US 12,468,612 B2
Creating method of a classification model about hard disk efficiency problem, analysis method of hard disk efficiency problem and classification model creating system about hard disk efficiency problem
Yi-Ju Liao, Taipei (TW); Jen-Yuan Chang, Taipei (TW); Po-Hsiu Chen, Taipei (TW); and Hsieh-Liang Tsai, Taipei (TW)
Assigned to INVENTEC (PUDONG) TECHNOLOGY CORPORATION, Shanghai (CN); and INVENTEC CORPORATION, Taipei (TW)
Filed by Inventec (Pudong) Technology Corporation, Shanghai (CN); and INVENTEC CORPORATION, Taipei (TW)
Filed on Jun. 15, 2022, as Appl. No. 17/840,774.
Claims priority of application No. 202210194415.1 (CN), filed on Mar. 1, 2022.
Prior Publication US 2023/0281093 A1, Sep. 7, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 11/30 (2006.01); G06F 3/0346 (2013.01); G06F 11/34 (2006.01); G06F 18/23213 (2023.01); G06N 5/01 (2023.01)
CPC G06F 11/3058 (2013.01) [G06F 3/0346 (2013.01); G06F 11/3409 (2013.01); G06F 18/23213 (2023.01); G06N 5/01 (2023.01)] 11 Claims
OG exemplary drawing
 
1. A creating method of a classification model about a hard disk efficiency problem, comprising: by an analyzing device, performing:
obtaining a plurality of pieces of measurement data of a plurality of hard disk devices each of which comprises a plurality of values of a plurality of vibration parameters;
discretizing the plurality of pieces of measurement data based on a k-means algorithm; and
obtaining the classification model about the hard disk efficiency problem based on the plurality of pieces of discretized measurement data and a decision tree algorithm;
wherein obtaining the classification model about the hard disk efficiency problem based on the plurality of pieces of discretized measurement data and the decision tree algorithm comprises:
determining a first decision node based on one or both of cross entropy and information gain to divide at least one part of the plurality of pieces of discretized measurement data into two measurement data sets, wherein the first decision node is associated with one of the plurality of vibration parameters;
defining that a value of i is a positive integer and its initial value is 2 and performing a classification operation, with the classification operation comprising:
determining a ith decision node based on one or both of the cross entropy and the information gain to divide one of the two measurement data sets classified by a (i−1)th decision node into another two measurement data sets, wherein the ith decision node is associated with another one of the plurality of vibration parameters;
determining whether the value of i is equal to a number of the plurality of vibration parameters or not, wherein the number of the plurality of vibration parameters is greater than or equal to 2;
if the value of i is not equal to the number of the plurality of vibration parameters, adding 1 to the value of i and performing the classification operation again; and
if the value of i is equal to the number of the plurality of vibration parameters, constituting the classification model about the hard disk efficiency problem based on the first decision node and one or more results generated by performing the classification operation one or more times.