CN115018357A - Farmer portrait construction method and system for production performance improvement - Google Patents
Farmer portrait construction method and system for production performance improvement Download PDFInfo
- Publication number
- CN115018357A CN115018357A CN202210717234.2A CN202210717234A CN115018357A CN 115018357 A CN115018357 A CN 115018357A CN 202210717234 A CN202210717234 A CN 202210717234A CN 115018357 A CN115018357 A CN 115018357A
- Authority
- CN
- China
- Prior art keywords
- farmer
- production performance
- index
- portrait
- production
- 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
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 138
- 238000010276 construction Methods 0.000 title claims abstract description 21
- 230000006872 improvement Effects 0.000 title claims description 21
- 238000000034 method Methods 0.000 claims abstract description 60
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 49
- 238000007781 pre-processing Methods 0.000 claims abstract description 24
- 230000011218 segmentation Effects 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims description 28
- 238000011156 evaluation Methods 0.000 claims description 21
- 238000011160 research Methods 0.000 claims description 20
- 238000005070 sampling Methods 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 13
- 230000000694 effects Effects 0.000 claims description 12
- 230000002068 genetic effect Effects 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 9
- 238000005192 partition Methods 0.000 claims description 8
- 238000007621 cluster analysis Methods 0.000 claims description 7
- 238000002922 simulated annealing Methods 0.000 claims description 7
- 238000012800 visualization Methods 0.000 claims description 6
- 238000011835 investigation Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 4
- 230000002441 reversible effect Effects 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 3
- 238000010438 heat treatment Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 abstract description 12
- 230000003993 interaction Effects 0.000 abstract description 4
- 238000004140 cleaning Methods 0.000 description 16
- 238000010586 diagram Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000007418 data mining Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 238000012271 agricultural production Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 230000014509 gene expression Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 230000035772 mutation Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 210000000349 chromosome Anatomy 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 108090000623 proteins and genes Proteins 0.000 description 3
- 238000013139 quantization Methods 0.000 description 3
- 238000000137 annealing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000012248 genetic selection Methods 0.000 description 2
- 238000013450 outlier detection Methods 0.000 description 2
- 230000002028 premature Effects 0.000 description 2
- 238000010992 reflux Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012272 crop production Methods 0.000 description 1
- 238000013479 data entry Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000005429 filling process Methods 0.000 description 1
- 238000011905 homologation Methods 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000013173 literature analysis Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000005477 standard model Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Agronomy & Crop Science (AREA)
- Mining & Mineral Resources (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a farmer portrait construction method and a farmer portrait construction system for improving production performance, wherein the method comprises the following steps: acquiring a basic data set; the basic data set is used for characterizing the characteristics and production performance of a producer; determining a characteristic index of production performance, and performing data preprocessing on the basic data set based on the characteristic index to obtain a complete sample set to be analyzed; designing a farmer label system according to the complete sample set based on an optimized fuzzy clustering algorithm, and performing clustering segmentation on farmer groups to obtain a farmer segmentation result based on production performance characteristics; and constructing a farmer portrait according to the farmer subdivision result. The invention can realize the functions of dynamically displaying data, performing real-time interaction with a user and the like, can solve the practical problem of agriculture and broaden the application of user portrait in the field of agriculture.
Description
Technical Field
The invention relates to the technical field of intersection of data mining and agricultural technology popularization, in particular to a farmer portrait construction method and system for production performance improvement.
Background
In the current agricultural production of China, small-scale producers, namely small farmers, are used as main parts, and different farmers have greater heterogeneity in the aspects of personal characteristics, production conditions, technical level and the like. Heterogeneity among farmers inevitably influences technical adoption and application, information acquisition and utilization in production, and finally influences production performance and production quality. The heterogeneity characteristics of the peasant household are analyzed, the relation between the characteristic attributes of the peasant household and the production performance of the peasant household is excavated, and the method is beneficial for an agricultural technology popularization department to provide targeted, personalized and accurate technical training, information transmission and other technical services for different peasant household groups, so that the agricultural technology popularization effect is improved, the production performance of the peasant household is improved, and the high-quality development of the industry is promoted.
Farmer analysis and agricultural production performance research has traditionally generally been based on classical agricultural production measurement models. With the rapid development of information technology, new research ideas and methods are brought to agricultural production management research by emerging technologies such as artificial intelligence and data mining.
At present, researches combining a data mining technology and the production performance of farmers are few, and the researches of the image of the farmers for improving the production performance of the farmers are not retrieved yet.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a farmer portrait construction method and system for improving production performance.
In order to achieve the purpose, the invention provides the following scheme:
a farmer portrait construction method for production performance improvement comprises the following steps:
acquiring a basic data set; the basic data set is used for characterizing the characteristics and production performance of a producer;
determining a characteristic index of production performance, and performing data preprocessing on the basic data set based on the characteristic index to obtain a complete sample set to be analyzed;
designing a farmer label system according to the complete sample set based on an optimized fuzzy clustering algorithm, and performing clustering segmentation on farmer groups to obtain a farmer segmentation result based on production performance characteristics;
and constructing a farmer portrait according to the farmer subdivision result.
Preferably, the acquiring the basic data set comprises:
sampling by a multi-stage sampling method by combining the regional layout of various crops and the actual situation of the production process to obtain a sample farmer;
designing a farmer investigation scheme according to the sample farmer;
acquiring basic data according to the farmer research scheme; the basic data is used for characterizing the characteristics and the production performance of a producer;
and carrying out data preprocessing on the basic data to obtain the basic data set.
Preferably, the data preprocessing includes outlier detection and missing value processing.
Preferably, the determining a characteristic indicator of production performance and performing data preprocessing on the basic data set based on the characteristic indicator to obtain a complete sample set to be analyzed includes:
the preset characteristic indexes of the production performance can be divided into a first index based on output effect evaluation and a second index based on technical efficiency evaluation;
the first index can be obtained by algebraic operation on the basic data set. The second index can be obtained by measuring and calculating through a DEA-BCC model, and comprises a technical efficiency index based on value output and a technical efficiency index based on physical output.
