CN113570413A - Method and device for generating advertisement keywords, storage medium and electronic equipment - Google Patents
Method and device for generating advertisement keywords, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN113570413A CN113570413A CN202110858976.2A CN202110858976A CN113570413A CN 113570413 A CN113570413 A CN 113570413A CN 202110858976 A CN202110858976 A CN 202110858976A CN 113570413 A CN113570413 A CN 113570413A
- Authority
- CN
- China
- Prior art keywords
- commodity
- information
- keyword
- keywords
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012545 processing Methods 0.000 claims abstract description 23
- 238000013507 mapping Methods 0.000 claims description 27
- 238000003062 neural network model Methods 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 15
- 230000015654 memory Effects 0.000 claims description 14
- 238000007781 pre-processing Methods 0.000 claims description 10
- 238000012163 sequencing technique Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 5
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 235000006040 Prunus persica var persica Nutrition 0.000 description 1
- 240000006413 Prunus persica var. persica Species 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a method and a device for generating advertisement keywords, a storage medium and electronic equipment, wherein the method comprises the following steps: acquiring a keyword set, and searching on a platform to be released through each keyword in the keyword set to obtain first commodity information corresponding to each keyword; acquiring second commodity information corresponding to the target commodity, and calculating according to the first commodity information and the second commodity information to obtain the Euclidean distance between each commodity corresponding to the keyword set and the target commodity; and determining similar commodities corresponding to the target commodity based on the Euclidean distance, and then determining the advertisement keywords of the target commodity based on the keywords matched with the similar commodities. By collecting the information of the target commodity and automatically analyzing and processing the information in the platform to be launched, the advertisement keywords of the target commodity can be automatically generated, and the problems of low efficiency and low accuracy of manually determining the advertisement keywords are solved.
Description
Technical Field
The present invention relates to the field of information processing technologies, and in particular, to a method and an apparatus for generating advertisement keywords, a storage medium, and an electronic device.
Background
The advertisement keyword is a core parameter of an advertisement putting service provided by the e-commerce platform for a merchant. The merchant sets related advertisement keywords and a putting strategy for the commodity, and the e-commerce platform can display the commodity to the customers who search for the keywords according to a certain strategy. In the advertisement putting process, merchants expect to be capable of generating advertisement keywords with strong pertinence, so that customers can obtain corresponding matched commodities when searching through the keywords, and the advertisement putting effect is improved.
At present, a general method for determining advertisement keywords by a merchant is to label keywords of related products manually, but with the increase of types of goods, the workload for obtaining the advertisement keywords is increased, the generation efficiency of the keywords is reduced by a manual labeling mode, and meanwhile, the keywords are only labeled from the perspective of the merchant goods, so that more putting scenes cannot be matched, and the accuracy of the keywords is reduced.
Disclosure of Invention
In order to solve the above problems, the invention provides a method and a device for generating advertisement keywords, which improve the efficiency and accuracy of generating advertisement keywords.
In order to achieve the purpose, the invention provides the following technical scheme:
a method for generating advertisement keywords comprises the following steps:
acquiring a keyword set, wherein the keyword set comprises a plurality of keywords, and each keyword is derived from a search word of a platform to be launched;
searching on the platform to be released through each keyword in the keyword set to obtain first commodity information corresponding to each keyword;
acquiring second commodity information corresponding to the target commodity;
calculating according to the first commodity information and the second commodity information to obtain Euclidean distances between each commodity corresponding to the keyword set and the target commodity;
determining similar commodities corresponding to the target commodity based on the Euclidean distance;
and determining advertisement keywords of the target commodity based on the keywords matched with the similar commodities.
Optionally, the obtaining the first commodity information corresponding to each of the keywords includes:
determining a commodity corresponding to each keyword;
and acquiring commodity detail data corresponding to the commodities, and determining the commodity detail data corresponding to each commodity as first commodity information.
Optionally, the calculating, according to the first commodity information and the second commodity information, an euclidean distance between each commodity corresponding to the keyword set and the target commodity includes:
preprocessing the first commodity information and the second commodity information respectively to obtain first lexical element information corresponding to the first commodity information and second lexical element information corresponding to the second commodity information;
mapping the first lemma information to obtain a first primary semantic code, and mapping the second lemma information to obtain a second primary semantic code;
mapping the first primary semantic code and the second primary semantic code to an abstract semantic space respectively to obtain a first abstract semantic code and a second abstract semantic code;
and calculating the Euclidean distance of the first abstract semantic code and the second abstract semantic code.
Optionally, the mapping the first lemma information to obtain a first primary semantic code, and mapping the second lemma information to obtain a second primary semantic code, includes:
inputting the first lexical information and the second lexical information into a target neural network model respectively, so that the target neural network model outputs a first primary semantic code and a second primary semantic code respectively; the target neural network model is a neural network model which is obtained based on training of a word element sample and can output a mapping relation between a word element and a primary semantic code.
