Data integration method, system and medium for multi-classification sports events

Abstract

The present application relates to the technical field of data classified storage, in particular to a data integration method and system for multi-classification sports events and a medium. First, the time sequence collection frequency of each professional data is subjected to cluster analysis to identify the multi-modal frequency characteristics thereof. Second, the collaborative index between the frequency cluster of different professional data is calculated to divide the high-correlation data cluster and reduce the mixed storage of low-correlation data and the null value rate of the relational database. Further, the write demand of each cluster is dynamically calculated in combination with the frequency matching strength and storage space occupied in the data cluster, so that the professional data is integrated and stored. In summary, the present application clusters the different types of professional data according to the collaboration of the frequency variation, reduces the null value problem in the storage process of the relational database, analyzes the storage space occupied by the professional data, and reduces the write pressure of the server.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data classification storage, in particular to a data integration method and system for multi-classification sports events and a medium. BACKGROUND

[0002] In the current field of sports event data analysis, with the rapid development of sensor technology, Internet of Things devices and data collection systems, multi-classification sports events (such as track and field, swimming, ball games, cycling, etc.) have generated massive, multi-dimensional time series data. These multi-dimensional professional data (such as cadence, speed, athlete heart rate, etc. in cycling events) are widely collected, stored and applied in training and competition, which can serve scenarios such as event management, audience experience optimization, business decision-making and athlete performance improvement.

[0003] In the prior art, a relational database is usually used to store multi-classification sports event data. However, due to the large difference in the collection frequency of professional data at different stages of the sports event, a large number of redundant field null values are generated in the database mode design. Especially in the face of high-concurrency real-time data updates, the relational database may have bottlenecks in scalability and response speed, making it difficult to efficiently process massive and variable data types. SUMMARY

[0004] In order to solve the technical problem that the collection frequency of professional data at different stages of the sports event is large, which leads to a large number of redundant field null values in the database mode design, especially in the face of high-concurrency real-time data updates, the relational database may have bottlenecks in scalability and response speed, making it difficult to efficiently process massive and variable data types, the purpose of the present application is to provide a data integration method and system for multi-classification sports events and a medium, the technical solution adopted is as follows:

[0005] A data integration method for multi-classification sports events, comprising:

[0006] In each sports event, the event data set of each athlete is obtained, which includes the collection frequency time series data of multiple professional data and the storage space occupied by each professional data;

[0007] In each event data set, the frequency values at all time points in each collection frequency time series data are subjected to cluster analysis to obtain frequency clusters. In the frequency clusters of different professional data, the similarity between the frequency clusters and the number of time points in the frequency clusters are comprehensively considered to determine the coordination index between any two professional data in each event data set, which is used for cluster analysis of professional data to obtain data clusters;

[0008] In the data clustering cluster of each event data set, the writing demand of each data clustering cluster is determined based on the matching between the frequency clustering clusters of the professional data and in combination with the storage space occupied by the professional data;

[0009] In all event data sets of all kinds of sports events, all event data sets are integrated and stored based on the writing demand of the data clustering cluster.

[0010] Further, the method for obtaining the coordination index comprises:

[0011] In each event data set, the frequency clustering clusters in any two kinds of professional data are combined in pairs to obtain all non-repeating combination clusters;

[0012] In each combination cluster, the similarity of the frequency distribution and the time between the frequency clustering clusters is analyzed to determine the frequency overlap index between the two frequency clustering clusters;

[0013] In each event data set, an optional professional data is selected as a target data, and in the target data and any remaining professional data, the maximum value of the frequency overlap index in each frequency clustering cluster in the target data is taken as the frequency domain overlap degree of each frequency clustering cluster in the target data;

[0014] Based on the number of times in the frequency clustering cluster in the target data, the frequency domain overlap degree corresponding to the frequency clustering cluster is weighted and fused to obtain the similarity index between the target data and the professional data;

[0015] In any two kinds of professional data, the normalized value of the average of the similarity index between the two kinds of professional data is taken as the coordination index between the two kinds of professional data.

[0016] Further, the method for obtaining the frequency overlap index comprises:

[0017] In each combination cluster, the average of all frequency values in each frequency clustering cluster is taken as the average characteristic value;

[0018] The absolute value of the difference between the average characteristic values of the two frequency clustering clusters in each combination cluster is negatively correlated and normalized to obtain the first similarity factor between the two frequency clustering clusters in each combination cluster;

[0019] The times of the frequency values in the two frequency clustering clusters in each combination cluster are compared, and the proportion of the same time is taken as the second similarity factor between the two frequency clustering clusters in each combination cluster;

[0020] The normalized value of the product of the first similarity factor and the second similarity factor is taken as the frequency overlap index between two frequency clustering clusters in each combined cluster.

[0021] Further, the method for obtaining the write demand amount comprises:

[0022] In each data clustering cluster, the product of the mean value of all frequency values in each frequency clustering cluster in each professional data and the storage occupation space of each professional data is taken as the write factor of each frequency clustering cluster in each professional data.

