Advertisement business information mining method and device, electronic equipment and storage medium
By using automated advertising business information mining methods and pre-configured business point sets and similarity models, the problem of inaccurate business information in advertising promotion is solved, achieving efficient and accurate advertising promotion and reducing labor costs.
Patent Information
- Application Number
- CN202111322665.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-11-09
- Publication Date
- 2026-02-06
- Estimated Expiration
- 2041-11-09
AI Technical Summary
In advertising and promotion scenarios, advertisers' failure to provide or providing inaccurate business information leads to the inability of advertisements to be promoted accurately, affecting the service experience. Furthermore, manually editing advertising business information consumes a lot of manpower and time and has a low accuracy rate.
By acquiring the description content of the target advertisement and utilizing a pre-configured set of business points, candidate business points that match the description content are automatically filtered. Combined with attribute comparison and similarity models, the target business point is determined, reducing manual costs and improving accuracy.
It has achieved automated information mining for advertising businesses, reduced labor costs, improved the accuracy and efficiency of advertising promotion, and ensured that advertisements can be promoted efficiently and accurately.
Smart Images
Figure CN114048376B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present disclosure relates to the technical field of artificial intelligence, in particular to the technical field of data mining, and can be applied to the business information mining of advertisements and the like. BACKGROUND
[0002] In the advertisement promotion business scenario, the advertiser does not provide the business information of the advertisement or the provided business information is not accurate enough when putting the advertisement to the advertisement promotion service provider in some cases, which can cause the advertisement to be unable to be accurately promoted and affect the service experience of the advertiser. SUMMARY
[0003] The present disclosure provides a business information mining method and device of an advertisement, an electronic device, and a storage medium.
[0004] According to a first aspect of the present disclosure, a business information mining method of an advertisement is provided, comprising:
[0005] obtaining at least one target description content of a target advertisement;
[0006] selecting at least one first candidate business point matching the semantic of the target description content from a business point set, wherein the business points in the business point set represent the business scope of the advertisement;
[0007] obtaining an attribute comparison result of the attribute information of each first candidate business point and the corresponding target description content;
[0008] determining a target business point set from the at least one first candidate business point according to the attribute comparison result corresponding to each first candidate business point.
[0009] According to a second aspect of the present disclosure, a business information mining device of an advertisement is provided, comprising:
[0010] a description content obtaining module configured to obtain at least one target description content of a target advertisement;
[0011] a candidate business determining module configured to select at least one first candidate business point matching the semantic of the target description content from a business point set, wherein the business points in the business point set represent the business scope of the advertisement;
[0012] an attribute comparison module configured to obtain an attribute comparison result of the attribute information of each first candidate business point and the corresponding target description content;
[0013] a target business determining module configured to determine a target business point set from the at least one first candidate business point according to the attribute comparison result corresponding to each first candidate business point.
[0014] According to a third aspect of the present disclosure, an electronic device is provided, comprising:
[0015] at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above-mentioned business information mining method for advertisement.
[0016] According to a fourth aspect of the present disclosure, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to perform the above-mentioned business information mining method for advertisement.
[0017] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the above-mentioned business information mining method for advertisement.
[0018] It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS
[0019] The accompanying drawings are used to better understand the present scheme, and do not constitute a limitation on the present disclosure. Among them:
[0020] Figure 1 A flowchart of a business information mining method for advertisement provided by an embodiment of the present disclosure is shown;
[0021] Figure 2 A flowchart of another business information mining method for advertisement provided by an embodiment of the present disclosure is shown;
[0022] Figure 3 A schematic diagram of a business information mining device for advertisement provided by an embodiment of the present disclosure is shown;
[0023] Figure 4 A schematic block diagram of an example electronic device that can be used to implement the business information mining method for advertisement provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0024] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to help understanding, and should be considered as merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, in order to be clear and concise, descriptions of well-known functions and structures are omitted in the following description.
[0025] In the advertising promotion business scenario, the advertiser does not provide the business information of the advertisement in some cases or the provided business information is not accurate enough when putting the advertisement to the advertising promotion service party, which can cause the advertisement to be unable to be accurately promoted and affect the service experience of the advertiser. Generally, the advertising promotion service party has a large number of advertisements to promote, and the promoted advertisements involve various industries. If the business information of the advertisement is manually edited by artificial, a large amount of manpower and time cost will be consumed, and the accuracy of the business information of the advertisement cannot be completely guaranteed.