Optionally, the determining, based on the euclidean distance, a similar product corresponding to the target product includes:
determining a candidate commodity set based on the Euclidean distance;
acquiring association information of commodities, and sequencing the commodities in the candidate commodity set according to the association information to obtain a sequencing result;
based on the ranking results, similar goods are determined.
Optionally, the method further comprises:
and determining classification information of the target commodity based on the second commodity information and the similar commodities.
Optionally, the pre-processing comprises: word segmentation processing, stem extraction processing and byte pair encoding processing.
An apparatus for generating advertisement keywords, comprising:
the system comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is used for acquiring a keyword set, the keyword set comprises a plurality of keywords, and each keyword is derived from a search word of a platform to be launched;
the retrieval unit is used for retrieving on the platform to be released through each keyword in the keyword set so as to obtain first commodity information corresponding to each keyword;
the second acquisition unit is used for acquiring second commodity information corresponding to the target commodity;
the calculation unit is used for calculating the Euclidean distance between each commodity corresponding to the keyword set and the target commodity according to the first commodity information and the second commodity information;
the first determining unit is used for determining similar commodities corresponding to the target commodity based on the Euclidean distance;
and the second determining unit is used for determining the advertisement keywords of the target product based on the keywords matched with the similar products.
A storage medium storing executable instructions which, when executed by a processor, implement a method of generating advertising keywords as in any one of the above.
An electronic device, comprising:
a memory for storing a program;
a processor configured to execute the program, where the program is specifically configured to implement the method for generating an advertisement keyword according to any one of the above items.
Compared with the prior art, the invention provides a method and a device for generating advertisement keywords, a storage medium and electronic equipment, wherein the method comprises the following steps: acquiring a keyword set, and searching on a platform to be released through each keyword in the keyword set to obtain first commodity information corresponding to each keyword; acquiring second commodity information corresponding to the target commodity, and calculating according to the first commodity information and the second commodity information to obtain the Euclidean distance between each commodity corresponding to the keyword set and the target commodity; and determining similar commodities corresponding to the target commodity based on the Euclidean distance, and then determining the advertisement keywords of the target commodity based on the keywords matched with the similar commodities. By collecting the information of the target commodity and automatically analyzing and processing the information in the platform to be launched, the advertisement keywords of the target commodity can be automatically generated, and the problems of low efficiency and low accuracy of manually determining the advertisement keywords are solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic flowchart of a method for generating advertisement keywords according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for calculating euclidean distances between each commodity and a target commodity according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a two-dimensional projection of a commodity according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a product recommendation system according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an apparatus for generating advertisement keywords according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first" and "second," and the like in the description and claims of the present invention and the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not set forth for a listed step or element but may include steps or elements not listed.
The embodiment of the invention provides a method for generating advertisement keywords, wherein the advertisement keywords can also be called advertisement putting words and are one of important parameters of a merchant about commodities in a platform to be put, and if the advertisement keywords are put successfully, a user can obtain corresponding matched commodities when searching the advertisement keywords. Specifically, referring to fig. 1, a flow chart of a method for generating an advertisement keyword according to an embodiment of the present invention is shown, where the method may include the following steps:
s101, acquiring a keyword set.
The keyword set comprises a plurality of keywords, and each keyword is derived from a search term of the platform to be launched. The to-be-released platform refers to a platform on which a merchant needs to release advertisement keywords, that is, the merchant needs to output commodities of the merchant through the platform, and specifically, the to-be-released platform generally refers to each e-commerce platform. The keywords in the keyword set refer to all the words that may be used for searching in the e-commerce platform, and may be chinese words, or english words, word combinations, phrases, and the like. For example, the basic word may be "shirt", "laptop", etc., a combination of words such as "shirt baby", etc., a phrase such as "shirt with a logo on the front", etc.
The keywords have flow attributes, different keyword search volumes are different, the flow of common words is larger than that of rare words in general, and the flow of short words is larger than that of long words or phrases. The traffic attributes of the keywords, in turn, have seasonal (variations appear periodic), popularity (sudden fluctuations outside the period). According to the embodiment of the invention, the keyword flow data which is relatively comprehensive and has real-time characteristics can be acquired by integrating the sub-modules for acquiring various keyword flow data and accessing the API, the crawler and the like. The data can provide keyword flow visualization for the merchants and also be a decision basis for the merchants to recommend advertisement putting words.
S102, retrieving on the platform to be released through each keyword in the keyword set to obtain first commodity information corresponding to each keyword.
