of multi-cores and distributed computers without largely rewriting code for algorithms.
Abstract: A number of online machine learning techniques based on tree model have been studied in order to cope with today's requirements of quickly processing large scale data-sets. We present a design pattern for incremental tree data processing as gradually constructing on-demand tree-model on memory. Our approach adopts the actor model as making use of multi-cores and distributed computers without largely rewriting code for algorithms. The pattern basically defines a node in the tree as an actor which is the unit of asynchronous processes and each data instance flows between actor nodes as a message. We study concrete two machine learning algorithms, VFDT for decision tree's top-down growth and BIRCH for hierarchical clustering's bottom up growth. For supporting VFDT, we propose an extension mechanism of replicating root nodes so that it can address bottleneck as starting of inputs. For supporting BIRCH, we split processes of recursive construction into asynchronous steps with correcting target node by traversing extra horizontal links between sibling nodes. We carried out machine learning tasks with our implementation on top of Akka Java, and we confirmed reasonable performance for the tasks with large scale data-sets.
15 Dec 1993
TL;DR: An upper bound is obtained for the communication complexity problem used by Karchmer and Wigderson to derive the depth version of the lower bound of Khrapchenko, showing that their method, as it is, cannot give better lower bounds.
Abstract: We study the computation of threshold functions using formulas over the basis {AND, OR, NOT}, with the aim of unifying the lower bounds of Hansel, Krichevskii, and Khrapchenko For this we consider communication complexity problems related to threshold function computation
We obtain an upper bound for the communication complexity problem used by Karchmer and Wigderson to derive the depth version of the lower bound of Khrapchenko This shows that their method, as it is, cannot give better lower bounds
We show how better lower bounds can be obtained if the referee (who was ignored in the Karchmer-Wigderson method) is involved in the argument
We show that the difficulty of the communication task persists even if the parties are required to operate correctly only for certain special inputs
21 Jul 2013
TL;DR: The research suggested that MNF3 is an optimal band to discriminate saline land from other object-grounds on condition of MNF<;-1.795, both of which show the effectiveness and feasibility of decision tree approach for monitoring and mapping spatial distribution of soil salinization.
Abstract: The dynamic monitoring and mapping of soil salinization is a practical significance work at present. In this paper, the middle reaches of Heihe River, China, was taken as a study case to discuss the effectiveness of extracting saline land information applying decision tree approach, based on Landsat TM data acquired on Sep.23, 2007. Through visual interpretation and statistical analysis of spectral characteristic associated with field survey and Google Earth image with higher resolution, finally five feature variables: thermal infrared band (TM6), Normalized Difference Vegetation Index (NDVI), Modified Normalized Difference Water Index (MNDWI), the third component of MNF rotation (MNF3) and the wetness of K-T transformation (TC3) were selected to construct decision tree model by setting the proper threshold values. The research suggested that MNF3 is an optimal band to discriminate saline land from other object-grounds on condition of MNF<;-1. The water body and vegetation district can be extracted by NDVI and MNDWI, respectively. Combining MNF3, TC3 and TM6 can well obtain sandy land and farmland information. The overall accuracy of classification results achieves 85.34% and Kappa Coefficient is 0.795, both of which show the effectiveness and feasibility of decision tree approach for monitoring and mapping spatial distribution of soil salinization.
TL;DR: A router-combined data fusion algorithm based on broadcast tree that can make data fusion effectively, which also has higher performance in delay, energy and accuracy is proposed.
Abstract: To keep the data synchronisation and fidelity of each node in wireless sensor network (WSN), we propose a router-combined data fusion algorithm based on broadcast tree. The approximation algorithm is used to transform the optimum aggregation tree into NP-C problem to solve the smallest Steiner tree. Then the cluster-based optimising directed diffusion routing is adopted to construct aggregation tree for data fusion. At the fusion phase, we introduce consensus factor to support degree matrix, to replace original constant factor and provide more weight effect for reliable sensor. The improved support degree matrix will measure the support degree of each sensor to form a novel weighted fusion algorithm. The analyses on simulation show that our fusion algorithm can make data fusion effectively, which also has higher performance in delay, energy and accuracy.