Collaborative mining method of traffic accident data based on decision tree and association rules

被引：0

作者：

Liu F.H. ^{[1
]}

机构：

[1] College of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou

来源：

Advances in Transportation Studies | 2023年 / 1卷 / Special Issue期

关键词：

association rules; collaborative mining; data classification; decision tree; map reduce computing framework; traffic accident data;

D O I：

10.53136/97912218061517

中图分类号：

学科分类号：

摘要：

It is of great significance to scientifically analyze and excavate the internal relationship between traffic accident data, find out the potential laws hidden in the different attributes of the accident data, and thus provide suggestions and theoretical and technical support for the management decisions of relevant departments. In order to overcome the low recall rate, low precision rate and long task completion time in traditional traffic accident data collaborative mining methods, a traffic accident data collaborative mining method using decision tree and association rules is proposed. First of all, use the Map Reduce computing framework to build a traffic accident data collaborative mining framework; Then, combining with the weighted model to improve the association rules, the traffic accident data is preliminarily mined; Finally, the attribute with the largest information gain is calculated, and the traffic accident data is classified using the decision tree to realize the collaborative mining of traffic accident data. The experimental results show that the recall and precision of this method can reach 98.75% and 98.74% respectively, and the mining task completion time is only 0.2s at the fastest, which has good application effect. © 2023, Aracne Editrice. All rights reserved.

引用

页码：73 / 86

页数：13

共 17 条

[1]

Alkheder S., Alrukaibi F., Aiash A., Location and time factors' effect on types of traffic accident in Kuwait, International Journal of Data Analysis Techniques and Strategies, 14, 1, pp. 22-31, (2022)

[2]

Baruah A.J., Baruah S., Data Augmentation and Deep Neuro-fuzzy Network for Student Performance Prediction with MapReduce Framework, International Journal of Automation and Computing, 18, 6, pp. 981-992, (2021)

[3]

Borah A., Nath B., An Efficient Method for Mining Rare Association Rules: A Case Study on Air Pollution, International Journal on Artificial Intelligence Tools, 30, 4, pp. 1-35, (2021)

[4]

Chen Z., Zhang J., Zhang Y., Huang Z., Traffic Accident Data Generation Based on Improved Generative Adversarial Networks, Sensors (Basel, Switzerland), 21, 17, pp. 5767-5778, (2021)

[5]

Fister I., Information Cartography in Association Rule Mining, IEEE Transactions on Emerging Topics in Computational Intelligence, 99, pp. 1-17, (2021)

[6]

Huang D.X., Research on historical traffic accident data modeling based on state observer, Advances in Transportation Studies, Special Issue, 1, pp. 35-44, (2021)

[7]

Jagannadha R., Distributed Frequent Subgraph Mining Using Gaston and MapReduce, International Journal on Semantic Web and Information Systems (IJSWIS), 17, 2, pp. 41-58, (2021)

[8]

Li Z., Li X., Tang R., Zhang L., Apriori Algorithm for the Data Mining of Global Cyberspace Security Issues for Human Participatory Based on Association Rules, Frontiers in Psychology, 11, 1, pp. 1-12, (2021)

[9]

Mao Y.P., Yu F.Q., Sun Y.Y., Tang Z.G., Road Traffic Accident Data Mining and Application, Traffic & Transportation, 33, pp. 106-111, (2020)

[10]

Narayana S., Chandanapalli S.B., Rao M.S., Srinivas K., Ant Cat Swarm Optimization-Enabled Deep Recurrent Neural Network for Big Data Classification Based on Map Reduce Framework, The Computer Journal, 25, 12, pp. 12-28, (2021)

← 1 2 →