Carrying out syntropy processing on reverse indexes in the first indexes and the second indexes; the formula of the index homodromous treatment is Y ij ＝max(X j )-X ij (ii) a Wherein, max (X) j ) Is the maximum value of the index j in all samples, X ij Is the original value of the index j on the sample i, Y ij Taking a value of an index j of the sample i after the homotropic treatment;
index number for index homodromous processingCarrying out standardization treatment; the formula of the normalization process is
Wherein mu and sigma are respectively a sample mean value and a sample mean square error; the normalized data is the complete sample set.
Preferably, the method for designing a farmer label system according to the complete sample set based on the optimized fuzzy clustering algorithm and performing cluster segmentation on the farmer population to obtain a farmer segmentation result based on the production performance characteristics includes:
determining the number of peasant household clusters based on the production performance characteristic index by using a fuzzy clustering algorithm and a clustering effectiveness index;
determining a partition matrix according to the cluster number and the complete sample set, respectively finding out R samples which are nearest to the same class and different classes from the partition matrix, and calculating to obtain the weight of each performance characteristic according to the distinguishing capability of the characteristic on close-range samples;
and performing genetic iteration and evolution on each performance characteristic according to a genetic algorithm fusion simulated annealing algorithm to obtain the segmentation result of the peasant household.
Preferably, the constructing a farmer portrait according to the farmer segmentation result includes:
constructing a farmer label system;
according to the formula
Quantifying the labels in the farmer label system; wherein G is i,s Scoring the labels under the S dimensionality of the ith type of farmers;
the average value of the jth characteristic belonging to the ith class of farmers; d jmax 、d jmin And
the j characteristic maximum, the j characteristic minimum and the average value are respectively; wherein the content of the first and second substances,t features, G, in the S-th dimension i,s Is in the interval [0,1 ]]Performing the following steps;
and carrying out visualization processing on the farmer portrait according to the quantized indexes.
Preferably, the farmer label system comprises a personal information label, a production condition label and a performance evaluation label; the personal information type label is used for describing farmer personal attributes of grower types, production specialization degree, age and education level; the production condition type label is used for reflecting planting information of a farmer cultivation area, a cultivation mode and a production type and is also used for reflecting the production behavior state of the farmer; the performance evaluation class labels are obtained from the clustering and subdividing results, and are used for subdividing and optimizing a farmer performance index system to obtain farmer groups with different production performances.
Preferably, the visualization process comprises an intra-class farmer portrait and an inter-class farmer portrait; the intra-class farmer portrait is used for displaying planting information of various farmers through a word cloud picture and a regional heating power map; the inter-class farmer portrait is used for comparing radar maps with grouping column maps so as to show the comparison effect of various farmers on the same label quantity.
A farmer portrait construction system for production performance improvement comprises:
the data acquisition module is used for acquiring a basic data set; the basic data set is used for characterizing the characteristics and production performance of a producer;
the data processing and measuring and calculating module is used for determining a characteristic index of production performance and carrying out data preprocessing on the basic data set based on the characteristic index to obtain a complete sample set to be analyzed;
the cluster analysis module is used for designing a farmer label system according to the complete sample set based on an optimized fuzzy clustering algorithm and performing cluster subdivision on farmer groups to obtain a farmer subdivision result based on production performance characteristics;
and the image generation module is used for constructing a farmer image according to the farmer subdivision result.
Preferably, the data acquisition module specifically includes:
the sampling unit is used for sampling by a multi-stage sampling method according to the regional layout and the actual production process conditions of various crops to obtain a sample farmer;
the scheme design unit is used for designing a farmer research scheme according to the sample farmer;
the data acquisition unit is used for acquiring basic data according to the farmer research scheme; the basic data is used for characterizing the characteristics and the production performance of a producer;
and the preprocessing unit is used for preprocessing the basic data to obtain the basic data set.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention provides a farmer portrait construction method and a farmer portrait construction system for improving production performance, wherein the method comprises the following steps: acquiring a basic data set; the basic data set is used for characterizing the characteristics and production performance of a producer; determining a characteristic index of production performance, and performing data preprocessing on the basic data set based on the characteristic index to obtain a complete sample set to be analyzed; in consideration of the characteristics that the traditional fuzzy clustering algorithm lacks feature weighting and randomness of an initial clustering center, the fuzzy clustering algorithm is optimized and instantiated to a farmer data set to obtain a farmer subdivision result based on production performance; and constructing a farmer portrait according to the farmer subdivision result. The invention can realize the functions of dynamically displaying data, performing real-time interaction with a user and the like, can solve the practical problem of agriculture and broaden the application of user portrait in the field of agriculture.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow chart of a farmer portrait construction method in an embodiment of the present invention;
FIG. 2 is a schematic diagram of the method steps in an embodiment provided by the present invention;
FIG. 3 is a schematic diagram of an abnormal value identification and detection process in an embodiment of the present invention;
FIG. 4 is a schematic diagram of missing value processing and filling flow in an embodiment provided by the present invention;
FIG. 5 is a schematic flow chart of an optimized fuzzy clustering algorithm in an embodiment provided by the present invention;
FIG. 6 is a schematic diagram of a farmer label system group-like image structure in an embodiment provided by the present invention;
fig. 7 is a system module connection diagram in an embodiment provided by the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a farmer portrait construction method and system for production performance improvement, which can realize functions of dynamically displaying data, performing real-time interaction with a user and the like, solve the practical problems of agriculture and widen the application of the farmer portrait in the agricultural field.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a flowchart of a farmer portrait construction method in an embodiment provided by the present invention, and as shown in fig. 1, the present invention provides a farmer portrait construction method for production performance improvement, which includes:
step 100: acquiring a basic data set; the basic data set is used for characterizing the characteristics and production performance of a producer;
step 200: determining a characteristic index of production performance, and performing data preprocessing on the basic data set based on the characteristic index to obtain a complete sample set to be analyzed;
step 300: designing a farmer label system according to the complete sample set based on an optimized fuzzy clustering algorithm, and performing clustering segmentation on farmer groups to obtain a farmer segmentation result based on production performance characteristics;
step 400: and constructing a farmer portrait according to the farmer subdivision result.
