Community operation data analysis method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of community operation, and discloses a community operation data analysis method and device, computer equipment and a storage medium. According to the community operation data analysis method, to-be-analyzed attribute data of a to-be-analyzed community and sample attribute data of sample communities are obtained and subjected to clustering analysis processing, and similar clustering communities and street clustering communities corresponding to the to-be-analyzed community in the sample communities are determined; obtaining similar operation index data of the similar clustering communities and street operation index data of the street clustering communities in a preset time window, and determining an index early warning threshold according to the similar operation index data and the street operation index data; and obtaining to-be-analyzed operation index data in a preset time window, judging whether the to-be-analyzed community is an abnormal community according to the to-be-analyzed operation index data and the index early warning threshold, and generating abnormal early warning information when it is confirmed that the to-be-analyzed community is the abnormal community. According to the invention, the comprehensiveness of anomaly monitoring is ensured, and the monitoring accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of community operation technology, and in particular to a community operation data analysis method, device, computer equipment and storage medium. Background Art

[0002] In urban life, population distribution and living patterns are highly organized, forming a community lifestyle, and community operations have emerged. Community operations are a bridge between enterprises and users. Through community operations, enterprises can better understand user needs and promote business development. Abnormal data warning is an important part of community operations. Community operators can detect abnormal data and trigger warnings by monitoring the fluctuations of business data, so as to make effective operational adjustments.

[0003] Existing community operations mainly rely on manual monitoring of business data, which has some defects. On the one hand, manual monitoring methods have a lag and cannot capture abnormal fluctuations in business data in a timely manner, which can easily lead to the risk of decline and loss in operating conditions. On the other hand, the business data itself will also fluctuate to varying degrees. For example, it may be affected by the overall fluctuations of the market data, and may be affected by factors such as off-seasons, holidays, and major promotions. It may also be related to the natural changes of the community itself. Manual monitoring methods can often only conduct longitudinal comparisons in a certain aspect, and cannot conduct comprehensive horizontal comparisons, which makes abnormal monitoring results prone to deviations. Summary of the invention

[0004] Based on this, it is necessary to provide a community operation data analysis method, device, computer equipment and storage medium to address the above technical issues, so as to solve the problems of untimely and incomplete abnormal monitoring of community operation data.

[0005] A community operation data analysis method, comprising: Acquire the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, perform cluster analysis on the attribute data to be analyzed and the sample attribute data, and determine similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed; Obtain similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determine an indicator warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data; The operation indicator data to be analyzed of the community to be analyzed within the preset time window is obtained, and whether the community to be analyzed is an abnormal community is determined according to the operation indicator data to be analyzed and the indicator warning threshold, and abnormal warning information is generated when it is confirmed that the community to be analyzed is an abnormal community.

[0006] A community operation data analysis device, comprising: A cluster analysis processing module, used to obtain the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, perform cluster analysis on the attribute data to be analyzed and the sample attribute data, and determine similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed; An early warning threshold determination module is used to obtain similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determine the indicator early warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data; The abnormal community judgment module is used to obtain the operation indicator data to be analyzed of the community to be analyzed within the preset time window, judge whether the community to be analyzed is an abnormal community according to the operation indicator data to be analyzed and the indicator warning threshold, and generate abnormal warning information when it is confirmed that the community to be analyzed is an abnormal community.

[0007] A computer device includes a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the above-mentioned community operation data analysis method is implemented.

[0008] A computer-readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the community operation data analysis method as described above.

[0009] In the above-mentioned community operation data analysis method, device, computer equipment and storage medium, the community operation data analysis method uses a machine learning algorithm to perform cluster analysis on the attribute data to be analyzed and the sample attribute data of the sample community on the basis of obtaining the attribute data to be analyzed of the community to be analyzed and the sample attribute data, and determines the similar cluster community and street cluster community corresponding to the community to be analyzed in the sample community. The similar cluster community can reflect the market fluctuation of a certain type of community, and the street cluster community can reflect the influence of external factors of the street on the community operation data, thereby improving the comprehensiveness and rationality of abnormal monitoring. In addition, the present invention also obtains similar operation index data of similar cluster communities and street operation index data of street cluster communities within a preset time window, and uses a statistical method to determine the indicator warning threshold corresponding to the community to be analyzed within the preset time window based on the similar operation index data and the street operation index data, so as to use the indicator warning threshold under different time windows to more finely judge whether the community to be analyzed is an abnormal community. The present invention combines machine learning algorithms and statistical methods, which can not only timely perform intelligent analysis on operation data, but also ensure the comprehensiveness of abnormal monitoring and improve the accuracy of monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0011] Figure 1 is a flow chart of a community operation data analysis method in one embodiment of the present invention; Figure 2 is a structural diagram of a community operation data analysis device in one embodiment of the present invention; Figure 3 is a schematic diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] The community operation data analysis method provided in this embodiment can be applied to application scenarios where product merchants or service providers conduct business on a community basis, and aims to identify and warn potential abnormal communities by counting and analyzing data. For example, when a smart cabinet operator operates a smart cabinet in a community, community operation data analysis can be to monitor the deposit and withdrawal volume of the smart cabinet to determine whether there is any abnormality in the deposit and withdrawal volume. In actual applications, the smart cabinet operation service end can trigger the process of the community operation data analysis method at regular intervals according to a fixed period, such as collecting data statistics and analyzing the deposit and withdrawal volume of the smart cabinet at 24 hours a day. Community operators can also actively trigger the community operation data analysis method process of the smart cabinet operation service end as needed.

[0014] In one embodiment, if Figure 1 As shown, a community operation data analysis method is provided, comprising the following steps S10-S30: S10, obtaining attribute data to be analyzed of the community to be analyzed and sample attribute data of the sample community, performing cluster analysis on the attribute data to be analyzed and the sample attribute data, and determining similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed.

