Search

The usual community detection methods are not applicable to bipartite networks due to their special 2 -mode structure. To identifying the community structure of bipartite networks，this paper proposed a novel algorithm based on Kullback-Leibler (KL) divergence between the 2-mode nodes. According to the connecting conditions between user set and object set， the algorithm obtained the link probability distribution on user set of bipartite networks，and developed K L similarity as a metric to evaluate the difference of node link patterns，and then detected the communities in bipartite networks overcoming the limitation of the 2-mode structure on nodes clustering. The experimental results and analysis in compute-generated and real network all show that this algorithm can effectively mine the meaningful community structures in bipartite networks，and improves the performance of community identification in the accuracy and efficiency. [ABSTRACT FROM AUTHOR]

For the investment of the variable-speed-limit boards was lack of the quantitative analysis and their locations were uncertain,this paper built an investment model,which implemented stair-like control strategies of variable-speed-limit based on the improved cell transmission model. First of all, it described the improved cell transmission model based on speed limit control strategies. Secondly,to implement stair-like control strategies,it introduced 0-1 integer programming algorithm and the related speed limit constraints. Finally,it optimized the locations of variable-speed-limit boards and dynamic speed limit values by genetic algorithm when the investment was constant. Simulation example results show that the average standard deviation of the critical density and saturation will decrease and the vehicle speed will increase as the investment increased,but the throughput will decrease and the average system delay will increase as well. The results show that the proposed model can effectively solve the problem of investment optimization of the variable-speed-limit boards. [ABSTRACT FROM AUTHOR]

To solve the poor performance of present algorithm in mining frequent weighted itemsets (FWI) from weighted items transaction databases, this paper proposed a fast mining algorithm for FWI based on weighted itemsets-Tidset trees. Firstly, it proposed a WIT-tree structure. Then, it used minimized weighted itemsets threshold value and closed down nature to trim infrequent nodes. Finally, it used Diffset strategy to allow memory rapidly calculate the weighted support itemsets with efficient way. The experimental results show that the mining time of the proposed algorithms is significantly less than several advanced mining algorithms in mining FWI. It can save time consumption with 99.37% comparing with the Apriori-based algorithm. It can save time consumption with 99.06% comparing with weighted frequent item sets generation algorithm based on bit matrix, which indicates that proposed algorithm has clearly improved the efficiency of frequent weighted itemsets mining. [ABSTRACT FROM AUTHOR]