期刊名称：International Journal of Data Mining & Knowledge Management Process
印刷版ISSN：2231-007X
电子版ISSN：2230-9608
出版年度：2014
卷号：4
期号：2
页码：1
DOI：10.5121/ijdkp.2014.4201
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：This paper aims to predict quality features of Iberian hams by using non-destructive methods of analysisand data mining. Iberian hams were analyzed by Magnetic Resonance Imaging (MRI) and ComputerVision Techniques (CVT) throughout their ripening process and physico-chemical parameters from themwere also measured. The obtained data were used to create an initial database. Deductive techniques ofdata mining (multiple linear regression) were used to estimate new data, allowing the insertion of newrecords in the database. Predictive techniques of data mining were applied (multiple linear regression) onMRI-CVT data, achieving prediction equations of weight, moisture and lipid content. Finally, data fromprediction equations were compared to data determined by physical-chemical analysis, obtaining highcorrelation coefficients in most cases. Therefore, data mining, MRI and CVT are suitable tools to estimatequality traits of Iberian hams. This would improve the control of the ham processing in a non-destructiveway.
关键词：Data mining; MRI; Active Contours; Iberian ham; Quality traits prediction