[0015] Understandably, when conducting business on a community basis, the operation service end can make abnormal judgments on one or more communities among all the communities where business has been conducted. For example, the operation service end of the smart cabinet can only analyze the deposit and withdrawal volume of the smart cabinet in one living community, or it can simultaneously analyze the deposit and withdrawal volume of the smart cabinets in multiple living communities. At this time, all communities where business has been conducted are divided into two types of communities: communities to be analyzed and sample communities. Communities to be analyzed refer to communities that need to use data statistics and analysis to make abnormal judgments, and sample communities refer to communities that exclude communities to be analyzed from all communities where business has been conducted. The operation service end can use the data of sample communities as a reference to conduct a more comprehensive analysis of the communities to be analyzed.

[0016] The attribute data to be analyzed refers to the community attribute data of the community to be analyzed, and the sample attribute data refers to the community attribute data of the sample community. Each community has corresponding community attribute data, which is a data set used to describe the various characteristics and conditions of the community that are different from other communities. Community attribute data includes community code, community name, community geographical location, community city level, community venue type, community business data, community management data, community development type and other data.

[0017] These community attribute data may affect the results of data analysis in different ways. For different communities, community attribute data have a certain degree of similarity, and business fluctuations also have mutual reference value. Cluster analysis is performed on the attribute data to be analyzed and the sample attribute data. The communities to be analyzed and the sample communities can be grouped according to the degree of similarity to obtain several cluster community sets. Cluster analysis is an unsupervised learning method in the field of machine learning. Its main purpose is to divide the data set into multiple groups or clusters so that the data points in the same cluster are similar to each other, while the data points between different clusters are quite different.

[0018] According to different cluster analysis methods (such as direct clustering method and street clustering method), different cluster community sets can be obtained, and then similar cluster communities and street cluster communities corresponding to the community to be analyzed in the sample community can be determined according to different cluster community sets. Similar cluster communities refer to sample communities that belong to the same cluster as the community to be analyzed under the direct clustering method, and street cluster communities refer to sample communities that belong to the same cluster as the community to be analyzed under the street clustering method. When the direct clustering method is adopted, cluster analysis is performed on the attribute data to be analyzed and the sample attribute data directly based on all communities that have carried out business, and a direct cluster community set can be obtained. At this time, the sample community that belongs to the same direct cluster community set as the community to be analyzed is determined as a similar cluster community corresponding to the community to be analyzed. When the street clustering method is adopted, all communities that have carried out business are first divided according to streets, and then cluster analysis is performed on the attribute data to be analyzed and the sample attribute data based on the communities that have carried out business in each street, and a street cluster community set can be obtained. At this time, the sample communities that belong to the same street cluster community set as the community to be analyzed are determined as the street cluster communities corresponding to the community to be analyzed. From the perspective of geographical administrative planning, streets are one level higher than communities, and one street includes multiple communities. In addition, the community to be analyzed may include only one community or multiple communities. When the community to be analyzed includes multiple communities, it is necessary to determine the similar cluster communities and street cluster communities corresponding to each community to be analyzed.

[0019] S20. Obtain similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determine the indicator warning threshold corresponding to the community to be analyzed within the preset time window based on the similar operation index data and the street operation index data.

[0020] Understandably, the preset time window is a specific time span pre-set to measure whether the community operation is normal, such as 1 week, half a month and 1 month. Operation indicator data refers to business quantitative data used to measure whether the community operation is normal, such as the number of items deposited and withdrawn in smart cabinets in units of communities. Similar operation indicator data refers to the operation indicator data of similar clustered communities, and street operation indicator data refers to the operation indicator data of street clustered communities. Similar operation indicator data and street operation indicator data are analyzed using statistical methods (such as mean, standard deviation, percentile, etc.), and fluctuation abnormal values ​​in similar operation indicator data and street operation indicator data are found and combined, and finally the indicator warning threshold corresponding to the community to be analyzed within the preset time window is obtained. That is, the indicator warning threshold corresponding to the community to be analyzed within the preset time window is determined based on the similar operation indicator data and the street operation indicator data. The indicator warning threshold refers to the critical value used to measure whether the operation indicator data of the community to be analyzed is abnormal.

[0021] S30, obtaining the operation indicator data to be analyzed of the community to be analyzed within the preset time window, judging whether the community to be analyzed is an abnormal community according to the operation indicator data to be analyzed and the indicator warning threshold, and generating abnormal warning information when confirming that the community to be analyzed is an abnormal community.

[0022] Understandably, the operational indicator data to be analyzed refers to the operational indicator data of the community to be analyzed, such as the number of items deposited and withdrawn in the smart cabinets of the community to be analyzed. The operational indicator data to be analyzed is compared with the indicator warning threshold to determine whether the operational indicator data to be analyzed exceeds the indicator warning threshold. When the operational indicator data to be analyzed exceeds the indicator warning threshold, it indicates that the fluctuation of the operational indicator data to be analyzed exceeds the normal level. At this time, the community to be analyzed is an abnormal community. That is, whether the community to be analyzed is an abnormal community is determined based on the operational indicator data to be analyzed and the indicator warning threshold. When it is confirmed that the community to be analyzed is an abnormal community, abnormal warning information is generated. The abnormal warning information is information used to prompt that the operation of the community to be analyzed is abnormal. The abnormal warning information may include the name of the abnormal community and the corresponding operational indicator data value, possible causes, and recommended adjustment measures. Furthermore, the abnormal warning information is pushed to the preset warning notification party, which may be the terminal of the community operator.

[0023] In one embodiment, the smart cabinet operation service end triggers the process of the community operation data analysis method according to a fixed period. For example, if the data statistics and analysis of the deposit and withdrawal volume of the smart cabinet are performed 24 hours a day, the smart cabinet operation service end can count the deposit and withdrawal volume of the smart cabinets in each community on a daily basis. The preset time window is 1 week. The smart cabinet operation service end obtains the deposit and withdrawal volume of the smart cabinets in similar clustered communities within the past week as similar operation indicator data, and obtains the deposit and withdrawal volume of the smart cabinets in street clustered communities within the past week as street operation indicator data. The indicator warning threshold corresponding to the community to be analyzed within 1 week is determined based on the similar operation indicator data and the street operation indicator data. The deposit and withdrawal volume of the smart cabinets in the community to be analyzed within 1 week is obtained as the operation indicator data to be analyzed, and the operation indicator data to be analyzed includes the deposit and withdrawal volume of the smart cabinets on that day. It is judged whether the community to be analyzed is an abnormal community based on the operation indicator data to be analyzed and the indicator warning threshold, and abnormal warning information is generated when the community to be analyzed is confirmed to be an abnormal community.

