Data clustering method, device and computer-readable storage medium

Abstract

The embodiment of the present application discloses a data clustering method, which includes: obtaining indicator data of an advertisement to be analyzed; determining the number of categories and the minimum number of indicator data included in each category; wherein the number of categories is the number of categories into which the indicator data is to be divided; screening the indicator data of the advertisement to be analyzed based on the number of categories to determine multiple first cluster centers; calculating the distance between each indicator data of the advertisement to be analyzed and each of the first cluster centers; based on the distance, the number of data, and the multiple first cluster centers, clustering the indicator data of the advertisement to be analyzed to obtain a target clustering result. The embodiment of the present application also discloses a data clustering device and a computer-readable storage medium.

Description

Technical Field

[0001] The present application relates to data processing technology in the computer field, and in particular to a data clustering method, device, and computer-readable storage medium. Background Art

[0002] Currently, with the rapid development of information flow advertising, various network products have emerged. In particular, online advertising has become a major form of advertising. Analyzing and evaluating the effectiveness of advertising can not only save costs but also bring more revenue to the company, thereby improving the company's return on investment. Online advertising is characterized by its complex and diverse forms, and the advertising itself is highly business-oriented. Its analysis requires strong business knowledge. Therefore, related technologies have proposed using the k-means clustering algorithm to cluster advertising effects, and then analyze the advertising effects based on the clustering results. However, the advertising effect clustering method provided by this technology can only use the algorithm's own computational logic to determine the sample type and convergence conditions. If the overall distribution of the classification results is not very uniform or there are abnormal advertising effects, the clustering results will be inaccurate, which will affect the analysis of advertising effects. Summary of the Invention

[0003] In order to solve the above technical problems, the embodiments of the present application hope to provide a data clustering method, device and computer-readable storage medium, which solves the problem of inaccurate clustering results when clustering advertising effects in related technologies, improves the accuracy of clustering results, and avoids affecting the analysis of advertising delivery effects.

[0004] The technical solution of this application is achieved as follows:

[0005] A data clustering method, comprising:

[0006] Obtain the indicator data of the advertisement to be analyzed;

[0007] Determine the number of categories and the minimum number of indicator data included in each category; wherein the number of categories is the number of categories into which the indicator data is to be divided;

[0008] screening the indicator data of the advertisement to be analyzed based on the number of categories to determine a plurality of first cluster centers;

[0009] Calculating the distance between each indicator data of the advertisement to be analyzed and each first cluster center;

[0010] Based on the distance, the data quantity and the multiple first cluster centers, the indicator data of the advertisement to be analyzed is clustered to obtain a target clustering result; wherein the quantity of indicator data included in each type of clustered indicator data meets a preset quantity.

[0011] In the above solution, obtaining the indicator data of the advertisement to be analyzed includes:

[0012] Obtaining basic indicator data of the advertisement to be analyzed;

[0013] The basic indicator data is subjected to data standardization processing to obtain the indicator data of the advertisement to be analyzed.

[0014] In the above solution, the step of performing data standardization on the basic indicator data to obtain the indicator data of the advertisement to be analyzed includes:

[0015] The basic indicator data is subjected to data standardization processing using a deviation standardization algorithm to obtain the indicator data of the advertisement to be analyzed.

[0016] In the above solution, the index data of the advertisement to be analyzed is screened based on the number of categories to determine a plurality of first cluster centers, including:

[0017] determining a target quantile value based on the number of categories;

[0018] The indicator data corresponding to the position of the target percentile value is determined from the indicator data of the advertisement to be analyzed to obtain the multiple first cluster centers.

[0019] In the above solution, clustering the indicator data of the advertisement to be analyzed based on the distance, the data quantity, and the multiple first cluster centers to obtain a target clustering result includes:

[0020] For each indicator data, comparing the distance between the indicator data and each first cluster center;

[0021] Based on the size relationship and the data quantity, the indicator data of the advertisement to be analyzed is clustered to obtain the target clustering result.

[0022] In the above solution, clustering the indicator data of the advertisement to be analyzed based on the size relationship and the data quantity to obtain the target clustering result includes:

[0023] For each indicator data, determining a first cluster center with the smallest distance from the multiple first cluster centers;

[0024] For each indicator data, classify the indicator data and the indicator data corresponding to the first cluster center with the smallest distance into one category to obtain classified first indicator data;

[0025] When it is determined that the difference in the number of indicator data included in each two categories of the first indicator data meets the preset number threshold, and the number of indicator data included in each category of the first indicator data is greater than the data number, the classified first indicator data is determined to be the target clustering result.