The final purpose of the advertisement keywords is to assist merchants to put advertisements for own commodities at lower cost, so that the advertisements are adopted by the platform to be put and finally displayed to users searching for the keywords. Therefore, in the embodiment of the present invention, after the keyword of the platform to be delivered is obtained, the keyword is used to perform a search on the platform to be delivered, so as to obtain the corresponding commodity information. Specifically, keywords and commodity result data can be acquired by accessing an API (application programming interface) or a crawler and the like, wherein the simplified process of the crawler can search keywords for a simulation user and record all commodities and detailed information thereof shown by an e-commerce platform to be launched. The first commodity information refers to an information set of all commodities, and includes commodity detail data corresponding to each commodity, such as a title, product parameters and other information. In order to obtain accurate commodity information, in the embodiment of the present invention, a (T +1) -level commodity detail data may be obtained by accessing an API, and correspondingly, the commodity detail data includes main fields: the system comprises a data acquisition unit, a data processing unit and a data processing unit, wherein the data acquisition unit is used for acquiring commodity detail data, and the data acquisition unit is used for acquiring a title, a brand, a file description, category information, a size, a material, evaluation information, related recommended product information, price information and the like, so that the acquired commodity detail data can also provide reference for user selection. The T + 1-level information is adopted to mean that the acquired information lags behind the information of the source for 1 day, so that the acquired information can be more accurate.
And S103, acquiring second commodity information corresponding to the target commodity.
The target commodity is a commodity to be released by a merchant, namely a commodity applying finally obtained advertisement keywords, and the title, detailed description data and the like of the target commodity are mainly collected and determined as second commodity information of the target commodity.
And S104, calculating according to the first commodity information and the second commodity information to obtain the Euclidean distance between each commodity corresponding to the keyword set and the target commodity.
And S105, determining similar commodities corresponding to the target commodity based on the Euclidean distance.
After the first commodity information and the second commodity information are obtained, the commodity information needs to be processed, so that the association relationship among the commodities can be obtained to determine the similar commodity corresponding to the target commodity. In order to facilitate calculation and improve calculation accuracy, the euclidean distance is used as a calculation parameter in the embodiment of the invention to determine the similar commodity corresponding to the target commodity. It should be noted that, in the embodiment of the present invention, besides referring to the euclidean distance, other calculation parameters, such as price parameters, brand comprehensive parameters, and the like, may also be used to perform comprehensive calculation to determine similar products, and a specific processing procedure will be described in detail in the subsequent embodiments of the present application, which is not described in detail herein.
Therefore, similar commodities corresponding to the target commodity are determined, the finally generated advertisement keywords can better accord with a search strategy of a platform to be launched, and then a merchant can launch relatively less advertisement cost to achieve the purpose of popularizing the commodity.
S106, determining advertisement keywords of the target product based on the keywords matched with the similar product.
After the similar product is determined, the keywords of which the similar product can be retrieved can be obtained in the keyword set. And then determines advertisement keywords based on the differences between the similar goods and the target goods or related information.
Because the invention realizes intelligent generation of the advertisement keywords, in order to enable the output advertisement keywords to meet the actual requirements of merchants, the advertisement keywords output in a possible implementation mode can be intelligently sequenced, for example, the obtained advertisement keywords can be sequenced according to cost, flow, search frequency and the like, so that the merchants can directly determine the required advertisement keywords based on the generated result, and further the efficiency of generating the advertisement keywords is improved.
The embodiment of the invention provides a method for generating advertisement keywords, which comprises the following steps: acquiring a keyword set, and searching on a platform to be released through each keyword in the keyword set to obtain first commodity information corresponding to each keyword; acquiring second commodity information corresponding to the target commodity, and calculating according to the first commodity information and the second commodity information to obtain the Euclidean distance between each commodity corresponding to the keyword set and the target commodity; and determining similar commodities corresponding to the target commodity based on the Euclidean distance, and then determining the advertisement keywords of the target commodity based on the keywords matched with the similar commodities. By collecting the information of the target commodity and automatically analyzing and processing the information in the platform to be launched, the advertisement keywords of the target commodity can be automatically generated, and the problems of low efficiency and low accuracy of manually determining the advertisement keywords are solved.
In a possible implementation manner of the embodiment of the present invention, after a corresponding product is obtained by searching based on a keyword in a keyword set, product detail information corresponding to the product is obtained by means of a crawler or the like, so as to obtain first product information. Namely, the obtaining of the first commodity information corresponding to each of the keywords includes: determining a commodity corresponding to each keyword; and acquiring commodity detail data corresponding to the commodities, and determining the commodity detail data corresponding to each commodity as first commodity information. In another embodiment, when a search is performed based on a keyword, a description information field of a product corresponding to the keyword may be directly extracted to obtain corresponding product information.