Fig. 2 is a schematic diagram of the method steps in the embodiment provided by the present invention, and as shown in fig. 2, the present invention further provides an implementation process of a farmer portrait method for improving production performance, which specifically includes:
s1: acquiring farmer characteristics and production data, and cleaning the data to obtain a basic data set;
s2: determining characteristic indexes of production performance of farmers, quantitatively measuring and calculating the production performance, and carrying out syntropy and standardization treatment on the indexes to obtain a complete sample set to be analyzed;
s3: and (5) carrying out production performance group segmentation on the basis of the optimized fuzzy clustering. Considering the defects that the traditional fuzzy clustering initial center selects the local optimum and the contribution distribution of the features is uneven, the contribution of the effective features to the clustering is strengthened and the feature dimension is reduced by assigning differentiated weights to the features; and a heuristic algorithm is utilized to realize global search of the sample set, and the performance of a farmer population clustering algorithm is improved.
S4: and generating a farmer portrait based on the production performance clustering segmentation result. And (3) obtaining a farmer group with different production performance expressions according to the clustering and subdividing results, and constructing a three-dimensional farmer label system: the system comprises a personal information label, a production condition label and a performance evaluation label, and is characterized by depicting inter-class images and intra-class images of farmer groups with different performance expressions, and visually displaying the portrait result.
Preferably, the step 100 comprises:
sampling by a multi-stage sampling method by combining the regional layout of various crops and the actual situation of the production process to obtain a sample farmer;
designing a farmer investigation scheme according to the sample farmer;
acquiring basic data according to the farmer research scheme; the basic data is used for characterizing the characteristics and the production performance of a producer;
and carrying out data preprocessing on the basic data to obtain the basic data set.
Specifically, in the embodiment, high yield is considered to be no longer the only index of agricultural attention, and yield cannot comprehensively measure the yield quality and benefit level of farmlands. By combining the universality of the research of farmers and the particularity of crop production, the attribute characteristics of the farmers, which comprehensively reflect the heterogeneity of the farmers, are screened through literature analysis and field investigation.
Preferably, the data preprocessing includes outlier detection and missing value processing.
Further, in this embodiment, the step S1 further includes:
s11: sampling by a multi-stage sampling method in combination with the regional layout and the actual production process conditions of various crops, wherein a focus sampling method is adopted in province and county villages and town villages, and village farmers adopt a random sampling mode to finally select sample farmers nationwide;
s12: designing a farmer research scheme, wherein the research content comprises the following steps: characteristic information of farmer attributes (geographical position of farmer, cultivation area, cultivation mode, agricultural product usage, type of grower, whether to join cooperative, education level, etc.) and production performance data (variety and unit yield, quantity and price of each material cost and labor cost input, yield, sale price, etc.);
s13: carrying out farmer research to obtain basic data capable of representing characteristics and production performance of a producer;
s14: performing data cleaning such as abnormal value detection, missing value processing and the like on the acquired original data; wherein the abnormal value detection mainly comprises physical discrimination and model discrimination; in the filling process of the missing value, the type of the missing data is distinguished firstly, and then the missing data is processed in three conditions; filling the vacancy values by adopting constant filling, mean filling or maximum expectation and other possible filling methods aiming at the vacancy values needing to be filled; a base data set is obtained.
Preferably, the step 200 comprises:
the preset characteristic indexes of the production performance can be divided into a first index based on output effect evaluation and a second index based on technical efficiency evaluation;
the first index can be obtained by algebraic operation on the basic data set. The second index can be obtained by measuring and calculating through a DEA-BCC model, and comprises a technical efficiency index based on value output and a technical efficiency index based on physical output.
Carrying out syntropy processing on reverse indexes in the first indexes and the second indexes; the formula of the index homodromous treatment is Y ij ＝max(X j )-X ij (ii) a Wherein, max (X) j ) Is the maximum value of the index j in all samples, X ij Is the original value of the index j on the sample i, Y ij Taking a value of an index j of the sample i after the homotropic treatment;
carrying out standardization processing on the index data subjected to index syntropy processing; the formula of the normalization process is
Wherein mu and sigma are respectively a sample mean value and a sample mean square error; the normalized data is the complete sample set.
Specifically, in S2 of this embodiment, the data preprocessing is completed by performing abnormal value detection, missing value processing, and index homologation processing in S1, so as to obtain a complete and effective data set, which is called a complete data set; the structure of the abnormal value identification and detection method of the present invention is shown in fig. 3 and 4. In order to ensure the data mining effect from a data source, the method carries out cleaning pretreatment on the farmer data acquired by investigation before data mining, and mainly comprises two aspects of abnormal value detection and vacancy value treatment. The abnormal value detection mainly comprises two methods of physical discrimination and model discrimination; the deletion value processing scheme is as follows: firstly, distinguishing missing data types, and then processing the missing data types in three conditions; and aiming at the vacancy values needing to be filled, filling the vacancy values by adopting methods which can be filled by constant filling, mean filling or maximum expectation and the like.
The step S2 further includes:
s21: the production performance characteristic indexes are divided into two categories of characteristic indexes based on output effect evaluation and characteristic indexes based on technical efficiency evaluation, and are further subdivided into 19 characteristic indexes of five dimensional layers of cost input indexes, income output indexes, benefit indexes, output value efficiency and output efficiency;
s22: the peasant household production performance measurement and calculation based on the output effect evaluation can be obtained through algebraic operation of research data, and for the purpose that performance indexes of different peasant households are comparable, the following indexes are all converted into output in unit area.
In order to make the production performance indexes of different farmers comparable, all the indexes are converted into index values corresponding to unit area:
the cost input indexes embody the input conditions in the process of planting crops by farmers, the cost input indexes comprise 6 secondary indexes, and fixed input, land input, material input, manual input and total input are selected;
the income output index reflects the output condition of farmers in the crop planting process, and comprises 3 secondary indexes, wherein the net value input ratio of agricultural products, the average yield of crops and the cultivation income per unit area are selected.
