[0066] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, in which the same or similar reference numerals indicate the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary, and are intended to explain the present invention, but should not be construed as limiting the present invention.
[0067] The following describes the method and device for processing advertisement billing based on entity clustering in embodiments of the present invention with reference to the accompanying drawings.
[0068] Generally, online advertising systems charge fees based on the effectiveness of advertisers’ advertisements. When advertisers publish advertisements, they usually need to purchase multiple ad words. For example, for shower gel advertisers, they need to buy advertisement words "shower gel" and "water". "Run", "Whitening" and so on.
[0069] The reserve price of advertising words is usually based on the corresponding level of the advertising account and the advertising words. Generally, the reservation price of the advertising words corresponding to the newly opened advertising accounts is relatively low. For example, for the newly opened advertising account 1, the advertising words are purchased The reserve price of "shower gel" is lower than the reservation price of the old advertising account 2 for buying the ad word "shower gel".
[0070] Therefore, in order to prevent some advertisers from opening new advertising accounts to circumvent the strategy of advertising reserve prices, the present invention proposes an entity clustering-based advertising billing processing method, which is specifically as follows:
[0071] figure 1 It is a flowchart of an advertising billing processing method based on entity clustering according to an embodiment of the present invention.
[0072] Such as figure 1 As shown, the advertising billing processing method based on entity clustering may include:
[0073] S101: Perform aggregation processing on advertisement history data, and determine a first mapping relationship between an advertising account identifier and an advertising entity.
[0074] It is understandable that even if the account names of the A advertising account and the B advertising account are different, the A advertising account and the B advertising account correspond to the same advertising entity, for example, the Internet advertising account A and the Internet advertising account B both correspond to the same Internet company. Among them, the advertising entity is a specific enterprise or individual that publishes an advertisement.
[0075] Since even if the advertising accounts are different, the target products promoted by the advertisements they publish are the same, or their corresponding advertising account IDs, registered email addresses and other account identifiers are the same, therefore, historical data can be aggregated and processed. Among them, historical data It may include the advertising account identifier and the corresponding advertising content, the name of the advertising account, the advertising ID, the advertising location, etc.
[0076] Furthermore, the first mapping relationship between the advertising account identifier and the advertising entity is determined, and in the mapping relationship, the advertising account identifier is classified and stored according to the corresponding advertising entity.
[0077] For example, in the first mapping relationship, the advertising account corresponding to advertising entity A includes advertising account A1, advertising account A2, and so on.
[0078] It should be noted that, according to different specific application scenarios, different methods can be used to aggregate the advertising history data to determine the first mapping relationship between the advertising account identifier and the advertising entity. Examples are as follows:
[0079] In the first example, the advertising account can be aggregated according to the advertising account identifier, that is, the historically displayed advertising account identifiers are analyzed, and the advertising account identifiers with the same entity identifier are grouped to determine the second between the advertising account identifier and the advertising entity A mapping relationship.
[0080] In this example, for example, if it is resolved that the user name of advertising account A is "Internet 1***" and the user name of advertising account B is "*** Internet 1***", then advertising account A It is divided into the same group as the advertising account B, and the first mapping relationship corresponding to the advertising entity is determined. In the first mapping relationship, the advertising account A and the advertising account B both belong to the Internet company 1.
[0081] In the second example, advertising accounts can be aggregated according to the content of advertising webpages, that is, the content of historically displayed advertising webpages can be detected, and advertising account IDs corresponding to advertising webpages with the same entity keywords can be grouped to determine the advertising account ID and the advertisement The first mapping relationship between entities.
[0082] In this example, for example, if the advertisement webpage content C contains the entity keywords "Internet 1", "Browser 1", etc., and the advertisement webpage content D also contains the entity keywords "Internet 1", " Browser 1", etc., the advertising account identifiers corresponding to the advertising webpage content C and the advertising webpage content D can be divided into the same group, and the first mapping relationship corresponding to the advertising entity can be determined. In the first mapping relationship, the advertising webpage The advertising account identifiers corresponding to the content C and the advertising webpage content D correspond to the same advertising entity.