[0024] Based on the acquisition of the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, this embodiment uses a machine learning algorithm to perform cluster analysis on the attribute data to be analyzed and the sample attribute data, and determines the similar cluster communities and street cluster communities corresponding to the community to be analyzed in the sample community. The similar cluster communities can reflect the market fluctuations of a certain type of community, and the street cluster communities can reflect the impact of external factors of the street on the community operation data, thereby improving the comprehensiveness and rationality of abnormal monitoring. In addition, this embodiment also obtains similar operation index data of similar cluster communities and street operation index data of street cluster communities within a preset time window, and uses statistical methods to determine the indicator warning threshold corresponding to the community to be analyzed within the preset time window based on the similar operation index data and the street operation index data, so as to use the indicator warning threshold under different time windows to more finely judge whether the community to be analyzed is an abnormal community. This embodiment combines machine learning algorithms and statistical methods, which can not only timely perform intelligent analysis on operation data, but also ensure the comprehensiveness of abnormal monitoring and improve the accuracy of monitoring results.

[0025] In one embodiment, in step S10, that is, obtaining the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, performing cluster analysis on the attribute data to be analyzed and the sample attribute data, and determining similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed, including: S101. Find out the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community from the community attribute data table, wherein the attribute data to be analyzed includes the city level to be analyzed, the venue type to be analyzed, the development type to be analyzed, the business data to be analyzed, and the operating data to be analyzed, and the sample attribute data includes the sample city level, the sample venue type, the sample development type, the sample business data, and the sample operating data; S102, performing a cluster analysis between the city level to be analyzed and the sample city level, and between the place type to be analyzed and the sample place type, and dividing all the communities to be analyzed and the sample communities into a plurality of spatial cluster community sets according to the result of the cluster analysis; S103, performing a secondary cluster analysis between the development type to be analyzed and the sample development type in the same spatial cluster community set, and dividing all the communities to be analyzed and the sample communities into a plurality of development cluster community sets according to the result of the secondary cluster analysis; S104, performing three cluster analyses on the business data to be analyzed and the sample business data in the same development cluster community set, and dividing all the communities to be analyzed and the sample communities into a plurality of business cluster community sets according to the results of the three cluster analyses; S105, performing four cluster analyses on the business data to be analyzed and the sample business data in the same business cluster community set, and dividing all the communities to be analyzed and the sample communities into a plurality of business cluster community sets according to the results of the four cluster analyses; S106: Determine a sample community that belongs to the same business cluster community set as the community to be analyzed as a similar cluster community corresponding to the community to be analyzed.

[0026] Understandably, before cluster analysis is performed on the attribute data to be analyzed and the sample attribute data, it is necessary to find the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community from the community attribute data table. The community attribute data table is a pre-constructed data table for recording the association between different communities and the community attribute data corresponding to each community. After the attribute data to be analyzed and the sample attribute data are divided according to the community attribute type, the community attribute data includes community code, community name, community geographical location, community city level, community venue type, community business data, community operation data, community development type and other data, but the community they belong to is different. Among them, the community geographical location is used to characterize the address information of the province, city, and district where the community is located; the community city level is used to characterize the information that the city where the community is located is first-tier, second-tier, third-tier or fourth-tier; the community venue type is used to characterize whether the community is a school district community centered on a school or a business district community centered on a shopping mall; the community development type is used to characterize whether the community is a newly developed community or a backward tail community; the community business data refers to the community user data that the operation is concerned about, such as order volume, sales volume, storage and retrieval volume, etc.; the community operation data is used to characterize the complaint rate and turnover rate of the community. Preferably, the attribute data to be analyzed include city levels to be analyzed, place types to be analyzed, development types to be analyzed, business data to be analyzed and operating data to be analyzed, and the sample attribute data include sample city levels, sample place types, sample development types, sample business data and sample operating data.

[0027] When the direct clustering method is used, in the process of clustering the attribute data to be analyzed and the sample attribute data, it is necessary to perform four progressive clustering analyses according to different types of community attribute data, and finally obtain a direct clustering community set.

[0028] First, a cluster analysis is performed between the city level to be analyzed and the sample city level, and between the place type to be analyzed and the sample place type. Based on the results of the cluster analysis, all communities to be analyzed and sample communities are divided into multiple spatial cluster community sets. A cluster analysis is a classification process used to compare whether the spatial factors (city level and place type) between the communities to be analyzed and the sample communities are the same. The spatial cluster community set refers to the cluster community set obtained by grouping the communities to be analyzed and the sample communities based on a cluster analysis.

[0029] Secondly, a secondary cluster analysis is performed between the development types to be analyzed and the sample development types in the same spatial cluster community set, and all the communities to be analyzed and the sample communities are divided into multiple development cluster community sets based on the results of the secondary cluster analysis. Secondary cluster analysis is a classification process used to compare whether the development types between the communities to be analyzed and the sample communities are the same. The operating conditions of newly developed communities are in the growth stage and are not easily affected, while the operating conditions of tail communities are easily affected by many factors, which is reflected in the fact that the fluctuations in operating data are greater than those in newly developed communities. Therefore, different standards need to be used for abnormal data monitoring. The development cluster community set refers to the cluster community set obtained by further grouping the communities to be analyzed and the sample communities in the spatial cluster community set based on the secondary cluster analysis.

[0030] Then, three clustering analyses are performed between the business data to be analyzed and the sample business data in the same development clustering community set, and all the communities to be analyzed and the sample communities are divided into multiple business clustering community sets according to the results of the three clustering analyses. Three-dimensional clustering analysis is a classification process used to compare whether the business data fluctuations between the communities to be analyzed and the sample communities are similar. For example, features that characterize the inherent fluctuations in the community operation process are extracted as the basis for similarity measurement (including mean, variance, skewness, kurtosis, etc., time domain features, including trends, seasonal fluctuations, autocorrelation, etc.), and then various clustering algorithms (including but not limited to Kmeans, SOM, etc.) are used to cluster these features to find clusters with similar fluctuations. The business clustering community set refers to the clustering community set obtained by further grouping the communities to be analyzed and the sample communities in the development clustering community set based on three-dimensional clustering analysis.

