CN113918785A - Enterprise data analysis method based on cluster ensemble learning - Google Patents
Enterprise data analysis method based on cluster ensemble learning Download PDFInfo
- Publication number
- CN113918785A CN113918785A CN202111182846.8A CN202111182846A CN113918785A CN 113918785 A CN113918785 A CN 113918785A CN 202111182846 A CN202111182846 A CN 202111182846A CN 113918785 A CN113918785 A CN 113918785A
- Authority
- CN
- China
- Prior art keywords
- data
- analyzed
- clustering
- enterprise
- cluster
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000007405 data analysis Methods 0.000 title claims abstract description 16
- 239000011159 matrix material Substances 0.000 claims abstract description 20
- 230000010354 integration Effects 0.000 claims abstract description 15
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 6
- 230000009193 crawling Effects 0.000 claims abstract description 5
- 238000005192 partition Methods 0.000 claims description 22
- 239000013598 vector Substances 0.000 claims description 8
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 2
- 238000009360 aquaculture Methods 0.000 description 2
- 244000144974 aquaculture Species 0.000 description 2
- 235000014347 soups Nutrition 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/231—Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Image Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an enterprise data analysis method based on cluster ensemble learning, which comprises the following steps: s1, obtaining data of an industry to be analyzed, and finding out a plurality of major enterprises to be analyzed in the industry to be analyzed; this example analyzes the marine industry in the Guangzhou region. S2, crawling relevant data of an enterprise to be analyzed; s3, preprocessing the crawled data, and sorting the preprocessed data into a data set; s4, adopting KMeans as a base clustering device, and performing clustering ensemble learning on the data set to obtain a basic clustering result; s5, constructing a joint matrix by using the basic clustering result; and S6, processing the combined matrix by adopting single-link hierarchical clustering to obtain a final clustering integration result of the enterprise to be analyzed. The method has the characteristics of accuracy, stability and robustness, convenience in calculation and high practicability.
Description
Technical Field
The invention relates to the technical field of information processing, in particular to an enterprise data analysis method based on cluster ensemble learning.
Background
In current enterprise data analysis methods, data clustering or unsupervised learning is an important but extremely difficult problem. Structures in data are organized or discovered by dividing a set of unlabeled objects into homogeneous groups or clusters, such as data mining, information retrieval image segmentation, and machine learning. In real-world problems, clusters may appear in different shapes, sizes, data sparsity, and degrees of separation. In addition, noise in the data can mask the true underlying structures present in the data. Clustering techniques require defining similarity measures between patterns, which are not easily specified without any a priori knowledge of the cluster shape. Clustering integration is an algorithm for improving the accuracy, stability and robustness of clustering results, and a better result can be generated by integrating a plurality of base clustering results.
For example, chinese patent publication No. CN 112800165B, patent publication No. 2021.08.27, discloses an industry cluster positioning method based on a clustering algorithm, which obtains real-time enterprise data of a plurality of enterprises in a preset area, and classifies the plurality of enterprises according to the industry label information, so as to solve the problem that the traditional industry cluster positioning method cannot accurately position an industry cluster in a large range. However, the results analyzed by the clustering algorithm have poor accuracy, stability and robustness, and cannot well reflect the clustering results among enterprises.
Disclosure of Invention
In order to solve the problem that the analysis result of the prior art has poor accuracy, stability and robustness, the invention provides an enterprise data analysis method based on cluster ensemble learning, which has the characteristics of accuracy, stability and robustness, convenience in calculation and strong practicability.
In order to achieve the purpose of the invention, the technical scheme is as follows:
an enterprise data analysis method based on cluster ensemble learning comprises the following steps:
s1, obtaining data of an industry to be analyzed, and finding out a plurality of major enterprises to be analyzed in the industry to be analyzed;
s2, crawling relevant data of an enterprise to be analyzed;
s3, preprocessing the crawled data and sorting the processed data into a data set;
s4, adopting KMeans as a base clustering device, and performing clustering ensemble learning on the data set to obtain a basic clustering result;
s5, constructing a joint matrix by using the basic clustering result;
and S6, processing the combined matrix by adopting single-link hierarchical clustering to obtain a final clustering integration result of the enterprise to be analyzed.
Preferably, step S7 is further included after step S6: and (4) visualizing the clustering integration result by combining the related data of the enterprise to be analyzed.
