CN109669939A - Object information processing method - Google Patents
Object information processing method Download PDFInfo
- Publication number
- CN109669939A CN109669939A CN201811300541.0A CN201811300541A CN109669939A CN 109669939 A CN109669939 A CN 109669939A CN 201811300541 A CN201811300541 A CN 201811300541A CN 109669939 A CN109669939 A CN 109669939A
- Authority
- CN
- China
- Prior art keywords
- data source
- value
- subobject
- attribute value
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The application provides a kind of object information processing method.This method comprises: obtaining the subobject of data source and the attribute value of each subobject, the first matrix is created with the corresponding subobject of data source, the second matrix is created with the attribute value of each subobject of data source, weight coefficient corresponding to each subobject for calculating data source according to first matrix calculates weight coefficient corresponding to each attribute value in each subobject of data source according to second matrix；The same sex value in data source with common attribute value is obtained, dominant score value of the data source in each attribute value is determined according to the same sex value；It weights weight coefficient and dominant score value to obtain the assessed value of data source.The accuracy of data source assessment can be improved in the scheme that the application proposes.
Description
Technical field
This application involves technical field of information processing, more particularly to a kind of object information processing method.
Background technique
In data processing field, value assessment need to be carried out to related data sources, usual way is that data source is worth
Quantitative evaluation is carried out from single dimension；Or data source is simply integrated from several dimensions into its value of assessment, however existing number
Absolute magnitude is all based on according to the scoring in source or relative quantity be calculated its same sex value, and different data source quantity, unit
Quantify having differences, another different data sources show different effects in each attribute value, if simply to each attribute
Value scores after only assigning a weighted value weighting, or a certain attribute value is taken to compare, then it is inadequate to will lead to final evaluation
Accurately.
Summary of the invention
The purpose of the application is, provides a kind of object information processing method, can improve the accurate of data source value assessment
Property.
It the purpose of the application and solves its technical problem and adopts the following technical solutions to realize.
A kind of object information processing method, this method comprises:
Obtain the subobject of data source and the attribute value of each subobject；
According to the subobject collection of total subobject building data source of data source, according to the attribute of each subobject of data source
The property set of value building data source, the subobject collection includes the corresponding subobject of data source, and the property set includes data source
Each subobject attribute value；
The first matrix is created with the corresponding subobject in subobject intensive data source, with data source in the property set
The attribute value of each subobject creates the second matrix, and the matrix element of first matrix is the weight coefficient of each subobject, the
Matrix element in two matrixes is the weight coefficient of the attribute value of each subobject；
Weight coefficient in first matrix and the second matrix is normalized by column, to normalized
First matrix and the summation of the second matrix by rows afterwards, obtain normalizing characteristic value；
By the normalizing characteristic value divided by the total quantity of homography element, obtain corresponding to each subobject of data source
Weight coefficient and each subobject in each attribute value corresponding to weight coefficient；
The same sex value in data source with common attribute value is obtained, determines data source in each attribute according to the same sex value
Dominant score value in value；
Weight coefficient corresponding to each subobject by data source, data source each subobject in each attribute value
Dominant score value of the corresponding weight coefficient and data source in each attribute value weights to obtain the assessed value of data source.
Preferably, when the numerical value of the same sex value in the data source with common attribute value is more than the first preset threshold,
Dominant score value of the data source in each attribute value is by standardization formula Sij_new=Sij/(Si_max+ 1) it determines；
When the numerical value of the same sex value in the data source with common attribute value is lower than the second preset threshold, data source exists
Dominant score value in each attribute value is by standardization formula Sij_new=1-Sij/(Si_max+ 1) it determines；
Wherein, Sij_newIndicate dominant score value, SijWith the same sex value of common attribute value, S in data sourcei_maxIndicate maximum
Same sex value, 1≤i≤n, 1≤j≤n, wherein i, j, n be natural number, n indicate each subobject attribute value quantity.