[0023] In each data clustering cluster, an optional professional data is taken as the test data, and an optional frequency clustering cluster in the test data is taken as the test clustering cluster.

[0024] In each data clustering cluster, between the test data and the remaining any professional data, the frequency clustering cluster with the maximum frequency overlap index between the test clustering cluster and the professional data is taken as the matching clustering cluster of the test clustering cluster, and the normalized value of the sum of the write factors of the test clustering cluster and the matching clustering cluster and the product of the frequency overlap index between the test clustering cluster and the matching clustering cluster is taken as the write index between the test clustering cluster and the matching clustering cluster.

[0025] Between the test data and the remaining any professional data, the maximum value of the write indices of all frequency clustering clusters in the test data is taken as the simultaneous write demand factor between the test data and the professional data.

[0026] In each data clustering cluster, the mean value of the simultaneous write demand factors between the test data and the remaining all professional data is taken as the write demand index corresponding to the test data.

[0027] In each data clustering cluster, the maximum value of all write demand indices of all professional data is taken as the write demand amount of each data clustering cluster.

[0028] Further, the method for obtaining the frequency clustering cluster comprises:

[0029] In the collection frequency time series data of each professional data of each event data set, DBSCAN clustering algorithm is used to cluster analyze the frequency values at all times to obtain the frequency clustering cluster; wherein the neighborhood radius and the minimum number of points in the neighborhood are both preset values.

[0030] Further, the method for obtaining the data clustering cluster comprises:

[0031] In each event data set, the value of the difference index between any two professional data after negative correlation mapping of the cooperation index between the two professional data is taken as the difference index between the two professional data.

[0032] Cluster all professional data in each sports data set based on the K-means clustering algorithm and the optimal K value to obtain data clustering clusters; wherein the distance measure is the difference index between the professional data.

[0033] Further, based on the write demand amount of the data clustering clusters in all sports data sets of all kinds of sports events, all sports data sets are integrated and stored, including:

[0034] Sort the data clustering clusters in all data sets of all sports events in ascending order according to the write demand amount to obtain a sorting sequence;

[0035] In the sorting sequence, the write demand amount of the data clustering clusters is calculated by first-order difference to obtain a difference sequence, and the latter data clustering cluster of the two data clustering clusters corresponding to the maximum value in the difference sequence in the sorting sequence is taken as a breakpoint;

[0036] In the sorting sequence, the data clustering clusters after the breakpoint are taken as distributed storage data, and the data clustering clusters before the breakpoint are taken as overall storage data;

[0037] Based on the K-means clustering algorithm and the preset K value, the data clustering clusters in the distributed storage data are clustered and analyzed to obtain distributed clustering clusters, wherein the distance measure is the absolute value of the difference in write demand amount between the data clustering clusters, and the preset K value is the number of servers;

[0038] The sum of the write demand amounts of all data clustering clusters in each distributed clustering cluster is taken as the storage demand value of each distributed clustering cluster, all distributed clustering clusters are sorted in descending order according to the storage demand value to obtain a storage demand sequence, all servers are sorted in descending order according to the storage capacity to obtain a storage capacity sequence, and each distributed clustering cluster in the storage demand sequence is stored in the server at the same position in the storage capacity sequence;

[0039] Update the storage capacity of each server, and store the overall storage data in the server with the maximum storage capacity.

[0040] Further, the method for obtaining the optimal K value includes:

[0041] The optimal K value is obtained based on the elbow method.

[0042] A data integration system for multi-classification sports events includes a processor and a memory, the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to realize the steps of a data integration method for multi-classification sports events.

[0043] A computer readable storage medium, having stored thereon a computer program, the computer program being executed by a processor to implement steps of a data integration method for multi-classification sports events.

[0044] The present application has the following advantages:

[0045] In view of the fact that the collection frequency of professional data of different events, or even the professional data of the same event at different progress stages, varies, there will be null value problems when storing the data using a relational database, and the null value problems are mainly caused by the mismatch of data collection frequency. Therefore, first, the event data set of each athlete is obtained under each sports event, including the collection frequency time series data of multiple professional data and the storage space occupied by each professional data. Then, in each event data set, the frequency values in the time series data of each professional data are subjected to cluster analysis to obtain frequency clusters, which can identify the collection frequency characteristics of each professional data. Further, based on the similarity and time characteristics of the frequency clusters, a coordination index is calculated in the frequency clusters of different professional data, and the professional data is subjected to cluster analysis to obtain data clusters, which accurately quantifies the correlation between the professional data in each event data set, reduces the mixed storage of low correlation data in the subsequent process, and optimizes the null value problem in the relational database storage process. Further, in order to avoid the problem of excessive server pressure caused by the simultaneous writing of a large amount of data in a certain stage of a data set with high coordination, the present application combines the matching between the frequency clusters of professional data and the storage space occupied by professional data in the data clusters of the event data set to determine the writing demand of each data cluster, which is used to avoid the problem of local storage overload in the subsequent storage process. Finally, the integration of all event data sets of all kinds of sports events is driven by the writing demand of the data clusters, and all event data sets are stored. In summary, the present application analyzes the coordination of the collection frequency changes of different kinds of professional data during the event process, and clusters the professional data, thereby reducing the null value problem in the relational database storage process, and analyzing the storage space occupied by the professional data, effectively reducing the writing pressure of the server. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor based on these drawings.

