Jiemin Zheng’s research while affiliated with Xiamen University and other places

In this paper, we develop an extended guided tabu search (EGTS) and a new heuristic packing algorithm for the two-dimensional loading vehicle routing problem (2L-CVRP). The 2L-CVRP is a combination of two well-known NP-hard problems, the capacitated vehicle routing problem, and the two-dimensional bin packing problem. It is very difficult to get a good performance solution in practice for these problems. We propose a meta-heuristic methodology EGTS which incorporates theories of tabu search and extended guided local search (EGLS). It has been proved that tabu search is a very good approach for the CVRP, and the guiding mechanism of the EGLS can help tabu search to escape effectively from local optimum. Furthermore, we have modified a collection of packing heuristics by adding a new packing heuristic to solve the loading constraints in 2L-CVRP, in order to improve the cost function significantly. The effectiveness of the proposed algorithm is tested, and proven by extensive computational experiments on benchmark instances.

In recent years, more and more people, especially young people, begin to use credit card with the changing of consumption concept in China so that the business on credit cards is growing fast. Therefore, it is significative that some effective tools such as credit-scoring models are created to help those decision makers engaged in credit cards. A novel credit-scoring model, called vertical bagging decision trees model (abbreviated to VBDTM), is proposed for the purpose in this paper. The model is a new bagging method that is different from the traditional bagging. The VBDTM model gets an aggregation of classifiers by means of the combination of predictive attributes. In the VBDTM model, all train samples and just parts of attributes take part in learning of every classifier. By contrast, classifiers are trained with the sample subsets in the traditional bagging method and every classifier has the same attributes. The VBDTM has been tested by two credit databases from the UCI Machine Learning Repository, and the analysis results show that the performance of the method proposed by us is outstanding on the prediction accuracy.

This paper addresses the capacitated vehicle routing problem with two-dimensional loading constraints (2L-CVRP). The 2L-CVRP is a combination of the two most important problems in distribution logistics, which are loading of freight into vehicles, and the successive routing of the vehicles to satisfy customer demand. The objective is to minimize the transportation cost. All vehicles must start and terminate at a central depot, and the transported items carried by the vehicles must be feasibly packed into the loading surfaces of the vehicles. A simulated annealing algorithm to solve the problem is presented, in which the loading component of the problem is solved through a collection of packing heuristics. A novel approach to plan packing is employed. An efficient data structure (Trie) is used to accelerate the algorithm. The extensive computational results prove the effectiveness of the algorithm. KeywordsVehicle routing–Loading constraints–Simulated annealing

Mining frequent itemsets is one of the most investigated fields in data mining. It is a fundamental and crucial task. Apriori is among the most popular algorithms used for the problem but support count is very time-consuming. In order to improve the efficiency of Apriori, a novel algorithm, named BitApriori, for mining frequent itemsets, is proposed. Firstly, the data structure binary string is employed to describe the database. The support count can be implemented by performing the Bitwise "And" operation on the binary strings. Another technique for improving efficiency in BitApriori presented in this paper is a special equal-support pruning. Experimental results show the effectiveness of the proposed algorithm, especially when the minimum support is low.

Association rule mining is to find association relationships among large data sets. Mining frequent patterns is an important aspect in association rule mining. In this paper, an efficient algorithm named Apriori-Growth based on Apriori algorithm and the FP-tree structure is presented to mine frequent patterns. The advantage of the Apriori-Growth algorithm is that it doesn't need to generate conditional pattern bases and sub- conditional pattern tree recursively. Computational results show the Apriori-Growth algorithm performs faster than Apriori algorithm, and it is almost as fast as FP-Growth, but it needs smaller memory.