Further, in step S2, the relevant data of the enterprise to be analyzed includes the operation area, the geographical location, and the registered capital of the enterprise.
Further, the preprocessing includes removing noise data, removing duplicate data, and then performing feature screening.
Further, in step S3, the step of sorting the processed data into a data set includes the following specific steps:
s301, dividing industries to be analyzed into a plurality of industry categories according to the operation field of the industries to be analyzed;
s302, coding the enterprise to be analyzed by using a one-hot method according to the corresponding industry type of the enterprise to be analyzed;
and S303, arranging the codes of the enterprises to be analyzed into a data set.
Furthermore, in S4, a KMeans clustering algorithm is used as a basis clustering device to perform cluster ensemble learning on the data set, and the specific steps are as follows:
s401, randomly selecting a value from the interval [2, 2c ], assigning the value to K, wherein c is the number of real clusters;
s402, randomly selecting K samples from a data set as initial K centroid vectors:
{μ1,μ2,…,μK};
s403, iterating the K samples to obtain a family partition C;
further, in S403, the iteration is performed on the K samples, and the specific steps are as follows:
m1. initializing Cluster partition C toWherein t is { t belongs to N |1 and is not more than t and not more than K };
m2. for i ═ 1, 2, …, m, sample x is calculatediAnd each centroid vector mujIs a distance of j is 1, 2, …, K, xiMinimum mark is dijCorresponding class λi(ii) a At this time, update is performed
M3. for j ═ 1, 2, …, K, for CjAveraging all the sample points in the image, and recalculating new centroid
M4. if all K centroid vectors have not changed, the output cluster division C ═ C1,C2,…,CK}。
Further, the step S5 specifically includes:
s501, generating N basic partitions P through N iterations by using a KMeans clustering algorithm:
s502, constructing a matrix of m by m, namely C, according to the number m of samplesm×m=0;
And S503, obtaining a final joint matrix by using a voting method.
Further, the voting formula in S503 is:
wherein n isijIndicating the assignment of sample (i, j) to base partition PiNumber of clusters in which P isiRefer toTo a certain partition in the group.
Further, the step S6 is to process the association matrix by single-link hierarchical clustering, specifically including the steps of;
s601, calculating the distance between the data point of each category and all the data points according to the joint matrix to determine the similarity between the data points;
s602, combining two data points or categories with the nearest distance to generate a clustering tree;
s603, hierarchical clustering is carried out according to a threshold value to obtain a final partition
The invention has the following beneficial effects:
on the basis of the existing clustering algorithm, the KMeans is used as a base clustering device to perform clustering ensemble learning on enterprise related data to obtain a clustering ensemble result of an enterprise.
Drawings
FIG. 1 is a schematic flow diagram of the present invention.
Fig. 2 is a sample point of a cluster integration result.
Fig. 3 is a sample gray image of the cluster integration result.
FIG. 4 is a sample dendrogram of cluster integration results.
Fig. 5 is a visual image of the cluster integration result.
Detailed Description
The invention is described in detail below with reference to the drawings and the detailed description.
Example 1
As shown in fig. 1, an enterprise data analysis method based on cluster ensemble learning includes the following steps:
s1, obtaining data of an industry to be analyzed, and finding out a plurality of major enterprises to be analyzed in the industry to be analyzed; the embodiment analyzes the marine industry in Guangdong province by crawling the operation field, registered capital and other data related to the marine enterprises in Guangdong province on the internet by using the popular python libraries beautiiful Soup, urllib and the like.
S2, crawling relevant data of an enterprise to be analyzed;
s3, preprocessing the crawled data and sorting the processed data into a data set;
s4, adopting KMeans as a base clustering device, and performing clustering ensemble learning on the data set to obtain a basic clustering result;
s5, constructing a joint matrix by using the basic clustering result;
and S6, processing the combined matrix by adopting single-link hierarchical clustering to obtain a final clustering integration result of the enterprise to be analyzed.
At step S2, the relevant data of the enterprise to be analyzed includes the operation area, the geographical location, and the registered capital of the enterprise.