[0026] The business information mining method, device, electronic equipment and storage medium of the advertisement provided by the embodiments of the present disclosure aim to solve at least one of the above technical problems in the prior art.
[0027] Figure 1 A flowchart of a business information mining method of an advertisement provided by an embodiment of the present disclosure is shown, as shown in Figure 1 The method can mainly include the following steps:
[0028] S110: Obtain at least one target description content of a target advertisement.
[0029] In order to facilitate understanding and description, the advertisement to be mined for its business points is defined as a target advertisement, and the related content for describing the target advertisement is defined as target description content in the embodiments of the present disclosure. Specifically, the database of the advertising promotion service party stores the related information of the advertisement, and the target description content of the target advertisement can be retrieved from the database.
[0030] Here, the advertising promotion service party can provide a corresponding application or website, and the user can input search information in the application or website to obtain corresponding advertisement content. For example, the user can input search information in the information search website and the short video application to obtain corresponding advertisement content.
[0031] Optionally, the target description content can be at least one of the keywords and the introduction information of the target advertisement. The keywords are the search words purchased by the advertiser for the target advertisement, and the introduction information includes the title and the abstract of the target advertisement, etc. The embodiments of the present disclosure can obtain the keywords, the title and the abstract of the target advertisement when obtaining at least one target description content of the target advertisement.
[0032] S120: Select at least one first candidate business point matched with the semantics of the target description content from a business point set.
[0033] The database of the advertisement promotion service provider stores at least one type of advertisement-related business point set, which contains business points of advertisements of various industries, and the business points represent the business scope of the advertisements. Taking an advertisement for language training as an example, the related business points can include business points such as “Chinese learning”, “English learning”, and “French learning”. The embodiments of the present disclosure can filter a business point matching the semantics of the target description content from the business point set, and define the filtered business point from the business point set as a first candidate business point.
[0034] As described above, the target description content can be at least one of the keywords and the introduction information of the target advertisement. In this step, at least one first candidate business point matching the semantics of the keywords can be filtered from the business point set, and at least one first candidate business point matching the semantics of the introduction information can also be filtered from the business point set.
[0035] In some cases, the actual business content of two business points with very relevant semantics can be different, for example, the semantics of the two business points “English training” and “English training franchise” are very similar, but the described business scope is different, so it is still necessary to further filter the accurate target business point from the first candidate business point. The specific filtering method can be referred to steps S130 and S140.
[0036] S130: Obtain an attribute comparison result of each first candidate business point and attribute information corresponding to the target description content.
[0037] In the embodiments of the present disclosure, the attribute information includes information of at least one dimension of brand, intent, region, and term. It can be understood that each target description content can have a corresponding first candidate business point. Step S130 can obtain information of at least one dimension of brand, intent, region, and term of the first candidate business point and the corresponding target description content, compare the information of each dimension of the two, and obtain the corresponding attribute comparison result. It can be understood that the attribute comparison result can indicate that the attributes of the first candidate business point and the corresponding target description content are the same or different.
[0038] S140: Determine a target business point set from the at least one first candidate business point according to the attribute comparison result corresponding to each first candidate business point.
[0039] It can be understood that for each first candidate business point, when the attribute comparison result indicates that the attributes of the first candidate business point and the corresponding target description content are the same, the first candidate business point can be determined as the target business point. Step S140 can group all the determined first candidate business points to form a target business point set of the target advertisement.
[0040] The business information mining method for advertisements provided by the embodiments of the present disclosure can pre-configure a business point set including business points of advertisements of various industries, can automatically determine the actual business point of an advertisement from the business point set based on the description content of the advertisement, can reduce the labor cost and time cost consumed for business mining of the advertisement, can improve the accuracy of the business information of the mined advertisement, and can also help to efficiently and accurately promote the advertisement based on the actual business point.