[0047] Figure 1A method flow chart of a data integration method for multi-classification sports events provided by an embodiment of the present application;

[0048] Figure 2 A method flow chart of a cooperative index acquisition method provided by an embodiment of the present application;

[0049] Figure 3 A system block diagram of a data integration system for multi-classification sports events provided by an embodiment of the present application;

[0050] Figure 4 A system structure schematic diagram of a data integration system for multi-classification sports events provided by an embodiment of the present application;

[0051] Figure 5 A schematic diagram of a computer readable storage medium provided by an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined purposes, the following describes in detail the specific implementation, structure, features and effects of a data integration method, system and medium for multi-classification sports events according to the present application, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0054] The following specifically describes the specific scheme of a data integration method, system and medium for multi-classification sports events provided by the present application in combination with the accompanying drawings.

[0055] Please refer to Figure 1 which shows a method flow chart of a data integration method for multi-classification sports events provided by an embodiment of the present application, which includes the following steps:

[0056] Step S1: Under each sports event, acquire the event data set of each athlete, and the event data set includes the acquisition frequency time sequence data of multiple professional data and the storage space occupation of each professional data.

[0057] The current sports events cover various sports types, such as bicycle, marathon, triathlon and run-bike, and the characteristics are that different events have significant differences in rules, data dimensions and technical requirements. The multi-classification sports event data includes data of various sports events, and when the data of various sports events is integrated and stored, the data of each sports event needs to be collected. The related data of sports events include project type, event name, holding time, holding place, group, athlete name, number, professional data of the athlete in the competition process, wherein the professional data of the bicycle event includes cadence, speed, power and the like. In the embodiment of the present application, the object of integrated storage is mainly the professional data generated under each sports event.

[0058] When the prior art stores professional data of bicycle, marathon, triathlon and run-bike sports events in a relational database, there will be a null value problem. The null value problem is mainly caused by the mismatch of data collection frequency. For example, the bicycle needs to collect power, cadence and speed at a high frequency, and the collection frequency is multiple times per second, while the marathon only needs to record the pace or heart rate periodically, for example, once per kilometer. If a high-frequency field such as "second-level power value" is preset in the statistics table, the low-frequency project such as marathon will inevitably have a large number of null values due to data vacancy. Conversely, if a sparse field such as "time per kilometer" is reserved for low-frequency projects, the original data of high sampling rate such as bicycle cannot be directly adapted, and needs to be split into multiple rows for storage or forced to reduce the frequency, further increasing data redundancy or information loss. At present, in sports events, in order to reduce invalid monitoring data during the event and at the same time ensure the data integrity of the main monitoring stage of the event, the sensor recording frequency of endurance events has realized "multi-dimensional scene driven dynamic adjustment", that is, the recording frequency of the sensor is automatically adjusted according to the stage scene of the event, for example, in the climbing section of the bicycle, the power meter sampling rate jumps from the regular 1Hz to 10Hz to capture the subtle fluctuations of pedaling force, and in the sprint period of the marathon, the foot pressure sensor switches from 5Hz to 50Hz to monitor the ground contact time change. This real-time change of collection frequency will cause the collection frequency of the same professional data to change significantly at different stages of the event. Therefore, in the embodiment of the present application, in order to reduce the null values in the relational database, it is necessary to judge the synergy between two kinds of professional data (cadence, power, speed, etc.) of the same event, and the synergy mainly refers to the consistency of the collection frequency and the consistency of the frequency change in the event process, so as to perform clustering analysis on the professional data and finally integrate and store the professional data.

[0059] Firstly, under each sports event, the event data set of each athlete is obtained, and the event data set includes multiple professional data acquisition frequency time series data and the storage space of each professional data, for example, in a bicycle race, the types of professional data can include the heart rate, pedal frequency, speed, power and the like of the athlete, so that relevant sensors can be deployed on the athlete and the racing car to collect the heart rate acquisition frequency time series data, the pedal frequency acquisition frequency time series data, the power acquisition frequency time series data, the speed acquisition frequency time series data and the like.

[0060] It should be noted that the data collection and acquisition in the embodiment of the application are authorized by relevant users, the process does not violate relevant laws and regulations, and does not violate public order and good customs.

[0061] Step S2: In each event data set, the frequency values at all time points in each acquisition frequency time series data are subjected to cluster analysis to obtain a frequency cluster; in the frequency clusters of different professional data, the similarity between the frequency clusters and the number of time points in the frequency clusters are comprehensively considered to determine the coordination index between any two professional data in each event data set, which is used for cluster analysis of the professional data to obtain a data cluster.

