Frequent Itemset Mining (FIM) is one of the most important data mining tasks. The characteristics of the mined datasets have a significant effect on the performance of FIM algorithms. Sparseness of datasets is one of the attributes that characterize the essential characteristics of datasets. Different types of FIM algorithms are very different in the scalability of dataset sparseness. Aiming at the measurement of sparseness of datasets and influence of sparsity on the performance of different types of FIM algorithms, the existing measurement methods were reviewed and discussed, then two methods were proposed to quantify the sparseness of the datasets:the measurement based on transaction difference and the measurement based on FP-Tree method, both of which considered the influence of the minimum support degree on the sparseness of the datasets in the background of the FIM task, and reflected the difference between the frequent itemsets of the transaction. The scalability of different types of FIM algorithms for sparseness of datasets was studied experimentally. The experimental results show that the sparseness of datasets is inversely proportional to the minimum support, and the FIM algorithm based on vertical format has the best scalability in three kinds of typical FIM algorithms.

Concerning that the conventional color image coding algorithm does not take advantage of the dependency of the RGB color components, a new algorithm of set-partitioning RGB color components synchronously based on the Set Partitioning In Hierarchical Trees (SPIHT) algorithm was proposed. In this algorithm, RGB color components were treated as a whole, partition was sorted and set at the same time by using the same list of LIS. The color embedded bit-stream generated by this algorithm can stop at any point of the bit-stream and reconstruct the color image. The simulation results show the Peak Signal-to-Noise Ratio (PSNR) of the new algorithm on test images is about 0.1dB to 0.70 dB higher than JPEG2000s.

In the traditional process of new-medicines’ preproduction, designing experiment schemes and analyzing data was usually manipulated artificially. It was too subjective. Aiming at these situations, a new-medicines’ preproduction system was presented, which was based on experiment-design and support-vector-machines methods. First, experiment-design method was introduced in order to gain scientific and logical experiment schemes. Then, support-vector-machines (SVM) method was adopted to establish regress models for those experimental data, which is an analyzing and modeling tool suiting small sample data. These models were used for predicting the experiments’ results and optimizing the schemes. In addition, an alternant optimizing method based on greedy search was introduced too. Finally, an example was given to validate that the whole set of methods was scientific and effective.