[0031] Finally, four cluster analyses are performed between the business data to be analyzed and the sample business data in the same business cluster community set. According to the results of the four cluster analyses, all the communities to be analyzed and the sample communities are divided into multiple business cluster community sets, namely direct cluster community sets. Four-fold cluster analysis is a classification process used to compare whether the fluctuation degree of business data between the community to be analyzed and the sample community is similar. For example, clustering algorithms (including but not limited to K-means, SOM, etc.) are used to perform more fine-grained clustering of indicators such as turnover rate to find clusters with similar fluctuation degrees. The business cluster community set refers to the cluster community set obtained by further grouping the communities to be analyzed and the sample communities in the business cluster community set based on the four cluster analyses. Furthermore, the sample community that belongs to the same business cluster community set as the community to be analyzed is determined as a similar cluster community corresponding to the community to be analyzed.

[0032] This embodiment gradually refines the dimensions of cluster analysis, starting from the spatial factors of the community, and gradually going deeper into the development stage, business fluctuations and operating indicators, and finally finds the sample community that is most similar to the community to be analyzed and uses it as a similar cluster community, thereby ensuring the rationality of the similar cluster community, improving the reference value of the similar cluster community, and helping to improve the accuracy of the monitoring results.

[0033] In one embodiment, step S101, that is, before searching the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community from the community attribute data table, includes: S1011. Obtain historical operation data of all operation communities; S1012, integrating and processing the historical operation data according to community attribute types to generate a community attribute data group corresponding to the operation community; S1013. Generate the community attribute data table according to all the community attribute data groups, so that when the community to be analyzed is found from the operating community, the community attribute data group corresponding to the community to be analyzed is determined as the attribute data to be analyzed, and when the sample community is found from the operating community, the community attribute data group corresponding to the sample community is determined as the sample attribute data.

[0034] Understandably, before searching for the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community from the community attribute data table, it is necessary to obtain the historical operation data of all operating communities from the database, and construct the community attribute data table with community attribute type and community dimension. The operating community refers to all communities that have launched services. Historical operation data refers to the operation data that have launched services within a specific historical time period (such as 1 month). Community attribute types include community code, community name, community geographical location, community city level, community venue type, community development type, community business type, community operation type, etc. Each piece of historical operation data includes data of all community attribute types, corresponding to different data table fields. The historical operation data is integrated and processed according to the community attribute type to generate a community attribute data group corresponding to each operating community. The community attribute data group is structured data used to record and manage the community and its corresponding community attribute data. For example, the data under the "community name" field is associated with the data under the "community city level, community venue type, community development type, community business type, community operation type" and other fields to generate a community attribute data group. Finally, the collection of all community attribute data groups is determined as a community attribute data table, so as to find out all community attribute data corresponding to each community. That is, when a community to be analyzed is found from an operating community, the community attribute data group corresponding to the community to be analyzed is determined as the attribute data to be analyzed, and when the sample community is found from an operating community, the community attribute data group corresponding to the sample community is determined as the sample attribute data. In addition, the community attribute data table can add new community attribute data groups when needed, which is easy to use and extensible.

[0035] This embodiment constructs a community attribute data table based on historical operation data according to community attribute types and community dimensions. The community attribute data table can clearly display the community attribute data of each community to facilitate subsequent query, comparison and analysis.

[0036] In one embodiment, step S10, i.e., obtaining the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, performing cluster analysis on the attribute data to be analyzed and the sample attribute data, and determining similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed, further includes: S107, determining a street sample community corresponding to the community to be analyzed in the sample community according to the street operation data table associated with the community to be analyzed, and obtaining attribute data to be analyzed of the community to be analyzed and street sample attribute data of the street sample community; wherein the attribute data to be analyzed includes a place type to be analyzed and a development type to be analyzed, and the street sample attribute data includes a street sample place type and a street sample development type; S108, performing cluster analysis between the place types to be analyzed and the street sample place types, and between the development types to be analyzed and the street sample development types, and dividing all the communities to be analyzed and the street sample communities into a plurality of street cluster community sets according to the results of the cluster analysis; S109: Determine a street sample community that belongs to the same street cluster community set as the community to be analyzed as a street cluster community corresponding to the community to be analyzed.

[0037] It is understandable that before cluster analysis is performed on the attribute data to be analyzed and the sample attribute data, it is necessary to determine the street sample community corresponding to the community to be analyzed in the sample community from the street operation data table associated with the community to be analyzed, and obtain the attribute data to be analyzed of the community to be analyzed and the street sample attribute data of the street sample community. The street operation data table is a pre-constructed data table for recording the association relationship between each community in the same street and the community attribute data corresponding to the community. The street sample community is a sample community that belongs to the same street as the community to be analyzed. Similar to the data in the above-mentioned community attribute data table, the attribute data to be analyzed and the street sample attribute data are only different in the communities they belong to. Preferably, the attribute data to be analyzed include the place type to be analyzed and the development type to be analyzed, and the street sample attribute data include the street sample place type and the street sample development type.

[0038] When the street clustering method is adopted, all communities that have started business are first divided according to streets, and then cluster analysis is performed on the attribute data to be analyzed and the sample attribute data based on the communities that have started business in each street. In the process of clustering the attribute data to be analyzed and the sample attribute data, street clustering analysis is performed between the place type to be analyzed and the street sample place type, and between the development type to be analyzed and the street sample development type. According to the results of the street clustering analysis, all communities to be analyzed and street sample communities are divided into multiple street clustering community sets. Street clustering analysis is a classification process used to compare whether the place type and development type between the community to be analyzed and the street sample community within the same street are the same. Furthermore, the street sample community that belongs to the same street clustering community set as the community to be analyzed is determined as the street clustering community corresponding to the community to be analyzed.

[0039] This embodiment clusters and groups the communities within the same street based on spatial factors and development stage dimensions, and ultimately finds the street sample community that is most similar to the community to be analyzed and uses it as the street cluster community, thereby improving the reference value of the street cluster community and helping to improve the accuracy of the monitoring results.

