Algorithm capable of effectively reducing cost of consistent attribute data

Abstract

The invention discloses an algorithm capable of effectively reducing the overhead of consistent attribute data, which is used for processing data with a large number of consistent decision attributes, and comprises the following steps of: dividing decision attribute values into various types of information, forming an initial set of a decision information system by using the various types of information; the method comprises the steps that a decision information system is established, a difference matrix is constructed through an initial set of the decision information system, key information in the difference matrix is extracted through decision attributes, attributes with the large frequency in the difference matrix are selected as candidate attributes, a candidate attribute set is a non-repeated set, and each candidate attribute is attribute data which cannot be decomposed again; the candidate attributes serve as anchor points, the difference matrix is converted, a conversion matrix based on the key attributes is obtained, and data overhead is reduced; according to the method, when a large number of consistent decision attributes exist in stored data, the decision attributes are effectively processed, and the purpose of simplifying a data set is achieved.

Description

technical field [0001] The invention relates to an algorithm capable of effectively reducing consistent attribute data overhead, and belongs to the technical field of data processing. Background technique [0002] In recent years, with the development of data mining, machine learning and big data and other fields, how to process data information at a high speed and effectively has become a direction with high commercial value. Rough set theory is a method that can effectively deal with inconsistency, incompleteness and The mathematical knowledge of imprecise information is widely used in some data processing directions. [0003] Rough set theory is a branch of knowledge reduction. Its common algorithms include: knowledge reduction algorithm based on positive region model, knowledge reduction algorithm based on difference matrix model, and knowledge reduction algorithm based on information entropy model. The commonly used method of using the difference matrix to simplify dat...

AI Technical Summary

Problems solved by technology

The commonly used method of using the difference matrix to simplify data first needs to transform the matrix, and use the kernel attributes as the initial set of the difference matrix, and iteratively solve the attribute with the highest frequency as the candidate set. It cannot be simplified, and the time complexity of the algorithm increases rapidly with the increase of the data set, and the storage space cannot be effectively optimized.

Images