[0041] Optionally, when the target business point set is determined from the at least one first candidate business point according to the attribute comparison result corresponding to each first candidate business point, the at least one second candidate business point with the same attribute as the target description content can be determined from the at least one first candidate business point according to the attribute comparison result corresponding to each first candidate business point. After the second candidate business point is determined, at least one reference description content of a reference advertisement related to the target advertisement is obtained. Based on the reference description content, the target business point set is determined from the at least one second candidate business point.
[0042] In the embodiments of the present disclosure, the other advertisements of the advertiser of the target advertisement can be put to the advertisement promotion service party as the reference advertisements. It can be understood that the business scope of the target advertisement and the reference advertisement is similar, and therefore the reference description content of the reference advertisement can be used to assist in screening the target business point of the target advertisement.
[0043] Optionally, when the target business point set is determined from the at least one second candidate business point based on the reference description content, the similarity between each second candidate business point and each reference description content can be calculated. Optionally, the reference description content can be at least one of the keywords and the introduction information of the reference advertisement, wherein the keywords are the search words purchased by the advertiser for the reference advertisement, and the introduction information includes the title and the abstract of the reference advertisement. The embodiments of the present disclosure can calculate the similarity between the second candidate business point and the search words and the introduction information of each reference advertisement. After obtaining at least one similarity corresponding to each second candidate business point, each similarity corresponding to each second candidate business point can be input into a pre-trained similarity model to output a fusion similarity corresponding to the second candidate business point. Based on the fusion similarity corresponding to the second candidate business point, the target business point set is determined from the at least one second candidate business point.
[0044] Optionally, in the case that the target service point set is determined from the at least one second candidate service point based on the fusion similarity corresponding to the second candidate service point, the target service point set can be determined from the at least one third candidate service point based on the fusion similarity corresponding to the third candidate service point. After the third candidate service point is determined, the intent information of each third candidate service point can be identified. The target service point set can be determined from the at least one third candidate service point based on the intent information corresponding to each third candidate service point. Specifically, the target intent information can be determined from the intent information corresponding to all third candidate service points, and the intent information with the largest proportion can be determined as the target intent information. The third candidate service point with the corresponding intent information being the target intent information can be determined as the target service point set.
[0045] Optionally, after the target service point set is determined, the target service point set can be removed and / or supplemented based on the service point blacklist and the service point whitelist. Specifically, the first service point corresponding to the target advertisement can be determined from the service point blacklist, and the first service point can be removed from the target service point set. The second service point corresponding to the target advertisement can be determined from the service point whitelist, and the second service point can be added to the target service point set.
[0046] Figure 2 A flowchart of another method for mining service information of an advertisement is shown, as shown in FIG. 6, the method can mainly include the following steps: Figure 2
[0047] S201: Obtain the keywords and introduction information of a target advertisement.
[0048] As described above, in order to facilitate understanding and description, the advertisement to be mined for its service point is defined as the target advertisement in the embodiments of the present disclosure. Specifically, the keywords and introduction information of the advertisement are stored in the database of the advertisement promotion service provider, and the keywords and introduction information of the target advertisement can be retrieved from the database. The keywords are the search words purchased by the advertiser for the target advertisement, and the introduction information includes the title and abstract of the target advertisement. In the embodiments of the present disclosure, at least one target description content of the target advertisement can be obtained, including the keywords, title and abstract of the target advertisement, and the service point of the target advertisement can be determined based on the information in multiple dimensions, thereby ensuring the accuracy of the service point result to a greater extent. It should be noted that after step S201, at least one of steps S202 and S203 is performed.
[0049] S202: Screen at least one first candidate service point matching the semantics of the keywords from the service point set.
[0050] In the embodiments of the present disclosure, the service points in the service point set and the keywords can be matched in various manners, so as to determine the corresponding first candidate service points.
[0051] Optionally, the embodiments of the present disclosure can perform literal matching on the keywords and the service points, and if the keywords contain the service points, the service points can be determined as the first candidate service points matched with the keywords.
[0052] Optionally, the embodiments of the present disclosure can obtain the cut words of the keywords and / or the service points, and perform literal matching on the keywords and the service points based on the obtained cut words. Specifically, if the keywords and the service points have the same cut word with the highest weight, and the core terms of the keywords are contained by the core terms of the service points and the title terms of the service points, or all the core terms of the service points are contained by the core terms of the keywords and the title terms of the keywords, the service points are determined as the first candidate service points matched with the keywords.