Compared with the existing technology, object information processing method provided in this embodiment is by decomposing the general objective of data source
For multiple subobjects (establishing subobject collection), and then each subobject is decomposed into multiple attribute values (establishing property set), structure again
The pairwise comparison matrix for building each layer obtains the weight coefficient of each subobject and each attribute value, then further according to data derived from same
The standardization same sex of attribute value is worth dominant score value of the data source in each attribute value, finally weights to obtain by result above
The assessed value of data source.Since the above method is based on analytic hierarchy process (AHP) on the whole, data source value assessment can be improved
Accuracy.
Above description is only the general introduction of technical scheme, in order to better understand the technological means of the application,
And it can be implemented in accordance with the contents of the specification, and in order to allow the above and other objects, features and advantages of the application can
It is clearer and more comprehensible, it is special below to lift preferred embodiment, and cooperate attached drawing, detailed description are as follows.
Detailed description of the invention
Fig. 1 is the flow diagram of object information processing method provided by the embodiments of the present application.
Specific embodiment
Further to illustrate that the application is the technical means and efficacy reaching predetermined goal of the invention and being taken, below in conjunction with
Attached drawing and preferred embodiment, the specific implementation to the object information processing method, device and browser that are proposed according to the application
Mode, method, step, feature and its effect, detailed description are as follows.
Aforementioned and other technology contents, feature and effect in relation to the application refer to the preferable reality of schema in following cooperation
Applying in the detailed description of example can clearly appear from.By the explanation of specific embodiment, when can be that reach predetermined mesh to the application
The technical means and efficacy taken be able to more deeply and it is specific understand, however institute's accompanying drawings are only to provide with reference to and say
It is bright to be used, not it is used to limit the application.
Fig. 1 is the flow diagram for the object information processing method that the application first embodiment provides.The method includes
Following steps:
A kind of object information processing method, this method comprises:
The attribute value of step 1, the subobject for obtaining data source and each subobject.
Step 2, the subobject collection that data source is constructed according to total subobject of data source, according to each subobject of data source
Attribute value building data source property set, the subobject collection includes the corresponding subobject of data source, and the property set includes
The attribute value of each subobject of data source.
Step 3 creates the first matrix with the corresponding subobject in subobject intensive data source, with number in the property set
The second matrix is created according to the attribute value of each subobject in source, the matrix element of first matrix is the weight system of each subobject
It counts, the matrix element in the second matrix is the weight coefficient of the attribute value of each subobject.
Weight coefficient in first matrix and the second matrix is normalized step 4 by column, to normalizing
Change treated first matrix and the summation of the second matrix by rows, obtains normalizing characteristic value.
Step 5, by the normalizing characteristic value divided by the total quantity of homography element, obtain each subobject of data source
Weight coefficient corresponding to each attribute value in corresponding weight coefficient and each subobject.
Step 6 obtains the same sex value in data source with common attribute value, determines data source each according to the same sex value
Dominant score value in a attribute value.
Weight coefficient corresponding to step 7, each subobject by data source, data source each subobject in it is each
The dominant score value of weight coefficient corresponding to attribute value and data source in each attribute value weights to obtain the assessment of data source
Value.
1, according to the subobject collection of total subobject building data source of data source, according to the category of each subobject of data source
Property value building data source property set, the subobject collection includes the corresponding subobject of data source, and the property set includes data
The attribute value of each subobject in source.
2, the building for carrying out hierarchy Model, related each matrix element is divided from top to down according to different attribute
Solution is subordinated to one layer of matrix element or to upper at general objective layer, subobject collection and property set, all matrix elements of same layer
Layer matrix element has an impact, while dominating next layer of matrix element or the effect by lower layer's matrix element again.It needs to illustrate
, according to actual needs, more layers can also be divided into according to the attribute of matrix element, the present invention is not limited thereto.
3, the first matrix is created with the corresponding subobject in subobject intensive data source, with data source in the property set
The attribute value of each subobject create the second matrix, the matrix element in first matrix and the second matrix is in matrix
The weight coefficient of middle weight coefficient.
Construct above-mentioned matrix, the subobject collection since the 3rd layer of hierarchy Model, that is, in the application, at
The matrix is constructed to comparison method, property set is also similarly constructed.
