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
••
01 Aug 2021
3 citations
•
08 May 2013
TL;DR: In this article, a method for building a software product feature tree model based on a demand cluster is presented, which includes a directed demand relational graph input module, a primitive feature model tree generating model, a feature model trees improving module, and a feature tree output module.
Abstract: The invention discloses a method for building a software product feature tree model based on a demand cluster. The method includes a directed demand relational graph input module, a primitive feature model tree generating model, a feature model tree improving module, and a feature model tree output module. The method includes that the a software product directed demand relational graph input module reads in composition information of a software product requirement, produces a software product directed demand relational graph, and sets relational weight of dependence intensity among representation requirements. The primitive feature model tree generating model first empties a feature tree, then produces primitive feature nodes of all layers of the feature tree according to the directed demand relational graph input module, and records mapping relation between the requirements and the feature nodes. The feature model tree improving module improves a primitive feature model tree through executing decomposition judgment, similarity judgment and exclusive judgment, and adds root nodes. According to the method for building the software product feature tree model based on the demand cluster, the degree of automation of building the product feature tree model is improved, and the automatic generation of the software product feature tree model based on the directed demand relational graph is achieved.
3 citations
•
26 Feb 2019
TL;DR: In this paper, a loan product matching method based on big data is proposed, where the tax data of the loan enterprise is input into the stochastic forest model, and the matching type label of loan product is obtained.
Abstract: The present application relates to a loan product matching method, apparatus, computer device and storage medium based on big data The method comprises: Get Historical Tax Data, Analyze historical tax data to get tax characteristics, the initial decision tree model is obtained by training tax characteristics, The initial decision tree model outputs a type label for the loan product, Obtain information on qualification requirements for each type of loan product, According to the qualification requirement information, the decision node is generated, the decision node is added into the initial decision tree model to obtain the matching decision tree model, and according to the matching decision tree model of each tax characteristic, the stochastic forest model is generated, the tax data of the loan enterprise is input into the stochastic forest model, and the matching type label of loan product is obtained The method can be used to recommend suitable loan products and reduce the labor cost
3 citations
••
3 citations
••
TL;DR: The concept of aregulated tree as a generalization of a regular tree which has the advantage of allowing the same lower bounds on the non-linear portion of the complexity is introduced.
Abstract: Andrew Yao proved some lower bounds for algebraic computation trees with integer inputs. In his key result he proved bounds on the number of components of the leaf space of a homogeneous decision tree derived from a computation tree. In this paper we present a shorter and more conceptual proof. We introduce the concept of aregulated tree as a generalization of a regular tree which has the advantage of allowing the same lower bounds on the non-linear portion of the complexity. The proof is an application of a result of Ben-Or.
3 citations