Advertiser client mining method and device, electronic device, and storage medium
The crawler obtains advertising data and uses optical character recognition and recognition models to automatically extract and filter new advertising brands, solving the problem of low mining efficiency of advertisers and customers in the existing technology, and achieving efficient new advertising brand discovery.
Patent Information
- Application Number
- CN202210095683.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-01-26
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-01-26
AI Technical Summary
Existing advertiser customer mining methods are inefficient and cannot effectively obtain information that meets needs. Relying on manual and paid ordering commercial information cannot guarantee the acquisition of new advertiser customer information.
Obtain advertising data through crawlers, identify advertising image text information through optical characters, extract advertising brands, use preset recognition models to identify brands, filter out new brands that are not in the brand library, and feedback and update them to the library.
The method of automatically mining new advertising brands has been realized, efficiency has been improved, information on advertisers and customers that meet their needs has been obtained, and new advertising brands have been easily discovered.
Smart Images

Figure CN114419614B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of information mining, and in particular to a method and device for mining advertiser clients, an electronic device, and a storage medium. Background Art
[0002] As the number of applications continues to increase, advertisers have more options for advertising, which makes advertising competition more and more intense. Therefore, in addition to maintaining existing advertisers, it is necessary to continuously expand new customers. Therefore, it is very important to explore new advertisers who have advertising needs.
[0003] At present, in order to obtain the required advertiser customer information, advertiser customer mining work requires manual work to understand various advertising product information in the market to obtain advertiser customer information. Secondly, it requires paid subscription to relevant commercial information to understand the latest advertising information of each advertiser.
[0004] However, the manual method is too inefficient, and the method of ordering commercial information cannot effectively guarantee that the information that meets one's needs is obtained, such as there is no guarantee that the information obtained is new advertiser customer information. Summary of the invention
[0005] Based on the above-mentioned deficiencies of the prior art, the present application provides a method and device for mining advertiser customers, an electronic device, and a storage medium to solve the problem that the existing methods are too inefficient or cannot guarantee the acquisition of required information.
[0006] In order to achieve the above objectives, this application provides the following technical solutions:
[0007] The first aspect of the present application provides a method for mining advertiser customers, comprising:
[0008] Get multiple advertisement data crawled by the crawler;
[0009] Based on each of the advertisement data, obtaining an advertisement picture corresponding to each of the advertisement data;
[0010] For each of the advertisement pictures, optical character recognition is performed on the advertisement pictures to obtain text information on each of the advertisement pictures;
[0011] Extracting the advertising brand corresponding to the advertising image from the text information on the advertising image;
[0012] If the advertising brand corresponding to the advertising picture is not extracted from the text information on the advertising picture, the advertising picture is recognized by using a preset recognition model to obtain the advertising brand corresponding to the advertising picture;
[0013] Filter out advertising brands that do not exist in the advertising brand library from the advertising brands corresponding to the advertising images to obtain multiple new advertising brands;
[0014] Provide feedback to each of the new advertising brands.
[0015] Optionally, in the advertiser client mining method provided above, the step of obtaining the advertisement image corresponding to each advertisement data based on each advertisement data includes:
[0016] Processing the advertisement data that are not in video format among the advertisement data into target format files according to preset rules; wherein the target format files refer to files in a format that supports playback;
[0017] The advertisement data in video format and the target format file are played respectively, and screenshots are taken according to a preset strategy during the playing process to obtain advertisement pictures corresponding to each advertisement data.
[0018] Optionally, in the advertiser client mining method provided above, the step of obtaining the advertisement image corresponding to each advertisement data based on each advertisement data includes:
[0019] For each piece of the advertisement data, determining whether the advertisement data is in a picture format;
[0020] If it is determined that the advertisement data is in a picture format, determining the advertisement data as the corresponding advertisement picture;
[0021] If it is determined that the advertisement data is not in a picture format, the advertisement data is played, and during the playing process, a screenshot is taken according to a preset strategy to obtain an advertisement picture corresponding to the advertisement data.
