Designing Language Of On-line Analytical Mining Process

Fierceness of market competition force the enterprise to look for tools for dealing with the complicated company affairs, the tools are more scientific, systematic, effective aid decision technology and whole solution. On-line analytical process and data mining are important technology among the decision technology and whole solution.On-line analytical process and data mining are different. Data mining is course of analyzing in depth the data automatically. Automation means that analyses that the course does not need the user to participate in, this is its merit, also insufficient. In fact, the user comes while hoping to participate in the course.. For example, they want to analyze itemset of the data, and the data is collecting differently, integrated level. They also want to choose algorithms of the course according to different needs, etc. Essence of On-line analytical process is a course of reasoning, if the variable among analyzing course reaches hundreds and thousands, it is very difficult. So people hope on-line analytical process and data mining that can be combined to a certain extent, help on the function each other.With the development of computing technique, some new application modes have been carried out, including data warehouse, orient-object, parsing language, querying multidimensional data, etc. These technology offers the foundation in order to solve combination between on-line analytical process and data mining. This thesis want to use these most new technology, seek mode who the two combine, and apply to reality, these kind of mode is called on-line analytical mining process.Main research measure and achievement:1,On-line analytical mining process is based on on-line analytical process. Through the analysis to the data warehouse technology, on-line analytical process set up on the data warehouse, which can help and analyze data, avoiding affairs operate to disturb data.. Through analyzing the characteristic of on-line analytical mining process and data mining, believed that the question proposed in data mining depends on the result of analyzing, It can't search knowledge blindly, on-line analytical process can just offer this kind of need that the data mining. On-line analytical mining process lie between the two, so data warehouse , on-line analytical process and on-line analytical mining process become the basic bottom-up relation.2,On-line analytical mining process builds on the basis of description of the mode, this kind of description has caused languages. It makes business abstract to become one a kind of data analysis modes, so can solve problem through computer. Through the study on technology at present, XML language can be regarded as the tool which describes the mode question. This gives the credit to database and making decision and analyzing products, the popularization energetically use XML languages of manufacturer. Becoming a kind of in common use description tools, XML languages can have been very convenient to use it for turning enterprise mode to decision analyze the transformation of the mode, have flexibility orient-object technical application. For having convenience flexible interfaces by user, liking SQL language come and describe user 's special requirement for on-line analytical mining process. XML describe mode of On-line analytical mining process with by language term, and kinds of liking SQL the designs of language become this text design and main method that realize, so this text is the design of On-line analytical mining process.3,On-line analytical mining process can build on the mode of combining on-line analytical process and relation rule technique in data mining. By analyzing we find that it is a bit more similar to on-line analytical process and relation rule technique on function and process, which can help each other.above analysis measure and result of study, probe into mode of on-line analytical mining process for offering greater and wider decision to user and analyzing the function. The mode is feasible to a extent.