Method for constructing product industry parallel data set by using incremental singular value decomposition method

A singular value decomposition and parallel data technology, applied in complex mathematical operations, instruments, character and pattern recognition, etc., can solve problems such as long time spans, increased batches, and difficulty in reproducing scene data

CN111178391AActive Publication Date: 2020-05-19PURPLE MOUNTAIN LAB
4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2020-05-19

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a method for constructing a product industry parallel data set by using an incremental singular value decomposition method. The method for constructing the data set is based oninfluence factor data and yield and quality data of a certain product production process. Parallel sample data similar to the target data are obtained from historical data samples, and process designand data analysis is performed by utilizing the parallel sample data. According to the method, an incremental singular value decomposition method is utilized, data dimension reduction and principal component analysis can be conducted on mass data in a library, data disturbance caused by random factors during data collection can be filtered out through the analysis, and the dimension and the sample number of imported data are controlled through a left singular matrix U and a right singular matrix. Meanwhile, the algorithm is based on a batch updating mode, and the complexity of the whole calculation is low.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of product information analysis and processing, in particular to a method for constructing parallel data sets of product industries using incremental singular value decomposition. Background technique

[0002] The production process of many industrial products is long and has many influencing factors, which are much more complicated than general industrial production. For example, in the process of agricultural planting and agricultural product production, there are common difficulties in data collection, including:

[0003] (1) The collected data has fewer dimensions;

[0004] (2) Fewer sample batches were collected;

[0005] (3) The time span of the whole collection process is long;

[0006] (4) The scene data is difficult to reproduce.

[0007] These data deficiencies have seriously affected the implementation of data mining and machine learning algorithms in the agricultural product industry. In the ...

Examples

Embodiment 1

[0109] In the process of growing grapes, the precipitation in a certain year is relatively high, the temperature is relatively low, and there is a danger of pests and diseases. I hope to modify the irrigation method of water and fertilizer, but I don’t know how to adjust it to ensure the optimal yield and quality.

[0110] Step 1: Establish a horizontal table data, link the daily precipitation, water and fertilizer irrigation, pest impact, water and fertilizer irrigation method and the grape yield and quality data of the previous years to form a wide table data Ω.

[0111] Step 2: Get the wide-table data set B corresponding to the output and quality of normal years in the wide-table data set Ω.

[0112] Step 3: Assuming that the water and fertilizer index in a normal year is x, according to the three water and fertilizer indexes of 1.5x, 2x, and 0.8x respectively, obtain parallel data candidate sets D1, D2, and D3 from the part of the data set Ω other than B;

[0113] Assuming...

Embodiment 2

[0120] During the brewing process of wine, a process engineer wants to increase the amount of an enzyme added, but they don't know if doing so will affect the quality of the final wine.

[0121] Step 1: Establish a horizontal table data, which includes indicators such as temperature, humidity, enzyme addition, fermentation time, acidity and sweetness of grape juice in previous production batches, and quality data such as color, taste and alcohol content of the produced wine Associated to form a wide table data Ω.

[0122] Step 2: Obtain the wide-table data set B corresponding to the current process result from the wide-table data set Ω.

[0123] Step 3: Assume that the enzyme index corresponding to the current industrial results is x, respectively, according to the assumption that the increased enzyme index is 2x, and obtain a parallel data candidate set D from the part of the data set Ω other than B;

[0124] Assuming that the wide table data set B has 200 attributes, perfor...