[0026] In the above solution, the method further includes:

[0027] When it is determined that the difference between the number of indicator data included in each two categories of the first indicator data does not meet a preset number threshold, or the number of indicator data included in any category of the first indicator data is less than or equal to the data number, the indicator data corresponding to the position of the preset quantile value is determined from each category of the first indicator data to obtain a second cluster center;

[0028] Calculate the distance between each first indicator data and each second cluster center, and for each first indicator data, classify the indicator data corresponding to the first indicator data and the second cluster center with the smallest distance into one category to obtain the classified Nth indicator data, until the difference in the number of indicator data included in each two categories of the Nth indicator data meets the preset quantity threshold, and the number of indicator data included in each category of the Nth indicator data is greater than the data quantity, determine that the classified Nth indicator data is the target clustering result.

[0029] In the above solution, the method further includes:

[0030] When it is determined that the difference between the number of indicator data included in each two categories of the first indicator data does not meet a preset number threshold, or the number of indicator data included in any category of the first indicator data is less than or equal to the data number, the indicator data corresponding to the position of the preset quantile value is determined from each category of the first indicator data to obtain a second cluster center;

[0031] Calculate the distance between each first indicator data and each second cluster center, and for each first indicator data, classify the indicator data corresponding to the first indicator data and the second cluster center with the smallest distance into one category to obtain the classified Nth indicator data, until the number of indicator data included in any category of the Nth indicator data is less than the number of data, and determine the classified N-1th indicator data as the target clustering result.

[0032] In the above solution, the method further includes:

[0033] The target clustering results are displayed in a graphical manner.

[0034] A data clustering method and device, the device comprising: a processor, a memory, and a communication bus;

[0035] The communication bus is used to realize the communication connection between the processor and the memory;

[0036] The processor is configured to execute the data clustering program stored in the memory to implement the following steps:

[0037] Obtain the indicator data of the advertisement to be analyzed;

[0038] Determine the number of categories and the minimum number of indicator data included in each category; wherein the number of categories is the number of categories into which the indicator data is to be divided;

[0039] screening the indicator data of the advertisement to be analyzed based on the number of categories to determine a plurality of first cluster centers;

[0040] Calculating the distance between each indicator data of the advertisement to be analyzed and each first cluster center;

[0041] Based on the distance, the data quantity and the multiple first cluster centers, the indicator data of the advertisement to be analyzed is clustered to obtain a target clustering result.

[0042] A computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the data clustering method described above.

[0043] Because the indicator data of the advertisement to be analyzed can be obtained, the number of categories and the minimum number of indicator data included in each category can be determined, the indicator data of the advertisement to be analyzed is screened based on the number of categories to determine multiple first cluster centers, and the distance between each indicator data of the advertisement to be analyzed and each first cluster center is calculated. Then, based on the distance, the number of data and the multiple first cluster centers, the indicator data of the advertisement to be analyzed is clustered to obtain the target clustering result. In this way, when clustering the indicator data of the advertisement (advertising effect), the number of categories to be divided and the minimum number of indicator data included in each category are taken into consideration, instead of simply using the calculation logic of the algorithm itself as in the relative technology. This solves the problem of inaccurate clustering results when clustering advertising effects in the related technology, improves the accuracy of the clustering results, and avoids affecting the analysis of the advertising effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of a data clustering method provided in an embodiment of the present application;

[0045] Figure 2A schematic diagram of a flow chart of another data clustering method provided in an embodiment of the present application;

[0046] Figure 3 A flowchart of another data clustering method provided in an embodiment of the present application;

[0047] Figure 4 A flowchart of a data clustering method provided in another embodiment of the present application;

[0048] Figure 5 A schematic diagram of the structure of a data clustering device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0050] The embodiment of the present application provides a data clustering method, which can be applied to a data clustering device. Figure 1 As shown, the method includes the following steps:

[0051] Step 101: Obtain indicator data of the advertisement to be analyzed.

