Method for recommending crop breeding varieties based on hybrid collaborative filtering algorithm

Abstract

The invention discloses a method for recommending crop breeding varieties based on a hybrid collaborative filtering algorithm. The method comprises the following steps: in step 1, data preprocessing is performed on data of check varieties and data of the varieties to be audited to obtain variety-property data tables; in step 2, clustering of the variety-property data tables is conducted so as to obtain multiple clusters; in step 3, a first cluster containing a check variety-property data table is searched; in step 4, a degree of similarity between the varieties to be audited in the first cluster and the check varieties is calculated, and varieties to be audited with a high degree of similarity are selected; in step 5, the selected varieties to be audited are taken as recommended varietiesfor crop breeding. The method disclosed in the invention can help effectively reduce calculation loss; especially in the commercial breeding, the method disclosed in the invention can help improve work efficiency of a staff, greatly reduce labor cost, and complete the work that a plurality of breeding experts need to do to meet requirements for large-scale breeding data analysis.

Description

technical field [0001] The invention relates to the technical field of crop breeding, and more specifically, the invention is a method for recommending crop breeding varieties based on a hybrid collaborative filtering algorithm. Background technique [0002] Since the rapid development of next-generation sequencing technology, crop breeding research has produced massive data, and integrating and maximizing the use of these biological data is undoubtedly of inestimable importance to modern breeding research. However, because the breeding industry is relatively professional and has a high demand for professional background, the existing breeding data analysis methods are still limited to methods based on statistics, and most of them rely on expert experience for data analysis. Due to limitations and other issues, it cannot be widely adapted to the field of breeding industry, let alone meet the needs of commercialization. [0003] Therefore, how to avoid the dependence of bree...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problems of strong subjectivity and inability to meet the needs of commercialization in the traditional breeding data analysis method, the present invention innovatively proposes a crop breeding variety recommendation method based on a hybrid collaborative filtering algorithm, which combines the collaborative filtering algorithm and the K-means algorithm Applied to the evaluation of breeding varieties, combined with the actual needs of breeding work, the evaluation of crop breeding varieties is more effective and reliable, thus solving many problems existing in the existing technology

Images