Referring to fig. 2, a flowchart of a method for calculating euclidean distances between each commodity and a target commodity provided in an embodiment of the present invention is shown, where the method may include:
s201, preprocessing the first commodity information and the second commodity information respectively to obtain first lexical element information corresponding to the first commodity information and second lexical element information corresponding to the second commodity information;
s202, mapping the first lemma information to obtain a first primary semantic code, and mapping the second lemma information to obtain a second primary semantic code;
s203, mapping the first primary semantic code and the second primary semantic code to an abstract semantic space respectively to obtain a first abstract semantic code and a second abstract semantic code;
and S204, calculating the Euclidean distance between the first abstract semantic code and the second abstract semantic code.
In step S201, the commodity information is mainly preprocessed, that is, token of the keyword text is realized, and the adopted technical means mainly includes one or more of word segmentation processing, word stem extraction processing, and BPE (byte pair) encoding processing, and these preprocessing means mainly divide the text into words with fine granularity, such as lemmas, which can be understood as the minimum units of the words. Correspondingly, the method can also be divided into language units such as words, punctuations, numbers, letters and the like, and some interference information and the like which do not influence the final result can be filtered.
In step S202, primary semantic mapping is completed, and the lemma information obtained after preprocessing in step S201 is mapped to a primary semantic space of a fixed dimension by an embedding method to obtain a primary semantic code, which essentially realizes mapping from a word to a vector, that is, it can be understood that text information is converted into a word vector. Inputting the first lexical information and the second lexical information into a target neural network model for processing, so that the target neural network model outputs a first primary semantic code and a second primary semantic code respectively; the target neural network model is a neural network model which is obtained based on training of a word element sample and can output a mapping relation between a word element and a primary semantic code.
In the embodiment of the invention, single or continuous words are input into the neural network model obtained by pre-training, and points or continuous points corresponding to the words in the primary semantic space are output. When the neural network model is trained through the samples, the training is to obtain a mapping relation between a word and a vector, wherein the vector can be understood as a specific point in the primary semantic space.
And then mapping the primary semantic code to an abstract semantic space to obtain an abstract semantic code. The abstract semantic code contains the abstract semantics of the text. Points in the abstract semantic space better keep the semantic relation of the original space, for example, words with far distance are far away in the depth semantic space.
For different commodity information, such as the title and description information of the commodity, the abstract semantic codes can be respectively calculated in the above mode, the Euclidean distance is calculated between the abstract semantic codes, and the closer the distance is to the corresponding commodity, the shorter the distance is, the more the abstract semantic codes are.
In an embodiment of the present invention, the determining, based on the euclidean distance, a similar commodity corresponding to a target commodity includes: determining a candidate commodity set based on the Euclidean distance; acquiring association information of commodities, and sequencing the commodities in the candidate commodity set according to the association information to obtain a sequencing result; based on the ranking results, similar goods are determined.
The related information can be business configurable policy information, such as related price, brand comprehensive information and the like, and the Euclidean distance calculation process adopts a text semantic distance model and sorts similar commodities by calculating text semantic distances between titles and descriptions of the commodities. The text semantic model is a set of deep neural network models which map text depth coding to semantic space.
The embodiment of the invention also comprises the following steps: and determining classification information of the target commodity based on the second commodity information and the similar commodities.
Namely, in the embodiment of the invention, not only the advertisement keywords can be generated, but also the product broad-class recommendation can be carried out on the target product. Because in the e-commerce platform, there may be more than one category selection into which the merchant's products may be divided. For example, products such as "cameras" can be classified either within the home electronics category or the outdoor sports category, and if an inappropriate category is selected, they will face more intense competition in advertising resulting in higher advertising costs. Therefore, the recommended product category is convenient for putting the commodity advertisement in the embodiment of the invention.
Referring to fig. 3, a schematic projection diagram of a two-dimensional space of a commodity according to an embodiment of the present invention is shown. In fig. 3, the titles and details of camera products that have been acquired by the product detail data acquisition module are visualized by projecting the titles and details of the camera products in a depth semantic space into a two-dimensional space, where the light gray dots are of the outdoor sports category (and the description of the outdoor sports category may be more biased toward the stability, endurance, etc. of the products), the dark gray dots are of the electronic product category (the description of the category may be more biased toward science, color, etc.), and the calculation of the distance from the depth semantic code projection of the user product (e.g., the point 42 in fig. 3, which may be classified into the electronic product category by the k-nearest neighbor algorithm) can obtain the most suitable category classification. The commodity distance calculation, the vocabulary distance calculation and other processes are mainly used for primarily screening the keywords to obtain fewer candidate keywords, so that the calculation amount of strategy application is conveniently reduced subsequently, and only the Euclidean distance needs to be calculated by adopting primary semantic coding. Correspondingly, the similar meaning words such as the commodity is red, when pink and peach red can be put into the candidate keyword.