[0052] In an embodiment of the present application, the indicator data of the advertisement to be analyzed may be indicator data that can characterize the effectiveness of the advertisement to be analyzed; wherein the indicator data may be indicator data corresponding to the effectiveness of the advertisement to be analyzed after it has been actually delivered. In a feasible implementation, the indicator data may include one or more of the following: cash consumption, click-through rate (CTR), cost per click (CPC), effective cost per mile (ECPM), conversion rate (CVR), return on investment (ROI), etc.

[0053] It should be noted that the indicator data may be the indicator data corresponding to the effect of the advertisement to be analyzed after being delivered in different channels; wherein, the indicator data may have a unified data standard.

[0054] Step 102: Determine the number of categories and the minimum number of indicator data included in each category.

[0055] The number of categories is the number of categories into which the indicator data is to be divided.

[0056] In the embodiment of the present application, the number of categories may be the number of categories into which the indicator data of the advertisement to be analyzed needs to be divided; it should be noted that the number of categories and the amount of data may be determined based on the advertisement to be analyzed. Of course, the number of categories and the amount of data may also be determined based on the advertisement to be analyzed and actual needs.

[0057] Step 103: Filter the indicator data of the advertisements to be analyzed based on the number of categories to determine a plurality of first cluster centers.

[0058] The first cluster center may be a plurality of indicator data selected from all indicator data, that is, the first cluster center refers to the indicator data itself. In an embodiment of the present application, the first cluster center may be determined by equally dividing all indicator data according to the number of categories; in one feasible implementation, the first cluster center is the indicator data at the equal division point of the equal division.

[0059] Step 104: Calculate the distance between each indicator data of the advertisement to be analyzed and each first cluster center.

[0060] In the embodiment of the present application, the distance between each first cluster center of the advertising slogan to be analyzed may refer to the distance between the indicator data corresponding to the first cluster center of the advertising slogan to be analyzed. The distance between each indicator data and each first cluster center may be calculated using the Euclidean distance formula.

[0061] Step 105: Based on the distance between each indicator data and each first cluster center, the amount of data, and multiple first cluster centers, cluster the indicator data of the advertisement to be analyzed to obtain a target clustering result.

[0062] Among them, for each indicator data, the distance between the indicator data and each first cluster center can be compared and analyzed, and the indicator data and the indicator data of the corresponding first cluster center can be classified into one category according to the result of the comparison and analysis, so as to finally obtain the target clustering result.

[0063] The data clustering method provided in the embodiments of the present application can obtain the indicator data of the advertisement to be analyzed, determine the number of categories of the advertisement to be analyzed and the minimum number of indicator data included in each category, screen the indicator data of the advertisement to be analyzed based on the number of categories to determine multiple first cluster centers, calculate the distance between each indicator data of the advertisement to be analyzed and each first cluster center, and then cluster the indicator data of the advertisement to be analyzed based on the distance, the number of data and the multiple first cluster centers to obtain the target clustering result. In this way, when clustering the indicator data of the advertisement (advertising effect), the number of categories to be divided and the minimum number of indicator data included in each category are taken into account, instead of simply using the calculation logic of the algorithm itself as in the relative technology. This solves the problem of inaccurate clustering results when clustering advertising effects in the related technology, improves the accuracy of the clustering results, and avoids affecting the analysis of the advertising delivery effect.

[0064] Based on the above embodiments, the embodiments of the present application provide a data clustering method, referring to Figure 2 As shown, the method includes the following steps:

[0065] Step 201: The data clustering device obtains basic indicator data of the advertisement to be analyzed.

[0066] The basic indicator data may be indicator data that does not have a unified data standard, that is, the numerical values ​​of the indicator data have different dimensions.

[0067] Step 202: The data clustering device performs data standardization processing on the basic indicator data to obtain indicator data of the advertisement to be analyzed.

[0068] It should be noted that performing data standardization on basic indicator data may refer to performing standardization processing on the dimensions of the basic indicator data, thereby ultimately unifying the dimensions of the indicator data of the advertisement to be analyzed.

[0069] Step 202 can be implemented in the following ways:

[0070] Step 202a: The data clustering device uses a deviation normalization algorithm to perform data normalization processing on the basic indicator data to obtain indicator data of the advertisement to be analyzed.