[0053] Optionally, the embodiments of the present disclosure can determine the vectors of the keywords and the service points, and determine the first candidate service points matched with the semantics of the keywords by comparing the vectors of the keywords and the service points.
[0054] Optionally, the embodiments of the present disclosure can expand the keywords to obtain the expanded words of the keywords. The expanded words and the service points are subjected to literal matching, and if the expanded words contain the service points, the service points are determined as the first candidate service points matched with the keywords.
[0055] S203: Selecting at least one first candidate service point matched with the semantics of the introduction information from the service point set.
[0056] In the embodiments of the present disclosure, the service points in the service point set and the introduction information can be matched in various manners, so as to determine the corresponding first candidate service points.
[0057] Optionally, the embodiments of the present disclosure can perform literal matching on the introduction information and the service points, and if the introduction information contains the service points, the service points can be determined as the first candidate service points matched with the introduction information.
[0058] Optionally, the embodiments of the present disclosure can obtain the cut words of the introduction information and / or the service points, and perform literal matching on the introduction information and the service points based on the obtained cut words. Specifically, if the introduction information and the service points have the same cut word with the highest weight, and the core terms of the introduction information are contained by the core terms of the service points and the title terms of the service points, or all the core terms of the service points are contained by the core terms of the introduction information and the title terms of the introduction information, the service points are determined as the first candidate service points matched with the introduction information.
[0059] Optionally, the embodiment of the present disclosure can determine the profile information and the vector of the business points, and determine the first candidate business point matching the semantics of the profile information by comparing the profile information and the vector of the business points.
[0060] Optionally, the embodiment of the present disclosure can pre-train a text similarity calculation model (such as a model based on CNN-DSSM algorithm), calculate the similarity between the profile information and the business points by the text similarity calculation model, and determine the first candidate business point matching the semantics of the profile information based on the similarity.
[0061] In some cases, the actual business content of two business points with very relevant semantics can be different, for example, the semantics of the two business points of "English training" and "English training franchise" are very similar, but the described business scope is different, so it is still necessary to further screen the accurate target business point from the first candidate business point, and the specific screening method can be referred to steps S204 and S212.
[0062] S204: Obtain attribute comparison results of each first candidate business point and corresponding target description content attribute information.
[0063] In the embodiment of the present disclosure, the attribute information includes information of at least one dimension of brand, intention, region, and term. It can be understood that each target description content can have a corresponding first candidate business point, and step S204 can obtain information of at least one dimension of brand, intention, region, and term of the first candidate business point and the corresponding target description content, and compare the information of each dimension of the two to obtain the corresponding attribute comparison results. It can be understood that the attribute comparison results can indicate that the attributes of the first candidate business point and the corresponding target description content are the same or different.
[0064] S205: According to the attribute comparison results of each first candidate business point, at least one second candidate business point with the same attribute as the target description content is determined from at least one first candidate business point.
[0065] It can be understood that for each first candidate business point, when the attribute comparison result indicates that the attribute of the first candidate business point and the corresponding target description content is the same, the first candidate business point can be determined as the second candidate business point.
[0066] S206: Obtain at least one reference description content of a reference advertisement related to the target advertisement.
[0067] In the embodiments of the present disclosure, the advertiser of the target advertisement can put other advertisements of the advertisement promotion service party as reference advertisements. It can be understood that the business scope of the target advertisement is similar to that of the reference advertisement, and therefore the target business point of the target advertisement can be assisted in screening based on the reference description content of the reference advertisement. Specifically, the database of the advertisement promotion service party stores related information of the advertisement, and the reference description content of the reference advertisement can be retrieved from the database.
[0068] Optionally, the reference description content can be at least one of a keyword and an introduction information of the reference advertisement. The keyword is a search word purchased by the advertiser for the reference advertisement, and the introduction information includes a title and an abstract of the reference advertisement. In the embodiments of the present disclosure, when at least one target description content of the reference advertisement is obtained, the keyword, the title and the abstract of the reference advertisement can be obtained.
[0069] S207: Calculate the similarity between each second candidate business point and each reference description content.