[0040] In one embodiment, step S107, that is, before determining the street sample community corresponding to the community to be analyzed in the sample community according to the street operation data table associated with the community to be analyzed, includes: S1071. Obtain historical operation data and street ownership information of all operating communities; S1072. Integrate the historical operation data according to the street affiliation information to generate a street operation data table associated with each of the operation communities, so that when the community to be analyzed is found from the operation community, the operation community with the same street affiliation information as the community to be analyzed is found from the street operation data table, and the found operation community is determined as the street sample community corresponding to the community to be analyzed.

[0041] Understandably, a community can only belong to one street. When the community and street planning is completed, each community has fixed and unique street ownership information. Street ownership information is information used to characterize the street to which the community belongs. Before determining the street sample community corresponding to the community to be analyzed in the sample community according to the street operation data table associated with the community to be analyzed, it is necessary to obtain the historical operation data and street ownership information of all operating communities from the database. On the basis of the same street ownership information, the street operation data table associated with each operating community is constructed based on the community attribute type and community dimension. One street corresponds to a street operation data table. Similar to the community attribute data table, the street operation data table consists of multiple community attribute data groups. Each piece of historical operation data includes data of all community attribute types, corresponding to different data table fields. The historical operation data is integrated and processed according to the community attribute type to generate a community attribute data group corresponding to each operating community. The community attribute data group is structured data used to record and manage the community and its corresponding community attribute data. For example, the data under the "community name" field is associated with the data under the fields such as "community city level, community venue type, community development type, community business type, and community operation type" to generate a community attribute data group. Finally, the set of community attribute data groups of the operating communities with the same street ownership information is determined as the street operating data table, so as to find out all community attribute data corresponding to each community. That is, when a community to be analyzed is found from the operating community, the community attribute data group corresponding to the community to be analyzed is determined as the attribute data to be analyzed. At the same time, a street sample community with the same street ownership information as the community to be analyzed is found from the street operating data table, and the community attribute data group corresponding to the street sample community is determined as the street sample attribute data.

[0042] Based on the same street ownership information, this embodiment constructs a street operation data table according to the community attribute type and community dimension based on the historical operation data. The street operation data table can clearly display the community attribute data of each community to facilitate subsequent query, comparison and analysis.

[0043] In one embodiment, in step S20, that is, obtaining similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determining the index warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data, includes: S201, obtaining similar operation index data of the similar clustering community within a preset time window, and determining a distribution index and a first fluctuation index according to the similar operation index data; S202, obtaining street operation index data of the street cluster community within the preset time window, and determining a second fluctuation index according to the street operation index data; S203: Determine an indicator warning threshold corresponding to the community to be analyzed within the preset time window according to the distribution indicator, the first fluctuation indicator, and the second fluctuation indicator.

[0044] Understandably, similar operation indicator data of similar clustered communities, street operation indicator data of street clustered communities, and operation indicator data of the to-be-analyzed communities can all be considered to obey normal distribution. Based on the 3sigma model, the indicator warning threshold corresponding to the to-be-analyzed community within the preset time window is determined based on similar operation indicator data and street operation indicator data. The 3sigma model is a statistical model that is used when the number is large enough and the data obeys normal distribution or approximate normal distribution. 3sigma first assumes that a set of data contains only random errors, calculates and processes these data to obtain the standard deviation, and determines an interval according to a certain probability, usually 3 times the standard deviation. If the error exceeds 3 times the standard deviation, it is not a random error but a gross error, and the data containing the error can be regarded as an outlier. For example, the probability of plus or minus 3 times the standard deviation (3sigma or 3σ) is 99.7%, so the probability of a value outside 3 times the standard deviation from the mean is 0.003, which is an extremely rare low-probability event.

[0045] In one embodiment, the preset time window is 1 month, and the operation index data is the smart cabinet access volume data of the community, and is counted on a daily basis. First, similar operation index data of similar clustered communities within the preset time window are obtained, and the distribution index and the first volatility index are determined based on the similar operation index data. The distribution index refers to 3 times the standard deviation of similar operation index data under normal distribution. The first volatility index refers to the month-on-month error of similar operation index data within two adjacent preset time windows. That is, the smart cabinet access volume data of similar clustered communities in the most recent 1 month (such as the 10th of the current month to the 10th of the previous month) is obtained with the current day as the starting point, and the smart cabinet access volume data is normally distributed to calculate 3 times the standard deviation and used as the distribution index, represented by 3σ. At the same time, the smart cabinet access volume data of similar clustered communities in the most recent 2 months (such as the 10th of the previous month to the 10th of the previous month) is obtained with the current day as the starting point, and the average month-on-month error of the smart cabinet access volume between the most recent 1 month and the most recent 2 months is calculated and used as the first volatility index, represented by u1. Secondly, the street operation index data of the street cluster community within the preset time window is obtained, and the second fluctuation index is determined based on the street operation index data. The second fluctuation index refers to the month-on-month error of the street operation index data within two adjacent preset time windows. That is, the smart cabinet access volume data of the street cluster community in the last 1 month and the last 2 months is obtained with the current day as the starting point, and the average month-on-month error of the smart cabinet access volume between the last 1 month and the last 2 months is calculated and used as the second fluctuation index, represented by u2. Finally, the indicator warning threshold corresponding to the community to be analyzed within the preset time window is determined based on the distribution index, the first fluctuation index and the second fluctuation index. Since the distribution index can reflect statistical fluctuations, the first fluctuation index can reflect the fluctuations of the market, and the second fluctuation index can reflect the fluctuations of the street, the distribution index, the first fluctuation index and the second fluctuation index are combined to obtain the indicator warning threshold, represented by threshold=3σ+(u2-u1). That is, the indicator warning threshold is the difference between the second volatility index and the first volatility index plus the distribution index. The difference "u2-u1" between the second volatility index and the first volatility index represents the difference in volatility between the street and the market.

[0046] Furthermore, when using the indicator warning threshold to judge whether the community to be analyzed is abnormal, the data of the smart cabinet deposit and withdrawal volume of the community to be analyzed in the first month and the second month are obtained from the current day as the starting point, and the average month-on-month error of the smart cabinet deposit and withdrawal volume of the community to be analyzed between the first month and the second month is calculated, represented by u3. When u3 is less than threshold=3σ+(u2-u1), the community to be analyzed is confirmed to be an abnormal community.