[0062] In each event data set, the time series frequency values of each professional data can be clustered first to divide different frequency modes (such as high frequency, low frequency or specific fluctuation stage), so as to reveal the state characteristics of each professional data in different time periods. By comparing the frequency clusters of different professional data, the similarity and the number of time points are evaluated, and the coordination between any two professional data in the event data set is quantified, for example, the heart rate and the speed may be synchronized to increase in the sprint stage to form a coordination cluster. Then, the professional data is clustered into a data cluster based on the coordination index, so as to reduce redundant information and provide a structured multi-source data grouping for subsequent analysis.

[0063] Firstly, in each event data set, the frequency values of the acquisition frequency time series data of each professional data are clustered to identify the frequency mode of each professional data to obtain a frequency cluster.

[0064] Preferably, in an embodiment of the application, the acquisition method of the frequency cluster comprises:

[0065] In the acquisition frequency time series data of each professional data of each event data set, the frequency values at all time points are subjected to cluster analysis by using a DBSCAN clustering algorithm to obtain a frequency cluster; wherein the neighborhood radius and the minimum number of points in the neighborhood are both preset values.

[0066] It should be noted that in this embodiment of the present application, the neighborhood radius is set to 0.3, and the minimum number of points in the neighborhood is set to 5, and the specific values can be adjusted according to the implementation scene, and are not limited here; the DBSCAN clustering algorithm is a known technology, and the specific process is not described here.

[0067] After obtaining the frequency clustering cluster corresponding to each professional data, since the state of the athlete will dynamically change (such as warming up, sprinting, and recovering) as the event progresses, different professional data may be difficult to directly align due to differences in collection frequency or timestamp; therefore, based on the frequency pattern, the covariation pattern between different professional data can be analyzed, the implicit correlation between professional data is found, the coordination index is obtained to help judge which professional data has a more consistent frequency distribution pattern, the robustness of state recognition is enhanced, soft alignment can be realized at the pattern level, and the error of hard time matching is avoided.

[0068] Therefore, in the frequency clustering cluster of different professional data, the similarity between the frequency clustering clusters and the number of time points in the frequency clustering cluster are comprehensively considered to determine the coordination index between any two professional data in each event data.

[0069] Preferably, in an embodiment of the present application, the method for obtaining the coordination index comprises:

[0070] Please refer to Figure 2 which shows a method flowchart of the method for obtaining the coordination index in an embodiment of the present application, and the method comprises the following steps:

[0071] Step S201: In each event data set, for any two professional data, the frequency clustering clusters in the two professional data are combined two by two to obtain all non-repeating combination clusters.

[0072] For example, in a certain event set, professional data a has three frequency clustering clusters 1, 2, and 3, and professional data b has two frequency clustering clusters 4 and 5, then the final combination cluster has 6, which are (1, 4), (1, 5), (2, 4), (2, 5), (3, 4), and (3, 5).

[0073] Step S202: In each combination cluster, the similarity of the frequency distribution between the frequency clustering clusters and the similarity of the time points are analyzed to determine the frequency overlap index between the two frequency clustering clusters.

[0074] The similarity of the frequency pattern between professional data can be represented by the similarity between the frequency clustering clusters.

[0075] Firstly, if the distribution of frequency values between two frequency clustering clusters is closer, it can be intuitively reflected that the frequency patterns between the two frequency clustering clusters are more similar, so in each combination cluster, the mean value of all frequency values in each frequency clustering cluster is taken as a mean characteristic value for reflecting the average level of the distribution of frequency values in each frequency clustering cluster, the absolute value of the difference between the mean characteristic values of the two frequency clustering clusters in each combination cluster is calculated, the smaller the absolute value, the more similar the frequency distribution between the two frequency clustering clusters, so the absolute value is negatively correlated and normalized to correct the logical relationship, so as to obtain the first similarity factor between the two frequency clustering clusters in each combination cluster, the larger the first similarity factor, the higher the similarity of the frequency distribution between the two frequency clustering clusters. The negative correlation mapping and normalization processing here can use the formula exp(―x), where exp() represents the exponential function with natural constant e as the base, and x represents the independent variable.

[0076] Then, if the time distribution of frequency values between two frequency clustering clusters is more similar, it means that the two frequency clustering clusters are closer in time distribution, so the time overlap will be higher, so in each combination cluster, the proportion of the number of times that the frequency values in the two frequency clustering clusters are the same is calculated as the second similarity factor between the two frequency clustering clusters in each combination cluster. Specifically, the overlap of the times of the frequency values in the two frequency clustering clusters is compared, the number of times that overlap (the same) is taken as the numerator, the sum of the number of frequency values in the two frequency clustering clusters is taken as the denominator, and the ratio obtained is taken as the second similarity factor. The larger the second similarity factor, the higher the time overlap between the two frequency clustering clusters in the combination cluster.