[0083] S102: Training the historical account reserve price of the advertisement word according to the first mapping relationship to generate an advertisement value distribution model.
[0084] It is understandable that advertising accounts belonging to the same advertising entity may have different historical account reservation prices for the same advertising word. For example, for the same advertising word "browser", the corresponding reservation price may be different due to the time when the account is registered. Therefore, it is necessary to standardize the historical account reserve price of advertising words. Among them, the reserve price is the lowest bid price set for the ad word in the online advertising system.
[0085] Specifically, according to the first mapping relationship, the historical account reservation price of advertising words is trained, and the number of times, frequency, and display time of the advertising words are comprehensively considered to generate an advertising value distribution model, which is relatively objective and accurate storage The reserve price of the ad word.
[0086] S103: Calculate a second mapping relationship between the advertising entity and the entity reserve price of the advertising word according to the advertising value distribution model, so as to perform charging processing on the online advertisement according to the first mapping relationship and the second mapping relationship.
[0087] Specifically, after obtaining the advertising value distribution model, the second mapping relationship between the advertising entity and the entity reserve price of the advertising word is calculated, and the second mapping relationship is stored with the advertising entity and the corresponding advertising word Physical reserve price.
[0088] For example, for the advertising entity 1, the entity reserve price of the corresponding advertising term "browser" is x, and for the advertising entity 2, the entity reserve price of the corresponding advertising term "browser" is y, etc.
[0089] Furthermore, when charging an advertising entity, query the first mapping relationship to obtain the advertising entity corresponding to the advertising account, and query the second mapping relationship according to the advertising entity to obtain the reserve price of the advertising term corresponding to the advertising entity, so as to facilitate According to the reserve price, the advertisement issued by the advertising entity is billed.
[0090] In other words, by establishing the correspondence between the advertising entity and the entity reserve price of the advertising word, billing can be based on the advertising entity level. No matter which advertising account the advertiser uses, the billing is the same, regardless of the registration time. It effectively avoids advertising entities and avoids the reserve price strategy by registering new advertising accounts.
[0091] In summary, the entity clustering-based advertising billing processing method of the embodiment of the present invention aggregates advertising historical data, determines the first mapping relationship between the advertising account identifier and the advertising entity, and according to the first mapping relationship Train the historical account reserve price of advertising words to generate an advertising value distribution model, and then calculate the second mapping relationship between the advertising entity and the entity reserve price of the advertising word according to the advertising value distribution model, so as to use the first mapping relationship and the second The mapping relationship performs charging processing for online advertisements. In this way, based on the establishment of a relationship between the advertising entity and the advertising reserve price, the advertising is charged according to the advertising reserve price corresponding to the advertising entity, which improves the processing efficiency of advertising charging.
[0092] In order to more clearly describe the entity clustering-based advertising billing processing method of the embodiment of the present invention, the following describes in detail the specific implementation process of online advertising billing processing based on the first mapping relationship and the second mapping relationship. as follows:
[0093] figure 2 It is a flowchart of an advertising billing processing method based on entity clustering according to another embodiment of the present invention, such as figure 2 As shown, in the above step S103, after calculating the second mapping relationship between the advertising entity and the entity reserve price of the advertising word according to the advertising value distribution model, the method includes:
[0094] S201: Obtain a bid of an advertisement word corresponding to the first advertisement account to be processed.
[0095] Specifically, the bid of the advertisement word corresponding to the first advertisement account to be processed is obtained, and the bid is the highest purchase price of the first advertisement account for the advertisement word.
[0096] S202: Query the first mapping relationship to determine the first advertising entity corresponding to the identifier of the first advertising account.