[0047] Based on the 3sigma model, this embodiment comprehensively considers the fluctuations of similar cluster communities and street cluster communities, and proposes a clear indicator warning threshold calculation method, which improves the rationality of the indicator warning threshold and can effectively identify abnormal data fluctuations.

[0048] In one embodiment, the preset time window includes a first preset time window, a second preset time window and a third preset time window, the second preset time window is larger than the first preset time window and smaller than the third preset time window; the indicator warning threshold includes a first indicator warning threshold corresponding to the first preset time window, a second indicator warning threshold corresponding to the second preset time window and a third indicator warning threshold corresponding to the third preset time window; in step S30, that is, judging whether the community to be analyzed is an abnormal community according to the operation indicator data to be analyzed and the indicator warning threshold, includes: S301, calculating a first indicator value to be analyzed corresponding to the first preset time window, a second indicator value to be analyzed corresponding to the second preset time window, and a third indicator value to be analyzed corresponding to the third preset time window according to the operation indicator data to be analyzed; S302: When the first indicator value to be analyzed reaches the first indicator warning threshold, and / or the second indicator value to be analyzed reaches the second indicator warning threshold, and / or the third indicator value to be analyzed reaches the third indicator warning threshold, determine that the community to be analyzed is an abnormal community.

[0049] Understandably, in one embodiment, based on operational needs, operational data analysis and early warning analysis can be performed in multiple time windows at the same time, such as short-term abnormal early warning analysis, medium-term abnormal early warning analysis, and long-term abnormal early warning analysis. At this time, the first preset time window refers to a specific time span used to measure whether the community operation is normal in the short term, such as one week, that is, the short-term abnormal early warning analysis is used to warn the community that has obvious abnormal data fluctuations in the past week compared to the previous week. The second preset time window refers to a specific time span used to measure whether the community operation is normal in the medium term, such as half a month, that is, the medium-term abnormal early warning analysis is used to warn the community that has obvious abnormal data fluctuations in the past half month compared to the first half of the month. The third preset time window refers to a specific time span used to measure whether the community operation is normal in the long term, such as one month, that is, the long-term abnormal early warning analysis is used to warn the community that has obvious abnormal data fluctuations in the past month compared to the previous month. In addition, for different time windows, the values ​​of the distribution index, the first fluctuation index, and the second fluctuation index are different, so the indicator warning thresholds for each community in the short, medium, and long term are also different. The first indicator warning threshold is the indicator warning threshold for short-term abnormal warning analysis, the second indicator warning threshold is the indicator warning threshold for medium-term abnormal warning analysis, and the third indicator warning threshold is the indicator warning threshold for long-term abnormal warning analysis. Similarly, for different time windows, the indicator values ​​to be analyzed calculated by the community to be analyzed are also different. The first indicator value to be analyzed is the indicator value for short-term abnormal warning analysis (such as the month-on-month increase in the number of deposits and withdrawals of smart cabinets with a time span of one week), the second indicator value to be analyzed is the indicator value for medium-term abnormal warning analysis (such as the month-on-month increase in the number of deposits and withdrawals of smart cabinets with a time span of half a month), and the third indicator value to be analyzed is the indicator value for long-term abnormal warning analysis (such as the month-on-month increase in the number of deposits and withdrawals of smart cabinets with a time span of one month).

[0050] Compare the first indicator value to be analyzed with the first indicator warning threshold, the second indicator value to be analyzed with the second indicator warning threshold, and the third indicator value to be analyzed with the third indicator warning threshold, respectively, to determine whether the community to be analyzed triggers short-term, medium-term, and long-term abnormal warnings. The community to be analyzed may only trigger one of the abnormal warnings (such as a short-term abnormal warning), or may trigger two abnormal warnings at the same time (such as a short-term abnormal warning and a medium-term abnormal warning), or may trigger three abnormal warnings at the same time. At this time, the community to be analyzed can be determined as an abnormal community, and abnormal warning information of different levels can be generated. For example, when only one abnormal warning is triggered, it is a low-level warning, when two abnormal warnings are triggered at the same time, it is a medium-level warning, and when three abnormal warnings are triggered at the same time, it is a high-level warning. In addition, if the community to be analyzed does not trigger any of the short-term, medium-term, and long-term abnormal warnings, it indicates that the community to be analyzed is a normal community.

[0051] This embodiment can simultaneously examine abnormal data fluctuations in different time windows and effectively identify abnormal communities in short-term, medium-term and long-term trends, covering more data anomaly situations, and can effectively help community operators take different measures to make adjustments.

[0052] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0053] In one embodiment, a community operation data analysis device is provided, and the community operation data analysis device corresponds one-to-one to the community operation data analysis method in the above embodiment. Figure 2 As shown, the community operation data analysis device includes a cluster analysis processing module 10, a warning threshold determination module 20 and an abnormal community judgment module 30. The functional modules are described in detail as follows: The cluster analysis processing module 10 is used to obtain the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, perform cluster analysis on the attribute data to be analyzed and the sample attribute data, and determine similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed; The warning threshold determination module 20 is used to obtain similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determine the indicator warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data; The abnormal community judgment module 30 is used to obtain the operation indicator data to be analyzed of the community to be analyzed within the preset time window, judge whether the community to be analyzed is an abnormal community according to the operation indicator data to be analyzed and the indicator warning threshold, and generate abnormal warning information when it is confirmed that the community to be analyzed is an abnormal community.

