Patent
Method and system used for entity matching
Li Zhixu,Yang Qiang,Jun Jiang +2 more
- 11 Nov 2015
5
TL;DR: In this paper, a pre-trained decision tree is used to obtain the attribute similarity and the confidence coefficient of each attribute of an instance pair to be matched, and then the confidence coefficients are combined with a regulation coefficient to calculate and output the entity similarity of the instance pair.
read more
Abstract: The invention provides a method and a system used for entity matching. The method comprises the following steps: beginning to access an instance pair to be matched from an attribute corresponding to a root node of a pre-trained decision tree to obtain the attribute similarity and the confidence coefficient of each attribute of the instance pair to be matched; combining the attribute similarity and the confidence coefficient with a regulation coefficient to calculate and output the entity similarity of the instance pair to be matched; and comparing the entity similarity with a preset entity similarity threshold value, and judging the similarity of the instance to be matched, wherein the decision tree is obtained from a common non-primary attribute set and/ or primary attribute set in the instance pair formed by known matched instances. The method obtains the decision tree in a way that the common non-primary attribute set and/ or primary attribute set of the two entities in the known instances, the function of the non-primary attribute is considered in an entity matching process, and the accuracy and the recall rate of the entity matching can be improved.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Structured entity recording method and device, server and storage medium
Xu Ye,Feng Zhifan,Lu Chao,Zhang Yang,Fang Zhou,Wang Shu,Zhu Yong,Li Ying +7 more
- 15 May 2018
TL;DR: In this article, a structured entity recording method is described, which comprises the steps that candidate entities related to structured entities are selected from a knowledge graph, the structured entities to be recorded are determined as associated entities according to priori attribute information of the type to which the candidate entities belong and a preset model, the associated entities and candidate entities are merged, and the associated entity are recorded into the knowledge graph.
6
Patent
Method and apparatus for combining different examples for describing same entity and equipment
Yang Yang,Mu Guanyu,Hua Nengwei,Wei Zhang,Wu Jia +4 more
- 20 Jul 2016
TL;DR: In this paper, a method and an apparatus for combining different examples for describing the same entity and equipment is described, where the connection drawing including multiple examples is acquired, and the connecting lines among the nodes represent the example relation of the corresponding examples.
5
Patent
Multi-source data integration method and device
Xu Zhe-Hao
- 10 Nov 2017
TL;DR: In this paper, a multi-source data integration method for acquiring data belonging to the same entity from a data set, at least one association attribute of the entity attribute of an entity can be respectively acquired with regard to an arbitrary entity, the attribute similarity of the association attributes of two entities can be acquired, the two entities could be determined to be the same entities if the attributed similarity is larger than a similarity threshold, and the entity attributes of the two entity attributes are associated to a same entity.
3
Patent
Method, device and apparatus for combining different instances describing same entity
Yang Yang,Mu Guanyu,Hua Nengwei,Wei Zhang,Wu Jia +4 more
- 17 Aug 2017
TL;DR: In this paper, the authors propose a method for combining different instances describing the same entity, where different nodes in the connection diagram indicate different instances, and connecting lines between the nodes indicate an instance relationship between the instances corresponding to the nodes.
Patent
Mining method and system of similar entities
Luo Jie
- 10 Aug 2018
TL;DR: In this article, a mining method and system of similar entities is described, which consists of the following steps: acquiring text description information corresponding to sample entities; summarizing the acquired text description and extracting characteristic information; calculating weights respectively corresponding to various characteristics in the extracted characteristic information to acquire corresponding decision formulas of category entities; and deciding description texts corresponding to other entities by utilizing the acquired decision formulas and finding out the entities of which the categories are similar to the categories of the sample entities.
References
Patent
Method and device for combining entities in knowledge map
Hu Shiwen,Xiang Bibo +1 more
- 01 Apr 2015
TL;DR: In this paper, a method and a device for combining entities in a knowledge map is presented. But the method comprises the following steps of generating a first-grade feature vector according to structural data corresponding to the entities in the knowledge map; generating a second-grade descriptor according to terms in documents corresponding to entities; determining the similarity of the different entities according to the descriptor and the feature vector; and determining whether the entity IDs with the same name are the same object or not.
15
Patent
Method and system based on encyclopedia data for classifying entities
Gong Yingkun,Hu Shiwen,Xiang Bibo +2 more
- 01 Apr 2015
TL;DR: In this paper, the authors proposed a method and a system based on encyclopedia data for classifying entities. But the method is not suitable for the classification of the data of which the similarity is lower than a threshold value, and thus the purpose of classifying the data is realized.
14
Patent
Data model optimization
Gunther Stuhec,Florian Gessner,Jens Lemcke +2 more
- 12 Nov 2008
TL;DR: In this paper, the semantic meaning of each of the names and one or more attributes of each entity class of the data model was determined by comparing the semantic meanings of the attributes of the first entity class to the semantic ones of the second entity class.
8
Patent
Extensible access control markup language (XACML) strategy assessment engine system based on various optimization mechanisms
Niu Dehua,Ma Jianfeng,Ma Zhuo,Wang Lei,Li Chennan +4 more
- 10 Jul 2013
TL;DR: In this paper, an extensible access control markup language (XACML) strategy assessment engine system based on various optimization mechanisms is presented. But the system cannot make a correct decision on access requests sent by a large number of users at the same time.
6