Clustering analysis can uncover hidden interconnections between data and segment the data according to multiple indicators， which can facilitate personalized and refined operations. However， data fragmentation and isolation caused by data islands seriously affects the effectiveness of cluster analysis applications. To solve data island problem and protect data privacy， an Equivalent Local differential privacy Federated K-means （ELFedKmeans） algorithm was proposed. A grid-based initial cluster center selection method and a privacy budget allocation scheme were designed for the horizontal federation learning model. To generate same random noise with lower communication cost， all organizations jointly negotiated random seeds， protecting local data privacy. The ELFedKmeans algorithm was demonstrated satisfying differential privacy protection through theoretical analysis， and it was also compared with Local Differential Privacy distributed K-means （LDPKmeans） algorithm and Hybrid Privacy K-means （HPKmeans） algorithm on different datasets. Experimental results show that all three algorithms increase F-measure and decrease SSE （Sum of Squares due to Error） gradually as privacy budget increases. As a whole， the F-measure values of ELFedKmeans algorithm was 1.794 5% to 57.066 3% and 21.245 2% to 132.048 8% higher than those of LDPKmeans and HPKmeans algorithms respectively； the Log（SSE） values of ELFedKmeans algorithm were 1.204 2% to 12.894 6% and 5.617 5% to 27.575 2% less than those of LDPKmeans and HPKmeans algorithms respectively. With the same privacy budget， ELFedKmeans algorithm outperforms the comparison algorithms in terms of clustering quality and utility metric.

Aiming at the problems that the dimensions of vectors generated by Traditional Vector Space Model （TVSM） are high， and the vector dot product operation to calculate the correlation between the documents and the queried keywords is time-consuming， a Dictionary Partition Vector Space Model （DPVSM） for ciphertext ranked search in cloud environment was proposed. Firstly， the specific definition of DPVSM was given， and it was proved that the relevance score between the queried keywords and the documents in DPVSM was exactly the same as that in TVSM. Then， by adopting the equal-length dictionary partition method， an encrypted vector generation algorithm and a relevance score calculation algorithm between documents and queried keywords were proposed. Experimental results show that the space occupation of document vectors of DPVSM is much lower than that of TVSM， and the more the number of documents， the greater the occupation reduction. In addition， the space occupation of query vectors and the time consumption of relevance score calculation are also much lower than those of TVSM. Obviously， DPVSM is superior to TVSM in both the space efficiency of generated vectors and the efficiency cost of relevance score calculation.

The Programmable Logic Controller (PLC) program needs modeling the program manually in the NuSMV model testing, which is not only a waste of manpower but also an error-prone procedure. In order to solve the problems, an automatic construction method of PLC program model based on state transition was proposed. Firstly, the Structured Text (ST) language features were analyzed and the ST program was parsed as an abstract grammar tree. Secondly, according to the abstract grammar tree, the control flow graph was generated based on different grammatical structure by control flow analysis. And then the program dependency graph was obtained by data flow analysis. Finally, the NuSMV input model was generated according to the program dependency graph. The experimental results shows that, the proposed method achieves the automatic construction from ST program to NuSMV input model, and the constructed NuSMV input model not only retains the original characteristics of ST program but also conforms to the input standard of NuSMV model detection tool. Compared with the traditional manual model construction method, the proposed method improves the efficiency and accuracy of model generation.

To improve the preprocessing performance for anti-spam and to search for phrases more efficiently, this paper creatively constructed an indexing dictionary based on hash algorithm, and designed a method of Chinese phrase segmentation based on this indexing dictionary aiming at anti-Chinese-spam. Through the study of the experimental data, this method is proved to be more efficient and accurate compared with the traditional mechanical classification, and it does improve the preprocessing performance and can be widely utilized in the field of Chinese phrase segmentation.