[0077] Finally, in each combination cluster, the product of the first similarity factor and the second similarity factor between the two frequency clustering clusters is normalized as the frequency overlap index between the two frequency clustering clusters, the larger the frequency overlap index, the more similar the frequency distribution between the two frequency clustering clusters in the combination cluster. The normalization is a technique known to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0078] At this point, in each event data set, for any two professional data, there is a frequency overlap index between any two frequency clustering clusters of the two professional data.

[0079] Step S203: In each event data set, optionally select one professional data as target data, and in the target data and the remaining any one professional data, determine the similarity index between the target data and the professional data based on the frequency overlap index between the frequency clustering clusters in the target data and the data and in combination with the number of times in the frequency clustering clusters in the target data.

[0080] Based on the analysis of the preceding steps, the greater the frequency overlap index between the two cluster clusters, the more similar the frequency changes, the more likely it is that the change in the sensor acquisition requirement caused by the change in the scene of the event stage, and then the sensor acquisition frequency has changed, so the frequency overlap index between the frequency cluster clusters can be further used to determine the similarity index between the professional data, which is used to reflect the degree of collaborative change.

[0081] For ease of explanation and illustration, in each event data set, optionally one professional data is taken as the target data, and in the target data and any other professional data, each frequency cluster in the target data corresponds to a plurality of combination clusters. For example, in step S201, professional data a is the target data, and in the target data a and professional data b, the combination clusters corresponding to the frequency cluster 1 in the target data a are (1, 4) and (1, 5).

[0082] Then in all combination clusters corresponding to each frequency cluster in the target data, the maximum value of the frequency overlap index is taken as the frequency domain overlap degree of each frequency cluster in the target data, which is used to reflect the maximum degree of similar frequency changes. At this time, each frequency cluster in the target cluster corresponds to a frequency domain overlap degree.

[0083] Finally, based on the number of time points in the frequency cluster in the target data, the frequency domain overlap degree corresponding to the frequency cluster is weighted and fused to obtain the similarity index between the target data and the professional data. Specifically, the normalized value of the number of time points in each frequency cluster in the target cluster is taken as the weight factor. The greater the weight factor, the more time points in the cluster, and the reference degree of the analysis and acquisition of the similarity index will be higher based on this, and thus the accuracy will be improved. The product of the weight factor and the frequency domain overlap degree of each frequency cluster is taken as the weighted overlap degree. The greater the weighted overlap degree, the higher the similarity between the target data and the professional data in the frequency cluster. Then, the normalized value of the mean of the weighted overlap degree of all frequency clusters in the target data is taken as the similarity index between the target data and the professional data. It is easy to understand that the greater the similarity index, the greater the similarity between the frequency distribution, change, and other characteristics between the target data and the professional data. The normalization is a well-known technical means for those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, and the specific normalization method is not limited here.

[0084] Step S204: In any two professional data, analyze the similarity index between the two professional data to determine the collaborative index between the two professional data.

[0085] Based on the aforementioned steps, it is known that for any two types of professional data, there will be two similar indicators. Therefore, the mean of the similar indicators between the two types of professional data is normalized and used as the collaborative indicator between them. Normalization is a technique well-known to those skilled in the art, and the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0086] Thus, in each event dataset, a collaborative index between professional data can be obtained based on the frequency variation of professional data. Then, cluster analysis can be performed on the professional data based on this index to obtain data clusters, so that professional data with the same frequency variation characteristics are grouped into one data cluster, which facilitates subsequent integration and storage.

[0087] Preferably, in one embodiment of the present invention, the method for obtaining data clusters includes:

[0088] In each event dataset, the value of the synergy index between any two professional data sets is negatively correlated and mapped, and used as the difference index between the two professional data sets. The negative correlation mapping and normalization can be performed using the formula exp(―x), where exp() represents an exponential function with the natural constant e as the base, and x represents the independent variable.

[0089] Then, based on the K-means clustering algorithm and the optimal K value, all professional data in each event dataset are clustered to obtain data clusters; where the distance metric is the difference index between professional data; the optimal K value can be obtained based on the elbow method.

[0090] It should be noted that the K-means clustering algorithm and the elbow method are both well-known techniques, and the specific process will not be elaborated here.

[0091] Step S3: In each event dataset, based on the matching between the frequency clusters of professional data and the storage space occupied by the professional data, determine the write requirement of each data cluster.

[0092] Based on the aforementioned steps, highly collaborative professional data can be clustered into one category. However, when highly collaborative professional data is stored together, due to changes in the event scenario, the collection frequency of these highly collaborative fields may surge simultaneously, leading to excessively high server bandwidth usage at certain times, or even failure to meet normal write requirements. Therefore, when merging and storing professional data, the write demand of the professional data in the data cluster during the event should be considered, and the professional data in the data cluster should be split, integrated, and stored based on the write demand.

[0093] Therefore, in each data cluster of each event data set, the writing demand of each data cluster is determined based on the matching between the frequency clusters of the professional data and in combination with the storage space of the professional data.