The embodiment can use the libraries such as the Beautiful Soup and the url lib which are popular nowadays to crawl data such as the operation field, the registered capital, the geographic position and the like related to the ocean enterprises in Guangdong province on the network. The data crawled from the internet comprises a plurality of fields such as the operation field, registered capital, geographical position and the like related to the ocean enterprises in Guangdong province, and the data have the defect of more redundancy. Therefore, a certain rule is required to be set for preprocessing data according to project requirements, wherein the preprocessing comprises the steps of removing noise data and removing repeated data, and then a data set of the Guangdong province ocean enterprise is finally constructed through feature screening. In step S3, the step of sorting the processed data into a data set includes the following specific steps:
s301, dividing industries to be analyzed into a plurality of industry categories according to the operation field of the industries to be analyzed; in this embodiment, enterprises are classified into 11 categories according to the fields of aquaculture industry, cruise industry, energy industry, engineering industry, equipment industry, aquaculture industry, logistics industry, ship industry, trade industry, transportation industry, and others.
S302, coding the enterprise to be analyzed by using a one-hot method according to the corresponding industry type of the enterprise to be analyzed;
and S303, arranging the codes of the enterprises to be analyzed into a data set.
In S4, a KMeans clustering algorithm is used as a basis clustering device to perform cluster ensemble learning on the data set, as shown in fig. 2, 3 and 4Clustering integrationIn order to construct an EAC clustering integration model, firstly, a KMeans clustering method is adopted to generate a basic clustering result, and for a given data sample set, the KMeans clustering algorithm divides the sample set into K clusters according to the distance between samples, namely, K clustering centers are selected. The points in the clusters are connected together as closely as possible, and the distance between the clusters is as large as possible; in general, K can be determined by elbow rules or contour coefficients, but in order to increase diversity in cluster integration, K is { K ∈ N |2 ≦ K ≦ 2c }, and c is the number of true class clusters. Assume that input sample set D ═ x1,x2,…,xmThe output cluster division C ═ C1,C2,…,CKWherein m is the number of samples, and K is the number of clusters, the specific steps are as follows:
s401, randomly selecting a value from the interval [2, 2c ], assigning the value to K, wherein c is the number of real clusters;
s402, randomly selecting K samples from a data set as initial K centroid vectors:
{μ1,μ2,…,μK};
s403, iterating the K samples to obtain a family partition C;
in S403, iterating the K samples, specifically including:
m1. initializing Cluster partition C toWherein t is { t belongs to N |1 and is not more than t and not more than K };
m2. for i ═ 1, 2, …, m, sample x is calculatediAnd each centroid vector mujIs a distance of j is 1, 2, …, K, xiMinimum mark is dijCorresponding class λi(ii) a At this time, update is performed
M3. for j ═ 1, 2, …, K, for CjAveraging all the sample points in the image, and recalculating new centroid
M4. if all K centroid vectors have not changed, the output cluster division C ═ C1,C2,…,CK}。
Step S5, specifically:
s502, constructing a matrix of m by m, namely C, according to the number m of samplesm×m=0;
And S503, obtaining a final joint matrix by using a voting method.
The voting formula in S503 is:
wherein n isiiIndicating the assignment of sample (i, j) to base partition PiNumber of clusters in which P isiRefer toTo a certain partition in the group.
Step S6 is to process the joint matrix by single-link hierarchical clustering, and the single-link hierarchical clustering algorithm combines two most similar data points of all data points by calculating the similarity between the two types of data points, and iterates this process repeatedly. The merging algorithm of single-link hierarchical clustering determines the similarity between the data points of each category and all the data points by calculating the distance between the data points, wherein the smaller the distance is, the higher the similarity is, and two data points or categories with the closest distance are combined to generate a clustering tree; in the experiment, from Cm×mAs input, hierarchical clustering is performed according to a threshold to obtain a final partitionThe specific steps are as follows;
s601, calculating the distance between the data point of each category and all the data points according to the joint matrix to determine the similarity between the data points;
s602, combining two data points or categories with the nearest distance to generate a clustering tree;
s603, hierarchical clustering is carried out according to a threshold value to obtain a final partition
Step S7 is also included after step S6: the cluster integration result is visualized by combining the relevant data of the enterprise to be analyzed, and the cluster integration result is visualized as shown in fig. 5 in this embodiment. In the embodiment, on the basis of the existing clustering algorithm, the relevant data of the marine enterprise to be analyzed is crawled, the relevant data is subjected to denoising processing and is divided into 11 types, and the data sets are sorted by a one-hot method; according to the method, KMeans are used as a base clustering device, clustering ensemble learning is carried out on enterprise related data of a marine enterprise, N times of iteration is carried out on K samples, a final partition is finally obtained through single-link hierarchical clustering, a clustering ensemble result of the enterprise is obtained, and the clustering ensemble result is visualized; the method is convenient to calculate and high in practicability, and solves the problem that the result analyzed by the prior art is poor in accuracy, stability and robustness.