The benefit indexes reflect the operation and management conditions in the agricultural product planting process, and comprise 4 secondary indexes, namely the input-output rate of the agricultural product physical quantity, the input-output rate of the net value of the agricultural product, the labor productivity of farmers and the return on investment.
S23: the technical efficiency performance indexes are measured and calculated by a data analysis method based on a front-edge surface, specifically, a DEA-BCC model based on scale reward change is adopted for measuring and calculating, so that two sets of technical efficiency indexes based on yield output and yield output are obtained, wherein the two sets of technical efficiency indexes respectively comprise technical efficiency, scale efficiency and pure technical efficiency;
considering that a non-parameter method in the efficiency measuring and calculating method can avoid result deviation caused by the deviation of a previously assumed variable relation, the performance measuring and calculating of farmers adopt a data envelope analysis method of a front surface. The invention adopts a DEA-BCC model based on scale change, and is represented as follows:
wherein j is selected each production unit DMU j (j ═ 1,2, 3.., n), m, s represent the number of input and output indices, respectively, v and u represent the weight of each input and output term i (i ═ 1,2,. multidot., m) and u r (r ═ 1, 2.., s) denotes the weight of each input/output term. Converting the model into a mathematical programming problem, then expressing as:
in this model, X j And Y j Respectively representing the input vector and the output vector of the jth decision unit, lambda j A weight value representing a decision unit is used,
and
for the dual variable added, θ represents the technical efficiency of the decision unit, and when θ is 1, it indicates that the production unit satisfies the technical validity, whereas when it is less than 1, it does not satisfy the technical validity. The method comprises the steps of taking manual input, land input, garden building input and material data input as input vectors of a decision unit, taking output values and output values as output vectors respectively, calculating production technical efficiency of farmers, and obtaining two sets of technical efficiency indexes based on output and output value output, wherein the two sets of technical efficiency indexes respectively comprise technical efficiency, scale efficiency and pure technical efficiency.
S24: the method comprises the following steps of (1) carrying out equidirectional standardization processing on indexes, wherein in a farmer production performance index system, most of indexes are in positive correlation with production performance, and the larger the index value is, the higher the production performance is, the indexes are called as positive indexes, such as labor productivity, technical efficiency and the like; cost indicators such as labor costs and material costs are negatively related to production performance, and the smaller the index value is, the higher the production performance is, and these indicators are called reverse indicators. In order to ensure that all data are in a proper interval and are not influenced by the index dimension and the sample correlation, the index is subjected to homodromous normalization processing between data analysis, and the method comprises the following steps:
in the formula, cor (X) j ) The correlation between the representative index j and the production performance, X ij For the original value of the index j on the sample i, μ and σ represent the mean and variance, max (X), respectively, after the sample is homologated j ) Represents the maximum value of the index j in all samples, Y ij And taking a value of the index j for the sample i after data preprocessing.
Preferably, the step 300 comprises:
determining the number of peasant household clusters based on the production performance characteristic index by using a fuzzy clustering algorithm and a clustering effectiveness index;
determining a partition matrix according to the cluster number and the complete sample set, respectively finding out R samples of the same type and different types of nearest neighbors from the partition matrix, and calculating to obtain the weight of each performance characteristic according to the distinguishing capability of the characteristic on the close-range samples;
and performing genetic iteration and evolution on each performance characteristic according to a genetic algorithm fusion simulated annealing algorithm to obtain the segmentation result of the peasant household.
Specifically, the conventional fuzzy clustering algorithm in this embodiment is a clustering algorithm for determining the degree of each sample point belonging to a certain cluster based on the membership degree, and the basic idea is as follows: n sample points X (X) 1 ,x 2 ,...,x n ) Dividing into c (c is more than or equal to 2 and less than or equal to n) fuzzy classes, calculating the cluster center of each class, and updating the cluster center continuously by iterationThe heart and membership matrix minimizes the objective function. Wherein an objective function is defined:
in the formula: u ═ U ik ]Is a membership matrix, u ik The membership degree of the kth sample belonging to the ith class; v ═ V i ]As a cluster center matrix, v i Is the i-th class center; m is a weighting index; d ik ＝||x k -v i I is sample x k And the clustering center v i Distance paradigm of (d). The fuzzy membership u can be solved by using a Lagrangian function ik And a clustering center v i And (4) matrix.
However, the traditional fuzzy clustering algorithm has the following defects in practical application: the influence of different characteristics of the sample on classification is not considered, and the importance of the characteristics in the clustering process cannot be highlighted; meanwhile, the defects that the traditional fuzzy clustering initial center selects the easily-brought local optimum and the distribution of the feature contribution degree is not uniform are not considered.
Therefore, the invention optimizes the problems on the basis of the traditional fuzzy clustering algorithm, provides an optimized fuzzy clustering algorithm, and has the flow shown in fig. 5.
Further, in this embodiment, the step S3 further includes:
s31: and determining the initial optimal clustering number and determining the K value through a clustering effectiveness function. The method adopts four clustering effectiveness indexes of DB, CH, XB and contour coefficient, and preliminarily determines the number (K value) of the farmers based on the production performance characteristic index by using a fuzzy clustering algorithm and the clustering effectiveness indexes.
S32: the weighting of the performance characteristic attributes, the importance of each index in a farmer production performance characteristic index system is different, and the contribution degree to the clustering result is different, so that a corresponding weight needs to be introduced into each characteristic, different weights are given to the characteristics according to the relevance of the characteristics and the categories, and the contribution of the important characteristics to the clustering result is highlighted. Selecting a sample x according to the clustering result preliminarily determined in S31 i And respectively find x and x from the partition matrix i And calculating the weight of each performance characteristic according to the distinguishing capability of the close-range samples by the R samples of the same type and different types, realizing the weighting of the performance characteristic attribute, and then carrying out optimized clustering.
There are several samples of different classes, called X for each class of samples n . The process of entitlements is described as follows:
1. from all the samples, one sample a was randomly drawn.
