Mining The Miner The Model Jzx-design Research, Based On The Incremental Updating Of Sequential Patterns

An amount of data have been generated for the recent years with the increasing capacity in gathering data, the popularization of the computer in a number of businesses and governments' transactions and rapid development in data collecting tools. Millions of databases have been used in business management, government administration, scientific and engineering data management, and many other application fields. Under this situation, it is urgently demanded that a tool or a technology can be developed to help us automatically extract the useful knowledge and information from myriad data. Thereby, data mining, as such a kind of the technology, is given the unprecedented attentions.Sequential Pattern, an important sub-branch of data mining technology, is used to capture the typical behaviors related to the time, that is, to mine the reoccurring regularity. It occupies a lot of researchers due to its stronger practicality.The key to data mining lies in the technology of dealing with amounts of data, managing and maintaining rules. Therefore, we have to resolve two issues as follows:1. Designing an efficient algorithm to mine sequential patterns2. Design an efficient algorithm to update and maintain the found sequential patternsThe second issue is as important as the first issue. Now we have made the further research on the second issue and propose some maintenance algorithms. However, these algorithms are based on the constant value of the minimal support to resolve the problem pertaining to the updating of sequential patterns. As a matter of fact, the value of the minimal support needs to be often regulated during the process of the mining so as to find statistically efficient sequential patterns and to meet user's need. At this point, the updating of the sequential patterns also exists when the database keep static. The paperpresent a incremental sequential pattern, IASP. When the new value of support is greater than the old one, IASP avoid scanning database by repeatedly keeping its candidate patterns and their support in a database, which reducing the frequency of scanning database and recover the defect of the traditional algorithms. The new algorithm is designed from the aspect to greatly improve the running efficiency.Combined with the sale of container manufacturer, the paper research and design a data mining Model (JZX-MINER). The Model contains the following characteristic:1. Constructed on the data warehouse. The paper introduces general theory of data warehouse and developing means and improves ands optimize the information package and snowflake model, and proposes some new concepts, like level entity, sort entity and level table, sort table, detailed information tables.2. Design a data mining model according IASP algorithm and obtain some information about sale tendency of products and special rules.JZX-MINER setting the large-scale container manufacturing company as the trade background can be applied to the retail trade of different fields.