[0070] The reference description content includes a keyword, a title and an abstract. In the embodiments of the present disclosure, the keyword, the title, the abstract and the second candidate business point can be respectively converted into vectors by an encoder, and the vectors can be pooled. Based on the corresponding vectors, the cosine distance between the second candidate business point and each keyword, title and abstract is calculated respectively, and the cosine distance is taken as the similarity. In addition, the LD distance between the second candidate business point and each keyword can also be calculated based on the corresponding vectors, and the LD distance is taken as the similarity. It can be understood that for each second candidate business point, step S207 can obtain a plurality of similarities corresponding to the second candidate business point.
[0071] S208: Input each similarity corresponding to each second candidate business point into a pre-trained similarity model, and output a fusion similarity corresponding to the second candidate business point.
[0072] It can be understood that for each second candidate business point, the second candidate business point and the plurality of keywords of the reference advertisement respectively have corresponding similarities, the second candidate business point and the plurality of titles of the reference advertisement respectively have corresponding similarities, and the second candidate business point and the plurality of abstracts of the reference advertisement respectively have corresponding similarities. The corresponding similarities of the second candidate business point are input into the similarity model as features, and the fusion similarity corresponding to the second candidate business point is output.
[0073] S209: Based on the fusion similarity corresponding to the second candidate business point, at least one third candidate business point is determined from the at least one second candidate business point.
[0074] Optionally, after determining the fusion similarity corresponding to each second candidate business point, the second candidate business points can be sorted in descending order of the fusion similarity, and the second candidate business points ranked in the top N positions are determined as third candidate business points.
[0075] Optionally, after determining the fusion similarity corresponding to each second candidate business point, the second candidate business points with a fusion similarity greater than a preset similarity threshold value are determined as third candidate business points, where the similarity threshold value can be determined according to actual design needs.
[0076] S210: Identify the intent information of each third candidate business point.
[0077] Optionally, the embodiments of the present disclosure can identify the intent information of each third candidate business point through an intent recognition tool. Here, intent recognition refers to recognizing the theme and intent implied in the text, and is a natural language understanding task biased towards the application layer. The intent recognition tool can be a pre-trained intent recognition model. For example, the intent recognition model can be a model including multiple classifiers, and the present disclosure does not make specific limitations on the specific form of the intent recognition model.
[0078] S211: Determine the intent information with the largest proportion from the intent information corresponding to all third candidate business points, and determine the intent information with the largest proportion as target intent information.
[0079] In the embodiments of the present disclosure, the intent information corresponding to each third candidate business point can be the same or different. The embodiments of the present disclosure can statistically analyze the intent information corresponding to all third candidate business points, determine the proportion of the same intent information, and determine the intent information with the largest proportion as target intent information.
[0080] S212: Determine the third candidate business point with the corresponding intent information being the target intent information as a target business point set.
[0081] In the embodiments of the present disclosure, when the corresponding intent information of a third candidate business point is the target intent information, the third candidate business point can be determined as a target business point. Step S212 can group all the determined target business points into a target business point set of a target advertisement.
[0082] Optionally, after determining the target business point set, the embodiment of the present disclosure can further eliminate and / or supplement the target business point set based on a business point blacklist and a business point whitelist. It can be understood that the business point blacklist and the business point whitelist each contain at least one business point. In this way, it can be avoided that the target business point determined is inaccurate or a business point is missed due to insufficient accuracy of the algorithm. For the convenience of understanding and description, the embodiment of the present disclosure defines the business point in the business point blacklist as a first business point, and defines the business point in the business point whitelist as a second business point.
[0083] The embodiment of the present disclosure can determine the first business point corresponding to the target advertisement in the business point blacklist, and eliminate the first business point from the target business point set. Specifically, when one target business point of the target business point set is the same as one first business point in the business point blacklist, the target business point can be eliminated from the target business point set.
[0084] The embodiment of the present disclosure can determine the second business point corresponding to the target advertisement in the business point whitelist, and add the second business point to the target business point set. Specifically, when the second business point corresponding to the target advertisement exists in the business point whitelist, and the target business point set contains the second business point, the second business point can be supplemented to the target business point set as a new target business point.
