Guojun Lv
Guojun Lv has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Jianjun Cao
National University of Defense Technology
2 shared publications - Chang Chen
The Department of Electronics Science and Technology University of Science and Technology of China Hefei China
1 shared publication Chen Chang
Command and Control Engineering College, Army Engineering University of PLA, Nanjing, China
1 shared publication- Hongmei Li
Academy of Military Science, Beijing, China
1 shared publication Junnan Yao
Communications Engineering College, Army Engineering University of PLA, Nanjing, China
1 shared publicationNianfeng Weng
The Sixty-third Research Institute, National University of Defense Technology
1 shared publication
Latest Publications
- 1 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 448-455
Ant Colony Optimization Parameters Control Based on Evolutionary Strength
Ant Colony Optimization (ACO) was developed to be a metaheuristic and widely used for combinatorial optimization problems. It is confirmed that the parameters setting is of vital importance for performance of ACO. This paper proposed a strategy to perform ACO Parameters Control based on Evolutionary Strength (ACOP_ES). First, we transform four parameters setting problem to decision making problem. Then, by learning the evolutionary strength curve, we set sum, peak, and kurtosis of evolutionary strength curve as the objective function, such that the continuous ant colony algorithm can be used to optimize this decision problem. Further, via characterizing an excellent evolutionary curve, a relatively optimal parameters setting can be found. Compared with existing methods, our proposed method focuses on the performance of the running process rather than the results only. To the best of our knowledge, we are among the first to introduce process performance evaluation of ACO to guide parameters control, and determine the relatively optimal parameters of ACO. At last, we apply our method to Multi-dimensional Knapsack Problem (MKP), the results look promising that parameters returned by ACOP_ES are better than the ones returned by other methods on datasets: WEISH*.DAT and WEING*.DAT.
- 1 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 448-455
Ant Colony Optimization Parameters Control Based on Evolutionary Strength
Ant Colony Optimization (ACO) was developed to be a metaheuristic and widely used for combinatorial optimization problems. It is confirmed that the parameters setting is of vital importance for performance of ACO. This paper proposed a strategy to perform ACO Parameters Control based on Evolutionary Strength (ACOP_ES). First, we transform four parameters setting problem to decision making problem. Then, by learning the evolutionary strength curve, we set sum, peak, and kurtosis of evolutionary strength curve as the objective function, such that the continuous ant colony algorithm can be used to optimize this decision problem. Further, via characterizing an excellent evolutionary curve, a relatively optimal parameters setting can be found. Compared with existing methods, our proposed method focuses on the performance of the running process rather than the results only. To the best of our knowledge, we are among the first to introduce process performance evaluation of ACO to guide parameters control, and determine the relatively optimal parameters of ACO. At last, we apply our method to Multi-dimensional Knapsack Problem (MKP), the results look promising that parameters returned by ACOP_ES are better than the ones returned by other methods on datasets: WEISH*.DAT and WEING*.DAT.
- 8 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 31 (3) , 927-937
From Whole to Part: Reference-Based Representation for Clustering Categorical Data
Dissimilarity measures play a crucial role in clustering and, are directly related to the performance of clustering algorithms. However, effectively measuring the dissimilarity is not easy, especially for categorical data. The main difficulty of the dissimilarity measurement for categorical data is that its representation lacks a clear space structure. Therefore, the space structure-based representation has been proposed to provide the categorical data with a clear linear representation space. This representation improves the clustering performance obviously but only applies to small data sets because its dimensionality increases rapidly with the size of the data set. In this paper, we investigate the possibility of reducing the dimensionality of the space structure-based representation while maintaining the same representation ability. A lightweight representation scheme is proposed by taking a set of representative objects as the reference system (called the reference set) to position other objects in the Euclidean space. Moreover, a preclustering-based strategy is designed to select an appropriate reference set quickly. Finally, the representation scheme together with the $k$ -means algorithm provides an efficient method to cluster the categorical data. The theoretical and the experimental analysis shows that the proposed method outperforms state-of-the-art methods in terms of both accuracy and efficiency.- 8 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 31 (3) , 927-937
From Whole to Part: Reference-Based Representation for Clustering Categorical Data
Dissimilarity measures play a crucial role in clustering and, are directly related to the performance of clustering algorithms. However, effectively measuring the dissimilarity is not easy, especially for categorical data. The main difficulty of the dissimilarity measurement for categorical data is that its representation lacks a clear space structure. Therefore, the space structure-based representation has been proposed to provide the categorical data with a clear linear representation space. This representation improves the clustering performance obviously but only applies to small data sets because its dimensionality increases rapidly with the size of the data set. In this paper, we investigate the possibility of reducing the dimensionality of the space structure-based representation while maintaining the same representation ability. A lightweight representation scheme is proposed by taking a set of representative objects as the reference system (called the reference set) to position other objects in the Euclidean space. Moreover, a preclustering-based strategy is designed to select an appropriate reference set quickly. Finally, the representation scheme together with the $k$ -means algorithm provides an efficient method to cluster the categorical data. The theoretical and the experimental analysis shows that the proposed method outperforms state-of-the-art methods in terms of both accuracy and efficiency.