[0094] Preferably, in one embodiment of the present application, the method for obtaining the writing demand comprises:

[0095] In each data cluster, the product of the mean of all frequency values in each frequency cluster of each professional data and the storage space of each professional data is taken as the writing factor of each frequency cluster of each professional data, the mean of the frequency values is used to reflect the data acquisition density, and the storage space is used to reflect the data amount of single writing, and the writing obtained by the combination of the two directly reflects the unit storage pressure of the frequency cluster.

[0096] In each data cluster, an optional professional data is taken as the test data, and an optional frequency cluster in the test data is taken as the test cluster.

[0097] In each data cluster, for an arbitrary frequency cluster of a certain professional data, the more similar frequency clusters in other professional data to the frequency cluster are in the same event stage, the greater the probability of simultaneous writing of the two is, and the writing pressure is also increased.

[0098] Therefore, in each data cluster, in the test data and the remaining arbitrary professional data, the frequency cluster with the maximum frequency overlap index between the test cluster and the professional data is taken as the matching cluster of the test cluster, in the test data and the professional data, the sum of the writing factors of the test cluster and the matching cluster is taken as the writing sum, the greater the writing sum is, the greater the bandwidth of the simultaneous writing of the test cluster and the matching cluster is, and the greater the writing pressure is; the greater the frequency overlap index between the test cluster and the matching cluster is, the more likely the two clusters are in the same event stage, and the more likely the simultaneous writing occurs, and the writing pressure is also increased; therefore, the product of the writing sum and the frequency overlap index between the test cluster and the matching cluster after normalization is taken as the writing index between the test cluster and the matching cluster, the greater the writing index is, the greater the writing demand of the simultaneous writing of the test cluster and the matching cluster is. The normalization is a technical means familiar to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, and the specific normalization method is not limited herein.

[0099] In the whole competition process, the frequency of data collection is determined by the moment when the data writing bandwidth is maximum, if the maximum bandwidth of professional data writing can be met, the normal writing requirement can be met, so the maximum value of the writing index of the frequency clustering cluster of the two kinds of professional data is used to measure the simultaneous writing requirement of the two kinds of professional data. Therefore, under the to-be-tested data and any kind of professional data, the maximum value of the writing index of all frequency clustering clusters in the to-be-tested data is used as the simultaneous writing requirement factor between the to-be-tested data and the professional data.

[0100] At this time, in each data clustering cluster, the to-be-tested data has a simultaneous writing requirement factor with other professional data, and then the average value of the simultaneous writing requirement factors between the to-be-tested data and all kinds of professional data is used as the writing requirement index corresponding to the to-be-tested data. The writing requirement index reflects the collaborative writing requirement between the to-be-tested data and all kinds of professional data with high collaboration, and the greater the value, the greater the simultaneous writing pressure.

[0101] Up to now, in the data clustering cluster, each kind of professional data has a writing requirement index, and based on the aforementioned theory that if the maximum bandwidth of professional data writing can be met, the normal writing requirement can be met, in each data clustering cluster, the maximum writing requirement index of all professional data is used as the writing requirement amount of each data clustering cluster.

[0102] Step S4: In all kinds of sports event data sets, based on the writing requirement amount of the data clustering cluster, all the event data sets are integrated and stored.

[0103] Based on the foregoing steps, the writing requirement amount of each data clustering cluster can be obtained, and in this step, based on the writing requirement amount of the data clustering cluster, all the event data sets of all kinds of sports events can be integrated and stored.

[0104] Preferably, in an embodiment of the present application, the process of integrating and storing all the event data includes:

[0105] Since the overall writing requirement amount of the professional data in part of the data clustering cluster is small, it does not need to be distributed and stored to meet the normal writing requirement, so it is necessary to divide all the data clustering clusters into clustering clusters that need to be distributed and stored and clustering clusters that do not need to be distributed and stored.

[0106] Firstly, data in all data sets of all sports events is clustered into data clusters, and the data clusters are sorted in ascending order according to write demand, to obtain a sorted sequence. In the sorted sequence, the write demand of the data clusters is calculated by first-order difference, to obtain a difference sequence, and the values in the difference sequence represent the difference between the write demands of adjacent two data clusters. The maximum value in the difference sequence is selected, and the position of the maximum value represents the position where the write demand of the data cluster changes most. Therefore, the data cluster after the position corresponding to the maximum value in the sorted sequence is selected as the breakpoint.

[0107] In the sorted sequence, the write demand of the data cluster after the breakpoint is larger, so the data cluster is stored as distributed storage data; the write demand of the data cluster before the breakpoint is smaller, so the data cluster is stored as whole storage data.

[0108] Then, the data clusters in the distributed storage data are analyzed by clustering based on the K-means clustering algorithm and a preset K value, to obtain distributed clusters, wherein the distance measurement is the absolute value of the difference in write demand between the data clusters, and the preset K value is the number of servers.