[0054] In one embodiment, the cluster analysis processing module 10 includes: A community attribute data table search unit, used to search for attribute data to be analyzed of the community to be analyzed and sample attribute data of the sample community from the community attribute data table, wherein the attribute data to be analyzed includes the city level to be analyzed, the venue type to be analyzed, the development type to be analyzed, the business data to be analyzed and the management data to be analyzed, and the sample attribute data includes the sample city level, the sample venue type, the sample development type, the sample business data and the sample management data; A primary cluster analysis unit, used to perform a primary cluster analysis between the city level to be analyzed and the sample city level, and between the place type to be analyzed and the sample place type, and divide all the communities to be analyzed and the sample communities into a plurality of spatial cluster community sets according to the result of the primary cluster analysis; A secondary cluster analysis unit, used for performing secondary cluster analysis between the development type to be analyzed and the sample development type in the same spatial cluster community set, and dividing all the communities to be analyzed and the sample communities into a plurality of development cluster community sets according to the result of the secondary cluster analysis; A tertiary clustering analysis unit, configured to perform a tertiary clustering analysis on the business data to be analyzed and the sample business data in the same development clustering community set, and divide all the communities to be analyzed and the sample communities into a plurality of business clustering community sets according to the results of the tertiary clustering analysis; A four-time cluster analysis unit, used to perform four-time cluster analysis on the business data to be analyzed and the sample business data in the same business cluster community set, and divide all the communities to be analyzed and the sample communities into a plurality of business cluster community sets according to the results of the four-time cluster analysis; The similar cluster community determination unit is used to determine the sample community that belongs to the same business cluster community set as the community to be analyzed as the similar cluster community corresponding to the community to be analyzed.

[0055] In one embodiment, the cluster analysis processing module 10 further includes: A historical operation data acquisition unit is used to acquire historical operation data of all operation communities; A community attribute data group generating unit, used for integrating the historical operation data according to the community attribute type, and generating a community attribute data group corresponding to each of the operation communities; A community attribute data table generating unit is used to generate the community attribute data table according to all the community attribute data groups, so that when the community to be analyzed is found from the operating community, the community attribute data group corresponding to the community to be analyzed is determined as the attribute data to be analyzed, and when the sample community is found from the operating community, the community attribute data group corresponding to the sample community is determined as the sample attribute data.

[0056] In one embodiment, the cluster analysis processing module 10 further includes: A street operation data table search unit, used to determine the street sample community corresponding to the community to be analyzed in the sample community according to the street operation data table associated with the community to be analyzed, and obtain the attribute data to be analyzed of the community to be analyzed and the street sample attribute data of the street sample community; wherein the attribute data to be analyzed includes the place type to be analyzed and the development type to be analyzed, and the street sample attribute data includes the street sample place type and the street sample development type; A street cluster analysis unit, used for performing cluster analysis between the place type to be analyzed and the street sample place type, and between the development type to be analyzed and the street sample development type, and dividing all the communities to be analyzed and the street sample communities into a plurality of street cluster community sets according to the result of the cluster analysis; The street cluster community determination unit is used to determine the street sample communities that belong to the same street cluster community set as the community to be analyzed as the street cluster community corresponding to the community to be analyzed.

[0057] In one embodiment, the cluster analysis processing module 10 further includes: A street ownership information acquisition unit is used to acquire historical operation data and street ownership information of all operating communities; A street operation data table generating unit is used to integrate and process the historical operation data according to the street affiliation information, and generate a street operation data table associated with each of the operation communities, so that when the community to be analyzed is found from the operation community, the operation community with the same street affiliation information as the community to be analyzed is found from the street operation data table, and the found operation community is determined as the street sample community corresponding to the community to be analyzed.

[0058] In one embodiment, the warning threshold determination module 20 includes: A similar cluster community fluctuation analysis unit, used to obtain similar operation index data of the similar cluster community within a preset time window, and determine a distribution index and a first fluctuation index according to the similar operation index data; A street cluster community fluctuation analysis unit, used to obtain street operation index data of the street cluster community within the preset time window, and determine a second fluctuation index according to the street operation index data; The indicator warning threshold determination unit is used to determine the indicator warning threshold corresponding to the community to be analyzed within the preset time window according to the distribution indicator, the first fluctuation indicator and the second fluctuation indicator.

[0059] In one embodiment, the abnormal community determination module 30 includes: an indicator determination unit to be analyzed, configured to calculate a first indicator value to be analyzed corresponding to the first preset time window, a second indicator value to be analyzed corresponding to the second preset time window, and a third indicator value to be analyzed corresponding to the third preset time window according to the operation indicator data to be analyzed; The abnormal community determining unit is used to determine that the community to be analyzed is an abnormal community when the first indicator value to be analyzed reaches the first indicator warning threshold, and / or the second indicator value to be analyzed reaches the second indicator warning threshold, and / or the third indicator value to be analyzed reaches the third indicator warning threshold.

[0060] For the specific definition of the community operation data analysis device, please refer to the definition of the community operation data analysis method above, which will not be repeated here. Each module in the above-mentioned community operation data analysis device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0061] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The readable storage medium stores an operating system, computer-readable instructions and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The database of the computer device is used to store data involved in the community operation data analysis method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, a community operation data analysis method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0062] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor executes the computer-readable instructions to implement the following steps: Acquire the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, perform cluster analysis on the attribute data to be analyzed and the sample attribute data, and determine similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed; Obtain similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determine an indicator warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data; The operation indicator data to be analyzed of the community to be analyzed within the preset time window is obtained, and whether the community to be analyzed is an abnormal community is determined according to the operation indicator data to be analyzed and the indicator warning threshold, and abnormal warning information is generated when it is confirmed that the community to be analyzed is an abnormal community.

[0063] In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided. The readable storage media provided in this embodiment include non-volatile readable storage media and volatile readable storage media. The readable storage media store computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the following steps are implemented: Acquire the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, perform cluster analysis on the attribute data to be analyzed and the sample attribute data, and determine similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed; Obtain similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determine an indicator warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data; The operation indicator data to be analyzed of the community to be analyzed within the preset time window is obtained, and whether the community to be analyzed is an abnormal community is determined according to the operation indicator data to be analyzed and the indicator warning threshold, and abnormal warning information is generated when it is confirmed that the community to be analyzed is an abnormal community.

[0064] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions. The computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0065] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0066] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A community operation data analysis method, characterized in that: include: Acquire the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, perform cluster analysis on the attribute data to be analyzed and the sample attribute data, and determine similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed; Obtain similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determine an indicator warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data; The operation indicator data to be analyzed of the community to be analyzed within the preset time window is obtained, and whether the community to be analyzed is an abnormal community is determined according to the operation indicator data to be analyzed and the indicator warning threshold, and abnormal warning information is generated when it is confirmed that the community to be analyzed is an abnormal community.