Referring to fig. 4, a schematic diagram of a product recommendation system provided by an embodiment of the invention is shown, where the product recommendation system includes a statistics device, a vocabulary distance calculator, a product distance calculator, a configurable policy, a recommendation broad category, or an advertisement putting word. From fig. 4, it can be derived that the product category recommendation module can obtain the category categories of recommendations and also obtain advertisement keywords after the merchant only inputs the title and detailed description of the product. The calculation of the vocabulary distance and the calculation of the commodity distance are already described in the above embodiments, and are not described herein again. Wherein, the statistics ware will make statistics of the information mainly have: the frequency and the proportion of the keywords searched in the current month, the current week and the current day, and the frequency and the proportion of the keywords and the search results occurring together; similarity between the commodity to be processed and other commodities; the contribution frequency and KL divergence of the key words and the large categories of the commodities.
The embodiment of the present invention further provides a device for generating advertisement keywords, referring to fig. 5, including:
the system comprises a first obtaining unit 10, a second obtaining unit, a third obtaining unit and a fourth obtaining unit, wherein the first obtaining unit is used for obtaining a keyword set, the keyword set comprises a plurality of keywords, and each keyword is derived from a search word of a platform to be launched;
the retrieval unit 20 is configured to retrieve, on the platform to be released, each keyword in the keyword set to obtain first commodity information corresponding to each keyword;
a second obtaining unit 30, configured to obtain second commodity information corresponding to a target commodity;
the calculating unit 40 is configured to calculate, according to the first commodity information and the second commodity information, an euclidean distance between each commodity corresponding to the keyword set and the target commodity;
a first determining unit 50, configured to determine, based on the euclidean distance, a similar product corresponding to the target product;
and a second determining unit 60, configured to determine an advertisement keyword of the target product based on the keyword matched with the similar product.
Further, the retrieval unit includes:
the first determining subunit is used for determining the commodities corresponding to each keyword;
and the second determining subunit is used for acquiring the commodity detail data corresponding to the commodities and determining the commodity detail data corresponding to each commodity as the first commodity information.
Further, the calculation unit includes:
the preprocessing subunit is configured to respectively preprocess the first commodity information and the second commodity information to obtain first lemma information corresponding to the first commodity information and second lemma information corresponding to the second commodity information;
the first mapping subunit is used for mapping the first lemma information to obtain a first primary semantic code, and mapping the second lemma information to obtain a second primary semantic code;
the second mapping subunit is used for mapping the first primary semantic code and the second primary semantic code to an abstract semantic space respectively to obtain a first abstract semantic code and a second abstract semantic code;
and the first calculating subunit is used for calculating the Euclidean distance between the first abstract semantic code and the second abstract semantic code.
Further, the second mapping subunit is specifically configured to:
inputting the first lexical information and the second lexical information into a target neural network model respectively, so that the target neural network model outputs a first primary semantic code and a second primary semantic code respectively; the target neural network model is a neural network model which is obtained based on training of a word element sample and can output a mapping relation between a word element and a primary semantic code.
Further, the first determining unit is specifically configured to:
determining a candidate commodity set based on the Euclidean distance;
acquiring association information of commodities, and sequencing the commodities in the candidate commodity set according to the association information to obtain a sequencing result;
based on the ranking results, similar goods are determined.
Optionally, the apparatus further comprises:
a classification determination unit configured to determine classification information of the target product based on the second product information and the similar product.
Further, the preprocessing in the preprocessing unit includes one or more of word segmentation processing, stem extraction processing, and byte pair encoding processing.
The embodiment of the invention provides a device for generating advertisement keywords, which comprises: acquiring a keyword set, and searching on a platform to be released through each keyword in the keyword set to obtain first commodity information corresponding to each keyword; acquiring second commodity information corresponding to the target commodity, and calculating according to the first commodity information and the second commodity information to obtain the Euclidean distance between each commodity corresponding to the keyword set and the target commodity; and determining similar commodities corresponding to the target commodity based on the Euclidean distance, and then determining the advertisement keywords of the target commodity based on the keywords matched with the similar commodities. By collecting the information of the target commodity and automatically analyzing and processing the information in the platform to be launched, the advertisement keywords of the target commodity can be automatically generated, and the problems of low efficiency and low accuracy of manually determining the advertisement keywords are solved.
Based on the foregoing embodiments, embodiments of the present application provide a storage medium storing executable instructions that, when executed by a processor, implement the method for generating advertisement keywords according to any one of the above. .
An embodiment of the present invention further provides an electronic device, including:
a memory for storing a program;
a processor configured to execute the program, where the program is specifically configured to implement the method for generating an advertisement keyword as described in any one of the above.
The Processor or the CPU may be at least one of an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a Central Processing Unit (CPU), a controller, a microcontroller, and a microprocessor. It is understood that the electronic device implementing the above-mentioned processor function may be other electronic devices, and the embodiments of the present application are not particularly limited.