2. Within a group of samples classified identically to sample a, k nearest neighbor samples are taken.
3. In all other sample groups classified differently from sample a, k nearest neighbor samples are also taken, respectively.
4. The weight of each feature is calculated. The weights for each feature are:
in general, the weighting process is adapted to feature-weight supervised learning methods with class labels, whereas in the cluster analysis of the present invention, the class label of each sample is unknown. We can getFirstly, clustering samples for one time, and selecting a sample x with large membership degree i And respectively find x and x from the partition matrix i R samples of similar and different nearest neighbors are calculated according to the method. And after the final weight is obtained, weighting each dimension characteristic by using the weight, and then carrying out optimized clustering.
After obtaining the optimal feature subset vector weight value, carrying out weighting processing on the Euclidean distance formula, and correcting the formula (4) to obtain a new objective function:
in the formula: ". x represents an array multiplication, i.e. multiplication by the corresponding element of the weft array. The fuzzy membership u can be solved again by applying a Lagrangian function ik And a clustering center v i And (4) matrix.
S33: introducing a genetic algorithm to optimize a fuzzy clustering algorithm, optimizing a population through the genetic algorithm, encoding possible values in a space to be solved into chromosomes, establishing an initial population, then carrying out genetic selection according to fitness, and carrying out cross and variation according to a certain probability. And (4) circularly executing genetic selection, crossing and mutation operations until the end condition is met.
In the embodiment, the initial clustering center is selected, the initial clustering center of the fuzzy clustering algorithm is determined to have randomness in consideration of fluctuation of performance of farmers, and if the initial clustering center is selected improperly, the initial clustering center is likely to fall into a local optimal condition or the iteration times are increased. Therefore, the initial cluster center of the fuzzy clustering algorithm is optimized by combining the simulated genetic annealing algorithm.
Optionally, the genetic algorithm is a computing model based on a natural evolution system, and is widely applied to optimization solution in the engineering field by combining biological evolution and random exchange theory to adaptively search an optimal solution. However, researches show that the basic genetic algorithm has the problems of poor global search capability, premature convergence and the like.
Further, the SA-based optimization algorithm idea is described as follows:
1. initializing parameters, and setting the size sizepop of the population, the maximum evolution times MAXGEN and the cross probability P c Probability of mutation P m 。
2. And calculating the fitness of the individual, wherein the higher the fitness is, the higher the possibility that the individual gene is inherited to the next generation is. We define that the better clustering result should be: the similarity between samples belonging to the same class is as high as possible, and the difference between samples of different classes is as large as possible. When Euclidean distance is taken as the basis for measuring "similarity", the above expression can be converted into the following criterion function:
the clustering problem is converted into a minimization problem of formula (1) or a maximization problem of formula (2). And (3) solving the clustering problem by using a genetic algorithm, wherein the formula (3) can be selected as the fitness of the individual, and the higher the fitness value is, the higher the probability that the individual gene enters the next generation is.
In the formula:
is the attribution class c j The kth sample, n j Is the number of samples in class j, m j Is a class c j M is the center of all samples.
3. Let the loop count variable gen equal to 0.
4. Selecting, crossing and mutating the initial population, and calculating the clustering center of newly generated individuals and the fitness f of each individual i '。
Preferably, the population crossing operation uses a real number crossing method, the kth chromosome a k And the l-th chromosome a l The method of the interleaving operation at j bit is as follows:
further, in the invention, the roulette method is selected by the group selection operation, namely based on the selection of fitness proportion, the probability of selecting the individual i is as follows:
further, the j gene a of the i individual ij The operation method for carrying out mutation comprises the following steps:
5. if gen is less than MAXGEN, then gen is gen +1, and go to step 4; otherwise, go to step 6.
S34: and a simulated annealing algorithm is fused, and a selection process of simulated annealing is applied after mutation operation, so that the number of attempts at a local optimal position is reduced, and the evolution speed is accelerated. Furthermore, the fuzzy clustering algorithm improvement considering the feature importance difference and the clustering algorithm performance at the same time is realized, and the fuzzy clustering algorithm improvement can be used for clustering segmentation of the farmer group to obtain the farmer segmentation result based on the production performance characteristics.
The simulated annealing algorithm compares the solving process of a certain type of optimization problem with the statistical thermodynamic equilibrium problem, and the optimal solution is obtained by jumping the algorithm from a local optimal trap by adopting a Metropolis acceptance criterion. However, the direct use of simulated annealing leads to a sharp increase in the amount of calculation. The optimization algorithm combining GA and SA provided by the invention can make up for deficiencies of each other, not only effectively overcomes the premature phenomenon of the traditional GA, but also can avoid the defect of long SA calculation time, so that the optimized fuzzy clustering algorithm can more effectively and more quickly converge to a global optimal solution.
Setting the initial temperature T of the annealing process 0 Temperature cooling coefficient k, termination temperature T end 。
In the calculation of the clustering center of the newly generated individuals and the fitness f of each individual i '. If f i '＞f i Replacing the old individual with the new individual; otherwise with probability
New individuals were accepted and old individuals were discarded.
Preferably, wherein the Metropolis acceptance criteria are expressed as follows:
if T i ＜T end If the algorithm is successfully ended, returning to the global optimal solution; otherwise, executing a cooling operation T i+1 ＝k.T i Go to step 3 in S33.
Preferably, the step 400 comprises:
constructing a farmer label system;
according to the formula
Quantifying the labels in the farmer label system; wherein G is i,s Scoring a label for the S dimension of the i-th farmer;
the average value of the jth characteristic belonging to the ith class of farmers; d jmax 、d jmin And
the j characteristic maximum, the j characteristic minimum and the average value are respectively; wherein, the S dimension contains t features, G i,s Is in the interval [0,1 ]]Performing the following steps;
and performing visualization processing on the farmer portrait according to the quantized indexes.
Specifically, in this embodiment S4, the user image is defined by tags based on artificially defined rules, and an accurate, fine-grained and structured tag system is the basis of the user image. The relationship structure between the farmer label system and the farmer population image is shown in FIG. 6.