One important feature of analytic hierarchy process (AHP) is exactly to represent two matrixes with the form of the ratio between importance degree two-by-two
The corresponding importance degree grade of element.Such as to a certain subobject (or link), each scheme under it is compared two-by-two, and
By its importance degree rating.It is denoted as the weight coefficient of i-th and jth matrix element.In this present embodiment, described relatively heavy
Want degree natural number 1 to 9 to indicate, 1 indicate one of matrix element in two matrix elements being compared to each other relative to
Relative importance is identical for another matrix element, and 9 indicate one of square in two matrix elements being compared to each other
Array element element relative importance for another matrix element is maximum, on the contrary then indicated with inverse.By comparing knot two-by-two
The matrix that fruit is constituted is referred to as pairwise comparison matrix (also referred to as judgment matrix).Pairwise comparison matrix has the property of positive reciprocal matrix,
That is (1) ai,,j>0；(2)ai,,j=1 (as i=j)；(3)ai,,j=1/xj,i(as i ≠ j), wherein ai,,jIn contrast with expression
Compared with matrix element corresponding with the i-th row, jth column matrix element in matrix, that is, important coefficient.Table two is right with the son
As the corresponding subobject in intensive data source (by taking three subobjects as an example) is the first matrix that matrix element constructs.Table three is with institute
The attribute value (by taking three attribute values as an example) for stating each subobject of data source in property set is the second square of matrix element building
Battle array.In general, in pairwise comparison matrix, corresponding matrix element be it is symmetrically arranged, as shown in table two and table three, lateral table
Head is symmetrically arranged with the matrix element in longitudinal gauge outfit.
4, the weight coefficient in first matrix and the second matrix is normalized by column.
5, to after normalized first matrix and the second matrix press capable summation again, obtain normalizing characteristic value.
6, by the normalizing characteristic value divided by the total quantity of homography element, each subobject institute for obtaining data source is right
Weight coefficient corresponding to each attribute value in weight coefficient and each subobject answered.
For the first matrix, above-mentioned steps can use formulaWherein, ωiIndicate certain data
The weight coefficient of i-th of source subobject, 1≤i≤n, 1≤j≤n, wherein i, j, n be natural number, n indicate each subobject
The quantity of attribute value.xi,,jIndicate important coefficient corresponding with the i-th row, jth column matrix element in the first matrix.At normalization
Reason can reduce the order of magnitude and unit to be influenced to assessment bring.
For the second matrix, above-mentioned steps can use formulaWherein, σiIndicate certain data source
The weight coefficient of i-th of subobject, 1≤i≤n, 1≤j≤n, 1≤k≤n, wherein i, j, k, n are natural number, and n indicates each height
Object attribute value quantity.yi,,jIndicate important coefficient corresponding with the i-th row, jth column matrix element in the first matrix.
xi,,jWith yi,,jIt can be provided for particular problem by experienced, good sense the expert of every field.
7, the same sex value in data source with common attribute value is obtained, and determines data source each according to the same sex value
Dominant score value in attribute value.
The step 7 can further include steps of
Judge the same sex value in data source with common attribute value for absolute number or ratio.
If it is absolute number, step is carried out: the same sex value in data source with common attribute value is ranked up, take out
Maximum same sex value is simultaneously standardized it, and the same sex value after the standardization is determined as data source in each attribute value
Dominant score value.
If it is ratio, step is carried out: the same sex value in data source with common attribute value is ranked up, take out most
Big same sex value simultaneously directly determines it as dominant score value of the data source in each attribute value.
Further, step 7 is further comprising the steps of: if the same sex value in data source with common attribute value is absolute
Number carries out step before being standardized first:
When the numerical value of the same sex value in the data source with common attribute value is more than the first preset threshold, data source exists
Dominant score value in each attribute value is by standardization formula Sij_new=Sij/(Si_max+ 1) it determines；
When the numerical value of the same sex value in the data source with common attribute value is lower than the second preset threshold, data source exists
Dominant score value in each attribute value is by standardization formula Sij_new=1-Sij/(Si_max+ 1) it determines；
Wherein, Sij_newIndicate dominant score value, SijWith the same sex value of common attribute value, S in data sourcei_maxIndicate maximum
Same sex value, 1≤i≤n, 1≤j≤n, wherein i, j, n be natural number, n indicate each subobject attribute value quantity.