[0085] Based on the same principle as the above-mentioned business information mining method of the advertisement, Figure 3 A schematic diagram of an advertisement business information mining device provided by the embodiment of the present disclosure is shown. As shown in Figure 3 The advertisement business information mining device 300 includes a description content acquisition module 310, a candidate business determination module 320, an attribute comparison module 330, and a target business determination module 340.
[0086] The description content acquisition module 310 is configured to acquire at least one target description content of a target advertisement.
[0087] The candidate business determination module 320 is configured to screen at least one first candidate business point matching the semantics of the target description content from a business point set, wherein the business points in the business point set represent the business scope of the advertisement.
[0088] The attribute comparison module 330 is configured to acquire an attribute comparison result of the attribute information of each first candidate business point and the corresponding target description content.
[0089] The target business determination module 340 is configured to determine a target business point set from the at least one first candidate business point according to the attribute comparison result corresponding to each first candidate business point.
[0090] The business information mining device for advertisements provided by the embodiments of the present disclosure is pre-configured with a business point set including business points of advertisements of multiple industries, can automatically determine the actual business point of an advertisement from the business point set based on the description content of the advertisement, can reduce the labor cost and time cost consumed for business mining of the advertisement, improve the accuracy of the mined business information of the advertisement, and is also helpful for efficiently and accurately promoting the advertisement based on the actual business point.
[0091] In the embodiments of the present disclosure, the description content acquisition module 310, when used for acquiring at least one target description content of a target advertisement, is specifically used for:
[0092] acquiring at least one of the keywords and the introduction information of the target advertisement;
[0093] The keywords are search words purchased by an advertiser for the target advertisement, and the introduction information includes at least one of the title and the abstract of the target advertisement.
[0094] In the embodiments of the present disclosure, the candidate business determination module 320, when used for screening at least one first candidate business point matched with the semantics of the target description content from the business point set, is specifically used for:
[0095] screening at least one first candidate business point matched with the semantics of the keywords from the business point set, and / or screening at least one first candidate business point matched with the semantics of the introduction information from the business point set.
[0096] In the embodiments of the present disclosure, the attribute information includes information of at least one dimension of brand, intention, region, and term.
[0097] In the embodiments of the present disclosure, the target business determination module 340, when used for determining a target business point set from at least one first candidate business point according to the attribute comparison result corresponding to each first candidate business point, is specifically used for:
[0098] determining at least one second candidate business point identical with the attribute of the target description content from the at least one first candidate business point according to the attribute comparison result corresponding to each first candidate business point;
[0099] acquiring at least one reference description content of a reference advertisement related to the target advertisement;
[0100] determining the target business point set from the at least one second candidate business point based on the reference description content.
[0101] In the embodiments of the present disclosure, the target business determination module 340, when used for determining the target business point set from the at least one second candidate business point based on the reference description content, is specifically used for:
[0102] calculate a similarity between each second candidate business point and each reference description content;
[0103] input each similarity corresponding to each second candidate business point into a pre-trained similarity model, and output a fusion similarity corresponding to each second candidate business point;
[0104] determine a target business point set from at least one second candidate business point based on the fusion similarity corresponding to each second candidate business point.
[0105] In the embodiments of the present disclosure, the target business determination module 340, when used to determine a target business point set from at least one second candidate business point based on the fusion similarity corresponding to each second candidate business point, is specifically used for:
[0106] determine at least one third candidate business point from at least one second candidate business point based on the fusion similarity corresponding to each second candidate business point;
[0107] identify the intent information of each third candidate business point;
[0108] determine a target business point set from at least one third candidate business point based on the intent information corresponding to each third candidate business point.
[0109] In the embodiments of the present disclosure, the target business determination module 340, when used to determine a target business point set from at least one third candidate business point based on the intent information corresponding to each third candidate business point, is specifically used for:
[0110] determine the intent information with the largest proportion from the intent information corresponding to all third candidate business points, and determine the intent information with the largest proportion as target intent information;
[0111] determine the third candidate business point with the corresponding intent information being the target intent information as the target business point set.
[0112] In the embodiments of the present disclosure, the target business determination module 340 is further used to perform at least one of the following:
[0113] determine a first business point corresponding to the target advertisement in the business point blacklist, and remove the first business point from the target business point set;
[0114] determine a second business point corresponding to the target advertisement in the business point whitelist, and add the second business point to the target business point set.