It should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. An enterprise data analysis method based on cluster ensemble learning is characterized in that: the method comprises the following steps:
s1, obtaining data of an industry to be analyzed, and finding out a plurality of major enterprises to be analyzed in the industry to be analyzed;
s2, crawling relevant data of an enterprise to be analyzed;
s3, preprocessing the crawled data, and sorting the preprocessed data into a data set;
s4, adopting KMeans as a base clustering device, and performing clustering ensemble learning on the data set to obtain a basic clustering result;
s5, constructing a joint matrix by using the basic clustering result;
and S6, processing the combined matrix by adopting single-link hierarchical clustering to obtain a final clustering integration result of the enterprise to be analyzed.
2. The method of analyzing enterprise data based on cluster ensemble learning of claim 1, wherein: step S7 is also included after step S6: and (4) visualizing the clustering integration result by combining the related data of the enterprise to be analyzed.
3. The method of analyzing enterprise data based on cluster ensemble learning of claim 2, wherein: at step S2, the relevant data of the enterprise to be analyzed includes the operation area, the geographical location, and the registered capital of the enterprise.
4. The method of cluster ensemble learning based enterprise data analysis according to claim 3, wherein: the preprocessing comprises the steps of firstly removing noise data and repeated data, and then carrying out feature screening.
5. The method of cluster ensemble learning based enterprise data analysis according to claim 4, wherein: in step S3, the sorting of the preprocessed data into a data set includes the following specific steps:
s301, dividing industries to be analyzed into a plurality of industry categories according to the operation field of the industries to be analyzed;
s302, coding the enterprise to be analyzed by using a one-hot method according to the corresponding industry type of the enterprise to be analyzed;
and S303, arranging the codes of the enterprises to be analyzed into a data set.
6. The method of cluster ensemble learning based enterprise data analysis according to claim 5, wherein: in the step S4, a KMeans clustering algorithm is used as a basis clustering device to perform cluster ensemble learning on the data set, and the method specifically comprises the following steps:
s401, randomly selecting a value from the interval [2, 2c ], assigning the value to K, wherein c is the number of real clusters;
s402, randomly selecting K samples from the data set as initial K centroid directionsQuantity: { mu. }1,μ2,…,μK};
And S403, iterating the K samples to obtain the family division C.
7. The method of cluster ensemble learning based enterprise data analysis according to claim 6, wherein: in S403, iterating the K samples, specifically including:
m1. initializing Cluster partition C toWherein t is { t belongs to N |1 and is not more than t and not more than K };
m2. for i ═ 1, 2, …, m, sample x is calculatediAnd each centroid vector mujIs a distance of X is to beiMinimum mark is dijCorresponding class λi(ii) a At this time, update is performed
M3. for j ═ 1, 2, …, K, for CjAveraging all the sample points in the image, and recalculating new centroid
M4. if all K centroid vectors have not changed, the output cluster division C ═ C1,C2,…,CK}。
8. The method of cluster ensemble learning based enterprise data analysis according to claim 7, wherein: step S5, specifically:
s502, constructing a matrix of m by m, namely C, according to the number m of samplesm×m=0;
And S503, obtaining a final joint matrix by using a voting method.