[0022] Optionally, in the advertiser client mining method provided above, extracting the advertising brand corresponding to the advertising image from the text information on the advertising image includes:
[0023] Segmenting the text information on the advertisement image to obtain a plurality of words;
[0024] Matching each of the words with a brand database to obtain the words belonging to the brand;
[0025] The words belonging to the brand are marked as the advertising brand corresponding to the advertising image.
[0026] Optionally, in the advertiser client mining method provided above, the step of providing feedback on each of the new advertising brands includes:
[0027] Summarize the new advertising brands and merge the repeated new advertising brands;
[0028] Generate an advertising brand information table based on each of the new advertising brands after the merging process;
[0029] The advertising brand information table is pushed to a designated address.
[0030] Optionally, in the above-mentioned advertiser client mining method, after filtering out advertising brands that do not exist in the advertising brand library from the advertising brands corresponding to the respective advertising images to obtain a plurality of new advertising brands, the method further includes:
[0031] Each of the new advertising brands is updated into the advertising brand library.
[0032] A second aspect of the present application provides a device for mining advertiser clients, including:
[0033] A data acquisition unit, used to acquire multiple advertisement data crawled by the crawler;
[0034] A picture acquisition unit, configured to acquire, based on each of the advertisement data, an advertisement picture corresponding to each of the advertisement data;
[0035] A text recognition unit, used for performing optical character recognition on each of the advertisement pictures to obtain text information on each of the advertisement pictures;
[0036] An extraction unit, used to extract the advertising brand corresponding to the advertising picture from the text information on the advertising picture;
[0037] The picture recognition unit is used to recognize the advertisement picture by using a preset recognition model to obtain the advertisement brand corresponding to the advertisement picture when the advertisement brand corresponding to the advertisement picture is not extracted from the text information on the advertisement picture;
[0038] A screening unit, used to screen out advertising brands that do not exist in the advertising brand library from the advertising brands corresponding to the respective advertising images, to obtain a plurality of new advertising brands;
[0039] The feedback unit is used to provide feedback on each of the new advertising brands.
[0040] Optionally, in the advertiser client mining device provided above, the image acquisition unit includes:
[0041] A format conversion unit, used for processing the advertisement data that are not in the video format in each of the advertisement data into a target format file according to a preset rule; wherein the target format file refers to a file in a format that supports playback;
[0042] The first screenshot unit is used to play the advertisement data in video format and the target format file respectively, and take screenshots according to a preset strategy during the playing process to obtain advertisement pictures corresponding to each advertisement data.
[0043] Optionally, in the advertiser client mining device provided above, the image acquisition unit includes:
[0044] A judging unit, configured to judge, for each piece of the advertisement data, whether the advertisement data is in a picture format;
[0045] A picture determination unit, configured to determine the advertisement data as a corresponding advertisement picture if it is determined that the advertisement data is in a picture format;
[0046] The second screenshot unit is used to play the advertisement data if it is determined that the advertisement data does not belong to the picture format, and to take screenshots according to a preset strategy during the playing process to obtain an advertisement picture corresponding to the advertisement data.
[0047] Optionally, in the advertiser client mining device provided above, the extraction unit includes:
[0048] A word segmentation unit, used for segmenting the text information on the advertisement image to obtain a plurality of words;
[0049] A matching unit, used for matching each of the words with a brand database to obtain the words belonging to the brand;
[0050] The marking unit is used to mark the words belonging to the brand as the advertising brand corresponding to the advertising image.
[0051] Optionally, in the advertiser client mining device provided above, the feedback unit includes:
[0052] A merging unit, used to aggregate the new advertising brands and merge the repeated new advertising brands;
[0053] A generating unit, configured to generate an advertising brand information table based on each of the new advertising brands after the merging process;
[0054] The push unit is used to push the advertising brand information table to a designated address.
[0055] Optionally, the advertiser client mining device provided above further includes:
[0056] An updating unit is used to update each of the new advertising brands into the advertising brand library.