The computer storage medium/Memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read Only Memory (EPROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a magnetic Random Access Memory (FRAM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical Disc, or a Compact Disc Read-Only Memory (CD-ROM); but may also be various terminals such as mobile phones, computers, tablet devices, personal digital assistants, etc., that include one or any combination of the above-mentioned memories.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all functional units in the embodiments of the present application may be integrated into one processing module, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit. Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media capable of storing program codes, such as a removable Memory device, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, and an optical disk.
The methods disclosed in the several method embodiments provided in the present application may be combined arbitrarily without conflict to obtain new method embodiments.
Features disclosed in several of the product embodiments provided in the present application may be combined in any combination to yield new product embodiments without conflict.
The features disclosed in the several method or apparatus embodiments provided in the present application may be combined arbitrarily, without conflict, to arrive at new method embodiments or apparatus embodiments.
The above description is only an embodiment of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for generating advertisement keywords is characterized by comprising the following steps:
acquiring a keyword set, wherein the keyword set comprises a plurality of keywords, and each keyword is derived from a search word of a platform to be launched;
searching on the platform to be released through each keyword in the keyword set to obtain first commodity information corresponding to each keyword;
acquiring second commodity information corresponding to the target commodity;
calculating to obtain Euclidean distances between each commodity corresponding to the keyword set and the target commodity according to the first commodity information and the second commodity information;
determining similar commodities corresponding to the target commodity based on the Euclidean distance;
and determining advertisement keywords of the target commodity based on the keywords matched with the similar commodities.
2. The method according to claim 1, wherein the obtaining of the first commodity information corresponding to each of the keywords comprises:
determining a commodity corresponding to each keyword;
and acquiring commodity detail data corresponding to the commodities, and determining the commodity detail data corresponding to each commodity as first commodity information.
3. The method according to claim 1, wherein the calculating, according to the first commodity information and the second commodity information, an euclidean distance between each commodity corresponding to the keyword set and the target commodity comprises:
preprocessing the first commodity information and the second commodity information respectively to obtain first lexical element information corresponding to the first commodity information and second lexical element information corresponding to the second commodity information;
mapping the first lemma information to obtain a first primary semantic code, and mapping the second lemma information to obtain a second primary semantic code;
mapping the first primary semantic code and the second primary semantic code to an abstract semantic space respectively to obtain a first abstract semantic code and a second abstract semantic code;
and calculating the Euclidean distance of the first abstract semantic code and the second abstract semantic code.
4. The method of claim 3, wherein mapping the first lemma information to obtain a first primary semantic code and mapping the second lemma information to obtain a second primary semantic code comprises:
inputting the first lexical information and the second lexical information into a target neural network model respectively, so that the target neural network model outputs a first primary semantic code and a second primary semantic code respectively; the target neural network model is a neural network model which is obtained based on training of a word element sample and can output a mapping relation between a word element and a primary semantic code.
5. The method of claim 1, wherein the determining similar products corresponding to the target product based on the Euclidean distance comprises:
determining a candidate commodity set based on the Euclidean distance;
acquiring association information of commodities, and sequencing the commodities in the candidate commodity set according to the association information to obtain a sequencing result;
based on the ranking results, similar goods are determined.
6. The method of claim 1, further comprising:
and determining classification information of the target commodity based on the second commodity information and the similar commodities.
7. The method of claim 3, wherein the pre-processing comprises: word segmentation processing, stem extraction processing and byte pair encoding processing.
8. An apparatus for generating advertisement keywords, comprising:
the system comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is used for acquiring a keyword set, the keyword set comprises a plurality of keywords, and each keyword is derived from a search word of a platform to be launched;
the retrieval unit is used for retrieving on the platform to be released through each keyword in the keyword set so as to obtain first commodity information corresponding to each keyword;
the second acquisition unit is used for acquiring second commodity information corresponding to the target commodity;
the calculation unit is used for calculating and obtaining Euclidean distances between each commodity corresponding to the keyword set and the target commodity according to the first commodity information and the second commodity information;
the first determining unit is used for determining similar commodities corresponding to the target commodity based on the Euclidean distance;
and the second determining unit is used for determining the advertisement keywords of the target product based on the keywords matched with the similar products.
9. A storage medium storing executable instructions which, when executed by a processor, implement a method of generating advertising keywords according to any of claims 1-7.