When the numerical value of the same sex value in the data source with common attribute value is more than the first preset threshold, the same sex
Value is known as positive attribute value, when the numerical value of the same sex value in the data source with common attribute value is lower than the second preset threshold
When, the same sex value is known as reverse attribute value, and it is generation that the forward direction attribute value, which is also referred to as profit evaluation model attribute value or hopes large attribute value,
Upwards or the attribute value that advances, increase, these attribute value values are bigger to be evaluated better table, and reverse attribute value is then opposite.Mark
Nondimensionalization when standardization can be absolute number to avoid attribute value.
8, each attribute in each subobject of weight coefficient, data source corresponding to each subobject by data source
The dominant score value of value corresponding weight coefficient and data source in each attribute value weights to obtain the assessed value of data source.
The process specifically weighted for example can be by the dominant score value of each attribute value of each data source multiplied by its attribute
The weight coefficient of value is multiplied by the weight coefficient of its corresponding subobject, and then the assessment point of the data source can be obtained in aggregation
Number.
In the present embodiment, the method also includes: sort the assessed value of data source to obtain best source.Namely
Assessed value is ranked up, the corresponding data source of assessment point to make number one, as best source are found.
Compared with the existing technology, object information processing method provided in this embodiment is by decomposing the general objective of data source
For multiple subobjects (establishing subobject collection), and then each subobject is decomposed into multiple attribute values (establishing property set), structure again
The pairwise comparison matrix for building each layer obtains the weight coefficient of each subobject and each attribute value, then further according to data derived from same
The standardization same sex of attribute value is worth dominant score value of the data source in each attribute value, finally weights to obtain by result above
The assessed value of data source.Since the above method is based on analytic hierarchy process (AHP) on the whole, assessment result can be more scientific, quasi-
Really.
The above is only the preferred embodiment of the application, not makes any form of restriction to the application, though
Right the application has been disclosed in a preferred embodiment above, however is not limited to the application, any technology people for being familiar with this profession
Member, is not departing within the scope of technical scheme, when the technology contents using the disclosure above are modified or are modified
For the equivalent embodiment of equivalent variations, but all technical spirits pair without departing from technical scheme content, according to the application
Any simple modification, equivalent change and modification made by above embodiments, in the range of still falling within technical scheme.
Claims (2)
1. a kind of object information processing method, which is characterized in that this method comprises:
Obtain the subobject of data source and the attribute value of each subobject；
According to the subobject collection of total subobject building data source of data source, according to the attribute value structure of each subobject of data source
The property set of data source is built, the subobject collection includes the corresponding subobject of data source, and the property set includes each of data source
A sub- object's property value；
Create the first matrix with the corresponding subobject in subobject intensive data source, in the property set data source it is each
The attribute value of subobject creates the second matrix, and the matrix element of first matrix is the weight coefficient of each subobject, the second square
Matrix element in battle array is the weight coefficient of the attribute value of each subobject；
Weight coefficient in first matrix and the second matrix is normalized by column, after normalized
First matrix and the summation of the second matrix by rows, obtain normalizing characteristic value；
By the normalizing characteristic value divided by the total quantity of homography element, power corresponding to each subobject of data source is obtained
Weight coefficient corresponding to each attribute value in weight coefficient and each subobject；
The same sex value in data source with common attribute value is obtained, determines data source in each attribute value according to the same sex value
Dominant score value；
Weight coefficient corresponding to each subobject by data source, data source each subobject in each attribute value institute it is right
Dominant score value of the weight coefficient and data source answered in each attribute value weights to obtain the assessed value of data source.