Preferably, the farmer label system comprises a personal information label, a production condition label and a performance evaluation label; the personal information type label is used for describing farmer personal attributes of grower types, production specialization degree, age and education level; the production condition type label is used for reflecting planting information of the farmer cultivation area, cultivation mode and production type and reflecting the production behavior state of the farmer; the performance evaluation class labels are obtained from the clustering and subdividing results, and are used for subdividing and optimizing a farmer performance index system to obtain farmer groups with different production performances.
Preferably, the visualization process comprises an intra-class farmer portrait and an inter-class farmer portrait; the intra-class farmer portrait is used for displaying planting information of various farmers through a word cloud picture and a regional heating power map; the inter-class farmer portrait is used for comparing radar maps with grouping column maps so as to show the comparison effect of various farmers on the same label quantity.
Further, step S4 in this embodiment includes:
s41: the method comprises the following steps of (1) constructing a farmer label system, wherein the farmer label system consists of three parts: personal information class labels, production condition class labels and performance evaluation class labels. The personal information type label describes the individual attributes of farmers, such as the type of grower, the specialization degree of production, the age, the education level and the like; the production condition type labels reflect planting information such as the cultivation area, cultivation mode and production type of farmers and reflect the production behavior state of the farmers; and the performance evaluation type labels are obtained from the clustering analysis results, and the peasant household performance index system is subdivided and optimized to obtain the peasant household groups with different production performances.
S42: and quantifying the characteristic labels of farmers, wherein data of a farmer label system is mostly numerical data, and the data can be converted into labels which are convenient for business personnel to understand.
In the embodiment, normalization processing is adopted, and performance characteristics of farmers are measured according to scores of tag intervals under different dimensions by taking the performance evaluation tags of the farmers as an example. The calculation formula of each label score of each type of farmer is as follows:
in the formula: g i,s Scoring a label for the S dimension of the i-th farmer;
the average value of the jth characteristic belonging to the ith class of farmers; d jmax 、d jmin And
the j characteristic maximum, the j characteristic minimum and the average value are respectively; wherein, the S dimension contains t features, G i,s Is in the interval [0,1 ]]In (1).
S43: the image of peasant household is visual, and the visual display is divided into two parts: the intra-class farmer portrait and the inter-class farmer portrait. The intra-class farmer portrait shows the planting information of various farmers through a word cloud picture and a regional thermal map, and the inter-class farmer portrait comparison takes a radar map and a grouping column map as expression forms to highlight the comparison of various farmers on the same label quantity. The combination of the two is convenient for the service personnel to know the commonality and the personality of the image farmers more accurately and conveniently.
And finally, in S4, comprehensively evaluating the farmer portrait by combining the farmer performance index system, the quantitative label score and the fuzzy clustering optimization method, and deeply analyzing the personal attributes and the production behavior attributes of the farmer and the association relation between the production performance attributes of the farmer.
Corresponding to the method, the embodiment provides a farmer portrait construction system for production performance improvement, which includes:
the data acquisition module is used for acquiring a basic data set; the basic data set is used for characterizing the characteristics and production performance of a producer;
the data processing and measuring and calculating module is used for determining a characteristic index of production performance and carrying out data preprocessing on the basic data set based on the characteristic index to obtain a complete sample set to be analyzed;
the cluster analysis module is used for designing a farmer label system according to the complete sample set based on an optimized fuzzy clustering algorithm and performing cluster subdivision on farmer groups to obtain a farmer subdivision result based on production performance characteristics;
and the image generation module is used for constructing a farmer image according to the farmer subdivision result.
Preferably, the data acquisition module specifically includes:
the sampling unit is used for sampling by a multi-stage sampling method according to the regional layout and the actual production process conditions of various crops to obtain a sample farmer;
the scheme design unit is used for designing a farmer research scheme according to the sample farmer;
the data acquisition unit is used for acquiring basic data according to the farmer research scheme; the basic data is used for characterizing the characteristics and the production performance of a producer;
and the preprocessing unit is used for preprocessing the basic data to obtain the basic data set.
In addition, as shown in fig. 7, a schematic structural diagram of a farmer portrait system for production performance improvement according to another embodiment of the present invention is shown, which includes: the system comprises a data acquisition module 21, a data cleaning module 22, a production performance measuring and calculating module 23, a cluster analysis module 24 and a sketch generation module 25. The method specifically comprises the following steps:
the data acquisition module 21 is used for acquiring basic data of the farmer portrait, which mainly comprises data of the individual information, the production conditions and the input and output of the farmer.
The data cleaning module 22 is used for realizing the cleaning function of basic data, and mainly comprises two sub-modules, a data standard sub-module and a data cleaning sub-module.
And the production performance measuring and calculating module 23 is used for calculating the production performance characteristic indexes of the vineyard peasant households according to the cleaned input and output index data. The system mainly comprises a performance index management submodule and a performance index measuring and calculating submodule.
The cluster analysis module 24 is used for implementing the construction work of the cluster optimization model designed by the present invention, and can design a model structure and adjustment parameters, adjust and improve the model, and train the model by using the imported training data sample set.
And the portrait generation module 25 is used for describing the pictures of the clustered farmer samples, obtaining scores of different labels of the farmer according to characteristic quantization, and respectively describing the results of the farmer group portrait in the class and the farmer group portrait between the classes through a visual system interface.
Specifically, the contents of the farmer portrait survey mainly include various basic information of farmland growers and plantations, planting varieties and areas, the quantity and funds of various input and output items, and the like.
The data acquisition module 21 is specifically configured to:
a uniform webapi interface is provided through a web client/a mobile phone client, data of the peasant household in three aspects of personal information, production conditions and input and output are collected, and the collected data are stored in a database of a server.
The data acquisition module is mainly used for synchronously acquiring data of a farmer big data acquisition and analysis platform database and acquiring log data, wherein the data source in the database is that farmer basic information and cost and income data are imported into the platform database through questionnaire form or text data by management personnel of each test station. The two data acquisition modes are different, but the data are formatted and then uniformly stored in a database of the server.
Wherein the data cleansing module 22 is specifically configured to:
the data cleaning module comprises two submodules in total, a data standard submodule and a data cleaning submodule.