2. The community operation data analysis method according to claim 1, characterized in that: The step of obtaining the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, performing cluster analysis on the attribute data to be analyzed and the sample attribute data, and determining similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed, includes: Find out the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community from the community attribute data table, wherein the attribute data to be analyzed includes the city level to be analyzed, the venue type to be analyzed, the development type to be analyzed, the business data to be analyzed and the operating data to be analyzed, and the sample attribute data includes the sample city level, the sample venue type, the sample development type, the sample business data and the sample operating data; Performing a cluster analysis between the city level to be analyzed and the sample city level, and between the place type to be analyzed and the sample place type, and dividing all the communities to be analyzed and the sample communities into a plurality of spatial cluster community sets according to the result of the cluster analysis; Performing a secondary cluster analysis between the development type to be analyzed and the sample development type in the same spatial cluster community set, and dividing all the communities to be analyzed and the sample communities into a plurality of development cluster community sets according to the result of the secondary cluster analysis; Performing three clustering analyses on the business data to be analyzed and the sample business data in the same development cluster community set, and dividing all the communities to be analyzed and the sample communities into a plurality of business cluster community sets according to the results of the three clustering analyses; Performing four cluster analyses on the business data to be analyzed and the sample business data in the same business cluster community set, and dividing all the communities to be analyzed and the sample communities into a plurality of business cluster community sets according to the results of the four cluster analyses; The sample communities that belong to the same business cluster community set as the community to be analyzed are determined as similar cluster communities corresponding to the community to be analyzed.

3. The community operation data analysis method according to claim 2, characterized in that: Before searching the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community from the community attribute data table, the method includes: Obtain historical operational data for all operational communities; Integrate the historical operation data according to community attribute types to generate community attribute data groups corresponding to each of the operation communities; The community attribute data table is generated based on all the community attribute data groups, so that when the community to be analyzed is found from the operating community, the community attribute data group corresponding to the community to be analyzed is determined as the attribute data to be analyzed, and when the sample community is found from the operating community, the community attribute data group corresponding to the sample community is determined as the sample attribute data.

4. The community operation data analysis method according to claim 1, characterized in that: The method of obtaining the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, performing cluster analysis on the attribute data to be analyzed and the sample attribute data, and determining similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed, further includes: Determine a street sample community corresponding to the community to be analyzed in the sample community according to the street operation data table associated with the community to be analyzed, and obtain attribute data to be analyzed of the community to be analyzed and street sample attribute data of the street sample community; wherein the attribute data to be analyzed includes a place type to be analyzed and a development type to be analyzed, and the street sample attribute data includes a street sample place type and a street sample development type; Performing cluster analysis between the place type to be analyzed and the street sample place type, and between the development type to be analyzed and the street sample development type, and dividing all the communities to be analyzed and the street sample communities into a plurality of street cluster community sets according to the results of the cluster analysis; The street sample communities that belong to the same street cluster community set as the community to be analyzed are determined as the street cluster communities corresponding to the community to be analyzed.

5. The community operation data analysis method according to claim 4, characterized in that: Before determining the street sample community corresponding to the community to be analyzed in the sample community according to the street operation data table associated with the community to be analyzed, the method includes: Obtain historical operation data and street ownership information for all operating communities; The historical operation data is integrated and processed according to the street affiliation information to generate a street operation data table associated with each of the operation communities, so that when the community to be analyzed is found from the operation community, the operation community with the same street affiliation information as the community to be analyzed is found from the street operation data table, and the found operation community is determined as the street sample community corresponding to the community to be analyzed.

6. The community operation data analysis method according to claim 1, characterized in that: The obtaining of similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determining the index warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data, includes: Acquire similar operation index data of the similar clustering community within a preset time window, and determine a distribution index and a first fluctuation index according to the similar operation index data; Obtaining street operation index data of the street cluster community within the preset time window, and determining a second fluctuation index according to the street operation index data; An indicator warning threshold corresponding to the community to be analyzed within the preset time window is determined according to the distribution indicator, the first fluctuation indicator and the second fluctuation indicator.

7. The community operation data analysis method according to claim 1, characterized in that: The preset time window includes a first preset time window, a second preset time window and a third preset time window, the second preset time window is larger than the first preset time window and smaller than the third preset time window; the indicator warning threshold includes a first indicator warning threshold corresponding to the first preset time window, a second indicator warning threshold corresponding to the second preset time window and a third indicator warning threshold corresponding to the third preset time window; The determining whether the community to be analyzed is an abnormal community according to the operation indicator data to be analyzed and the indicator warning threshold comprises: Calculate, according to the to-be-analyzed operational indicator data, a first to-be-analyzed indicator value corresponding to the first preset time window, a second to-be-analyzed indicator value corresponding to the second preset time window, and a third to-be-analyzed indicator value corresponding to the third preset time window; When the first indicator value to be analyzed reaches the first indicator warning threshold, and / or the second indicator value to be analyzed reaches the second indicator warning threshold, and / or the third indicator value to be analyzed reaches the third indicator warning threshold, the community to be analyzed is determined to be an abnormal community.

8. A community operation data analysis device, characterized in that: include: A cluster analysis processing module, used to obtain the attribute data to be analyzed of the community to be analyzed and the sample attribute data of the sample community, perform cluster analysis on the attribute data to be analyzed and the sample attribute data, and determine similar cluster communities and street cluster communities in the sample community corresponding to the community to be analyzed; An early warning threshold determination module is used to obtain similar operation index data of the similar clustered communities and street operation index data of the street clustered communities within a preset time window, and determine the indicator early warning threshold corresponding to the community to be analyzed within the preset time window according to the similar operation index data and the street operation index data; The abnormal community judgment module is used to obtain the operation indicator data to be analyzed of the community to be analyzed within the preset time window, judge whether the community to be analyzed is an abnormal community according to the operation indicator data to be analyzed and the indicator warning threshold, and generate abnormal warning information when it is confirmed that the community to be analyzed is an abnormal community.

9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the community operation data analysis method as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by one or more processors, the one or more processors execute the community operation data analysis method as described in any one of claims 1 to 7.

Images