[0097] It can be understood that, in order to avoid that the first advertising account may be owned by the corresponding advertising entity, the newly applied account in order to circumvent the reserve price strategy, after obtaining the bid of the advertising word corresponding to the first advertising account, is not based on the first advertising account. The registration time corresponding to the advertising account is used to obtain the reserve price of the advertising word, but the first advertising entity to which it belongs must first be determined.
[0098] Specifically, query the first mapping relationship to determine the first advertising entity corresponding to the identifier of the first advertising account. For example, the first advertising entity may be determined according to the account name of the first advertising account, the entity keyword in the advertising content, etc. The first advertising entity corresponding to the advertising account.
[0099] S203: Query the second mapping relationship to determine the reserve price of the first entity corresponding to the advertisement term of the first advertising entity.
[0100] S204: Perform charging processing on the online advertisement according to the reserve price and the bid of the first entity.
[0101] Specifically, after the first advertising entity corresponding to the first advertising account is determined, the second mapping relationship is queried to determine the first entity reserve price corresponding to the advertising term of the first advertising entity, so as to determine the first entity reserve price and Bid charges for online advertising.
[0102] It should be noted that, according to different specific application scenarios, the method of charging online advertisements is different according to the reserve price and bid of the first entity. Examples are as follows:
[0103] The first example is to compare the reserve price of the first entity with the bid, and if the reserve price of the first entity is greater than the bid, then the bid of the advertising word corresponding to the first advertising account is filtered, and the first advertising account cannot obtain the corresponding advertising word.
[0104] If the reserve price of the first entity is less than or equal to the bid, the online advertisement is charged according to the bid.
[0105] In the second example, in actual applications, even if the bid of the first advertising account is higher than the reserve price of the first entity, the reserve price of the first entity can be used according to the click-through rate of the advertisement, and the reserve price of the first entity can be used. Charge processing between the price points.
[0106] In this example, when the bid of the ad word is higher than the reserve price of the first entity, the bid of the higher ad word is not charged, but the bid of the ad word is calculated with the reservation price of the first entity. The price is between the bid of the ad word and the reserve price of the first entity.
[0107] For example, if the bid of the advertisement is higher than the reserve price of the first entity, use the formula reserve price of the advertisement = bid * 30% + reserve price of the first entity * 70% to calculate the final reserve price of the billed advertisement .
[0108] In this example, when the bid of the advertisement word is lower than the reserve price of the first entity, the bid and billing of the advertisement word are filtered out, and the first advertisement account cannot obtain the corresponding advertisement word.
[0109] In summary, the entity clustering-based advertising billing processing method of the embodiment of the present invention obtains and after calculating the second mapping relationship between the advertising entity and the entity reserve price of the advertising word according to the advertising value distribution model To process the bid of the advertising word corresponding to the first advertising account, query the first mapping relationship to determine the first advertising entity corresponding to the identifier of the first advertising account, and query the second mapping relationship to determine the advertising word corresponding to the first advertising entity The reserve price of the first entity, and then charge for online advertising based on the reserve price and bid of the first entity. As a result, the processing efficiency of advertising billing is further improved.
[0110] In order to achieve the above objective, the present invention also proposes an advertisement billing processing device based on entity clustering.
[0111] image 3 It is a schematic structural diagram of an advertisement billing processing device based on entity clustering according to an embodiment of the present invention.
[0112] Such as image 3 As shown, the entity clustering-based advertising billing processing device may include: a first determining module 110, a training generating module 120, and a calculating module 130.
[0113] Wherein, the first determining module 110 is configured to aggregate historical advertisement data, and determine the first mapping relationship between the advertising account identifier and the advertising entity.
[0114] Even if the advertising accounts are different, the advertisements they publish are for the same products, that is, the corresponding advertising entities are the same. Therefore, in order to facilitate accurate charging of advertisers, the first determining module 110 aggregates historical data. The data may include advertising account identification and corresponding advertising content, advertising account name, advertising ID, advertising location, etc.