2. the method as described in claim 1, it is characterised in that:
When the numerical value of the same sex value in the data source with common attribute value is more than the first preset threshold, data source is each
Dominant score value in attribute value is by standardization formula Sij_new=Sij/(Si_max+ 1) it determines；
When the numerical value of the same sex value in the data source with common attribute value is lower than the second preset threshold, data source is each
Dominant score value in attribute value is by standardization formula Sij_new=1-Sij/(Si_max+ 1) it determines；
Wherein, Sij_newIndicate dominant score value, SijWith the same sex value of common attribute value, S in data sourcei_maxIndicate the maximum same sex
Value, 1≤i≤n, 1≤j≤n, wherein i, j, n be natural number, n indicate each subobject attribute value quantity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811300541.0A CN109669939A (en) | 2018-11-02 | 2018-11-02 | Object information processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811300541.0A CN109669939A (en) | 2018-11-02 | 2018-11-02 | Object information processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109669939A true CN109669939A (en) | 2019-04-23 |
Family
ID=66141724
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811300541.0A Pending CN109669939A (en) | 2018-11-02 | 2018-11-02 | Object information processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109669939A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112148822A (en) * | 2020-08-28 | 2020-12-29 | 中国地质大学（武汉） | Fine-grained attribute weighting method and system |
-
2018
- 2018-11-02 CN CN201811300541.0A patent/CN109669939A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112148822A (en) * | 2020-08-28 | 2020-12-29 | 中国地质大学（武汉） | Fine-grained attribute weighting method and system |
| CN112148822B (en) * | 2020-08-28 | 2024-04-19 | 中国地质大学（武汉） | Fine granularity attribute weighting method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200065619A1 (en) | Image processing method, processing apparatus and processing device | |
| Yadav et al. | Development and validation of standard area diagrams to aid assessment of pecan scab symptoms on fruit | |
| CN103942568B (en) | A kind of sorting technique based on unsupervised feature selection | |
| DE102017011260A1 (en) | Tagging large images using a picture-with-theme embedding | |
| DE102017011262A1 (en) | Theme linking and marking for dense images | |
| US20160132750A1 (en) | Local feature representation for image recognition | |
| CN107610123A (en) | A kind of image aesthetic quality evaluation method based on depth convolutional neural networks | |
| CN110032737A (en) | A kind of boundary combinations name entity recognition method neural network based | |
| CN110059716B (en) | Construction of CNN-LSTM-SVM network model and MOOC conjugate prediction method | |
| Sabrol et al. | Intensity based feature extraction for tomato plant disease recognition by classification using decision tree | |
| CN106446566A (en) | Elderly cognitive function classification method based on random forest | |
| DE112020001105T5 (en) | DETECTING A GENETIC MUTATION USING DEEP LEARNING | |
| CN112257025A (en) | Combined weighting method and system for comprehensive evaluation of driving performance of engineering vehicle | |
| WO2023125456A1 (en) | Multi-level variational autoencoder-based hyperspectral image feature extraction method | |
| CN109711401A (en) | A kind of Method for text detection in natural scene image based on Faster Rcnn | |
| CN104951987B (en) | Crop Breeding evaluation method based on decision tree | |
| CN107067182A (en) | Towards the product design scheme appraisal procedure of multidimensional image | |
| CN106033554A (en) | Big data processing method for two-stage depth learning model based on sectionalization | |
| CN106934000A (en) | A kind of automatic quality detecting method of the voice of calling system and system | |
| US11281938B2 (en) | Image processing method, processing apparatus and processing device | |
| CN105787488A (en) | Image feature extraction method and device realizing transmission from whole to local | |
| CN110781831A (en) | Hyperspectral optimal waveband selection method and device based on self-adaption | |
| CN111783516A (en) | Ploughing quality natural grade evaluation method based on deep learning | |
| CN113011386A (en) | Expression recognition method and system based on equally divided characteristic graphs | |
| Tanwar et al. | Red rot disease prediction in sugarcane using the deep learning approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190423 |
|
| WD01 | Invention patent application deemed withdrawn after publication |