[0071] In the embodiment of the present application, the dimension of the basic indicator data can be standardized by using a deviation standardization algorithm; in a feasible implementation method, a conversion function can be used. Each basic indicator data is converted so that the obtained indicator data value range is between [0, 1]. Assuming that the basic indicator data includes x1, x2...x n ; where x i It is the basic indicator data to be processed. It is the basic indicator data with the smallest value among all basic indicator data. It is the basic indicator data with the largest value among all basic indicator data; i is x i After processing, we get the indicator data; that is, for x1, x2...x n After processing with deviation standardization algorithm, we can get y1, y2...y n , and y1, y2......y n The value range is between [0, 1].

[0072] It should be noted that the numerical range of the processed indicator data is in a unified dimension, which is more conducive to visual analysis based on the indicator data.

[0073] Step 203: The data clustering device determines the number of categories and the minimum number of indicator data included in each category.

[0074] The number of categories is the number of categories into which the indicator data is to be divided.

[0075] In the embodiment of the present application, the number of categories can be represented by k, and the number of minimum indicator data included in each category can be represented by m; wherein m and k are both positive integers.

[0076] Step 204: The data clustering device determines a target quantile value based on the number of categories.

[0077] In an embodiment of the present application, all indicator data can be divided equally according to the number of categories, and the value corresponding to each equal part is the target percentile value; wherein, the target percentile value may include multiple percentile values; in a feasible implementation method, the formula (1 / (k+1))*n can be used to determine the target percentile value; wherein, n∈(1, k), and n is a positive integer.

[0078] Step 205: The data clustering device determines the indicator data corresponding to the position of the target percentile value from the indicator data of the advertisement to be analyzed, and obtains a plurality of first cluster centers.

[0079] Among them, the indicator data corresponding to each percentile value included in the target percentile value can be used as the first cluster center; in a feasible implementation method, if k = 3, we need to calculate the 0.25, 0.5, and 0.75 percentile values ​​of each dimension. After calculating these values, the indicator data at the corresponding positions of these values ​​can be used as the first cluster center.

[0080] Step 206: The data clustering device calculates the distance between each indicator data of the advertisement to be analyzed and each first cluster center.

[0081] In an embodiment of the present application, the indicator data can be regarded as a point, that is, the position coordinates of each point in the multidimensional space are determined, and then the position coordinates of each point in the multidimensional space are substituted into the Euclidean distance formula to calculate the distance between each indicator data and each first cluster center.

[0082] Step 207: The data clustering device compares the distance between each indicator data and each first cluster center.

[0083] Step 208: The data clustering device clusters the indicator data of the advertisement to be analyzed based on the size relationship and the amount of data to obtain a target clustering result.

[0084] In an embodiment of the present application, for each indicator data, the first cluster center with the smallest distance can be determined from all the first cluster centers based on the size relationship between the indicator data and each first cluster center, and the indicator data can be classified based on the first cluster center with the smallest distance and the amount of data, thereby achieving the target clustering result by classifying all the indicator data.

[0085] It should be noted that, for the description of the same steps and contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0086] The data clustering method provided in the embodiments of the present application takes into account the number of categories to be divided and the amount of data including at least indicator data in each category when clustering the indicator data of the advertisement (advertising effect), rather than simply using the calculation logic of the algorithm itself as in the relative technology. This solves the problem of inaccurate clustering results when clustering advertising effects in the related technology, improves the accuracy of the clustering results, and avoids affecting the analysis of the advertising effect.

[0087] Based on the above embodiments, the embodiments of the present application provide a data clustering method, referring to Figure 4 As shown, the method includes the following steps:

[0088] Step 301: The data clustering device obtains basic indicator data of the advertisement to be analyzed.

[0089] Step 302: The data clustering device uses a deviation normalization algorithm to perform data normalization processing on the basic indicator data to obtain indicator data of the advertisement to be analyzed.

[0090] Step 303: The data clustering device determines the number of categories and the minimum number of indicator data included in each category.

[0091] The number of categories is the number of categories into which the indicator data is to be divided.

[0092] Step 304: The data clustering device determines a target quantile value based on the number of categories.

[0093] Step 305: The data clustering device determines the indicator data corresponding to the position of the target percentile value from the indicator data of the advertisement to be analyzed, and obtains a plurality of first cluster centers.

[0094] Step 306: The data clustering device calculates the distance between each indicator data of the advertisement to be analyzed and each first cluster center.