[0115] Furthermore, the first determining module 110 determines the first mapping relationship between the advertising account identifier and the advertising entity, and in the mapping relationship, the user account is classified and stored according to the corresponding advertising entity.
[0116] It should be noted that, according to different specific application scenarios, the first determining module 110 may use different methods to aggregate historical advertising data to determine the first mapping relationship between the advertising account identifier and the advertising entity. Examples are as follows:
[0117] In the first example, the first determining module 110 can aggregate the advertising account according to the advertising account identifier, that is, analyze the historically displayed advertising account identifiers, and group the advertising account identifiers with the same entity identifier to determine the advertising account identifier and The first mapping relationship between advertising entities.
[0118] In this example, for example, if the first determining module 110 parses out that the user name of advertising account A is "Internet 1***", and the user name of advertising account B is "*** Internet 1***", Then, the advertising account A and the advertising account B are divided into the same group, and the first mapping relationship corresponding to the advertising entity is determined. In the first mapping relationship, the advertising account A and the advertising account B both belong to the Internet company 1.
[0119] In the second example, the first determining module 110 can aggregate the advertising accounts according to the content of the advertising webpage, that is, detect the content of the historically displayed advertising webpages, group the advertising account identifiers corresponding to the advertising webpages with the same entity keywords, and determine The first mapping relationship between the advertising account and the advertising entity.
[0120] In this example, for example, if the first determining module 110 detects that the advertisement webpage content C contains the entity keywords "Internet 1", "Browser 1", etc., and the advertisement webpage content D also contains the entity key The words "Internet 1", "Browser 1", etc., the advertising account identifiers corresponding to the advertising webpage content C and the advertising webpage content D can be divided into the same group, and the first mapping relationship corresponding to the advertising entity can be determined. In a mapping relationship, the advertising account identifiers corresponding to the advertising webpage content C and the advertising webpage content D correspond to the same advertising entity.
[0121] The training generating module 120 is configured to train the historical account reserve price of advertisement words according to the first mapping relationship to generate an advertisement value distribution model.
[0122] Specifically, the training generation module 120 trains the historical account reserve price of the advertisement word according to the first mapping relationship, and comprehensively considers the number of times, frequency, and display time of the advertisement word are clicked, and generates an advertisement value distribution model. The advertisement value distribution model is relatively Objectively and accurately store the reserve price of the advertisement.
[0123] The calculation module 130 is configured to calculate a second mapping relationship between the advertising entity and the entity reserve price of the advertising word according to the advertising value distribution model, so as to perform charging processing on the online advertisement according to the first mapping relationship and the second mapping relationship.
[0124] Specifically, after obtaining the advertising value distribution model, the calculation module 130 calculates the second mapping relationship between the advertising entity and the entity reserve price of the advertising word, and the second mapping relationship is stored with the advertising entity and the corresponding The physical reserve price of the ad.
[0125] Furthermore, when charging an advertising entity, query the first mapping relationship to obtain the advertising entity corresponding to the advertising account, and then query the second mapping relationship according to the advertising entity to obtain the reserve price of the advertising term corresponding to the advertising entity, so as to facilitate According to the reserve price, the advertisement issued by the advertising entity is charged.
[0126] It should be noted that the foregoing embodiment of the method for processing advertising billing based on entity clustering corresponds to the advertising billing processing device based on entity clustering in the embodiment of the present invention, and the advertising billing processing device based on entity clustering of the present invention Details that are not disclosed in the embodiments will not be repeated here.
[0127] In summary, the entity clustering-based advertising billing processing device of the embodiment of the present invention aggregates advertising historical data, determines the first mapping relationship between the advertising account identifier and the advertising entity, and according to the first mapping relationship Train the historical account reserve price of advertising words to generate an advertising value distribution model, and then calculate the second mapping relationship between the advertising entity and the entity reserve price of the advertising word according to the advertising value distribution model, so as to use the first mapping relationship and the second The mapping relationship performs charging processing for online advertisements. In this way, based on the establishment of a relationship between the advertising entity and the advertising reserve price, the advertising is charged according to the advertising reserve price corresponding to the advertising entity, which improves the processing efficiency of advertising charging.