10. An electronic device, comprising:
a memory for storing a program;
a processor for executing the program, the program being particularly adapted to implement the method of generating advertising keywords according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110858976.2A CN113570413B (en) | 2021-07-28 | 2021-07-28 | Advertisement keyword generation method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110858976.2A CN113570413B (en) | 2021-07-28 | 2021-07-28 | Advertisement keyword generation method and device, storage medium and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113570413A true CN113570413A (en) | 2021-10-29 |
| CN113570413B CN113570413B (en) | 2023-12-05 |
Family
ID=78168575
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110858976.2A Active CN113570413B (en) | 2021-07-28 | 2021-07-28 | Advertisement keyword generation method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113570413B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113724022A (en) * | 2021-11-03 | 2021-11-30 | 北京达佳互联信息技术有限公司 | Keyword determination method and device, computer equipment and medium |
| CN114154064A (en) * | 2021-12-01 | 2022-03-08 | 北京鸥鹭数据科技有限公司 | Commodity keyword optimization method and device |
| CN115082131A (en) * | 2022-07-26 | 2022-09-20 | 北京聚云数字信息技术有限公司 | Method and system for classified evaluation of advertisements based on specific standards |
| CN115470322A (en) * | 2022-10-21 | 2022-12-13 | 深圳市快云科技有限公司 | Keyword generation system and method based on artificial intelligence |
| CN116720004A (en) * | 2023-08-09 | 2023-09-08 | 腾讯科技（深圳）有限公司 | Recommendation reason generation method, device, equipment and storage medium |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050137939A1 (en) * | 2003-12-19 | 2005-06-23 | Palo Alto Research Center Incorporated | Server-based keyword advertisement management |
| JP2010044584A (en) * | 2008-08-12 | 2010-02-25 | Yahoo Japan Corp | Merchandise advertisement distribution device, merchandise advertisement distribution method, and merchandise advertisement distribution control program |
| CN103577432A (en) * | 2012-07-26 | 2014-02-12 | 阿里巴巴集团控股有限公司 | Method and system for searching commodity information |
| CN103870507A (en) * | 2012-12-17 | 2014-06-18 | 阿里巴巴集团控股有限公司 | Method and device of searching based on category |
| WO2014107488A1 (en) * | 2013-01-04 | 2014-07-10 | 24/7 Customer, Inc. | Determining product categories by mining chat transcripts |
| US20150254714A1 (en) * | 2014-03-10 | 2015-09-10 | Yahoo! Inc. | Systems and methods for keyword suggestion |
| JP2016004301A (en) * | 2014-06-13 | 2016-01-12 | ヤフー株式会社 | Advertisement creation device, advertisement creation method and advertisement creation program |
| CN105809458A (en) * | 2014-12-29 | 2016-07-27 | 苏宁云商集团股份有限公司 | Advertisement accurate delivery method and system in e-commerce site |
| CN106204156A (en) * | 2016-07-20 | 2016-12-07 | 天涯社区网络科技股份有限公司 | A kind of advertisement placement method for network forum and device |
| CN107909388A (en) * | 2017-10-30 | 2018-04-13 | 北京奇艺世纪科技有限公司 | A kind of crucial word association extended method and device that word behavior is bought based on advertiser |
| CN107944898A (en) * | 2016-10-13 | 2018-04-20 | 驰众信息技术（上海）有限公司 | The automatic discovery of advertisement putting building information and sort method |
| CN109388742A (en) * | 2017-08-09 | 2019-02-26 | 阿里巴巴集团控股有限公司 | A kind of searching method, search server and search system |
| CN110020880A (en) * | 2018-01-10 | 2019-07-16 | 腾讯科技（深圳）有限公司 | Advertisement placement method, device and equipment |
| CN110309114A (en) * | 2018-02-28 | 2019-10-08 | 腾讯科技（深圳）有限公司 | Processing method, device, storage medium and the electronic device of media information |
| CN111415196A (en) * | 2020-03-18 | 2020-07-14 | 深圳市卡牛科技有限公司 | Advertisement recall method, device, server and storage medium |
| CN112990986A (en) * | 2021-04-29 | 2021-06-18 | 广州宜推网络科技有限公司 | Advertisement service information publishing system based on block chain |
-
2021
- 2021-07-28 CN CN202110858976.2A patent/CN113570413B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050137939A1 (en) * | 2003-12-19 | 2005-06-23 | Palo Alto Research Center Incorporated | Server-based keyword advertisement management |
| JP2010044584A (en) * | 2008-08-12 | 2010-02-25 | Yahoo Japan Corp | Merchandise advertisement distribution device, merchandise advertisement distribution method, and merchandise advertisement distribution control program |
| CN103577432A (en) * | 2012-07-26 | 2014-02-12 | 阿里巴巴集团控股有限公司 | Method and system for searching commodity information |
| CN103870507A (en) * | 2012-12-17 | 2014-06-18 | 阿里巴巴集团控股有限公司 | Method and device of searching based on category |
| WO2014107488A1 (en) * | 2013-01-04 | 2014-07-10 | 24/7 Customer, Inc. | Determining product categories by mining chat transcripts |
| US20150254714A1 (en) * | 2014-03-10 | 2015-09-10 | Yahoo! Inc. | Systems and methods for keyword suggestion |