[0095] Step 307: The data clustering device compares the distance between each indicator data and each first cluster center.

[0096] Step 308: The data clustering device determines, for each indicator data, a first cluster center with the smallest distance from a plurality of first cluster centers.

[0097] Step 309: The data clustering device classifies each indicator data and the indicator data corresponding to the first cluster center with the smallest distance into one category to obtain classified first indicator data.

[0098] In an embodiment of the present application, each indicator data and the indicator data corresponding to the first cluster center closest to the indicator data are classified into one category. After completing the first classification of all indicator data using this classification method, the classified first indicator data can be obtained.

[0099] Step 310: When the data clustering device determines that the difference in the number of indicator data included in each two categories of first indicator data meets a preset number threshold, and the number of indicator data included in each category of first indicator data is greater than the number of data, the classified first indicator data is determined to be the target clustering result.

[0100] In an embodiment of the present application, after the first classification, if the number of indicator data included in each category of first indicator data obtained after classification is basically the same (that is, the indicator data is evenly distributed), and the number of indicator data included in each category of first indicator data is greater than the number of data, it is determined that the result presented by the first indicator data after classification is the target clustering result.

[0101] In other embodiments of this application, refer to Figure 4 As shown, the method further includes the following steps:

[0102] Step 311: When the data clustering device determines that the difference in the number of indicator data included in each two categories of first indicator data does not meet the preset number threshold, or the number of indicator data included in any category of first indicator data is less than or equal to the number of data, the indicator data corresponding to the position of the preset percentile value is determined from each category of first indicator data to obtain the second cluster center.

[0103] In an embodiment of the present application, if the number of indicator data included in each category of first indicator data obtained after classification is significantly different (i.e., the indicator data is unevenly distributed), or if the number of indicator data included in a certain category of first indicator data is less than or equal to the number of data, then it is determined that the classified first indicator data does not meet the requirements, and it is necessary to continue to determine the indicator data at the position of the preset quantile value from the included first indicator data for each category of first indicator data as the second cluster center; in this way, the cluster centers are re-determined for all categories of first indicator data to obtain multiple second cluster centers of the same number as the first cluster centers. The preset quantile value can be 0.5.

[0104] It should be noted that step 312 or step 313 may be executed after step 311;

[0105] Step 312: The data clustering device calculates the distance between each first indicator data and each second cluster center, and for each first indicator data, classifies the first indicator data and the indicator data corresponding to the second cluster center with the smallest distance into one category to obtain the classified Nth indicator data, until the difference in the number of indicator data included in each category of the Nth indicator data meets the preset quantity threshold, and the number of indicator data included in each category of the Nth indicator data is greater than the number of data, the classified Nth indicator data is determined to be the target clustering result.

[0106] Step 313: The data clustering device calculates the distance between each first indicator data and each second cluster center, and for each first indicator data, classifies the indicator data corresponding to the first indicator data and the second cluster center with the smallest distance into one category to obtain the classified Nth indicator data, until the number of indicator data included in any category of the Nth indicator data is less than the number of data, and determines that the classified N-1th indicator data is the target clustering result.

[0107] In an embodiment of the present application, for each first indicator data, the first indicator data and the indicator data corresponding to the second cluster center with the smallest distance can be classified into one category to obtain classified second indicator data. If the number of indicator data included in each category of the second indicator data obtained after classification is quite different (that is, the indicator data are unevenly distributed), or the number of indicator data included in a certain category of the second indicator data is less than or equal to the number of data, then it is determined that the classified second indicator data does not meet the requirements, and it is necessary to continue to determine, for each category of the second indicator data, the indicator data at the position of the preset quantile value from the included first indicator data as the third cluster center, calculate the distance between each second indicator data and each third cluster center, and for each second indicator data, classify the second indicator data and the indicator data corresponding to the third cluster center with the smallest distance into one category. Obtain the third indicator data after classification. If the third indicator data after classification still does not meet the requirements, continue to reclassify until the difference in the number of indicator data included in each category of N-th indicator data meets the preset quantity threshold, and the number of indicator data included in each category of N-th indicator data is greater than the number of data. At this time, the N-th indicator data after classification can be determined to be the target clustering result; of course, in actual applications, there will be uneven distribution of indicator data, and it is inevitable that the number of indicator data in a certain category is very small or the number of indicator data fails to meet actual business needs. Alternatively, if the third indicator data after classification still does not meet the requirements, continue to reclassify until the number of indicator data included in any category of N-th indicator data is less than the number of data. At this time, the N-1-th indicator data obtained last time can be determined to be the target clustering result.