[0109] Next, the sum of the write demands of all data clusters in each distributed cluster is taken as the storage demand value of each distributed cluster, and all distributed clusters are sorted in descending order according to the storage demand value, to obtain a storage demand sequence. At the same time, all servers are sorted in descending order according to the storage capacity, to obtain a storage capacity sequence. Each distributed cluster in the storage demand sequence is stored in the server at the same position in the storage capacity sequence.

[0110] Finally, the storage capacity of each server is updated, and the whole storage data is stored in the server with the maximum storage capacity.

[0111] In summary, in view of the change of the collection frequency of the professional data of different events, and even the change of the collection frequency of the professional data of the same event at different progress of the event, there is a null value problem when the professional data is stored by using the relational database, and the null value problem is obviously caused by the mismatch of the data collection frequency. Therefore, first, the data set of each athlete in each sports event is obtained, including the time sequence data of the collection frequency of various professional data and the storage space occupied by each professional data. Then, in each event data set, the frequency value in the time sequence data of each professional data is subjected to cluster analysis, and the frequency cluster is obtained, which can identify the collection frequency characteristics of each professional data. Further, in the frequency cluster of different professional data, the coordination index is calculated based on the similarity of the frequency cluster and the time number characteristics, and the professional data is subjected to cluster analysis, and the data cluster is obtained, which accurately quantifies the correlation between the professional data in each event data set, reduces the mixed storage of low correlation data in the subsequent process, and optimizes the null value problem in the storage process of the relational database. Further, in order to avoid the problem that the data set with high coordination is simultaneously written with a large amount of data at a certain stage, causing the problem of excessive server pressure, the matching between the frequency clusters of the professional data in the data cluster of the event data set is combined with the storage space occupied by the professional data, and the writing demand of each data cluster is determined, which is used to avoid the problem of local storage overload in the subsequent storage process. Finally, the writing demand of the data cluster drives the integration of all event data sets of all kinds of sports events, and all event data sets are stored. The embodiment of the application analyzes the coordination of the change of the collection frequency of different kinds of professional data in the process of the event, and clusters the professional data, thereby reducing the null value problem in the storage process of the relational database, and analyzing the storage space occupied by the professional data, effectively reducing the writing pressure of the server.

[0112] The embodiment of the application further provides a data integration system for multi-classification sports events. Figure 3 which shows a system block diagram, including a data acquisition module 301 for realizing the step S1 in the method embodiment; a cluster analysis module 302 for realizing the step S2 in the method embodiment; a writing demand analysis module 303 for realizing the step S3 in the method embodiment; and an integrated storage module 304 for realizing the step S4 in the method embodiment.

[0113] It should be noted that the system provided by the above embodiment is only exemplified by the division of the above functional modules, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the data integration system for multi-classification sports events and the data integration method for multi-classification sports events provided by the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be described here.

[0114] Please refer to Figure 4 , which shows a system structure schematic diagram of a data integration system for multi-classification sports events provided by an embodiment of the application, including a processor 400, a memory 401, a bus 402 and a communication interface 403, the processor 400, the communication interface 403 and the memory 401 are connected through the bus 402; wherein the memory 401 can contain a high-speed random access memory, the bus 402 can be an ISA bus, a PCI bus or an EISA bus, etc., the processor 400 can be an integrated circuit chip with signal processing capability; the memory 401 stores at least one instruction, at least one program, a code set or an instruction set, which is loaded and executed by the processor to realize the steps in the data integration method for multi-classification sports events.

[0115] The embodiment of the application also provides a computer readable storage medium corresponding to the method provided by the foregoing embodiments, please refer to Figure 5 , which shows that the storage medium is an optical disc, and a computer program (i.e. program product) is stored on the optical disc, which will execute the method provided by any of the foregoing embodiments when the processor runs.

[0116] It should be noted that examples of the computer readable storage medium can also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), read-only memory (ROM), other optical or magnetic storage media, which will not be described one by one here.

[0117] It should be noted that the above-mentioned sequence of the embodiments is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0118] The various embodiments described in this specification are presented by way of example, and each embodiment is not inherently more important than any other embodiment.

Claims

1. A data integration method for multi-classification sports events, characterized in that, The method comprises: In each sports event, obtain the event data set of each athlete, which comprises the acquisition frequency time series data of multiple professional data and the storage space occupation of each professional data; In each event data set, perform cluster analysis on the frequency values of all time points in each acquisition frequency time series data to obtain frequency clusters; in the frequency clusters of different professional data, determine the coordination index between any two professional data in each event data set based on the similarity between the frequency clusters and the number characteristics of the time points in the frequency clusters, so as to perform cluster analysis on the professional data to obtain data clusters; In each data cluster of each event data set, determine the write demand of each data cluster based on the matching between the frequency clusters of the professional data and the storage space occupation of the professional data; In all event data sets of all kinds of sports events, integrate and store all event data sets based on the write demand of the data cluster; The method for obtaining the coordination index comprises: In each event data set, for any two professional data, combine the frequency clusters in the two professional data in pairs to obtain all non-repeated combination clusters; In each combination cluster, analyze the similarity of the frequency distribution and the similarity of the time points between the frequency clusters to determine the frequency overlap index between the two frequency clusters; In each event data set, select one professional data as target data, and in the target data and any other professional data, in all combination clusters corresponding to each frequency cluster in the target data, take the maximum value of the frequency overlap index as the frequency domain overlap degree of each frequency cluster in the target data; Based on the number of time points in the frequency cluster in the target data, weight and fuse the frequency domain overlap degree corresponding to the frequency cluster to obtain the similarity index between the target data and the professional data; In any two professional data, take the normalized value of the average of the similarity index between the two professional data as the coordination index between the two professional data.