[0115] It can be understood that the above modules of the advertisement business information mining device in the embodiments of the present disclosure have the functions of realizing the corresponding steps of the advertisement business information mining method described above. The functions can be realized by hardware, or realized by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions. The above modules can be software and / or hardware, and the above modules can be realized individually or realized in an integrated manner. The function description of each module of the advertisement business information mining device can be referred to the corresponding description of the advertisement business information mining method described above, and will not be repeated here.
[0116] In the technical solutions of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solutions comply with relevant laws and regulations and do not violate public order and good customs.
[0117] According to the embodiments of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium and a computer program product.
[0118] Figure 4 A schematic block diagram of an example electronic device that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the present disclosure described and / or claimed in this document.
[0119] As Figure 4 shown, the device 400 includes a computing unit 401 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other through a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.
[0120] A plurality of components in the device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a magnetic disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the device 400 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.
[0121] The computing unit 401 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 401 performs various methods and processes described above, such as the business information mining method for advertisements. For example, in some embodiments, the business information mining method for advertisements can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded to the RAM 403 and executed by the computing unit 401, one or more steps of the business information mining method for advertisements described above can be performed. Alternatively, in other embodiments, the computing unit 401 can be configured to perform the business information mining method for advertisements by any other appropriate means, such as by means of firmware.
[0122] Various implementations of the systems and techniques described above herein can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.
[0123] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a standalone software package, or entirely on a remote machine or server.
[0124] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0125] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.
[0126] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.
[0127] The computer system can include clients and servers. This relationship can be. The servers are typically remote from the clients with the interactions between them occurring over a communication network. The relationship between a client and a server is one of client-server. The server can be a cloud server, a server of a distributed system, or a server incorporating a blockchain.
[0128] It should be understood that the various forms of flow shown above can be used with reordering, additions, or removals of steps. For example, each of the steps recited in the present disclosure can be performed in parallel, in series, or in a different order, unless otherwise specifically noted, and is not limited to the order recited in this document.
[0129] The specific embodiments described above are not intended to be limiting, and persons skilled in the art will appreciate that various modifications, combinations, sub-combinations and alternatives can be made to the specific embodiments without departing from the spirit and principles of the disclosure. Any further modifications, equivalents, alternatives, and / or improvements made to the specific embodiments described above are intended to fall within the scope of the disclosure.
Claims
1. A method for mining business information in advertising, comprising: Obtain at least one target description from the target ad; Filter at least one first candidate business point from the set of business points that semantically matches the target description content, wherein the business points in the set of business points represent the business scope of the advertisement; Information on each first candidate business point and its corresponding target description content in at least one dimension, including brand, intent, region, and terminology, is obtained, and the information in each dimension is compared to obtain the attribute comparison result of the attribute information of each first candidate business point and its corresponding target description content. Based on the attribute comparison results corresponding to each first candidate service point, at least one second candidate service point with the same attributes as the target description content is determined from at least one first candidate service point; Obtain at least one reference description of a reference advertisement related to the target advertisement, the reference description including at least one of the keywords and brief information of the reference advertisement; Based on the reference description, a set of target service points is determined from at least one of the second candidate service points.
2. The method according to claim 1, wherein, The acquisition of at least one target description of the target advertisement includes: Obtain at least one of the keywords and description information of the target advertisement; The keywords are the search terms that advertisers purchase for the target advertisement, and the description information includes at least one of the title and summary of the target advertisement.
3. The method according to claim 2, wherein, The step of filtering at least one first candidate business point from the set of business points that semantically matches the target description content includes: Filter at least one first candidate business point from the set of business points that matches the semantics of the keyword, and / or filter at least one first candidate business point from the set of business points that matches the semantics of the introduction information.
4. The method according to claim 1, wherein, The step of determining the target service point set from at least one second candidate service point based on the reference description content includes: Calculate the similarity between each of the second candidate service points and each of the reference description contents; Each similarity corresponding to the second candidate business point is input into a pre-trained similarity model, and the fused similarity corresponding to the second candidate business point is output. Based on the fusion similarity corresponding to the second candidate service point, a set of target service points is determined from at least one of the second candidate service points.