[0108] It should be noted that the conditions for stopping clustering in the embodiment of the present application can effectively filter out abnormal indicator data in the indicator data, thereby ensuring the accuracy of the target clustering results.

[0109] In other embodiments of the present application, the method further includes the following steps:

[0110] Step 314: Display the target clustering results in a graphical display manner.

[0111] In an embodiment of the present application, the target clustering results can be output in a graphical form through graphical tools such as Python and Excel to obtain highly readable graphics, which facilitates analysis of the effects of the advertisements to be analyzed, thereby achieving targeted optimization of the advertisements to be analyzed.

[0112] It should be noted that, for the description of the same steps and contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0113] The data clustering method provided in the embodiments of the present application takes into account the number of categories to be divided and the amount of data including at least indicator data in each category when clustering the indicator data of the advertisement (advertising effect), rather than simply using the calculation logic of the algorithm itself as in the relative technology. This solves the problem of inaccurate clustering results when clustering advertising effects in the related technology, improves the accuracy of the clustering results, and avoids affecting the analysis of the advertising effect.

[0114] Based on the above embodiments, the embodiments of the present application provide a data clustering device, which can be applied to Figures 1 to 4 In the data clustering method provided in the corresponding embodiment, refer to Figure 5 As shown, the device may include: a processor 41, a memory 42 and a communication bus 43, wherein:

[0115] The communication bus 43 is used to realize the communication connection between the processor 41 and the memory 42;

[0116] The processor 41 is configured to execute the data clustering program stored in the memory 42 to implement the following steps:

[0117] Obtain the indicator data of the advertisement to be analyzed;

[0118] Determine the number of categories and the minimum amount of indicator data to be included in each category;

[0119] Among them, the number of categories is the number of categories into which the indicator data is to be divided;

[0120] Filtering the indicator data of the advertisements to be analyzed based on the number of categories to determine a plurality of first cluster centers;

[0121] Calculating the distance between each indicator data of the advertisement to be analyzed and each first cluster center;

[0122] Based on the distance, the number of data and the multiple first cluster centers, the indicator data of the advertisement to be analyzed are clustered to obtain a target clustering result.

[0123] In other embodiments of the present application, the processor 41 is configured to execute the steps stored in the memory 42 to obtain the indicator data of the advertisement to be analyzed, so as to implement the following steps:

[0124] Obtain basic indicator data of the advertisement to be analyzed;

[0125] The basic indicator data is standardized to obtain the indicator data of the advertisement to be analyzed.

[0126] In other embodiments of the present application, the processor 41 is configured to execute data normalization processing on the basic indicator data stored in the memory 42 to obtain indicator data of the advertisement to be analyzed, so as to implement the following steps:

[0127] The deviation standardization algorithm is used to standardize the basic indicator data to obtain the indicator data of the advertisement to be analyzed.

[0128] In other embodiments of the present application, the processor 41 is configured to execute the method stored in the memory 42 to filter the indicator data of the advertisements to be analyzed based on the number of categories, and determine a plurality of first cluster centers to implement the following steps:

[0129] Determine target quantile values ​​based on the number of categories;

[0130] The indicator data corresponding to the position of the target quantile value is determined from the indicator data of the advertisement to be analyzed, and a plurality of first cluster centers are obtained.

[0131] In other embodiments of the present application, the processor 41 is configured to execute the process stored in the memory 42 to cluster the indicator data of the advertisement to be analyzed based on the distance, the number of data, and the plurality of first cluster centers to obtain a target clustering result, so as to implement the following steps:

[0132] For each indicator data, compare the distance between the indicator data and each first cluster center;

[0133] Based on the size relationship and data quantity, the indicator data of the advertisement to be analyzed is clustered to obtain the target clustering result.