[0128] Figure 4 It is a schematic structural diagram of an advertisement billing processing device based on entity clustering according to another embodiment of the present invention, such as Figure 4 As shown in image 3 Based on what is shown, the entity clustering-based advertising billing processing device further includes: an acquisition module 140, a second determination module 150, a third determination module 160, and a processing module 170.
[0129] Wherein, the obtaining module 140 is used to obtain the bid of the advertisement word corresponding to the first advertisement account to be processed.
[0130] Specifically, the obtaining module 140 obtains the bid of the advertisement word corresponding to the first advertisement account to be processed, and the bid may be determined according to the registration time of the first advertisement account.
[0131] The second determining module 150 is configured to query the first mapping relationship to determine the first advertising entity corresponding to the identifier of the first advertising account.
[0132] Specifically, the second determining module 150 queries the first mapping relationship to determine the first advertising entity corresponding to the identifier of the first advertising account. For example, the second determining module 150 may publish the advertising content according to the account name of the first advertising account. Keywords, etc., determine the first advertising entity corresponding to the first advertising account.
[0133] The third determining module 160 is configured to query the second mapping relationship to determine the first entity reserve price corresponding to the advertisement term of the first advertising entity.
[0134] The processing module 170 is configured to perform charging processing on the online advertisement according to the reserve price and the bid of the first entity.
[0135] Specifically, after the first advertising entity corresponding to the first advertising account is determined, the third determining module 160 queries the second mapping relationship to determine the first entity reserve price corresponding to the advertising term of the first advertising entity, thereby the processing module 170 The online advertisement is charged according to the reserve price and bid of the first entity.
[0136] It should be noted that, according to different specific application scenarios, the processing module 170 performs different charging processing methods for online advertisements according to the reserve price and bid of the first entity. Examples are as follows:
[0137] In an embodiment of the present invention, Figure 5 It is a schematic structural diagram of an advertising billing processing device based on entity clustering according to another embodiment of the present invention, such as Figure 5 As shown in Figure 4 Based on what is shown, the processing module 170 includes a comparison unit 171 and a processing unit 172.
[0138] Among them, the comparing unit 171 is used to compare the reserve price and the bid of the first entity.
[0139] The processing unit 172 is configured to filter the bid of the advertising word corresponding to the first advertising account when the reserve price of the first entity is greater than the bid.
[0140] In an embodiment of the present invention, the processing unit 172 is further configured to perform charging processing on the online advertisement according to the bid when the reserve price of the first entity is less than or equal to the bid.
[0141] It should be noted that the foregoing embodiment of the advertising billing processing method based on entity clustering corresponds to the advertising billing processing device based on entity clustering in the embodiment of the present invention, and the advertising billing processing device based on entity clustering of the present invention Details that are not disclosed in the embodiments will not be repeated here.
[0142] To sum up, the entity clustering-based advertising billing processing device of the embodiment of the present invention obtains and To process the bid of the advertising word corresponding to the first advertising account, query the first mapping relationship to determine the first advertising entity corresponding to the identifier of the first advertising account, and query the second mapping relationship to determine the advertising word corresponding to the first advertising entity The reserve price of the first entity, and then charge for online advertising based on the reserve price and bid of the first entity. As a result, the processing efficiency of advertising billing is further improved.
[0143] In the description of this specification, descriptions with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean specific features described in conjunction with the embodiment or example , Structure, materials or features are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples and the characteristics of the different embodiments or examples described in this specification without contradicting each other.
[0144] Although the embodiments of the present invention have been shown and described above, it can be understood that the above-mentioned embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can comment on the foregoing within the scope of the present invention. The embodiment undergoes changes, modifications, substitutions and modifications.