[0057] A third aspect of the present application provides an electronic device, including:
[0058] Memory and processor;
[0059] Wherein, the memory is used to store programs;
[0060] The processor is used to execute the program. When the program is executed, it is specifically used to implement the method for mining advertiser clients as described in any one of the above.
[0061] A fourth aspect of the present application provides a computer storage medium for storing a computer program, wherein when the computer program is executed, it is used to implement the method for mining advertiser clients as described in any one of the above.
[0062] The present application provides a method for mining advertiser clients, which can obtain advertising data that relatively meets the needs through multiple advertising data crawled by a crawler. Then, based on each advertising data, the advertising pictures corresponding to each of the advertising data are obtained, so as to perform optical character recognition on each advertising picture respectively, and obtain the text information on each advertising picture. Then, the advertising brand corresponding to the advertising picture is extracted from the text information on the advertising picture. If the advertising brand corresponding to the advertising picture is not extracted from the text information on the advertising picture, the advertising picture is recognized using a preset recognition model to obtain the advertising brand corresponding to the advertising picture, so as to ensure that the advertising brand corresponding to the advertising picture is obtained. Finally, the advertising brands that do not exist in the advertising brand library are screened out from the advertising brands corresponding to each advertising picture, and multiple new advertising brands are obtained, and each new advertising brand is fed back, thereby realizing a method for automatically mining new advertising brands, which can very conveniently mine new advertising brands that meet the requirements. BRIEF DESCRIPTION OF THE DRAWINGS
[0063] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0064] Figure 1 A flowchart of a method for mining advertiser clients provided in an embodiment of the present application;
[0065] Figure 2 A flowchart of a method for obtaining an advertisement picture corresponding to each advertisement data provided by another embodiment of the present application;
[0066] Figure 3 A flowchart of another method for obtaining an advertisement picture corresponding to each advertisement data provided by another embodiment of the present application;
[0067] Figure 4 A flowchart of a method for extracting an advertising brand from text information provided by another embodiment of the present application;
[0068] Figure 5 A flowchart of a method for providing feedback to each new advertising brand provided in another embodiment of the present application;
[0069] Figure 6 A schematic diagram of the structure of a device for mining advertiser clients provided by another embodiment of the present application;
[0070] Figure 7 A schematic diagram of the structure of an electronic device provided in another embodiment of the present application. DETAILED DESCRIPTION
[0071] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0072] In this application, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.
[0073] The present application embodiment provides a method for mining advertiser customers, such as Figure 1 As shown, the following steps are included:
[0074] S101. Acquire multiple advertisement data crawled by a crawler.
[0075] Optionally, a preset crawler program can be started regularly to crawl advertising data that meets the requirements from the Internet. The crawler program can be configured according to the type and source of the required advertising data, so that the advertising data that meets the requirements can be crawled.
[0076] Among them, since the current advertisements are mainly pictures and videos, the types of crawled advertisement data are mainly pictures and videos, and of course, they can also include text and other types of advertisements.
[0077] S102: Based on each piece of advertisement data, obtain an advertisement picture corresponding to each piece of advertisement data.
[0078] In the embodiment of the present application, it is necessary to perform optical character recognition on the image and the text information on the image to extract the advertiser's information from the text information, but some advertising data is not in image format. Therefore, it is necessary to first obtain the advertising image corresponding to each advertising data based on each advertising data.
[0079] Optionally, for picture ads, it can be used directly as the corresponding ad image; for video ads, it can be parsed to obtain each frame of the image, and then the image including the brand information can be extracted from it as the ad image corresponding to the video ad, or screenshots can be used. For text format ads, the format can be directly converted, or other methods can be used to convert it into an image, so as to obtain its corresponding ad image.
[0080] Optionally, in another embodiment of the present application, a specific implementation of step S102 is as follows: Figure 2 As shown, the following steps are included:
[0081] S201: Processing the advertisement data that are not in video format among the advertisement data into target format files according to preset rules.
[0082] The target format file refers to a file in a format that supports playback.
[0083] It should be noted that, considering that the amount of data crawled by the crawler may be relatively large, in the embodiment of the present application, all advertising data are taken by screenshots to obtain their corresponding advertising images, so that the amount of data can be reduced while obtaining the advertising images.
[0084] However, since some formats of advertising data, such as picture ads and text ads, cannot be played directly, it is necessary to process the advertising data that is not in the video format into a target format file according to preset rules.
[0085] Optionally, the image advertisement can be processed into PPT format, or each image advertisement can be used as a video frame to form a video file. For advertisements in text format or other formats, they can be converted into image format first, and then processed together with other image advertisements. It should be noted that if you want to get each image during the screenshot process, when generating the target format file, you can set the time interval between each frame of the image to be equal to the time interval of the screenshot.
[0086] S202: Play the advertisement data in video format and the target format file respectively, and take screenshots according to a preset strategy during the playback process to obtain advertisement pictures corresponding to each advertisement data.
[0087] Optionally, the time interval for taking screenshots may be set according to requirements, so as to obtain advertisement pictures corresponding to each advertisement data.
[0088] Optionally, in another embodiment of the present application, another implementation of step S102 is as follows: Figure 3 As shown, the following steps are included:
[0089] S301: For each piece of advertisement data, determine whether the advertisement data is in a picture format.
[0090] It should be noted that the advertising data in picture format can be directly used as the winning advertising picture, so in the embodiment of the present application, the video advertising data is only screenshoted.
[0091] It should also be noted that in the embodiment of the present application, in addition to advertising data in picture format and advertising data in video format, there may also be advertising data in text and other formats, but this part of the advertising data can be directly converted into a picture through format conversion, so this part of the advertising data can be converted into a picture format before executing step S301.
[0092] Therefore, the advertisement data only stores data in picture format and video format. Therefore, if it is determined that the advertisement data is in picture format, step S302 is executed, that is, the advertisement data is directly determined as its corresponding advertisement picture. If it is determined that the advertisement data is not in picture format, step S303 is executed.
[0093] S302: Determine the advertisement data as its corresponding advertisement picture.
[0094] S303: Play the advertisement data, and take screenshots according to a preset strategy during the playback process to obtain an advertisement image corresponding to the advertisement data.
[0095] S103 , performing optical character recognition on each advertisement picture to obtain text information on each advertisement picture.
[0096] Since there is usually brand information on the advertising pictures, all the words on the advertising pictures are recognized through optical character recognition to obtain the text information on the advertising pictures.
[0097] S104: extracting the advertising brand corresponding to the advertising image from the text information on the advertising image.
[0098] Since an advertisement image usually includes a lot of text information, it is necessary to extract the advertisement brand corresponding to the advertisement image. Optionally, the advertisement brand corresponding to the advertisement image can be matched from the text information by using regular rules or the like, or the advertisement brand corresponding to the advertisement image can be extracted from the text information by using a neural network model or other methods.
[0099] Optionally, in another embodiment of the present application, a specific implementation of step S104 is as follows: Figure 4 As shown, the following steps are included:
[0100] S401: Segment the text information on the advertisement image to obtain a plurality of words.
[0101] Optionally, the text information may be segmented using a word segmentation tool, such as Jieba word segmentation tool, so as to divide the text information into multiple words.
[0102] S402: Match each word with the brand database to obtain words belonging to the brand.
[0103] Specifically, in the embodiment of the present application, a large number of brands on the market are stored in a brand database in advance, so that brands with advertising needs can be matched therefrom, that is, words belonging to the brand can be matched from various words.
[0104] S403: Mark the words belonging to the brand as the advertising brand corresponding to the advertising image.
[0105] S105: Determine whether to extract the advertisement brand corresponding to the advertisement image from the text information on the advertisement image.
[0106] Considering the accuracy of optical character recognition and extracting advertising brands from text information, in the embodiment of the present application, if the advertising brand corresponding to the advertising picture cannot be extracted from the text information on the advertising picture, the model is further used for recognition, that is, if it is determined that the advertising brand corresponding to the advertising picture has not been extracted from the text information on the advertising picture, step S106 is executed. If it is determined that the advertising brand corresponding to the advertising picture has been extracted from the text information on the advertising picture, step S106 does not need to be executed, and step S107 is directly executed.
[0107] S106: Using a preset recognition model to recognize the advertisement image, and obtaining the advertisement brand corresponding to the advertisement image.
[0108] Optionally, the recognition model may be a pre-trained neural network model or other models that can recognize images. By directly inputting the advertisement image into the recognition model, the recognition model recognizes the text therein and determines the advertisement brand therein.
[0109] It should be noted that if the recognition model cannot be used to obtain the advertising brand corresponding to the advertising image, it means that there is a high probability that there is no advertising brand information on the advertising image, and you can choose to remove it at this time.
[0110] S107 , filtering out advertising brands that do not exist in the advertising brand library from the advertising brands corresponding to the various advertising images, to obtain a plurality of new advertising brands.
[0111] It should be noted that the embodiment of the present application is mainly to discover new advertisers, so in the embodiment of the present application, the existing advertising brands are stored in the advertising brand library, so that after obtaining the advertising brand corresponding to each advertising image, new advertising brands can be screened out based on the advertising brand library.
[0112] Accordingly, after obtaining a new advertising brand, in order not to use it as a new advertising brand in the next mining, in the embodiment of the present application, step S107 is executed to obtain multiple new advertising brands, and the following steps may be further executed:
[0113] Update each new advertising brand to the advertising brand library.
[0114] S108. Provide feedback on each new advertising brand.
[0115] Optionally, each new advertising brand can be directly fed back to the advertising business personnel. Of course, it is also possible to further query the relevant information of each new advertising brand and feed back the relevant information to the advertising business personnel.
[0116] Optionally, in another embodiment of the present application, a specific implementation of step S108 is as follows: Figure 5 As shown, including:
[0117] S501. Summarize all new advertising brands and merge duplicate new advertising brands.
[0118] It should be noted that since an advertiser may publish multiple advertisements at the same time, there may be identical advertising brands among the extracted new advertising brands. Therefore, it is necessary to merge the duplicate new advertising brands into one, that is, to eliminate other identical new advertising brands and keep only one.
[0119] S502: Generate an advertising brand information table based on each new advertising brand after the merging process.
[0120] In order to facilitate the use of advertising business personnel, in the embodiment of the present application, the information of each new advertising brand is filled into the table, thereby generating an advertising brand information table.
[0121] S503: Push the advertising brand information table to a designated address.
[0122] It should be noted that the email address of the advertising business personnel or the address of the subsystem used by them can be pre-configured.
[0123] The embodiment of the present application provides a method for mining advertiser clients, which can obtain advertising data that relatively meets the needs through multiple advertising data crawled by a crawler. Then, based on each advertising data, the advertising pictures corresponding to each of the advertising data are obtained, so as to perform optical character recognition on each advertising picture respectively, and obtain the text information on each advertising picture. Then, the advertising brand corresponding to the advertising picture is extracted from the text information on the advertising picture. If the advertising brand corresponding to the advertising picture is not extracted from the text information on the advertising picture, the advertising picture is recognized using a preset recognition model to obtain the advertising brand corresponding to the advertising picture, so as to ensure that the advertising brand corresponding to the advertising picture is obtained. Finally, the advertising brands that do not exist in the advertising brand library are screened out from the advertising brands corresponding to each advertising picture, and multiple new advertising brands are obtained, and each new advertising brand is fed back, thereby realizing a method for automatically mining new advertising brands, which can very conveniently mine new advertising brands that meet the requirements.
[0124] The embodiment of the present application provides a device for mining advertiser customers, such as Figure 6 As shown, it includes the following units:
[0125] The data acquisition unit 601 is used to acquire multiple advertisement data crawled by the crawler.
[0126] The picture acquisition unit 602 is used to acquire the advertisement picture corresponding to each advertisement data based on each advertisement data.
[0127] The text recognition unit 603 is used to perform optical character recognition on each advertisement picture to obtain text information on each advertisement picture.
[0128] The extraction unit 604 is used to extract the advertisement brand corresponding to the advertisement image from the text information on the advertisement image.
[0129] The picture recognition unit 605 is used to recognize the advertisement picture using a preset recognition model to obtain the advertisement brand corresponding to the advertisement picture when the advertisement brand corresponding to the advertisement picture is not extracted from the text information on the advertisement picture.
[0130] The screening unit 606 is used to screen out the advertising brands that do not exist in the advertising brand library from the advertising brands corresponding to the various advertising images, and obtain a plurality of new advertising brands.
[0131] The feedback unit 607 is used to provide feedback on each new advertising brand.
[0132] Optionally, in the advertising client mining device provided in another embodiment of the present application, the image acquisition unit includes:
[0133] The format conversion unit is used to process the advertisement data that is not in the video format among the advertisement data into a target format file according to a preset rule, wherein the target format file refers to a file in a format that supports playback.
[0134] The first screenshot unit is used to play the advertisement data in video format and the target format file respectively, and take screenshots according to a preset strategy during the playing process to obtain advertisement pictures corresponding to each advertisement data.
[0135] Optionally, in the advertising client mining device provided in another embodiment of the present application, the image acquisition unit includes:
[0136] The judging unit is used to judge, for each piece of advertisement data, whether the advertisement data is in a picture format.
[0137] The picture determination unit is used to determine the advertisement data as the corresponding advertisement picture if it is determined that the advertisement data is in picture format.
[0138] The second screenshot unit is used to play the advertisement data if it is determined that the advertisement data does not belong to the picture format, and to take screenshots according to a preset strategy during the playing process to obtain an advertisement picture corresponding to the advertisement data.
[0139] Optionally, in the device for mining advertiser clients provided in another embodiment of the present application, the extracting unit includes:
[0140] The word segmentation unit is used to segment the text information on the advertisement image to obtain multiple words.
[0141] The matching unit is used to match each word with the brand database to obtain the words belonging to the brand.
[0142] The marking unit is used to mark the words belonging to the brand as the advertising brand corresponding to the advertising image.
[0143] Optionally, in the device for mining advertiser clients provided in another embodiment of the present application, the feedback unit includes:
[0144] The merging unit is used to aggregate various new advertising brands and merge repeated new advertising brands.
[0145] The generating unit is used to generate an advertising brand information table based on each new advertising brand after the merging process.
[0146] The push unit is used to push the advertising brand information table to a specified address.
[0147] Optionally, in the device for mining advertiser clients provided in another embodiment of the present application, the device further includes:
[0148] The updating unit is used to update each new advertising brand into the advertising brand library.
[0149] It should be noted that the specific working process of each unit provided in the above embodiments of the present application can refer to the corresponding steps in the above method embodiments, which will not be repeated here.
[0150] Another embodiment of the present application provides an electronic device, such as Figure 7 As shown, including:
[0151] Memory 701 and processor 702 .
[0152] The memory 701 is used to store programs.
[0153] The processor 702 is used to execute the program stored in the memory 701, and when the program is executed, it is specifically used to implement the method for mining advertiser clients provided in any one of the above embodiments.
[0154] Another embodiment of the present application provides a computer storage medium for storing a computer program. When the computer program is executed, it is used to implement the method for mining advertiser clients provided in any of the above embodiments.
[0155] Computer storage media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.
[0156] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0157] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for mining advertiser customers, characterized in that: include: Acquire multiple advertisement data crawled by the crawler, wherein the crawler program is configured according to the type and source of the advertisement data; Based on each of the advertisement data, obtaining an advertisement picture corresponding to each of the advertisement data; For each of the advertisement pictures, optical character recognition is performed on the advertisement pictures to obtain text information on each of the advertisement pictures; Segmenting the text information on the advertisement image to obtain a plurality of words; Matching each of the words with a brand database to obtain the words belonging to the brand; Marking the words belonging to a brand as the advertising brand corresponding to the advertising image; If the advertising brand corresponding to the advertising picture is not extracted from the text information on the advertising picture, the advertising picture is recognized by using a preset recognition model to obtain the advertising brand corresponding to the advertising picture; Filter out advertising brands that do not exist in the advertising brand library from the advertising brands corresponding to the advertising images to obtain multiple new advertising brands; Summarize the new advertising brands and merge the repeated new advertising brands; Generate an advertising brand information table based on each of the new advertising brands after the merging process; The advertising brand information table is pushed to a designated address.
2. The method according to claim 1, characterized in that The acquiring, based on each of the advertisement data, advertisement pictures corresponding to each of the advertisement data, comprises: Processing the advertisement data that are not in video format among the advertisement data into target format files according to preset rules; wherein the target format files refer to files in a format that supports playback; The advertisement data in video format and the target format file are played respectively, and screenshots are taken according to a preset strategy during the playing process to obtain advertisement pictures corresponding to each advertisement data.
3. The method according to claim 1, characterized in that The acquiring, based on each of the advertisement data, advertisement pictures corresponding to each of the advertisement data, comprises: For each piece of the advertisement data, determining whether the advertisement data is in a picture format; If it is determined that the advertisement data is in a picture format, determining the advertisement data as the corresponding advertisement picture; If it is determined that the advertisement data is not in a picture format, the advertisement data is played, and during the playing process, a screenshot is taken according to a preset strategy to obtain an advertisement picture corresponding to the advertisement data.
4. The method according to claim 1, characterized in that: After the step of filtering out advertising brands that do not exist in the advertising brand library from the advertising brands corresponding to the respective advertising images to obtain a plurality of new advertising brands, the step further includes: Each of the new advertising brands is updated into the advertising brand library.
5. A device for mining advertiser clients, characterized in that: include: A data acquisition unit, used to acquire a plurality of advertisement data crawled by the crawler, wherein the crawler program is configured according to the type and source of the advertisement data; A picture acquisition unit, configured to acquire, based on each of the advertisement data, an advertisement picture corresponding to each of the advertisement data; A text recognition unit, used for performing optical character recognition on each of the advertisement pictures to obtain text information on each of the advertisement pictures; An extraction unit, used to extract the advertising brand corresponding to the advertising picture from the text information on the advertising picture; The picture recognition unit is used to recognize the advertisement picture by using a preset recognition model to obtain the advertisement brand corresponding to the advertisement picture when the advertisement brand corresponding to the advertisement picture is not extracted from the text information on the advertisement picture; A screening unit, used to screen out advertising brands that do not exist in the advertising brand library from the advertising brands corresponding to the respective advertising images, to obtain a plurality of new advertising brands; A feedback unit, used for providing feedback on each of the new advertising brands; The extraction unit comprises: A word segmentation unit is used to segment text information on the advertisement image to obtain multiple words; A matching unit, used to match each word with the brand database to obtain words belonging to the brand; A marking unit, used for marking the words belonging to the brand as the advertising brand corresponding to the advertising image; The feedback unit comprises: A merging unit is used to aggregate various new advertising brands and merge duplicate new advertising brands; A generating unit, used for generating an advertising brand information table based on each new advertising brand after the merging process; The push unit is used to push the advertising brand information table to a specified address.
6. The device according to claim 5, characterized in that The picture acquisition unit includes: A format conversion unit, used for processing the advertisement data that are not in the video format in each of the advertisement data into a target format file according to a preset rule; wherein the target format file refers to a file in a format that supports playback; The first screenshot unit is used to play the advertisement data in video format and the target format file respectively, and take screenshots according to a preset strategy during the playing process to obtain advertisement pictures corresponding to each advertisement data.
7. An electronic device, characterized in that: include: Memory and processor; Wherein, the memory is used to store programs; The processor is used to execute the program, and when the program is executed, it is specifically used to implement the method for mining advertiser clients as described in any one of claims 1 to 4.
8. A computer storage medium, characterized in that: Used to store a computer program, which, when executed, is used to implement the method for mining advertiser clients as described in any one of claims 1 to 4.
