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
Patent•
11 Feb 2020
TL;DR: In this article, an interactive modeling method and device for a decision tree model, equipment and a storage medium is presented. But the authors focus on the problem that the interactive modeling mode based on a single machine cannot carry out modeling by utilizing massive training data stored in a distributed manner.
Abstract: The invention discloses an interactive modeling method and device for a decision tree model, equipment and a storage medium. The method comprises the steps of acquiring the decision tree model to be operated and an operation task for operating the decision tree model; operating the decision tree model according to the operation task, and determining model information to be calculated according tothe operated decision tree model; distributing the calculation task of the model information to each distributed execution machine connected with the local equipment, so that each distributed execution machine executes the calculation task according to respective local data; and obtaining model information according to the calculation result of each distributed execution machine, and visually outputting the operated decision tree model and the model information. According to the invention, the interactive modeling of the decision tree model is carried out by combining the plurality of distributed actuators storing the training data of the decision tree model, so that the problem that the interactive modeling mode based on a single machine cannot carry out modeling by utilizing massive training data stored in a distributed manner is solved.
Patent•
15 Feb 2019
TL;DR: In this paper, a fault scene tree model for a multi-state multi-phase task system is proposed. But the model is not suitable for the case of multi-stage systems.
Abstract: The present application provides a fault scene tree modeling method for a multi-state multi-phase task system. The method comprises the following steps: firstly, the correlation relation of the faultmechanism in the fault scene tree is studied, and the expression method of each correlation relation in the fault scene tree is proposed; Secondly, the construction method of fault scene tree of multi-state system is studied, and the time sequence scene tree and fault sequence scene tree model are established. Thirdly, the multi-phase task system is studied, and the event sequence scene tree modelof the multi-phase task system is established by analogy of the event sequence scene tree of the two-state case of the multi-phase task system. Finally, the reliability of the mechanism model and fault scene tree model is simulated by Matlab, and the simulation results of two-state system, multi-stage two-state system and multi-stage multi-state system are obtained according to the need. The fault scenario tree proposed in this application can characterize the system from three dimensions: logic, time and probability, so as to dynamically depict the whole process of system fault occurrence and development.
01 Jan 2016
TL;DR: This chapter introduces the random forest supervised learning model, which uses the decision tree model for parametrization, but it integrates a sampling technique, a subspace method, and an ensemble approach to optimize the model building.
Abstract: The main objective of this chapter is to introduce you to the random forest supervised learning model. The random forest technique uses the decision tree model for parametrization, but it integrates a sampling technique, a subspace method, and an ensemble approach to optimize the model building. The sampling approach is called the bootstrap, which adopts a random sampling approach with replacement. The subspace method also adopts a random sampling approach, but it helps extract smaller subsets (i.e., subspaces) of features. It also helps build decision trees based on them and select decision trees for the random forest construction. The ensemble approach helps build classifiers based on the so-called bagging approach. The objectives of this chapter include detailed discussions on these approaches. The chapter also discusses the training and testing algorithms that are suitable for the random forest supervised learning. The chapter also presents simple examples and visual aids to better understand the random forest supervised learning technique.
20 Dec 2019
TL;DR: In order to deal with the context dilution problem introduced in the lossless compression of M-ary sources, a Lossless compression algorithm based on a context tree model is proposed that can achieve better compression results.
Abstract: In order to deal with the context dilution problem introduced in the lossless compression of M-ary sources, a lossless compression algorithm based on a context tree model is proposed. By making use of the principle that conditioning reduces entropy, the algorithm constructs a context tree model to make use of the correlation among adjacent image pixels. Meanwhile, the M-ary tree is transformed into a binary tree to analyze the statistical information of the source in more details. In addition, the escape symbol is introduced to deal with the zero-frequency symbol problem when the model is used by an arithmetic encoder. The increment of the description length is introduced for the merging of tree nodes. The experimental results show that the proposed algorithm can achieve better compression results.
29 Mar 2020
TL;DR: From the results of research with primary data of 2000-2003 graduate students in Amik PPMI Tangerang, it is explained that the particle swarm optimization method can increase accuracy by 87.56% and increase by 01.01% from the decision tree method with a value of 86.55%.
Abstract: Good accreditation results are the goal of the college. With good accreditation, prospective students can glance at and enter the tertiary institution. To achieve this, there are several aspects that affect good accreditation results, one of which is graduate students who play an important role in determining accreditation. Timely graduate students can benefit the college or a student. Graduates can be predicted before the final semester using a method one of which is the decision tree. Decision tree is a method that is simple and easy to understand by producing rules in the form of a decision tree, but using a decision tree model alone is not enough to produce optimal results. So we need a method for optimization that is particle swarm optimization with advantages can improve accuracy by eliminating unused features. From the results of research with primary data of 2000-2003 graduate students in Amik PPMI Tangerang explained that the particle swarm optimization method can increase accuracy by 87.56% and increase by 01.01% from the decision tree method with a value of 86.55%. From the particle swarm optimization method can also find out which unused attributes have no weight, so that way can improve accuracy. From the results of the increase, it can be used by the Amik University of Tangerang to prevent students from graduating on time.