2. The data integration method for multi-classification sports events according to claim 1, wherein, The method for obtaining the frequency overlap index comprises: In each combination cluster, take the average of all frequency values in each frequency cluster as the average characteristic value; Take the normalized value of the absolute value of the difference between the average characteristic values of the two frequency clusters in each combination cluster after negative correlation mapping as the first similarity factor between the two frequency clusters in each combination cluster; Compare the time points of the frequency values in the two frequency clusters in each combination cluster, and take the proportion of the same time points as the second similarity factor between the two frequency clusters in each combination cluster; Take the normalized value of the product of the first similarity factor and the second similarity factor as the frequency overlap index between the two frequency clusters in each combination cluster.

3. The data integration method for multi-classification sports events according to claim 1, wherein, The method for obtaining the write demand comprises: In each data cluster, take the product of the average of all frequency values in each frequency cluster in each professional data and the storage space occupation of each professional data as the write factor of each frequency cluster in each professional data; Within each data cluster, select one type of professional data as the test data, and within the test data, select one frequency cluster as the test cluster; In each data cluster, between the data to be tested and any other type of professional data, the frequency cluster with the largest frequency overlap index between the professional data and the cluster to be tested is taken as the matching cluster of the cluster to be tested. In the data to be tested and that type of professional data, the sum of the writing factors of the cluster to be tested and the matching cluster, and the product of the frequency overlap index between the cluster to be tested and the matching cluster, are normalized and taken as the writing index between the cluster to be tested and the matching cluster. Under the test data and any other professional data, the maximum value of the writing index of all frequency clusters in the test data is taken as the simultaneous writing demand factor between the test data and the professional data. Within each data cluster, the mean of the simultaneous writing demand factors between the data to be tested and all other professional data is used as the writing demand index corresponding to the data to be tested. Within each data cluster, the maximum value among all write demand metrics for all professional data is taken as the write demand for each data cluster.

4. The data integration method for multi-classification sports events according to claim 1, wherein, The method for obtaining the frequency clusters includes: In the time series data of each professional data collection frequency in each event dataset, the DBSCAN clustering algorithm is used to perform cluster analysis on the frequency values ​​at all times to obtain frequency clusters; where the neighborhood radius and the minimum number of points in the neighborhood are preset values.

5. The data integration method for multi-classification sports events according to claim 1, wherein, The method for obtaining the data clusters includes: In each event dataset, the value of the synergy index between any two types of professional data is negatively correlated and used as the difference index between the two types of professional data. Based on the K-means clustering algorithm and the optimal K value, all professional data in each event dataset are clustered to obtain data clusters; where the distance metric is the difference index between professional data.

6. The data integration method for multi-classification sports events according to claim 1, wherein, The process involves integrating and storing all event datasets across all types of sports events, based on the write requirements of data clusters, including: The data clusters from all datasets of all sports events are sorted in ascending order according to the write demand, resulting in a sorted sequence. In the sorting sequence, the write demand of the data clusters is calculated by first-order difference to obtain a difference sequence. The data cluster with the maximum value in the difference sequence corresponding to the next data cluster in the sorting sequence is taken as the breakpoint. In the sorting sequence, the data clusters after the breakpoint are stored as distributed data, and the data clusters before the breakpoint are stored as the whole data. Cluster analysis is performed on the data clusters in the distributed storage data based on the K-means clustering algorithm and the preset K value to obtain distributed clusters. The distance metric is the absolute value of the difference in write demand between data clusters, and the preset K value is the number of servers. Summing up the write demand values of all data cluster in each distribution cluster as the storage demand value of each distribution cluster, sorting all distribution clusters in descending order according to the storage demand value to obtain a storage demand sequence, sorting all servers in descending order according to the storage capacity to obtain a storage capacity sequence, and storing each distribution cluster in the storage demand sequence into the server at the same position in the storage capacity sequence; Updating the storage capacity of each server and storing the overall storage data into the server with the maximum storage capacity.

7. The data integration method for multi-classification sports events according to claim 5, wherein, The method for obtaining the optimal K value comprises: The optimal K value is obtained based on an elbow method.

8. A data integration system for multi-classification sports events, characterized by, The computer program is executed by the processor to implement the steps of the data integration method for the multi-classification sports event according to any one of claims 1-7.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the data integration method for the multi-classification sports event according to any one of claims 1-7.

Images