5. The method according to claim 4, wherein, The step of determining a target service point set from at least one of the second candidate service points based on the fusion similarity corresponding to the second candidate service point includes: Based on the fusion similarity corresponding to the second candidate service point, at least one third candidate service point is determined from at least one second candidate service point; Identify the intent information for each of the third candidate service points; Based on the intent information corresponding to each of the third candidate service points, a set of target service points is determined from at least one of the third candidate service points.
6. The method according to claim 5, wherein, The step of determining a target service point set from at least one third candidate service point based on the intent information corresponding to each of the third candidate service points includes: From the intent information corresponding to all the third candidate service points, determine the intent information with the largest proportion, and determine the intent information with the largest proportion as the target intent information; The third candidate service point whose corresponding intent information is the target intent information is determined as the target service point set.
7. The method according to any one of claims 1 to 6, after determining the target service point set from at least one of the first candidate service points based on the attribute comparison results corresponding to each of the first candidate service points, further comprising at least one of the following: Identify the first business point corresponding to the target advertisement from the business point blacklist, and remove the first business point from the target business point set; Identify the second business point corresponding to the target advertisement from the business point whitelist, and add the second business point to the target business point set.
8. A business information mining device for advertising, comprising: The description content acquisition module is used to acquire at least one target description content of the target advertisement; The candidate business determination module is used to filter at least one first candidate business point from the set of business points that semantically matches the target description content, wherein the business points in the set of business points represent the business scope of the advertisement. The attribute comparison module is used to obtain information on at least one dimension of brand, intent, region, and terminology for each first candidate business point and the corresponding target description content, and compare the information on each dimension of the two to obtain the attribute comparison result of the attribute information of each first candidate business point and the corresponding target description content. The target service determination module is used to determine at least one second candidate service point with the same attributes as the target description content from at least one first candidate service point based on the attribute comparison results corresponding to each first candidate service point. Obtain at least one reference description of a reference advertisement related to the target advertisement, the reference description including at least one of the keywords and brief information of the reference advertisement; Based on the reference description, a set of target service points is determined from at least one of the second candidate service points.
9. The apparatus according to claim 8, wherein, When the description content acquisition module is used to acquire at least one target description content of the target advertisement, it is specifically used for: Obtain at least one of the keywords and description information of the target advertisement; The keywords are the search terms that advertisers purchase for the target advertisement, and the description information includes at least one of the title and summary of the target advertisement.
10. The apparatus according to claim 9, wherein, When the candidate service determination module is used to filter at least one first candidate service point from the service point set that semantically matches the target description content, it is specifically used for: Filter at least one first candidate business point from the set of business points that matches the semantics of the keyword, and / or filter at least one first candidate business point from the set of business points that matches the semantics of the introduction information.
11. The apparatus according to claim 8, wherein, When the target service determination module is used to determine a set of target service points from at least one of the second candidate service points based on the reference description content, it is specifically used for: Calculate the similarity between each of the second candidate service points and each of the reference description contents; Each similarity corresponding to the second candidate business point is input into a pre-trained similarity model, and the fused similarity corresponding to the second candidate business point is output. Based on the fusion similarity corresponding to the second candidate service point, a set of target service points is determined from at least one of the second candidate service points.
12. The apparatus according to claim 11, wherein, When the target service determination module determines a set of target service points from at least one of the second candidate service points based on the fusion similarity corresponding to the second candidate service points, it is specifically used for: Based on the fusion similarity corresponding to the second candidate service point, at least one third candidate service point is determined from at least one second candidate service point; Identify the intent information for each of the third candidate service points; Based on the intent information corresponding to each of the third candidate service points, a set of target service points is determined from at least one of the third candidate service points.
13. The apparatus according to claim 12, wherein, When the target service determination module determines a set of target service points from at least one of the third candidate service points based on the intent information corresponding to each of the third candidate service points, it is specifically used for: From the intent information corresponding to all the third candidate service points, determine the intent information with the largest proportion, and determine the intent information with the largest proportion as the target intent information; The third candidate service point whose corresponding intent information is the target intent information is determined as the target service point set.
14. An electronic device comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.
15. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-7.
16. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-7.
Citation Information
Patent Citations
Advertisement classification method capable of automatic recognizing classified advertisement type
CN101097570A