[0134] In other embodiments of the present application, the processor 41 is configured to execute the clustering of the indicator data of the advertisement to be analyzed based on the size relationship and the amount of data stored in the memory 42 to obtain a target clustering result, so as to implement the following steps:

[0135] For each indicator data, determining the first cluster center with the smallest distance from multiple first cluster centers;

[0136] For each indicator data, classify the indicator data and the indicator data corresponding to the first cluster center with the smallest distance into one category to obtain the classified first indicator data;

[0137] Determine the indicator data corresponding to the position of the preset quantile value from each type of the first indicator data to obtain the second cluster center;

[0138] Calculating the distance between each first indicator data and each second cluster center, and for each first indicator data, classifying the first indicator data and the indicator data corresponding to the second cluster center with the smallest distance into one category to obtain classified second indicator data;

[0139] When it is determined that the difference in the number of indicator data included in each category of second indicator data meets the preset number threshold, and the number of indicator data included in each category of second indicator data is greater than the data number, the classified second indicator data is determined to be the target clustering result.

[0140] In other embodiments of the present application, the processor 41 is configured to execute the data clustering program stored in the memory 42, and may further implement the following steps:

[0141] When it is determined that the difference in the number of indicator data included in each category of the third indicator data does not meet the preset number threshold, or the number of indicator data included in any category of the first indicator data is less than or equal to the number of data, the indicator data corresponding to the position of the preset quantile value is determined from each category of the second indicator data to obtain the third cluster center;

[0142] Calculate the distance between each second indicator data and each third cluster center, and for each second indicator data, classify the second indicator data and the indicator data corresponding to the third cluster center with the smallest distance into one category to obtain the classified Nth indicator data, until the difference in the number of indicator data included in each category of the Nth indicator data meets the preset quantity threshold, and the number of indicator data included in each category of the Nth indicator data is greater than the number of data, determine that the classified Nth indicator data is the target clustering result.

[0143] In other embodiments of the present application, the processor 41 is configured to execute the data clustering program stored in the memory 42, and may further implement the following steps:

[0144] When it is determined that the difference in the number of indicator data included in each category of the third indicator data does not meet the preset number threshold, or the number of indicator data included in any category of the first indicator data is less than or equal to the number of data, the indicator data corresponding to the position of the preset quantile value is determined from each category of the second indicator data to obtain the third cluster center;

[0145] Calculate the distance between each second indicator data and each third cluster center, and for each second indicator data, classify the indicator data corresponding to the second indicator data and the third cluster center with the smallest distance into one category to obtain the classified Nth indicator data, until the number of indicator data included in any category of the Nth indicator data is less than the number of data, and determine the classified N-1th indicator data as the target clustering result.

[0146] In other embodiments of the present application, the processor 41 is configured to execute the data clustering program stored in the memory 42, and may further implement the following steps:

[0147] The target clustering results are displayed in a graphical manner.

[0148] It should be noted that the specific implementation process of the steps executed by the processor in this embodiment can be referred to Figures 1 to 4 The implementation process of the data clustering method provided in the corresponding embodiment will not be repeated here.

[0149] The data clustering device provided in the embodiments of the present application takes into account the number of categories to be divided and the amount of data including at least indicator data in each category when clustering the indicator data of the advertisement (advertising effect), rather than simply using the calculation logic of the algorithm itself as in the relative technology. This solves the problem of inaccurate clustering results when clustering advertising effects in the related technology, improves the accuracy of the clustering results, and avoids affecting the analysis of the advertising delivery effect.

[0150] Based on the above embodiments, the embodiments of the present application provide a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the following Figures 1 to 4 The corresponding embodiment provides steps of the data clustering method.

[0151] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not 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 embodiments of the present application. The above-mentioned serial numbers of the embodiments of the present application are for description only and do not represent the advantages and disadvantages of the embodiments.

[0152] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0154] In addition, all functional units in the embodiments of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the above-mentioned integrated units can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0155] Those skilled in the art will understand that all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM), magnetic disks or optical disks, and other media that can store program codes.

[0156] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods of each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0157] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A data clustering method, characterized in that: The method comprises: Obtain the indicator data of the advertisement to be analyzed; Determine the number of categories and the minimum number of indicator data included in each category; wherein the number of categories is the number of categories into which the indicator data is to be divided; screening the indicator data of the advertisement to be analyzed based on the number of categories to determine a plurality of first cluster centers; Calculating the distance between each indicator data of the advertisement to be analyzed and each first cluster center; Based on the distance, the data quantity and the multiple first cluster centers, the indicator data of the advertisement to be analyzed is clustered to obtain a target clustering result.