| JP2016004301A (en) * | 2014-06-13 | 2016-01-12 | ヤフー株式会社 | Advertisement creation device, advertisement creation method and advertisement creation program |
| CN105809458A (en) * | 2014-12-29 | 2016-07-27 | 苏宁云商集团股份有限公司 | Advertisement accurate delivery method and system in e-commerce site |
| CN106204156A (en) * | 2016-07-20 | 2016-12-07 | 天涯社区网络科技股份有限公司 | A kind of advertisement placement method for network forum and device |
| CN107944898A (en) * | 2016-10-13 | 2018-04-20 | 驰众信息技术（上海）有限公司 | The automatic discovery of advertisement putting building information and sort method |
| CN109388742A (en) * | 2017-08-09 | 2019-02-26 | 阿里巴巴集团控股有限公司 | A kind of searching method, search server and search system |
| CN107909388A (en) * | 2017-10-30 | 2018-04-13 | 北京奇艺世纪科技有限公司 | A kind of crucial word association extended method and device that word behavior is bought based on advertiser |
| CN110020880A (en) * | 2018-01-10 | 2019-07-16 | 腾讯科技（深圳）有限公司 | Advertisement placement method, device and equipment |
| CN110309114A (en) * | 2018-02-28 | 2019-10-08 | 腾讯科技（深圳）有限公司 | Processing method, device, storage medium and the electronic device of media information |
| CN111415196A (en) * | 2020-03-18 | 2020-07-14 | 深圳市卡牛科技有限公司 | Advertisement recall method, device, server and storage medium |
| CN112990986A (en) * | 2021-04-29 | 2021-06-18 | 广州宜推网络科技有限公司 | Advertisement service information publishing system based on block chain |
Non-Patent Citations (1)
| Title |
|---|
| 全巧梅;: "农产品电商中精准推荐算法应用研究", 科技广场, no. 09, pages 39 - 42 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113724022A (en) * | 2021-11-03 | 2021-11-30 | 北京达佳互联信息技术有限公司 | Keyword determination method and device, computer equipment and medium |
| CN114154064A (en) * | 2021-12-01 | 2022-03-08 | 北京鸥鹭数据科技有限公司 | Commodity keyword optimization method and device |
| CN115082131A (en) * | 2022-07-26 | 2022-09-20 | 北京聚云数字信息技术有限公司 | Method and system for classified evaluation of advertisements based on specific standards |
| CN115470322A (en) * | 2022-10-21 | 2022-12-13 | 深圳市快云科技有限公司 | Keyword generation system and method based on artificial intelligence |
| CN116720004A (en) * | 2023-08-09 | 2023-09-08 | 腾讯科技（深圳）有限公司 | Recommendation reason generation method, device, equipment and storage medium |
| CN116720004B (en) * | 2023-08-09 | 2023-12-15 | 腾讯科技（深圳）有限公司 | Recommendation reason generation method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113570413B (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113570413B (en) | Advertisement keyword generation method and device, storage medium and electronic equipment | |
| CN106649818B (en) | Application search intention identification method and device, application search method and server | |
| CN106919619B (en) | Commodity clustering method and device and electronic equipment | |
| CN106709040B (en) | Application search method and server | |
| TWI631474B (en) | Method and device for product identification label and method for product navigation | |
| CN107833082B (en) | Commodity picture recommendation method and device | |
| CN110347908B (en) | Voice shopping method, device, medium and electronic equipment | |
| CN110325986A (en) | Article processing method, device, server and storage medium | |
| CN103123633A (en) | Generation method of evaluation parameters and information searching method based on evaluation parameters | |
| CN111400507B (en) | Entity matching method and device | |
| CN104951468A (en) | Data searching and processing method and system | |
| CN112380349A (en) | Commodity gender classification method and device and electronic equipment | |
| CN113254711B (en) | Interactive image display method and device, computer equipment and storage medium | |
| CN111046225A (en) | Audio resource processing method, device, equipment and storage medium | |
| CN106484698A (en) | A kind of method for pushing of search keyword and device | |
| CN110597987A (en) | Search recommendation method and device | |
| Kiran et al. | User specific product recommendation and rating system by performing sentiment analysis on product reviews | |
| CN116894711A (en) | Commodity recommendation reason generation method and device and electronic equipment | |
| CN115147130A (en) | Problem prediction method, apparatus, storage medium, and program product | |
| CN116796027A (en) | Commodity picture label generation method and device, equipment, medium and product thereof | |
| CN113495987A (en) | Data searching method, device, equipment and storage medium | |
| CN111192111A (en) | Product sales data analysis method and terminal equipment | |
| CN111651590A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN115618126A (en) | Search processing method, system, computer readable storage medium and computer device | |
| CN115131108A (en) | E-commerce commodity screening system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |