Channel analysis method and device

Abstract

The present invention discloses a method and device for channel analysis, and relates to the field of computer technology. A specific implementation of the method includes: determining the conversion data of all channels corresponding to preset indicators based on a directed graph constructed based on channel data and indicator data of all channels; removing the channel to be analyzed from the directed graph, and determining the conversion data of the remaining channels corresponding to the preset indicators based on the removed directed graph; and performing channel analysis on the channel to be analyzed based on the conversion data of all channels and the conversion data of the remaining channels. This implementation overcomes the technical problems of single indicator data and ignoring the mutual influence of various channels, and achieves the horizontal expansion of indicator data, from a single channel behavior to the interactive relationship between multiple channels, which is more in line with complex actual conditions.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a channel analysis method and device. Background Art

[0002] When existing technologies conduct value analysis on multiple channels within a system, they often use methods such as linear uniform distribution, decay over time, first-touch channel, and last-touch channel.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0004] 1. When analyzing data for each channel, only a single channel is considered, while the mutual influence of each channel is ignored, which does not conform to the complex actual situation.

[0005] 2. When evaluating the value of each channel, its impact on the overall situation is ignored, which is quite different from the evaluation result that takes into account the impact of the actual channel on the overall situation. Summary of the invention

[0006] In view of this, an embodiment of the present invention provides a channel analysis method and device, which can consider the frequency and probability of mutual transfer between channels, expand from a single channel behavior to the interactive relationship between multiple channels, better conform to the complex actual situation, analyze the impact of each channel on the whole, make the evaluation method more comprehensive, and can be quickly deployed under large amounts of data and respond to complex data changes in a timely manner.

[0007] To achieve the above object, according to one aspect of an embodiment of the present invention, a channel analysis method is provided, comprising:

[0008] Determine conversion data of all channels corresponding to preset indicators according to a directed graph constructed based on channel data and indicator data of all channels;

[0009] Removing the channels to be analyzed from the directed graph, and determining conversion data of the remaining channels corresponding to the preset indicators according to the directed graph after the removal;

[0010] Perform channel analysis on the channel to be analyzed based on the conversion data of all channels and the conversion data of the remaining channels;

[0011] The remaining channels refer to a collection of channels among all the channels except the channel to be analyzed.

[0012] Optionally, performing channel analysis on the channel to be analyzed includes:

[0013] Determine the impact factor of the channel to be analyzed corresponding to the preset indicator according to the following formula;

[0014]

[0015] Among them, R i The influencing factors of the channel to be analyzed, G_ALL is the conversion data of all channels, and G_REMOVE is the conversion data of the remaining channels.

[0016] Optionally, determining the conversion data of all channels corresponding to the preset indicator includes: determining all valid paths in a directed graph constructed based on the channel data and indicator data of all channels; determining the conversion data of each valid path based on all the valid paths; and calculating the conversion data of all channels corresponding to the preset indicator based on the conversion data of each valid path;

[0017] Determining the conversion data of the remaining channels corresponding to the preset indicator according to the removed directed graph, including: determining all valid paths in the removed directed graph; determining the conversion data of each valid path; and calculating the conversion data of the remaining channels corresponding to the preset indicator according to the conversion data of each valid path;

[0018] The valid path refers to the path leading to the preset indicator;

[0019] The calculation formula for conversion data of each effective path is:

[0020]

[0021] In a directed graph, the conversion data calculation formula for all valid paths is:

[0022]

[0023] Among them, Q t is the conversion data of the tth effective path, m is the number of channels of the tth effective path, P ij is the transfer data from the i-th channel to the j-th channel in the t-th valid path, i is the original channel, j is the pointed channel, and G is the conversion data of all valid paths of the directed graph.

[0024] Optionally, the channel analysis includes: the channel analysis is also used to: obtain the weight of each channel according to the ratio of the influencing factor of the channel to be analyzed to the sum of the influencing factors of all channels.

[0025] Optionally, before determining the conversion data of all channels corresponding to the preset indicators, the method further includes:

[0026] Determine all channel paths formed based on all channels according to the channel data and indicator data of all channels;

[0027] Determine all channel pairs and transfer data of each channel pair according to all channel paths;

[0028] A directed graph is constructed according to all channel pairs and the transfer data of each channel pair.

[0029] Optionally, identify all channel pairs including:

[0030] For each of the channel paths, a vector consisting of each channel in the channel path and the channel it points to is taken as a channel pair.

[0031] Optionally, determining the transfer data for each of the channel pairs comprises:

[0032] For each of the channel pairs, determining the number of the channel pairs for which all channel paths exist;

[0033] According to the number of all channel paths existing in each channel pair, construct a matrix of the number of transfers between channels;

[0034] According to the inter-channel transfer frequency matrix, construct the inter-channel transfer probability matrix;

[0035] According to the values ​​of the inter-channel transition probability matrix, the transfer data of the channel pair is determined.

[0036] Optionally, constructing a directed graph includes:

[0037] A directed graph is constructed with the indicator and each channel as vertices and the transfer data of each channel pair as the edge length between the two vertices corresponding to the channel pair.

[0038] Optionally, the channel data includes: the time when the channel is accessed; determining all channel paths formed based on all the channels includes:

[0039] Group all channel data by preset identifiers;

[0040] For each group, arrange the channels in the group in ascending order according to the time of access to form a channel path.

[0041] To achieve the above object, according to another aspect of an embodiment of the present invention, a channel analysis device is provided, comprising: a directed graph construction module, a conversion data calculation module, and a channel analysis module, wherein:

[0042] A directed graph construction module, which determines conversion data of all channels corresponding to preset indicators according to a directed graph constructed based on channel data and indicator data of all channels;

[0043] A conversion data calculation module removes the channels to be analyzed from the directed graph, and determines the conversion data of the remaining channels corresponding to the preset indicators according to the directed graph after the removal;

[0044] A channel analysis module performs channel analysis on the channel to be analyzed based on the conversion data of all channels and the conversion data of the remaining channels;

[0045] The remaining channels refer to a collection of channels among all the channels except the channel to be analyzed.

[0046] Optionally, the channel analysis module performs channel analysis on the channel to be analyzed, including:

[0047] Determine the impact factor of the channel to be analyzed corresponding to the preset indicator according to the following formula:

[0048]

[0049] Among them, R i The influencing factors of the channel to be analyzed, G_ALL is the conversion data of all channels, and G_REMOVE is the conversion data of the remaining channels.

[0050] Optionally, the directed graph construction module is further used to determine the conversion data of all channels corresponding to the preset indicator, including: determining all valid paths in the directed graph according to the directed graph constructed based on the channel data and indicator data of all channels; determining the conversion data of each valid path according to all the valid paths; and calculating the conversion data of all channels corresponding to the preset indicator according to the conversion data of each valid path;

[0051] The conversion data calculation module is also used to determine the conversion data of the remaining channels corresponding to the preset indicators according to the removed directed graph, including: determining all valid paths in the removed directed graph; determining the conversion data of each valid path; and calculating the conversion data of the remaining channels corresponding to the preset indicators according to the conversion data of each valid path;

[0052] The valid path refers to the path leading to the preset indicator;

[0053] The calculation formula for conversion data of each effective path is:

[0054]

[0055] In a directed graph, the conversion data calculation formula for all valid paths is:

[0056]

[0057] Among them, Q tis the conversion data of the tth effective path, m is the number of channels of the tth effective path, P ij is the transfer data from the i-th channel to the j-th channel in the t-th valid path, i is the original channel, j is the pointed channel, and G is the conversion data of all valid paths of the directed graph.

[0058] Optionally, the channel analysis module is further used to obtain the weight of each channel according to the ratio of the influencing factor of the channel to be analyzed to the sum of the influencing factors of all channels.

[0059] Optionally, before determining the conversion data of all channels corresponding to the preset indicator, the directed graph construction module is further used to:

[0060] Determine all channel paths formed based on all channels according to the channel data and indicator data of all channels;

[0061] Determine all channel pairs and transfer data of each channel pair according to all channel paths;

[0062] A directed graph is constructed according to all channel pairs and the transfer data of each channel pair.

[0063] Optionally, the directed graph construction module determines that all channel pairs include:

[0064] For each of the channel paths, a vector consisting of each channel in the channel path and the channel it points to is taken as a channel pair.

[0065] Optionally, the directed graph construction module determines the transfer data of each channel pair, including:

[0066] For each of the channel pairs, determining the number of the channel pairs for which all channel paths exist;

[0067] According to the number of all channel paths existing in each channel pair, construct a matrix of the number of transfers between channels;

[0068] According to the inter-channel transfer frequency matrix, construct the inter-channel transfer probability matrix;

[0069] According to the values ​​of the inter-channel transition probability matrix, the transfer data of the channel pair is determined.

[0070] Optionally, the directed graph construction module is further used to:

[0071] A directed graph is constructed with the indicator and each channel as vertices and the transfer data of each channel pair as the edge length between the two vertices corresponding to the channel pair.

[0072] Optionally, the channel data includes: the time when the channel is accessed; the directed graph construction module determines that all channel paths formed based on all the channels include:

[0073] All channel data are grouped by preset identifiers, and each group of data is sorted in ascending order by access time;

[0074] For each group, arrange the channels in the group in ascending order according to the access time to form a channel path.

[0075] To achieve the above-mentioned purpose, according to another aspect of an embodiment of the present invention, there is provided an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the multi-channel analysis method provided by an embodiment of the present invention.

[0076] To achieve the above objective, according to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored, and when the program is executed by a processor, the multi-channel analysis method provided by an embodiment of the present invention is implemented.

[0077] One embodiment of the above invention has the following advantages or beneficial effects:

[0078] Because the technical means of removing the channels to be analyzed in the directed graph and then analyzing the channel conversion data is adopted, the technical problem of ignoring the impact of each channel on the whole when evaluating its value and the large gap between the complex process of the actual channel's impact on the final result is overcome, thereby achieving the technical effect of being able to analyze the impact of each channel on all channels, making the evaluation method more comprehensive.

[0079] Because of the technical means of constructing a directed graph, the problem of only considering a single channel while ignoring the mutual influence of each channel during channel data analysis is overcome, which is not in line with the complex actual technical problem. The result is the expansion from a single channel behavior to the interactive relationship between multiple channels, which is more in line with the complex actual technical effect.

[0080] Because all the technical means in this embodiment are combined, the technical problem of single indicator data is overcome, thereby achieving the technical effect of horizontal expansion of channel data and indicator numbers, and rapid deployment and timely response to complex situations under large data volumes.

[0081] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0083] Figure 1 is a schematic diagram of the main process of the channel analysis method according to an embodiment of the present invention;

[0084] Figure 2 is a directed graph constructed based on channel data and indicator data of all channels according to a specific embodiment of the present invention;

[0085] Figure 3 is a directed graph of remaining channels after removing the channels to be analyzed according to a specific embodiment of the present invention;

[0086] Figure 4 is a schematic diagram of main modules of a channel analysis device according to an embodiment of the present invention;

[0087] Figure 5 is an exemplary system architecture diagram that can be applied to channel analysis in an embodiment of the present invention;

[0088] Figure 6 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0089] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0090] In business practice, there are often many intermediate channels between products (or services, etc.) and users (or terminals). The more channels there are, the more difficult it is to analyze the role played by each channel. Therefore, objectively analyzing and judging the overall influence of each channel involves the method provided by the present invention.

[0091] Figure 1 Schematic diagram of the main process of the channel analysis method according to an embodiment of the present invention. Figure 1 As shown, a channel analysis method includes:

[0092] Step 101: Determine conversion data of all channels corresponding to preset indicators according to a directed graph constructed based on channel data and indicator data of all channels;

[0093] Step 102: remove the channels to be analyzed from the directed graph, and determine the conversion data of the remaining channels corresponding to the preset indicators according to the directed graph after the removal;

[0094] Step 103: Perform channel analysis on the channel to be analyzed based on the conversion data of all channels and the conversion data of the remaining channels;

[0095] The remaining channels refer to a collection of channels among all the channels except the channel to be analyzed.

[0096] The channel data is used to record the channel access situation in order to achieve the indicator data. In some optional embodiments, it can include at least one of the following: user identification, channel identification, and access time. The user identification is used to indicate the source of the data; the channel identification and the corresponding access time are conducive to subsequent time sorting.

[0097] The indicator data is used to record the data related to the indicator and to measure the size of the generated indicator data. The indicator data in this method can use one or more at the same time, thereby overcoming the technical problem of single indicator data, achieving the effect of horizontal expansion of indicators, and having the technical effect of rapid deployment and timely response to complex situations under large data volumes.

[0098] The directed graph is a method for representing the relationship between objects. In this embodiment, it is used to represent the relationship between channels.

[0099] By adopting the technical means of constructing a directed graph, we overcome the technical problem of not conforming to the complex actual situation caused by only considering a single channel and ignoring the mutual influence of each channel when conducting channel data analysis, thereby achieving the technical effect of expanding from a single channel behavior to the interactive relationship between multiple channels, which is more in line with the complex actual situation.

[0100] By adopting the technical means of removing the channels to be analyzed in the directed graph and thus analyzing the channel conversion data, we overcome the technical problem of ignoring the impact of each channel on all channels when evaluating its value, and the large gap between the complex process of the actual channel's impact on the indicator data. This allows us to analyze the impact of each channel on all channels, making the evaluation method more comprehensive.

[0101] By combining the above steps in this embodiment, the technical problem of single indicator data is overcome, thereby achieving the technical effect of horizontal expansion of channel data and indicator numbers, and rapid deployment and timely response to complex situations under large data volumes.

[0102] The channel analysis for the channel to be analyzed in step 103 may optionally include:

[0103] Determine the impact factor of the channel to be analyzed corresponding to the preset indicator according to the following formula;

[0104]

[0105] Among them, R i is the influencing factor of the i-th channel to be analyzed, G_ALL is the conversion data of all channels, and G_REMOVE is the conversion data of the remaining channels.

[0106] By using the technical means of calculating and removing the influencing factors of the channels, the influence of the channels to be analyzed can be quantified. Not only can the differences in influence be clearly compared through the quantified figures, but it can also overcome the technical problem of ignoring the influence of each channel on all channels when evaluating the value of each channel, and the large gap between the complex process of the influence of the actual channel on the indicator data. This can achieve a more accurate analysis of the impact of each channel on all channels and achieve a more comprehensive and beneficial effect of channel evaluation.

[0107] When determining the conversion data of all channels corresponding to the preset indicator, optionally, it includes: determining all valid paths in the directed graph based on the channel data and indicator data of all channels; determining the conversion data of each valid path based on all the valid paths; calculating the conversion data of all channels corresponding to the preset indicator based on the conversion data of each valid path;

[0108] The valid path refers to the path leading to the preset indicator;

[0109] The calculation formula for conversion data of each effective path is:

[0110]

[0111] In a directed graph, the conversion data calculation formula for all valid paths is:

[0112]

[0113] Among them, Q t is the conversion data of the tth effective path, m is the number of channels of the tth effective path, P ij is the transfer data from the i-th channel to the j-th channel in the t-th valid path, i is the original channel, j is the pointed channel, and G is the conversion data of all valid paths of the directed graph.

[0114] Through this step, the path in the directed graph that can reach the preset indicator is called an effective path. The conversion data of the effective path is calculated, and all the conversion data are accumulated to obtain the conversion data of all channels corresponding to the preset indicator. The conversion data of all channels obtained is the effect produced by all channels, and is also used as reference data for subsequent calculations to remove the conversion data of each channel to improve the accuracy of the evaluation.

[0115] Determining conversion data of remaining channels corresponding to preset indicators based on the removed directed graph includes: determining all valid paths in the removed directed graph; determining conversion data of each valid path; and calculating conversion data of remaining channels corresponding to the preset indicators based on the conversion data of each valid path.

[0116] By calculating the conversion data after removing each channel, the mutual influence between channels is taken into account. This overcomes the technical problem that the existing technology ignores the mutual influence of each channel on the other channels when evaluating the value of each channel, and the large gap between the complex process of actually evaluating the influence of channels on indicator data. This achieves the technical effect of being able to more accurately analyze the influence of each channel on indicator data and make the evaluation more comprehensive.

[0117] In step 103, the channel analysis further includes the following steps:

[0118] The channel analysis is also used to obtain the weight of each channel according to the ratio of the influencing factor of the channel to be analyzed to the sum of the influencing factors of all channels.

[0119] The weight of each channel can be calculated by using the influence factor of each channel and the sum of the influence factors of each channel. The weight of each channel can obtain a highly accurate evaluation result, ignoring the inaccurate evaluation problem caused by the different influence of each channel, thereby improving the accuracy of channel analysis and evaluation. By calculating the weight of each channel, the importance of each channel can be obtained, and the weight of each channel can be used for channel evaluation and redistribution.

[0120] Specifically, the implementation formula is:

[0121]

[0122] Where: W i is the weight of the ith channel.

[0123] R i is the influencing factor of the ith channel,

[0124] ΣR i It is the sum of all channel influencing factors.

[0125] In step 101, before determining the conversion data of all channels corresponding to the preset indicators, the following steps may also be included:

[0126] Determine all channel paths formed based on all channels according to the channel data and indicator data of all channels;

[0127] Determine all channel pairs and transfer data of each channel pair according to all channel paths;

[0128] A directed graph is constructed according to all channel pairs and the transfer data of each channel pair.

[0129] The channel path is constructed based on the channel data and indicator data of all channels, and then the channel path is converted into channel pairs, and the calculated transfer data is used to construct a directed graph. The directed graph is used to reflect the mutual relationship between channels and the transfer data between channels, which facilitates the analysis and evaluation of each channel. The establishment of a directed graph overcomes the shortcomings of the existing technology that ignores the interaction between channels and is more in line with reality.

[0130] In this embodiment, determining all channel pairs may optionally include:

[0131] For each of the channel paths, a vector consisting of each channel in the channel path and the channel it points to is taken as a channel pair.

[0132] Summarizing each channel pair can more clearly show the transfer relationship between channels. For example, if a channel path is "start > channel_1 > channel_2 > channel_3 > indicator data", it can be split into four "source-to" channel pairs, namely: "start > channel_1", "channel_1 > channel_2", "channel_2 > channel_3", "channel_3 > indicator data". Forming the channel transfer relationship into a vector form can more clearly show the channel transfer direction and improve the efficiency of channel analysis.

[0133] Determining transfer data for each of the channel pairs, optionally including:

[0134] For each of the channel pairs, determining the number of the channel pairs for which all channel paths exist;

[0135] According to the number of all channel paths existing in each channel pair, construct a matrix of the number of transfers between channels;

[0136] According to the inter-channel transfer frequency matrix, construct the inter-channel transfer probability matrix;

[0137] According to the values ​​of the inter-channel transition probability matrix, the transfer data of the channel pair is determined.

[0138] The transfer between channels is constructed in the form of a matrix, so that the channel transfer times are easy to use. The transfer times matrix is ​​converted into a transfer data matrix, and the transfer data of each channel pair can be directly used for the subsequent construction of a directed graph, thereby improving the efficiency of channel evaluation.

[0139] The constructing of the directed graph may include:

[0140] A directed graph is constructed with the indicator and each channel as vertices and the transfer data of each channel pair as the edge length between the two vertices corresponding to the channel pair.

[0141] A directed graph is a method of representing the relationship between objects. Each vertex represents each object, each edge represents a connection relationship, and the variable length of each edge represents transfer data. In this embodiment, the vertices include the starting point, all channels, and indicators; each edge is a transfer between channel pairs, and the length of each edge is the transfer data of the corresponding channel pair. Constructing a directed graph can better evaluate the changes in channel transfers, so that the changes in channels can be clearly utilized, and the accuracy of channel analysis can be improved.

[0142] Constructing channel data includes: the time when the channel is accessed; determining all channel paths formed based on all the channels, including:

[0143] Group all channel data by preset identifiers;

[0144] For each group, arrange the channels in the group in ascending order according to the time of access to form a channel path.

[0145] The collected channel data is grouped according to the specified identifier, and the data is sorted in chronological order within the group to clearly show the order of channels visited by each specified identifier and the corresponding time, while displaying the indicator data generated by visiting each channel. By constructing a grouping and sorting matrix, channel data and indicator data can be more convenient to use, and the order of channels can be clearer, paving the way for the subsequent construction of channel paths. Summarize all paths and the indicator data achieved to pave the way for splitting into channel pairs. Recording the changes in each step of the path can achieve the beneficial effect of not missing out on the evaluation of each channel that affects the indicator data.

[0146] In business practice, marketing activities are promoted through multiple channels, so that marketing activity information reaches users through multiple methods and channels. Users often learn about activity information through more than one channel, and each learning of activity information will have a potential impact on users, thereby completing the final purchase decision. Therefore, objectively and accurately evaluating the influence of multiple channels or marketing factors in the entire user decision chain or path, as well as the subsequent cost allocation of each channel, all involve the method provided by the present invention.

[0147] In this embodiment, the steps of this method are described in detail by taking the analysis of various modes or channels of marketing promotion in business practice as an example.

[0148] Get channel data and indicator data in the system.

[0149] In this embodiment, the channel data is used to record the visits brought by each channel after the costs such as advertising are placed on each channel in order to achieve marketing benefits. The user identifier is used to indicate the source of the channel data and indicator data; the channel identifier and the corresponding access time are conducive to subsequent sorting by time. This method is explained by taking three users (user_1, user_2, user_3) and three channels as an example, where channel_1, channel_2, and channel_3 are three channel identifiers.

[0150] The indicator data is used to record the data related to the indicator and to measure the size of the indicator data. For example, the order amount, order volume, profit, etc. are increased. In this specific embodiment, the indicator data is a revenue indicator, which is used to record revenue-related indicators. Here, whether to purchase is used as an example to illustrate the multi-channel analysis method, with "1" representing purchase and "0" representing non-purchase. The channel data and indicator data collected in step 1 are summarized in Table 1.

[0151] Table 1 Obtained channel data and indicator data record table

[0152] User ID Channel ID Access time Indicator data user_1 Channel_1 2017 / 1 / 1 0 user_1 Channel_2 2017 / 1 / 3 0 user_1 Channel_3 2017 / 1 / 4 1 user_2 Channel_1 2017 / 1 / 3 1 user_3 Channel_2 2017 / 1 / 3 0 user_3 Channel_3 2017 / 1 / 6 1

[0153] The acquired channel data is grouped by the specified identifier, and each group of data is sorted in ascending order by access time.

[0154] In this embodiment, the channel data and indicator data of all channels obtained have been grouped according to the user identifiers, and each user identifier is grouped into one group, so there is no need to repeat the process.

[0155] Within each group, a channel path is formed according to the arranged order.

[0156] According to the channel data and indicator data of the ordered grouping matrix, all channel paths in each channel identifier are summarized. In this embodiment, the designated identifier is a user identifier. Based on Table 1, a channel path table belonging to a user identifier can be obtained, as shown in Table 2.

[0157] Table 2 Channel path table by user ID

[0158]

[0159] Summarize the channel path and purchase times of each user ID to pave the way for subsequent splitting into channel pairs. Record the changes in each step of the channel path to achieve the beneficial effect of not missing out on the evaluation of each channel that affects the conversion index.

[0160] Based on the channel paths summarized for each channel, the channel pairs consisting of the channel and the channel it points to are broken down according to the access time, and the number of times each channel pair occurs is marked.

[0161] Take user_1 as an example, the corresponding channel path is "start>channel_1>channel_2>channel_3>purchase". It is divided into four channel pairs, namely: "start>channel_1", "channel_1>channel_2", "channel_2>channel_3", "channel_3>purchase". The number of times each channel transfer occurs and its indicator data value need to be recorded at the same time. All channel pairs of user_1 are shown in Table 3.

[0162] Table 3 Example table of all channel pairs for user_1

[0163] Original Channel Pointing channel Number of transfers start Channel_1 1 Channel_1 Channel_2 1 Channel_2 Channel_3 1 Channel_3 Buy 1

[0164] The channel pairs of the three groups of channel pairs in this embodiment are added together to obtain the total number of transfers of the channel pairs. As shown in Table 4,

[0165] Table 4 Summary of channel pairs for all users

[0166] Original Channel Pointing channel Number of transfers start Channel_1 2 start Channel_2 1 Channel_1 Channel_2 1 Channel_1 Do not buy 1 Channel_2 Channel_3 2 Channel_3 Buy 2

[0167] According to the number of times all channel pairs occur in all channel paths, a matrix of the number of transfers between channels is constructed. In this embodiment, the channel pairs and the number of transfers counted in Table 4 are converted into a matrix of the number of transfers with the original channels as rows and the pointed channels as columns. The number of transfers represents the number of times the original channels transfer to the pointed channels, and a column is added for the total number of original channel transfers. As shown in Table 5,

[0168] Table 5. Matrix of transfer times between channels

[0169]

[0170] Then, based on the inter-channel transfer frequency matrix, the inter-channel transfer probability matrix is ​​constructed; among which, the inter-channel transfer probability is the transfer data of the channel pair.

[0171] According to the inter-channel transfer frequency matrix in Table 5, each element in the matrix is ​​divided by the total number of times of each source channel to obtain the inter-channel transfer probability matrix, as shown in Table 6.

[0172] Table 6: Inter-channel transition probability matrix

[0173]

[0174] According to the inter-channel transfer probability matrix in Table 6, all channel pairs and the transfer data of each channel pair can be determined as the side length of the directed graph. In this specific embodiment, the end point includes purchase and non-purchase. Using the values ​​in the inter-channel transfer probability matrix in Table 6, the directed graph constructed is as follows: Figure 2 shown.

[0175] like Figure 2 As shown in the figure, all users were exposed to only two channels, Channel_1 and Channel_2, which accounted for 66.7% and 33.3% of the total, respectively. Among the users who were exposed to Channel_1, 50% subsequently exposed to Channel_2, and the remaining 50% did not place an order. All users who were exposed to Channel_2 were exposed to Channel_3. Among them, the number of users who were exposed to Channel_3 who placed an order and the number of users who did not place an order accounted for 50% each.

[0176] By constructing a directed graph, it is possible to clearly show the probability and order of users being affected by each channel.

[0177] According to the directed graph, all valid paths in the directed graph are determined. The valid path refers to the path that is finally purchased. Figure 2 , determine all valid paths in the directed graph. One conversion path is "start > channel_1 > channel_2 > channel_3 > purchase", named as valid path 1; the other conversion path is "start > channel_2 > channel_3 > do not purchase", named as valid path 2.

[0178] According to the effective path, the conversion data of each path is determined. Optionally, the calculation formula of the conversion data of each path is:

[0179]

[0180] Among them, Q t is the conversion data of each path, m is the number of channels in a path, P ij is the transfer data of two channels, i is the original channel, j is the directed channel. In this embodiment, the conversion data of effective path 1 is 0.333*1*0.5 (recorded as 16.7%), and the conversion data of effective path 2 is 0.667*0.5*1*0.5 (recorded as 16.7%).

[0181] According to the conversion data of each path, the conversion data of the entire directed graph is calculated. The conversion data calculation formula of all valid paths of the directed graph is:

[0182]

[0183] Among them, G is the overall transformation data of the directed graph.

[0184] Add up the conversion data of each path to calculate the conversion data of the entire directed graph. Using the formula:

[0185]

[0186] It can be obtained that in this embodiment, the conversion data of the directed graph is 0.333*1*0.5+0.667*0.5*1*0.5, that is, before the channel is removed, the overall conversion data (G_ALL) is: 33.3%.

[0187] like Figure 3 As shown, when Channel_1 is removed, only one valid conversion path "Start>Channel_2>Channel_3>Conversion" remains, and the conversion data of the corresponding path is 0.333*1*0.5, that is, the overall conversion data (G_REMOVE) of the entire directed graph after removing Channel_1 is 16.7%.

[0188] According to the conversion data of all channels and the conversion data of the remaining channels, channel analysis is performed on the channel to be analyzed, including the following steps:

[0189] Determine the impact factor of the channel to be analyzed corresponding to the preset indicator according to the following formula;

[0190]

[0191] Among them, R i is the influence factor of the i-th channel to be analyzed, G_ALL is the conversion data of all channels, and G_REMOVE is the conversion data of the remaining channels after removing the channel to be analyzed. Determine the influence factor of the channel to be analyzed corresponding to the preset indicator, that is, the influence factor of a single channel (R i ), so the impact factor R1 of “channel_1” is (1–0.167 / 0.333)=0.5.

[0192] Similarly, by executing the above steps, we can get the impact factors R2 and R3 of channel_2 and channel_3 as 1, 1 respectively.

[0193] By utilizing the technical means of calculating the removal effect, we overcome the technical problem of ignoring the impact of each channel on all channels when evaluating its value, and the large gap between the complex process of the actual channel's impact on the final result. This allows us to more accurately analyze the impact of each channel on all channels, making the evaluation method more comprehensive.

[0194] The weight of each channel is obtained based on the ratio of the channel's impact factor to the sum of all channel impact factors.

[0195] According to the influence factors of each channel, calculate the weight of each channel, that is, the proportion of each channel influence factor to the sum of all channel influence factors:

[0196]

[0197] Where: W iAssign weights to the final revenue of channel i.

[0198] R i is the removal effect of channel i.

[0199] According to the removal effects of the above three channels, they are 0.5, 1, and 1 respectively. The weights of the three channels are respectively based on the proportion of their respective removal effects to the total removal effects of all channels:

[0200] Channel_1: 0.5 / (0.5+1+1)=0.2

[0201] Channel_2: 1 / (0.5+1+1)=0.4

[0202] Channel_3: 1 / (0.5+1+1)=0.4

[0203] The above three channels are weighted according to the ratio of 0.2:0.4:0.4.

[0204] Figure 4 A channel analysis device 400 is provided, comprising:

[0205] Module 401, directed graph construction module, determines conversion data of all channels corresponding to preset indicators according to a directed graph constructed based on channel data and indicator data of all channels;

[0206] Module 402, a conversion data calculation module, removes the channels to be analyzed from the directed graph, and determines the conversion data of the remaining channels corresponding to the preset indicators according to the removed directed graph;

[0207] Module 403, a channel analysis module, performs channel analysis on the channel to be analyzed according to the conversion data of all channels and the conversion data of the remaining channels;

[0208] The remaining channels refer to a collection of channels among all the channels except the channel to be analyzed.

[0209] By constructing directed graph module 401, the channel analysis device is used to analyze the impact of all channels on the results, overcoming the technical problem of single indicator data, thereby achieving the technical effect of horizontal expansion of channel data and indicator numbers, and having rapid deployment and timely response to complex situations under large data volumes.

[0210] By constructing the conversion data calculation module 402, the technical problem of ignoring the impact of each channel on all channels when evaluating the value of each channel and the large gap between the complex process of the impact of the actual channel on the indicator data is overcome, thereby achieving the technical effect of being able to analyze the impact of each channel on all channels and making the evaluation method more comprehensive.

[0211] By constructing the channel analysis module 403, the problem of only considering a single channel and ignoring the mutual influence of each channel during channel data analysis, which is not in line with the complex actual technical problem, is overcome, thereby achieving the expansion from a single channel behavior to the interactive relationship between multiple channels, which is more in line with the complex actual technical effect.

[0212] In the embodiment of the present invention, the channel analysis module 403 performs channel analysis on the channel to be analyzed, including:

[0213] Determine the impact factor of the channel to be analyzed corresponding to the preset indicator according to the following formula;

[0214]

[0215] Among them, R i is the influencing factor of the channel to be analyzed for the i-th channel, G_ALL is the conversion data of all channels, and G_REMOVE is the conversion data of the remaining channels.

[0216] By calculating the impact factor, the impact size obtained from the channel analysis can be quantified, which makes it easier to compare the influence size of the channel and improves the accuracy of judging the influence of the channel. By introducing the concept of impact factor, the technical problem of ignoring the impact of each channel on all channels when evaluating the value of each channel is overcome, which is a complex process with a large gap between the actual channel's impact on the final result. This can achieve a more accurate analysis of the impact of each channel on all channels, making the evaluation method more comprehensive.

[0217] In the embodiment of the present invention, in the conversion data calculation module 402, the directed graph construction module is further used to include:

[0218] According to a directed graph constructed based on channel data and indicator data of all channels, all valid paths in the directed graph are determined; according to all valid paths, conversion data of each valid path is determined; according to the conversion data of each valid path, conversion data of all channels corresponding to preset indicators are calculated;

[0219] The conversion data calculation module 402 is also used to determine the conversion data of the remaining channels corresponding to the preset indicator according to the removed directed graph, including: determining all valid paths in the removed directed graph; determining the conversion data of each valid path; and calculating the conversion data of the remaining channels corresponding to the preset indicator according to the conversion data of each valid path;

[0220] The valid path refers to the path leading to the preset indicator;

[0221] The calculation formula for conversion data of each effective path is:

[0222]

[0223] In a directed graph, the conversion data calculation formula for all valid paths is:

[0224]

[0225] Among them, Q t is the conversion data of the tth effective path, m is the number of channels of the tth effective path, P ij is the transfer data from the i-th channel to the j-th channel in the t-th valid path, i is the original channel, j is the pointed channel, and G is the conversion data of all valid paths of the directed graph.

[0226] By determining the effective path in the conversion data calculation module 402, and then calculating the conversion data of the effective path, the conversion data of all channels of the directed graph can be obtained, and the proportion of the overall conversion to all channels can be determined. This module is used to compare and remove a certain channel, and the change in the conversion data of all channels can more accurately determine the proportion of the channel.

[0227] The channel analysis module 403 is further used to obtain the weight of each channel according to the ratio of the influencing factor of the channel to be analyzed to the sum of the influencing factors of all channels.

[0228] By using the obtained weights, a highly accurate evaluation result can be obtained, which can solve the problem of inaccurate evaluation caused by ignoring the different influences of each channel in the multi-channel analysis method in the prior art, thereby improving the accuracy of multi-channel analysis evaluation.

[0229] In the embodiment of the present invention, before determining the conversion data of all channels corresponding to the preset indicators, the directed graph construction module 401 is further used to:

[0230] Determine all channel paths formed based on all channels according to the channel data and indicator data of all channels;

[0231] Determine all channel pairs and transfer data of each channel pair according to all channel paths;

[0232] A directed graph is constructed according to all channel pairs and the transfer data of each channel pair.

[0233] The channel path is constructed based on the channel data and indicator data of all channels, and then the path is converted into channel pairs, and the calculated transfer data is used to construct a directed graph. The directed graph is used to reflect the mutual relationship between channels and the transfer data between channels, which facilitates the analysis and evaluation of each channel. The establishment of a directed graph overcomes the shortcomings of the existing technology that ignores the interaction between channels and is more in line with reality.

[0234] The directed graph construction module 401 determines that all channel pairs include:

[0235] For each of the channel paths, a vector consisting of each channel in the channel path and the channel it points to is taken as a channel pair.

[0236] By setting up channel pairs, the direction of the channel flow can be clearly obtained, and the path is split into channel pairs, avoiding the situation of missing data when counting the channel flow direction, and improving the accuracy of channel analysis. Use directed graphs to build the connection relationship between channels, and use directed graphs to clearly display the transfer changes between channels, so as to facilitate better evaluation of changes between channels, thereby improving the accuracy of channel evaluation.

[0237] In the embodiment of the present invention, in the directed graph construction module, determining the transfer data of each channel pair includes:

[0238] For each of the channel pairs, determining the number of the channel pairs for which all channel paths exist;

[0239] According to the number of all channel paths existing in each channel pair, construct a matrix of the number of transfers between channels;

[0240] According to the inter-channel transfer frequency matrix, construct the inter-channel transfer probability matrix;

[0241] According to the values ​​of the inter-channel transition probability matrix, the transfer data of the channel pair is determined.

[0242] The number of occurrences of each channel pair is determined by traversal, avoiding inaccurate statistics. Thus, constructing the inter-channel transfer matrix can make the transfer data of the channel pair convenient for subsequent modules to directly call, thus improving the efficiency of channel analysis.

[0243] In the embodiment of the present invention, the directed graph construction module 401 is further used for:

[0244] A directed graph is constructed with the indicator and each channel as vertices and the transfer data of each channel pair as the edge length between the two vertices corresponding to the channel pair.

[0245] By setting all channel data and indicator data as vertices and constructing a directed graph with channel transfer data as the edge length, the relationship between channels can be clearly obtained, making it convenient to remove the channels to be analyzed.

[0246] The channel data includes: the time when the channel is accessed; the directed graph construction module 401 determines that all channel paths formed based on all the channels include:

[0247] All channel data are grouped according to preset identifiers, and each group of channel data is sorted in ascending order by access time;

[0248] For each group, arrange the channels in the group in ascending order according to the access time to form a channel path.

[0249] Channel data is grouped according to the specified identifier and arranged in chronological order within the group, which can clearly show the order of the channels visited and the time of visit. By constructing group sorting, channel data and indicator data can be more convenient to use, and the channel sequence is clearer, paving the way for building channel paths.

[0250] By summarizing each path and indicator data, we can clearly obtain the flow of each step of the channel, paving the way for the subsequent splitting into channel pairs. Recording the changes in the path channel can achieve the beneficial effect of not missing any channel that has an impact on all channels. Summarizing the indicator data generated by each channel pair can more clearly obtain the number of times the channel pair is used.

[0251] According to another aspect of an embodiment of the present invention, there is provided an electronic device for querying data based on configuration information, comprising:

[0252] one or more processors;

[0253] a storage device for storing one or more programs,

[0254] When one or more programs are executed by one or more processors, the one or more processors implement the data query method based on configuration information of the present invention.

[0255] Figure 5 An exemplary system architecture 500 to which the channel analysis method or apparatus according to an embodiment of the present invention can be applied is shown.

[0256] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0257] Users can use terminal devices 501, 502, 503 to interact with server 505 through network 504 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 501, 502, 503, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0258] The terminal devices 501 , 502 , and 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0259] The server 505 may be a server that provides various services, such as a backend management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 501, 502, and 503. The backend management server may analyze and process the received data such as product information query requests, and feed back the processing results (such as target push information, product information - only an example) to the terminal device.

[0260] It should be noted that the channel analysis method provided in the embodiment of the present invention is generally executed by the server 505 , and accordingly, the channel analysis device is generally disposed in the server 505 .

[0261] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0262] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present invention. Figure 6 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0263] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0264] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0265] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0266] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0267] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0268] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes a sending module, an acquisition module, a determination module and a first processing module. The names of these modules do not constitute a limitation on the modules themselves in some cases, for example, the sending module may also be described as a "module for sending a data acquisition request to a connected server".

[0269] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently and not be assembled into the device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by a device, the device includes: step S101, determining the conversion data of all channels corresponding to the preset index according to the directed graph constructed based on the channel data and indicator data of all channels; step S102, removing the channel to be analyzed from the directed graph, and determining the conversion data of the remaining channels corresponding to the preset index according to the removed directed graph; step S103, performing channel analysis on the channel to be analyzed according to the conversion data of all channels and the conversion data of the remaining channels; wherein the remaining channels refer to the set of channels among all channels except the channel to be analyzed.

[0270] It can be seen from the technical solution of the embodiment of the present invention that the technical means for channel analysis based on the channel data and indicator data of all channels is adopted, so as to overcome the technical problem of single indicator data, thereby achieving the technical effect that the channel data and indicator number can be horizontally expanded, and the technical effect of rapid deployment and timely response to complex situations under large data volume is achieved;

[0271] By adopting the technical means of analyzing channel conversion data after removing the channels to be analyzed in the directed graph, the technical problem of ignoring the impact of each channel on the overall situation when evaluating the value of each channel and the large gap between the complex process of the actual channel's impact on the final result is overcome, thereby achieving the technical effect of being able to analyze the impact of each channel on the overall situation and making the evaluation method more comprehensive;

[0272] By utilizing the technical means of constructing a directed graph, we overcome the technical problem of only considering a single channel when conducting data analysis on the channel and ignoring the mutual influence of each channel, which is not in line with the complex actual situation. We then achieve the technical effect of expanding from a single channel behavior to the interactive relationship between multiple channels, which is more in line with the complex actual situation.

[0273] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A channel analysis method, characterized in that: include: Determine conversion data of all channels corresponding to preset indicators according to a directed graph constructed based on channel data and indicator data of all channels; The channels are used to promote marketing activity information, the channel data is used to record the visits brought by each channel after the advertising costs are placed on each channel in order to achieve marketing benefits, and the indicator data is the income indicator; Removing the channels to be analyzed from the directed graph, and determining conversion data of the remaining channels corresponding to the preset indicators according to the directed graph after the removal; Perform channel analysis on the channels to be analyzed based on the conversion data of all channels and the conversion data of the remaining channels to determine the impact of each channel on all channels; The remaining channels refer to the set of channels among all the channels except the channel to be analyzed; Determining conversion data of all channels corresponding to the preset indicators, including: determining all valid paths in the directed graph based on the channel data and indicator data of all channels; determining conversion data of each valid path based on all valid paths; calculating conversion data of all channels corresponding to the preset indicators based on the conversion data of each valid path; Determining the conversion data of the remaining channels corresponding to the preset indicator according to the removed directed graph, including: determining all valid paths in the removed directed graph; determining the conversion data of each valid path; and calculating the conversion data of the remaining channels corresponding to the preset indicator according to the conversion data of each valid path; The valid path refers to the path leading to the preset indicator; The calculation formula for conversion data of each effective path is: In a directed graph, the conversion data calculation formula for all valid paths is: Among them, Q t is the conversion data of the tth effective path, m is the number of channels of the tth effective path, P ij is the transfer data from the i-th channel to the j-th channel in the t-th valid path, i is the original channel, j is the pointed channel, and G is the conversion data of all valid paths of the directed graph.

2. The method according to claim 1, characterized in that Performing channel analysis on the channel to be analyzed, including: Determine the impact factor of the channel to be analyzed corresponding to the preset indicator according to the following formula; Among them, R i The influencing factors of the channel to be analyzed, G_ALL is the conversion data of all channels, and G_REMOVE is the conversion data of the remaining channels.

3. The method according to claim 2 or 1, characterized in that Channel analysis includes: obtaining the weight of each channel based on the ratio of the influencing factor of the channel to be analyzed to the sum of the influencing factors of all channels.

4. The method according to claim 1, characterized in that Before determining the conversion data of all channels corresponding to the preset indicators, the method further includes: Determine all channel paths formed based on all channels according to the channel data and indicator data of all channels; Determine all channel pairs and transfer data of each channel pair according to all channel paths; A directed graph is constructed according to all channel pairs and the transfer data of each channel pair.

5. The method according to claim 4, characterized in that Make sure all channel pairs include: For each of the channel paths, a vector consisting of each channel in the channel path and the channel it points to is taken as a channel pair.

6. The method according to claim 4, characterized in that Determine the transfer data for each of the channel pairs, including: For each of the channel pairs, determining the number of the channel pairs for which all channel paths exist; According to the number of all channel paths existing in each channel pair, construct a matrix of the number of transfers between channels; According to the inter-channel transfer frequency matrix, construct the inter-channel transfer probability matrix; According to the values ​​of the inter-channel transition probability matrix, the transfer data of the channel pair is determined.

7. The method according to any one of claims 4 to 6, characterized in that The construction of the directed graph comprises: A directed graph is constructed with the indicator and each channel as vertices and the transfer data of each channel pair as the edge length between the two vertices corresponding to the channel pair.

8. The method according to any one of claims 4 to 6, characterized in that The channel data includes: the time when the channel is accessed; determining all channel paths formed based on all the channels, including: Group all channel data by preset identifiers; For each group, arrange the channels in the group in ascending order according to the time of access to form a channel path.

9. A channel analysis device, characterized in that: include: Directed graph construction module, conversion data calculation module, channel analysis module, among which, A directed graph construction module determines conversion data of all channels corresponding to preset indicators based on a directed graph constructed based on channel data and indicator data of all channels; the channels are used to promote marketing activity information, the channel data are used to record the visits brought by each channel after the advertising costs are placed on each channel in order to achieve marketing benefits, and the indicator data are revenue indicators; A conversion data calculation module removes the channels to be analyzed from the directed graph, and determines the conversion data of the remaining channels corresponding to the preset indicators according to the directed graph after the removal; A channel analysis module performs channel analysis on the channel to be analyzed based on the conversion data of all channels and the conversion data of the remaining channels to determine the impact of each channel on all channels; The remaining channels refer to the set of channels among all the channels except the channel to be analyzed; Determining conversion data of all channels corresponding to the preset indicators, including: determining all valid paths in the directed graph based on the channel data and indicator data of all channels; determining conversion data of each valid path based on all valid paths; calculating conversion data of all channels corresponding to the preset indicators based on the conversion data of each valid path; Determining the conversion data of the remaining channels corresponding to the preset indicator according to the removed directed graph, including: determining all valid paths in the removed directed graph; determining the conversion data of each valid path; and calculating the conversion data of the remaining channels corresponding to the preset indicator according to the conversion data of each valid path; The valid path refers to the path leading to the preset indicator; The calculation formula for conversion data of each effective path is: In a directed graph, the conversion data calculation formula for all valid paths is: Among them, Q t is the conversion data of the tth effective path, m is the number of channels of the tth effective path, P ij is the transfer data from the i-th channel to the j-th channel in the t-th valid path, i is the original channel, j is the pointed channel, and G is the conversion data of all valid paths of the directed graph.

10. The device according to claim 9, characterized in that The channel analysis module performs channel analysis on the channel to be analyzed, including: Determine the impact factor of the channel to be analyzed corresponding to the preset indicator according to the following formula: Among them, R i The influencing factors of the channel to be analyzed, G_ALL is the conversion data of all channels, and G_REMOVE is the conversion data of the remaining channels.

11. The device according to claim 9, characterized in that The channel analysis module is also used to obtain the weight of each channel according to the ratio of the influencing factor of the channel to be analyzed to the sum of the influencing factors of all channels.

12. The device according to claim 9, characterized in that Before determining the conversion data of all channels corresponding to the preset indicators, the directed graph construction module is further used to: Determine all channel paths formed based on all channels according to the channel data and indicator data of all channels; Determine all channel pairs and transfer data of each channel pair according to all channel paths; A directed graph is constructed according to all channel pairs and the transfer data of each channel pair.

13. The device according to claim 12, characterized in that The directed graph building module determines all channel pairs including: For each of the channel paths, a vector consisting of each channel in the channel path and the channel it points to is taken as a channel pair.

14. The device according to claim 12, characterized in that The directed graph construction module determines the transfer data of each channel pair, including: For each of the channel pairs, determining the number of the channel pairs for which all channel paths exist; According to the number of all channel paths existing in each channel pair, construct a matrix of the number of transfers between channels; According to the inter-channel transfer frequency matrix, construct the inter-channel transfer probability matrix; According to the values ​​of the inter-channel transition probability matrix, the transfer data of the channel pair is determined.

15. The device according to any one of claims 12 to 14, characterized in that The directed graph building module is further used for: A directed graph is constructed with the indicator and each channel as vertices and the transfer data of each channel pair as the edge length between the two vertices corresponding to the channel pair.

16. The device according to any one of claims 12 to 14, characterized in that The channel data includes: the time when the channel is accessed; the directed graph construction module determines that all channel paths formed based on all the channels include: All channel data are grouped by preset identifiers, and each group of data is sorted in ascending order by access time; For each group, arrange the channels in the group in ascending order according to the access time to form a channel path.

17. A channel analysis electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8.

18. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images