The data standard submodule manages data cleaning rules, and the system provides basic data standards including data cleaning rules of abnormal value identification, missing value filling and the like;
and the user uses a proper data standard model according to the data characteristics and the self requirements. After the user uploads the data, the system normalizes the data according to user-defined data standards.
The data cleaning submodule is responsible for calling a data cleaning rule to preprocess data and converting dirty data into clean data meeting requirements.
A data analyst then defines a set of rules for data cleaning according to the requirements for data, the rules are generally defined based on the requirements of data analysis and according to the data types, and the rules generally include data entry constraints, abnormal value detection, missing value filling, duplicate record deletion, sorting, structured data and the like. And judging whether the data meet the data cleaning requirements or not according to the cost and income data of the peasant households. If so, cleaning the data, and then converting the clean data to reflux; if not, the reflux is direct.
The production performance measuring module 23 is specifically configured to:
the production performance measuring and calculating module comprises two submodules in total, an index management submodule and an index measuring and calculating submodule.
And the index management submodule manages production performance characteristic index rules and comprises an algebraic operation model based on output effect evaluation and a DEA-BCC model based on technical efficiency evaluation.
And the index measurement and calculation sub-module calls an index management rule to perform characteristic derivation on the data, and converts the data into a complete sample set suitable for modeling after data standardization and homodromous processing.
Wherein the cluster analysis module 24 is specifically configured to:
and packaging the optimized fuzzy clustering algorithm into a system module, designing a model structure and adjusting parameters, adjusting and improving the model, and performing model training by using the imported training data sample set.
Wherein, the portrait generation module 25 is specifically configured to:
the portrait generation module comprises three submodules in total, a portrait display submodule and a characteristic quantization submodule.
The characteristic quantization submodule is used for quantizing the characteristic values of the label of the peasant household after the clustering analysis, and measuring the performance characteristics of the peasant household by using the label interval scores under different dimensionalities.
The portrait display sub-module is used for displaying a model system of the farmer portrait and respectively describing results of the farmer portrait in the class and the farmer portrait between the classes through a visual system interface.
The invention has the following beneficial effects:
the farmer portrait system oriented to production performance improvement can realize functions of dynamically displaying data, performing real-time interaction with users and the like, so that the practical problem of agriculture is solved and the application of user portrait in the field of agriculture is widened.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (10)
1. A farmer portrait construction method for production performance improvement is characterized by comprising the following steps:
acquiring a basic data set; the basic data set is used for characterizing the characteristics and production performance of a producer;
determining a characteristic index of production performance, and performing data preprocessing on the basic data set based on the characteristic index to obtain a complete sample set to be analyzed;
designing a farmer label system according to the complete sample set based on an optimized fuzzy clustering algorithm, and performing clustering segmentation on farmer groups to obtain a farmer segmentation result based on production performance characteristics;
and constructing a farmer portrait according to the farmer subdivision result.
2. The method for constructing a farm household portrait with oriented production performance improvement as claimed in claim 1, wherein the acquiring of the basic data set comprises:
sampling by a multi-stage sampling method by combining the regional layout of various crops and the actual situation of the production process to obtain a sample farmer;
designing a farmer investigation scheme according to the sample farmer;
acquiring basic data according to the farmer research scheme; the basic data is used for characterizing the characteristics and the production performance of a producer;
and carrying out data preprocessing on the basic data to obtain the basic data set.
3. The method for constructing a farm household portrait with oriented production performance improvement as claimed in claim 2, wherein the data preprocessing comprises abnormal value detection and missing value processing.
4. The method for constructing a farmer portrait for improving production performance as claimed in claim 1, wherein the determining characteristic index of production performance and the preprocessing of the basic data set based on the characteristic index to obtain a complete sample set to be analyzed comprises:
the preset characteristic indexes of the production performance are divided into a first index based on output effect evaluation and a second index based on technical efficiency evaluation;
the first index can be obtained by algebraic operation on a basic data set; the second index can be obtained by measuring and calculating through a DEA-BCC model, and comprises a technical efficiency index based on value output and a technical efficiency index based on physical output;
carrying out syntropy processing on reverse indexes in the first indexes and the second indexes; the formula of the index homodromous treatment is Y ij ＝max(X j )-X ij (ii) a Wherein, max (X) j ) Is the maximum value of the index j in all samples, X ij Is the original value of the index j on the sample i, Y ij Taking a value of an index j of the sample i after the homotropic treatment;
carrying out standardization processing on the index data subjected to index syntropy processing; the formula of the normalization process is
Wherein mu and sigma are respectively a sample mean value and a sample mean square error; the normalized data is the complete sample set.
5. The method for constructing the farmer portrait facing production performance improvement according to claim 1, wherein the method for constructing the farmer portrait based on production performance improvement comprises the following steps of designing a farmer label system according to the complete sample set based on an optimized fuzzy clustering algorithm, and performing clustering segmentation on a farmer population to obtain a farmer segmentation result based on production performance characteristics:
determining the number of peasant household clusters based on the production performance characteristic index by using a fuzzy clustering algorithm and a clustering effectiveness index;
determining a partition matrix according to the cluster number and the complete sample set, respectively finding out R samples of the same type and different types of nearest neighbors from the partition matrix, and calculating to obtain the weight of each performance characteristic according to the distinguishing capability of the characteristic on the close-range samples;
and performing genetic iteration and evolution on each performance characteristic according to a genetic algorithm fusion simulated annealing algorithm to obtain the segmentation result of the peasant household.
6. The farmer portrait construction method oriented to production performance improvement of claim 1, wherein the construction of the farmer portrait according to the farmer segmentation result comprises:
constructing a farmer label system;
according to the formula
Quantifying the labels in the farmer label system; wherein G is i,s Scoring the labels under the S dimensionality of the ith type of farmers;
the average value of the jth characteristic belonging to the ith class of farmers; d is a radical of jmax 、d jmin And
the j-th characteristic maximum, minimum and mean values are respectively; wherein, the S dimension contains t features, G i,s Is in the interval [0,1 ]]Performing the following steps;
and performing visualization processing on the farmer portrait according to the quantized indexes.