2. The method according to claim 1, characterized in that The step of obtaining the indicator data of the advertisement to be analyzed includes: Obtaining basic indicator data of the advertisement to be analyzed; The basic indicator data is subjected to data standardization processing to obtain the indicator data of the advertisement to be analyzed.

3. The method according to claim 2, characterized in that The step of performing data standardization on the basic indicator data to obtain the indicator data of the advertisement to be analyzed includes: The basic indicator data is subjected to data standardization processing using a deviation standardization algorithm to obtain the indicator data of the advertisement to be analyzed.

4. The method according to claim 1, wherein The step of screening the indicator data of the advertisement to be analyzed based on the number of categories to determine a plurality of first cluster centers includes: determining a target quantile value based on the number of categories; The indicator data corresponding to the position of the target percentile value is determined from the indicator data of the advertisement to be analyzed to obtain the multiple first cluster centers.

5. The method according to claim 1, wherein The clustering of the indicator data of the advertisement to be analyzed based on the distance, the data quantity, and the plurality of first cluster centers to obtain a target clustering result includes: For each indicator data, comparing the distance between the indicator data and each first cluster center; Based on the size relationship and the data quantity, the indicator data of the advertisement to be analyzed is clustered to obtain the target clustering result.

6. The method according to claim 5, characterized in that The clustering of the indicator data of the advertisement to be analyzed based on the size relationship and the data quantity to obtain the target clustering result includes: For each indicator data, determining a first cluster center with the smallest distance from the multiple first cluster centers; For each indicator data, classify the indicator data and the indicator data corresponding to the first cluster center with the smallest distance into one category to obtain classified first indicator data; When it is determined that the difference in the number of indicator data included in each two categories of the first indicator data meets the preset number threshold, and the number of indicator data included in each category of the first indicator data is greater than the data number, the classified first indicator data is determined to be the target clustering result.

7. The method according to claim 6, characterized in that The method further comprises: When it is determined that the difference between the number of indicator data included in each two categories of the first indicator data does not meet a preset number threshold, or the number of indicator data included in any category of the first indicator data is less than or equal to the data number, the indicator data corresponding to the position of the preset quantile value is determined from each category of the first indicator data to obtain a second cluster center; Calculate the distance between each first indicator data and each second cluster center, and for each first indicator data, classify the first indicator data and the indicator data corresponding to the second cluster center with the smallest distance into one category to obtain the classified Nth indicator data, until the difference in the number of indicator data included in each two categories of the Nth indicator data meets the preset quantity threshold, and the number of indicator data included in each category of the Nth indicator data is greater than the data quantity, determine that the classified Nth indicator data is the target clustering result.

8. The method according to claim 6, characterized in that The method further comprises: When it is determined that the difference between the number of indicator data included in each two categories of the first indicator data does not meet a preset number threshold, or the number of indicator data included in any category of the first indicator data is less than or equal to the data number, the indicator data corresponding to the position of the preset quantile value is determined from each category of the first indicator data to obtain a second cluster center; Calculate the distance between each first indicator data and each second cluster center, and for each first indicator data, classify the indicator data corresponding to the first indicator data and the second cluster center with the smallest distance into one category to obtain the classified Nth indicator data, until the number of indicator data included in any category of the Nth indicator data is less than the number of data, and determine the classified N-1th indicator data as the target clustering result.

9. The method according to claim 1, characterized in that The method further comprises: The target clustering results are displayed in a graphical manner.

10. A data clustering device, characterized in that: The device includes: a processor, a memory and a communication bus; The communication bus is used to realize the communication connection between the processor and the memory; The processor is configured to execute the data clustering program stored in the memory to implement the following steps: Obtain the indicator data of the advertisement to be analyzed; Determine the number of categories and the minimum number of indicator data included in each category; wherein the number of categories is the number of categories into which the indicator data is to be divided; screening the indicator data of the advertisement to be analyzed based on the number of categories to determine a plurality of first cluster centers; Calculating the distance between each indicator data of the advertisement to be analyzed and each first cluster center; Based on the distance, the data quantity and the multiple first cluster centers, the indicator data of the advertisement to be analyzed is clustered to obtain a target clustering result.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the data clustering method according to any one of claims 1 to 9.

Images