10. The method of cluster ensemble learning based enterprise data analysis according to claim 8, wherein: step S6, the concrete steps of processing the association matrix by adopting single-link hierarchical clustering are as follows;
s601, calculating the distance between the data point of each category and all the data points according to the joint matrix to determine the similarity between the data points;
s602, combining two data points or categories with the nearest distance to generate a clustering tree;
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111182846.8A CN113918785A (en) | 2021-10-11 | 2021-10-11 | Enterprise data analysis method based on cluster ensemble learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111182846.8A CN113918785A (en) | 2021-10-11 | 2021-10-11 | Enterprise data analysis method based on cluster ensemble learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113918785A true CN113918785A (en) | 2022-01-11 |
Family
ID=79239110
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111182846.8A Pending CN113918785A (en) | 2021-10-11 | 2021-10-11 | Enterprise data analysis method based on cluster ensemble learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113918785A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105844303A (en) * | 2016-04-08 | 2016-08-10 | 云南大学 | Sampling type clustering integration method based on local and global information |
CN106056136A (en) * | 2016-05-20 | 2016-10-26 | 浙江工业大学 | Data clustering method for rapidly determining clustering center |
US20170316079A1 (en) * | 2016-04-27 | 2017-11-02 | Oracle International Corporation | Scalable tri-point arbitration and clustering |
CN111625576A (en) * | 2020-05-15 | 2020-09-04 | 西北工业大学 | Score clustering analysis method based on t-SNE |
CN111814153A (en) * | 2020-07-10 | 2020-10-23 | 辽东学院 | Commercial website security maintenance method based on big data |
CN112800165A (en) * | 2021-04-06 | 2021-05-14 | 北京智源人工智能研究院 | Industrial cluster positioning method and device based on clustering algorithm and electronic equipment |
-
2021
- 2021-10-11 CN CN202111182846.8A patent/CN113918785A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105844303A (en) * | 2016-04-08 | 2016-08-10 | 云南大学 | Sampling type clustering integration method based on local and global information |
US20170316079A1 (en) * | 2016-04-27 | 2017-11-02 | Oracle International Corporation | Scalable tri-point arbitration and clustering |
CN106056136A (en) * | 2016-05-20 | 2016-10-26 | 浙江工业大学 | Data clustering method for rapidly determining clustering center |
CN111625576A (en) * | 2020-05-15 | 2020-09-04 | 西北工业大学 | Score clustering analysis method based on t-SNE |
CN111814153A (en) * | 2020-07-10 | 2020-10-23 | 辽东学院 | Commercial website security maintenance method based on big data |
CN112800165A (en) * | 2021-04-06 | 2021-05-14 | 北京智源人工智能研究院 | Industrial cluster positioning method and device based on clustering algorithm and electronic equipment |
Non-Patent Citations (1)
Title |
---|
蒋俊钊;程良伦;李全杰;: "基于标签相关性的卷积神经网络多标签分类算法", 工业控制计算机, no. 07, 25 July 2018 (2018-07-25) * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Bespalov et al. | Scale-space representation of 3d models and topological matching | |
Stanford et al. | Finding curvilinear features in spatial point patterns: principal curve clustering with noise | |
Caelli et al. | An eigenspace projection clustering method for inexact graph matching | |
Unnikrishnan et al. | Toward objective evaluation of image segmentation algorithms | |
US8055677B2 (en) | Geometrization for pattern recognition data analysis, data merging and multiple criteria decision making | |
CN112580483B (en) | Remote sensing image feature discretization method based on rough fuzzy model | |
KR100824698B1 (en) | Region Weighting Mehtod Based on Spatial Location Informaiton for Relevance Feedback in Image Search Engine | |
Omran et al. | Particle swarm optimization for pattern recognition and image processing | |
CN115311502A (en) | Remote sensing image small sample scene classification method based on multi-scale double-flow architecture | |
Abouelaziz et al. | Learning graph convolutional network for blind mesh visual quality assessment | |
CN114565861A (en) | Airborne downward-looking target image positioning method based on probability statistic differential homoembryo set matching | |
Lee et al. | A genetic feature weighting scheme for pattern recognition | |
CN111597367B (en) | Three-dimensional model retrieval method based on view and hash algorithm | |
CN108921853A (en) | Image partition method based on super-pixel and clustering of immunity sparse spectrums | |
CN112734913A (en) | Three-dimensional model sphere expression calculation method based on multi-stage deformation reconstruction | |
CN113918785A (en) | Enterprise data analysis method based on cluster ensemble learning | |
CN115205496B (en) | Digital twin model light weight method and system | |
Borkowski et al. | Matching 2D image segments with genetic algorithms and approximation spaces | |
Patil et al. | Measurement-based quantum clustering algorithms | |
Azzawi et al. | Face recognition based on mixed between selected feature by multiwavelet and particle swarm optimization | |
Davidson | Visualizing clustering results | |
Ma et al. | Complex buildings orientation recognition and description based on vector reconstruction | |
CN112101405A (en) | Robust depth self-encoder and density peak value-based track clustering and abnormal value identification method | |
Li et al. | Sketching merge trees for scientific visualization | |
Beer et al. | Graph Ordering and Clustering: A Circular Approach |
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 |