7. The farmer portrait construction method oriented to production performance improvement of claim 6, characterized in that the farmer label system comprises a personal information label, a production condition label and a performance evaluation label; the personal information type label is used for describing farmer personal attributes of grower types, production specialization degree, age and education level; the production condition type label is used for reflecting planting information of the farmer cultivation area, cultivation mode and production type and reflecting the production behavior state of the farmer; the performance evaluation class labels are obtained from the clustering and subdividing results, and are used for subdividing and optimizing a farmer performance index system to obtain farmer groups with different production performances.
8. The method for constructing a farmer portrait oriented toward production performance improvement according to claim 6, wherein the visualization processing comprises an intra-class farmer portrait and an inter-class farmer portrait; the intra-class farmer portrait is used for displaying planting information of various farmers through a word cloud picture and a regional heating power map; the inter-class farmer portrait is used for comparing radar maps with grouping column maps so as to show the comparison effect of various farmers on the same label quantity.
9. A farmer portrait construction system for production performance improvement is characterized by comprising:
the data acquisition module is used for acquiring a basic data set; the basic data set is used for characterizing the characteristics and production performance of a producer;
the data processing and measuring and calculating module is used for determining a characteristic index of production performance and carrying out data preprocessing on the basic data set based on the characteristic index to obtain a complete sample set to be analyzed;
the cluster analysis module is used for designing a farmer label system according to the complete sample set based on an optimized fuzzy clustering algorithm and performing cluster subdivision on farmer groups to obtain a farmer subdivision result based on production performance characteristics;
and the image generation module is used for constructing a farmer image according to the farmer subdivision result.
10. The system for constructing a farm household representation for improving production performance of claim 9, wherein the data acquisition module specifically comprises:
the sampling unit is used for sampling by a multi-stage sampling method according to the regional layout and the actual production process conditions of various crops to obtain a sample farmer;
the scheme design unit is used for designing a farmer research scheme according to the sample farmer;
the data acquisition unit is used for acquiring basic data according to the farmer research scheme; the basic data is used for characterizing the characteristics and the production performance of a producer;
and the preprocessing unit is used for preprocessing the basic data to obtain the basic data set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210717234.2A CN115018357A (en) | 2022-06-23 | 2022-06-23 | Farmer portrait construction method and system for production performance improvement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210717234.2A CN115018357A (en) | 2022-06-23 | 2022-06-23 | Farmer portrait construction method and system for production performance improvement |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115018357A true CN115018357A (en) | 2022-09-06 |
Family
ID=83077681
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210717234.2A Pending CN115018357A (en) | 2022-06-23 | 2022-06-23 | Farmer portrait construction method and system for production performance improvement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115018357A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117314243A (en) * | 2023-09-28 | 2023-12-29 | 北京工商大学 | Method for evaluating efficiency of operators based on cluster analysis |
| CN117493817A (en) * | 2023-12-29 | 2024-02-02 | 中国西安卫星测控中心 | Method, system and device for evaluating benefit of processing satellite anomalies |
-
2022
- 2022-06-23 CN CN202210717234.2A patent/CN115018357A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117314243A (en) * | 2023-09-28 | 2023-12-29 | 北京工商大学 | Method for evaluating efficiency of operators based on cluster analysis |
| CN117493817A (en) * | 2023-12-29 | 2024-02-02 | 中国西安卫星测控中心 | Method, system and device for evaluating benefit of processing satellite anomalies |
| CN117493817B (en) * | 2023-12-29 | 2024-04-16 | 中国西安卫星测控中心 | Method, system and device for evaluating benefit of processing satellite anomalies |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107230113A (en) | A kind of house property appraisal procedure of multi-model fusion | |
| CN115018357A (en) | Farmer portrait construction method and system for production performance improvement | |
| CN110059852A (en) | A kind of stock yield prediction technique based on improvement random forests algorithm | |
| CN115906954A (en) | Multivariate time sequence prediction method and device based on graph neural network | |
| Chakrabarty | A regression approach to distribution and trend analysis of quarterly foreign tourist arrivals in India | |
| CN116109195B (en) | Performance evaluation method and system based on graph convolution neural network | |
| CN106682915A (en) | User cluster analysis method in customer care system | |
| CN115618021A (en) | Method and device for recommending suitable planting area of crop variety | |
| CN111062511B (en) | Aquaculture disease prediction method and system based on decision tree and neural network | |
| Sun | Real estate evaluation model based on genetic algorithm optimized neural network | |
| Kumar et al. | A Novel Cluster Analysis-Based Crop Dataset Recommendation Method in Precision Farming. | |
| Jethva et al. | A review on data mining techniques for fertilizer recommendation | |
| Ramdhani et al. | Feature selection optimization based on genetic algorithm for support vector classification varieties of raisin | |
| Vanitha | Decision support model for prioritization of cotton plant diseases using integrated fahp-topsis approach | |
| Manjula et al. | Efficient prediction of recommended crop variety through soil nutrients using deep learning algorithm | |
| Vanarase et al. | Crop Prediction Using Data Mining and Machine Learning Techniques | |
| CN114780599A (en) | Comprehensive analysis system based on wheat quality ratio test data | |
| Gawade et al. | AGROFERDURE: Intelligent Crop Recommendation System For Agriculture Crop Productivity Using Machine Learning Algorithm | |
| Jackson et al. | Robust Ensemble Machine Learning for Precision Agriculture | |
| CN113408895A (en) | Ecological quality index construction method and system based on pixel scale | |
| CN112070336A (en) | Manufacturing industry information quantitative analysis method and device based on analytic hierarchy process | |
| CN113570455A (en) | Stock recommendation method and device, computer equipment and storage medium | |
| Filipe | Using tree-based ensemble methods to improve the B2B customer acquisition process in the fashion industry | |
| CN117314266B (en) | Novel intelligent scientific and technological talent evaluation method based on hypergraph attention mechanism | |
| CN117556147B (en) | Electronic commerce data classification recommendation system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |