Material delivery method and device, electronic equipment and storage medium

Abstract

The invention provides a material delivery method and device, electronic equipment and a storage medium, and relates to the technical field of computers. The method comprises the steps of determining user side features of a target user to be subjected to material putting; determining a target commodity according to the user side features of the target user; determining a target material according to the user side feature of the target user and the commodity side feature of the target commodity, the target material including information corresponding to at least one of the user side feature and the commodity side feature; and putting the target material to the target user. According to the method and the device, personalized materials can be determined for different users, accurate putting can be realized, the accuracy of material putting is improved, and the content conversion effect and the user experience are ensured.

Description

Background Art

[0002] With the continuous development of computer technology, there are endless solutions for delivering materials in online interactive scenarios, such as personalized advertising, message notifications, etc. However, current delivery solutions for advertising materials often rely on online web pages, videos, and tweets.

[0003] However, for offline scenarios, it is often impossible to deliver personalized materials to customers. Currently, there is an urgent need for a material delivery method for offline scenarios that can determine personalized materials for different users and achieve precise delivery, improve the accuracy of material delivery, and ensure content conversion effects and user experience.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The present disclosure provides a material delivery method, device, electronic device and storage medium, which at least to a certain extent overcome the problem that the related technology cannot determine personalized materials for different users in offline scenarios and achieve accurate delivery, resulting in low accuracy of material delivery, poor content conversion effect and user experience.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a material delivery method is provided, including: determining user-side characteristics of a target user to whom material delivery is to be performed; determining a target product based on the user-side characteristics of the target user; determining a target material based on the user-side characteristics of the target user and product-side characteristics of the target product, wherein the target material includes information corresponding to at least one of the user-side characteristics and the product-side characteristics; and delivering the target material to the target user.

[0008] In some embodiments, the target material is determined based on the user-side characteristics of the target user and the product-side characteristics of the target product, including: generating a material prompt based on the user-side characteristics of the target user and the product-side characteristics of the target product, the material prompt being used to describe the information included in the target material; matching the material prompt with the material in the current material library; when the match is successful, obtaining the target material from the current material library; when the match fails, generating the target material based on the material prompt.

[0009] In some embodiments, material prompts are generated based on user-side features of the target user and product-side features of the target product, including: determining a target group index for each user-side feature and each product-side feature, respectively, the target group index being used to indicate the target user's degree of attention to the corresponding feature; screening the user-side features and product-side features based on the target group index to obtain at least one target feature; and generating material prompts based on the at least one target feature.

[0010] In some embodiments, after determining the target material, the method provided by the embodiments of the present disclosure also includes: identifying named entities in the target material; performing relationship extraction processing on each named entity to obtain the association relationship between each named entity; determining whether the association relationship between each named entity satisfies the preset logic; when the association relationship between each named entity satisfies the preset logic, storing the target material in the current material library.

[0011] In some embodiments, when the association relationships between various named entities satisfy the preset logic, the target material is stored in the current material library, including: when the association relationships between various named entities satisfy the preset logic, determining the degree of match between the target material and the target user; when the degree of match meets the preset matching conditions, storing the target material in the current material library.

[0012] In some embodiments, a target product is determined based on user-side features of a target user, including: determining product-side features of multiple products to be placed; determining a user feature vector based on the user-side features of the target user; determining corresponding product feature vectors based on the product-side features of the multiple products to be placed; calculating multiple similarities between the user feature vector and the multiple product feature vectors, wherein a similarity is determined between the user feature vector and a product feature vector of a product to be placed, and each similarity is used to indicate the possibility of the corresponding product to be placed being placed to the target user; and screening the multiple products to be placed based on the multiple similarities to obtain the target product.

[0013] In some embodiments, the target material also includes network address indication information, and the network address indication information is used for the target user to view the network details page of the target product. The method also includes: determining the delivery effect data corresponding to each material in the current material library, and the delivery effect data corresponding to any material is used to describe the possibility of viewing the corresponding network details page through the network address indication information in the material; deleting each material whose delivery effect data meets the preset low delivery effect condition; wherein, when the match is successful, obtaining the target material from the current material library, including: if the material prompt successfully matches multiple materials in the current material library, then screening the multiple materials according to the delivery effect data to obtain the target material.

[0014] In some embodiments, determining the delivery effect data corresponding to each material in the current material library includes: determining the explosive coefficient of each material in the current material library, the explosive coefficient is used to describe the possibility that the corresponding material is a explosive material; according to the explosive coefficient, determining the delivery effect data corresponding to each material in the current material library.

[0015] According to another aspect of the present disclosure, a material delivery device is also provided, including: a user-side feature determination module, used to determine the user-side features of a target user to whom material delivery is to be performed; a target product determination module, used to determine a target product based on the user-side features of the target user; a target material determination module, used to determine a target material based on the user-side features of the target user and the product-side features of the target product, wherein the target material includes information corresponding to at least one of the user-side features and the product-side features; and a delivery module, used to deliver the target material to the target user.

[0016] In some embodiments, the target material determination module is used to generate a material prompt based on the user-side characteristics of the target user and the product-side characteristics of the target product, where the material prompt is used to describe the information included in the target material; match the material prompt with the material in the current material library; when the match is successful, obtain the target material from the current material library; when the match fails, generate the target material based on the material prompt.

[0017] In some embodiments, the target material determination module is used to determine the target group index of each user-side feature and each product-side feature respectively, and the target group index is used to indicate the target user's attention level to the corresponding feature; the user-side features and product-side features are screened according to the target group index to obtain at least one target feature; and a material prompt is generated according to at least one target feature.

[0018] In some embodiments, the material delivery device provided by the embodiments of the present disclosure also includes: a named entity recognition module, which is used to identify named entities in the target material; perform relationship extraction processing on each named entity to obtain the association relationship between each named entity; determine whether the association relationship between each named entity satisfies the preset logic; and a target material storage module, which is used to store the target material in the current material library when the association relationship between each named entity satisfies the preset logic.

[0019] In some embodiments, a matching degree determination module is used to determine the matching degree between the target material and the target user when the association relationship between each named entity satisfies the preset logic; a target material storage module is used to store the target material in the current material library when the matching degree meets the preset matching conditions.

[0020] In some embodiments, a target product determination module is used to determine product-side features of multiple products to be placed; determine a user feature vector based on the user-side features of the target user; determine corresponding product feature vectors based on the product-side features of the multiple products to be placed; calculate multiple similarities between the user feature vector and the multiple product feature vectors, wherein a similarity is determined between the user feature vector and a product feature vector of a product to be placed, and each similarity is used to indicate the possibility of the corresponding product to be placed being placed to the target user; and screen the multiple products to be placed based on the multiple similarities to obtain the target product.

[0021] In some embodiments, the target material also includes network address indication information, and the network address indication information is used for the target user to view the network details page of the target product. The material delivery device provided by the embodiment of the present disclosure also includes: a delivery effect data module, which is used to determine the delivery effect data corresponding to each material in the current material library, and the delivery effect data corresponding to any material is used to describe the possibility of viewing the corresponding network details page through the network address indication information in the material; a material deletion module, which is used to delete each material whose delivery effect data meets the preset low delivery effect condition; a target material determination module, which is used to screen multiple materials according to the delivery effect data to obtain the target material if the material prompt successfully matches multiple materials in the current material library.

[0022] In some embodiments, the delivery effect data module is used to determine the explosive coefficient of each material in the current material library. The explosive coefficient is used to describe the possibility that the corresponding material is a explosive material; based on the explosive coefficient, the delivery effect data corresponding to each material in the current material library is determined.

[0023] According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned material delivery methods by executing the executable instructions.

[0024] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any one of the above-mentioned material delivery methods is implemented.

[0025] According to another aspect of the present disclosure, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the material delivery method provided in various optional ways in any embodiment of the present disclosure.

[0026] The technical solution provided by the embodiment of the present disclosure can determine the target product based on the user-side characteristics of the target user to be delivered the material, and can determine the personalized target material based on the user-side characteristics of the target user and the product-side characteristics of the target product. Therefore, the present disclosure can determine personalized materials for different users and achieve accurate delivery, improve the accuracy of material delivery, and ensure content conversion effect and user experience.

[0027] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0029] Figure 1 A schematic diagram showing a system architecture in an embodiment of the present disclosure;

[0030] Figure 2 A flow chart of a material delivery method in an embodiment of the present disclosure is shown;

[0031] Figure 3 A schematic diagram showing a similarity calculation using a double tower model in an embodiment of the present disclosure is shown;

[0032] Figure 4 A schematic diagram of a large model calling in an embodiment of the present disclosure is shown;

[0033] Figure 5 A schematic diagram of the architecture of material delivery in an embodiment of the present disclosure is shown;

[0034] Figure 6 A flow chart for determining a target material in an embodiment of the present disclosure is shown;

[0035] Figure 7 A schematic diagram showing a target material delivery in an embodiment of the present disclosure;

[0036] Figure 8 A schematic diagram of a material delivery device in an embodiment of the present disclosure is shown;

[0037] Fig. 9 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown;

[0038] Fig.10 A schematic diagram of a computer-readable storage medium in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0039] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0040] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0041] The specific implementation of the embodiment of the present disclosure is described in detail below with reference to the accompanying drawings.

[0042] Figure 1 FIG. 1 shows an exemplary application system architecture diagram to which the material delivery method in the embodiment of the present disclosure can be applied. Figure 1 As shown, the system architecture may include a terminal device 101 , a network 102 and a server 103 .

[0043] Among them, the terminal device 101 can send the user-side characteristics of the target user to be delivered to the server 103. Then, the server 103 can determine the user-side characteristics of the target user, and determine the target product according to the user-side characteristics of the target user. Then, the server 103 can determine the target material according to the user-side characteristics of the target user and the product-side characteristics of the target product, and the target material includes information corresponding to at least one of the user-side characteristics and the product-side characteristics. Finally, the delivery of the target material to the target user can be achieved.

[0044] Alternatively, the server 103 may send the user-side features of the target user to be delivered to the terminal device 101. Then, the terminal device 101 may determine the user-side features of the target user, and may determine the target product based on the user-side features of the target user. Then, the terminal device 101 may determine the target material based on the user-side features of the target user and the product-side features of the target product, wherein the target material includes information corresponding to at least one of the user-side features and the product-side features. Finally, the delivery of the target material to the target user may be achieved.

[0045] The network 102 is a medium for providing a communication link between the terminal device 101 and the server 103, and can be a wired network or a wireless network.

[0046] Optionally, the wireless network or wired network described above uses standard communication technology and / or protocol. The network is usually the Internet, but it can also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or any combination of a virtual private network). In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Mark-up Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPSec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0047] The terminal device 101 can be various electronic devices, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, smart speakers, smart watches, wearable devices, augmented reality devices, virtual reality devices, etc.

[0048] The server 103 may be a server that provides various services, such as a background management server that provides support for the device operated by the user using the terminal device 101. The background management server may analyze and process the received request and other data, and feed back the processing results to the terminal device.

[0049] Optionally, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0050] Those skilled in the art will know that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration, and any number of terminal devices, networks and servers may be provided according to actual needs, and the embodiments of the present disclosure do not limit this.

[0051] Under the above system architecture, a material delivery method is provided in an embodiment of the present disclosure, and the method can be executed by any electronic device with computing and processing capabilities.

[0052] In some embodiments, the material delivery method provided in the embodiments of the present disclosure can be executed by a terminal device of the above-mentioned system architecture; in other embodiments, the material delivery method provided in the embodiments of the present disclosure can be executed by a server in the above-mentioned system architecture; in other embodiments, the material delivery method provided in the embodiments of the present disclosure can be implemented by the terminal device and the server in the above-mentioned system architecture through interaction.

[0053] Figure 2 A flow chart of a material delivery method in an embodiment of the present disclosure is shown. Figure 2 As shown, the material delivery method provided in the embodiment of the present disclosure includes the following steps S202 to S208.

[0054] S202: Determine user-side characteristics of target users to whom material delivery is to be performed.

[0055] The embodiments of the present application do not limit the materials to be delivered and the target users, nor do they limit the types and quantities of the user-side features of the target users. Optionally, it can be determined according to the application scenario of the material delivery method. For example, when the material delivery method provided by the embodiments of the present disclosure is applied to the e-commerce platform scenario, at this time, the target user can be a user who consumes on the e-commerce platform, and the user-side features can include gender, age, occupation, marital status, city of residence, whether there are children, shopping preference categories, etc. The material to be delivered can be related information such as an advertisement for a certain product on the e-commerce platform.

[0056] For another example, the material delivery method can be applied to an enterprise management scenario, in which case the target user may be an employee of the enterprise, and the user-side characteristics may include gender, age, position, time of employment, etc. The material to be delivered may be personalized information sent to each employee. This disclosure is not limited to this.

[0057] Taking the e-commerce platform scenario as an example, in one possible implementation, the material can be printed on a logistics waybill with the recipient as the target user to achieve accurate delivery of the target material. At this time, the target user can be determined from the e-commerce platform users with logistics packages to be sent. Exemplarily, user-side features can be mined and user portraits can be constructed based on the user's browsing, adding to shopping carts, placing orders, and other behaviors on the e-commerce platform and the corresponding product information. For example, the user's gender can be judged based on the type of clothes that are frequently purchased. For another example, for users who frequently buy baby milk powder, it can be judged that the user has children. In an exemplary embodiment, after determining the target user, the user-side features of the target user can be obtained from the target user's terminal device and / or the cloud of the e-commerce platform.

[0058] S204: Determine the target product according to the user-side characteristics of the target user.

[0059] In some embodiments, the target product is determined based on the user-side characteristics of the target user, including: determining the product-side characteristics of multiple products to be placed; determining a user feature vector based on the user-side characteristics of the target user; determining corresponding product feature vectors based on the product-side characteristics of the multiple products to be placed; calculating multiple similarities between the user feature vector and the multiple product feature vectors, wherein a similarity is determined between the user feature vector and a product feature vector of a product to be placed, and each similarity is used to indicate the possibility of the corresponding product to be placed being placed to the target user; and screening the multiple products to be placed based on the multiple similarities to obtain the target product.

[0060] For example, product-side features may include categories, brands, price ranges, etc. Categories may also include primary categories, secondary categories, tertiary categories, etc. Primary categories include food, digital products, furniture, clothing, etc. Taking food as a primary category, secondary categories may be more detailed categories than food, such as beverages, snacks, etc. Taking beverages as a secondary category, tertiary categories may be more detailed categories than beverages, such as juice, yogurt, etc.

[0061] In the exemplary embodiment, the similarity between the user feature vector of the target user and the product feature vector of the first product is calculated as an example. The first product can be any one of the multiple products to be put on the market. At this time, the user-side features and the product-side features corresponding to the first product can be input into the double-tower model. Thus, the user feature vector corresponding to the target user and the product feature vector corresponding to the first product are obtained. After that, the similarity between the user feature vector corresponding to the target user and the product feature vector corresponding to the first product can be calculated.

[0062] It should be noted that the dual-tower model can be composed of two towers, wherein each tower can be an independent neural network, one tower (hereinafter referred to as the first tower) is used to process user-side features, and the other tower (hereinafter referred to as the second tower) is used to process product-side features. In a possible implementation, the dual-tower model can be a DSSM (Deep Structured Semantic Model). Exemplarily, both the user-side features and the product-side features can be in text form.

[0063] In an exemplary embodiment, after the first tower receives the user-side features, the user-side features can be converted into a low-dimensional vector form, i.e., a user feature vector, by feature embedding and other processing. Since the target user can include multiple user-side features, the first tower can also be used to concatenate multiple vectors corresponding to multiple user-side features to obtain the user feature vector.

[0064] Similarly, after the second tower receives the product-side features, it can convert the product-side features into a low-dimensional vector form, i.e., a product feature vector, through feature embedding and other processing. Among them, since the first product can contain multiple product-side features, the second tower can also be used to splice multiple vectors corresponding to multiple product-side features to obtain the product feature vector. Finally, the similarity can be calculated based on the user feature vector and the product feature vector, and the similarity is proportional to the target user's interest in the first product.

[0065] For example, the disclosed embodiment can calculate the similarity based on the user-side features of the target user and the product-side features corresponding to the multiple products. For example, after determining the similarity between the target user and each product to be placed, the product to be placed corresponding to the highest similarity can be used as the target product. Alternatively, a similarity threshold can be preset first, and any one of the multiple products to be placed whose similarity is greater than the similarity threshold can be designated as the target product.

[0066] In another possible implementation, after obtaining the user-side characteristics of the target user, the product category that the target user is interested in can be determined based on the user-side characteristics of the target user. Then, the categories of the current products to be placed are obtained, and if the category corresponding to any product to be placed is the same as the product category that the target user is interested in, the product to be placed can be used as the target product.

[0067] For example, if the categories corresponding to the multiple commodities to be placed are the same as the commodity categories that the target user is interested in, any one of them can be selected as the target commodity. Alternatively, in this case, the target commodity can be designated by the backend management personnel.

[0068] S206: Determine a target material according to the user-side feature of the target user and the product-side feature of the target product, wherein the target material includes information corresponding to at least one of the user-side feature and the product-side feature.

[0069] In some embodiments, the target material is determined based on the user-side characteristics of the target user and the product-side characteristics of the target product, including: generating a material prompt based on the user-side characteristics of the target user and the product-side characteristics of the target product, the material prompt being used to describe the information included in the target material; matching the material prompt with the material in the current material library; when the match is successful, obtaining the target material from the current material library; when the match fails, generating the target material based on the material prompt.

[0070] In an exemplary embodiment, the material prompt may be composed of a prompt word and a sentence template. For example, the sentence template may be, for example, "Please generate an advertisement copy for B who is interested in A", where A and B parts may be filled with corresponding prompt words to obtain the material prompt. Each prompt word may correspond to a dimension of commodity-side features or a dimension of user-side features. For example, A may be a car charcoal bag, and B may be at least one of having a child, having a car, and being a man.

[0071] Then, the material prompt can be matched with the current material library, and a plurality of materials can be included in the current material library. If the material prompt is successfully matched with any material in the current material library, the information included in the material is consistent with the prompt words in the material prompt. At this moment, the material can be used as the target material. Alternatively, if the material prompt is not matched with each material in the current material library, the material prompt can be input into the text generation interface and / or the picture generation interface of the large model, thereby obtaining corresponding text and / or pictures. It should be noted that the target material can be in the form of text, can be in the form of pictures, or can be in the form of a combination of text and pictures. If text and pictures are generated respectively according to the material prompt, then the text and pictures need to be combined to obtain the target material.

[0072] In a possible implementation, both the user-side features and the product-side features can be input into a large model, which is used to generate corresponding prompt words based on the user-side features and the product-side features, and obtain material prompts based on the prompt words, and finally obtain the corresponding target material.

[0073] It should be noted that, in the process of generating material prompts, if all the features of each dimension of the user-side features and product-side features are input into the big model, the big model may fail to obtain the key information, resulting in poor targeting of the generated target materials. Therefore, the user-side features and product-side features can be screened by dimension.

[0074] In some embodiments, a material prompt is generated based on the user-side characteristics of the target user and the product-side characteristics of the target product, including: determining a target group index for each user-side characteristic and each product-side characteristic, respectively, the target group index being used to indicate the target user's attention level to the corresponding characteristic; screening the user-side characteristics and the product-side characteristics according to the target group index to obtain at least one target feature; and generating the material prompt based on the at least one target feature.

[0075] It should be noted that the TGI (Target Group Index) can reflect the strength or weakness of the target group within a specific research scope. Taking the product-side features as an example, TGI can characterize the differences in the attention paid to various types of products by users with different user-side features. Among them, when TGI is equal to 100, it indicates the average level, and when TGI is higher than 100, it means that the level of attention paid by this type of user to a certain type of product is higher than the overall level. Therefore, the TGI can be calculated for each product-side feature, so as to select at least one product-side feature that meets the target group index conditions as the target feature of the input large model. Similarly, the TGI can be calculated for each user-side feature, so as to select at least one user-side feature that meets the target group index conditions as the target feature of the input large model.

[0076] Exemplarily, the target group index condition may be set by a preset target group index threshold. When the target group index corresponding to a feature is greater than the preset target group index threshold, the feature satisfies the target group index condition.

[0077] In an exemplary embodiment, the corresponding TGI can be obtained by dividing the proportion of the target group with a certain characteristic by the proportion of the population with the same characteristic, and then multiplying the result by the standard number 100.

[0078] In a possible implementation, taking the scenario of advertising on an e-commerce platform as an example, the above-mentioned target group refers to the users targeted by a certain category of goods, and the overall refers to the entire user group of the e-commerce platform. Taking the milk powder category as an example, among the users targeted by milk powder products, that is, users who are interested in milk powder products, 50% of the users have "children" in the user-side features, and 20% of the entire user group of the e-commerce platform have children. Therefore, the TGI value corresponding to the user-side feature of "having children" is relatively high, indicating that users with children pay great attention to milk powder products. Therefore, the user-side feature of "having children" can be input into the large model.

[0079] In some embodiments, after determining the target material, the material delivery method provided by the embodiments of the present disclosure may also include: identifying named entities in the target material; performing relationship extraction processing on each named entity to obtain the association relationship between each named entity; determining whether the association relationship between each named entity satisfies the preset logic; when the association relationship between each named entity satisfies the preset logic, storing the target material in the current material library.

[0080] It should be noted that, because the current materials may involve safety and compliance issues. For example, when the material is an advertisement, false propaganda will not only lead to a negative user experience, but may even lead to the risk of complaints. Therefore, the target material can be verified before delivery.

[0081] Exemplarily, for target materials of text type, double verification can be performed based on NER (Named Entity Recognition) technology and relationship extraction technology. Among them, for example, taking the text of "Dad drives, charcoal bag removes odors, and baby feels more at ease", the text can be used as a prompt for named entity recognition, and the named entities may include dad, car, charcoal bag, and baby. Then, the named entities can be processed by relationship extraction to obtain the association relationship between each named entity. In the above example, dad and baby can be a father-son relationship, car and charcoal bag can be a usage relationship, and dad and car can be a driving relationship. Then, it can be checked whether the association relationship between the above named entities meets the preset logic.

[0082] Exemplarily, the preset logic can be used to describe various association relationships that conform to common sense of human logic. The preset logic may include father-child relationship, usage relationship, and driving relationship. Therefore, the above text "Dad drives, charcoal bag removes odor, and baby feels more at ease" can pass double verification.

[0083] In some embodiments, when the association relationships between the various named entities satisfy the preset logic, the target material is stored in the current material library, including: when the association relationships between the various named entities satisfy the preset logic, determining the degree of match between the target material and the target user; when the degree of match meets the preset matching conditions, storing the target material in the current material library.

[0084] Exemplarily, the target material can be input into the big model to determine the degree of match between the target material and the target user. Taking the text "Dad drives, charcoal bag removes odor, baby is more at ease" as an example, the big model can be used to determine whether the text is an advertisement text for the car charcoal bag for the target user "male, with children, with cars". If the result output by the big model is "yes", it can be considered that the degree of match meets the preset matching conditions, and if the result output by the big model is "no", it can be considered that the degree of match does not meet the preset matching conditions.

[0085] In some embodiments, in the material delivery method provided by the embodiments of the present disclosure, after determining that the degree of matching meets the preset matching conditions, the target material can also be manually reviewed. For example, one or more groups of materials that have passed the double verification can be manually batch reviewed and selected to pass or fail. The reviewed materials will be stored in the material library, and no subsequent review is required, supporting rapid delivery and use.

[0086] In some embodiments, the target material also includes network address indication information, which is used for the target user to view the network details page of the target product. The material delivery method provided by the embodiment of the present disclosure may also include: determining the delivery effect data corresponding to each material in the current material library, the delivery effect data corresponding to any material is used to describe the possibility of viewing the corresponding network details page through the network address indication information in the material; deleting each material whose delivery effect data meets the preset low delivery effect condition.

[0087] In this case, when the match is successful, the target material is obtained from the current material library, including: if the material prompt successfully matches multiple materials in the current material library, multiple materials are screened according to the delivery effect data to obtain the target material.

[0088] The disclosed embodiments do not limit the content of the network address indication information. For example, the network address indication information may be a website address corresponding to the network details page or a product QR code, etc. The target user may enter the network details page by inputting the website address corresponding to the network details page into the terminal device, or the target user may scan the product QR code corresponding to the network details page through the terminal device to enter the network details page.

[0089] In an exemplary embodiment, if the system detects that the target user enters the network details page through the network address indication information, this can be recorded to determine the delivery effect data of the target material. Exemplarily, the target material can be delivered at least once, and the delivery effect data can be determined by the number of deliveries and the number of times the network details page is entered through the network address indication information.

[0090] In a possible implementation, the number of placements can be represented by exposure data, and the network address indication information can be a product QR code corresponding to the network details page. Therefore, the number of times the network details page is entered through the network address indication information can be represented by the scanning rate.

[0091] In an exemplary embodiment, any material in the current material library may include multiple fields, and the fields may include at least one of a user group ID (Identity), a commodity type ID, a text ID, and a picture ID. Among them, the user group ID can be used to uniquely identify a type of user group, and the commodity type ID can be used to identify a type of commodity that needs to be advertised. The text ID can be used to uniquely identify the corresponding text content in the material. The picture ID can be used to uniquely identify the corresponding picture content in the material.

[0092] In addition, any material in the current material library may also include a material verification feedback flag (identifier), a scan rate, and a traffic share. Among them, if the material passes the manual review, the material verification feedback flag may be 1; if the material fails the manual review, the material verification feedback flag may be 0. The scan rate is used to indicate the probability that a user scans a product QR code indicating information of a network address. The traffic share can be used to indicate the traffic delivery ratio of the corresponding material. For example, if the material is subsequently found to be a hot product, the traffic delivery volume can be expanded by increasing the value of the traffic share.

[0093] Exemplarily, the table structure corresponding to the current material library may be as shown in Table 1.

[0094] Table 1

[0095]

[0096] In an exemplary embodiment, one material in each category can be designated as a benchmark group, and all other materials can be used as experimental groups. The disclosed embodiment can determine the delivery effect of the materials in each experimental group by using a significance Z test, thereby discovering materials with high click-through rates in real time. The significance Z test can be calculated based on the following formula (1).

[0097]

[0098] In formula (1), Z can be used to represent the Z value of the significance Z test. The exposure data of the material of the reference group is n2, and the scanning rate of the material of the reference group is P2. The exposure data of the material of any experimental group is n1, and the scanning rate is P1.

[0099] In an exemplary embodiment, when the Z value is greater than 1.65, it means that there is a 90% confidence level that the experimental group effect is better than the baseline group effect. When the Z value is greater than 1.96, it means that there is a 95% confidence level that the experimental group effect is better than the baseline group effect. When the Z value is greater than 2.58, it means that there is a 99% confidence level that the experimental group effect is better than the baseline group effect. A preset low delivery effect condition can be set in advance, and the materials whose delivery effect data meet the preset low delivery effect condition can be automatically eliminated and the experimental group can be updated.

[0100] For example, the material library can be updated in real time according to the above method. When the target material needs to be determined, the material prompt can be used to match the current material library. If multiple materials match the material prompt, the one with the best delivery effect data can be used as the target material. Alternatively, one of the multiple materials matching the material prompt can be manually designated as the target material.

[0101] In some embodiments, determining the delivery effect data corresponding to each material in the current material library includes: determining the explosive coefficient of each material in the current material library, the explosive coefficient is used to describe the possibility that the corresponding material is a explosive material; based on the explosive coefficient, determining the delivery effect data corresponding to each material in the current material library.

[0102] Exemplarily, the hot product coefficient may be a multiple of the scanning rate of the benchmark group. For example, if a material having a scanning rate that is twice that of the benchmark group is a hot product material, then the hot product coefficient is 2.

[0103] In some embodiments, when considering the explosive product coefficient, the significance Z test can be calculated based on the following formula (2).

[0104]

[0105] Among them, Z can be used to represent the Z value of the significance Z test. The exposure data of the material of the benchmark group is n2, and the scanning rate of the material of the benchmark group is P2. The exposure data of the material of any experimental group is n1, and the scanning rate is P1. X is the hot product coefficient. It should be noted that XP2 can be regarded as the scanning rate of the hot product group.

[0106] Exemplarily, when the Z value is greater than 1.65, it means that there is a 90% confidence level, and the experimental group has better results than the hot product group. When the Z value is greater than 1.96, it means that there is a 95% confidence level, and the experimental group has better results than the hot product group. When the Z value is greater than 2.58, it means that there is a 99% confidence level, and the experimental group has better results than the hot product group. By setting a preset high delivery effect condition, the delivery flow of high-confidence materials can be increased, thereby achieving automatic conversion of hot products. For example, the preset high delivery effect condition can be a Z value greater than 2.5.

[0107] S208: delivering the target material to the target user.

[0108] In an exemplary embodiment, the target material can be printed on a logistics waybill with the recipient being the target user to achieve accurate delivery of the target material. Alternatively, it can also be delivered offline to the target user on a specific occasion, and the embodiments of the present disclosure do not limit the offline delivery method.

[0109] It should be noted that the acquisition, storage, use, and processing of data in the technical solution disclosed in this disclosure are in compliance with the relevant provisions of national laws and regulations, and various types of data such as personal identity data, operation data, behavioral data, etc. related to individuals, customers, and groups obtained in the embodiments of this disclosure have all been authorized.

[0110] The method provided by the embodiment of the present disclosure can determine the target product based on the user-side characteristics of the target user to be delivered the material, and can determine the personalized target material based on the user-side characteristics of the target user and the product-side characteristics of the target product. Therefore, the present disclosure can determine personalized materials for different users and achieve accurate delivery, improve the accuracy of material delivery, and ensure content conversion effect and user experience.

[0111] For example, a schematic diagram of similarity calculation using a twin-tower model provided in an embodiment of the present disclosure can be as follows: Figure 3 shown.

[0112] In this Figure 3 In the above example, user-side features and product-side features can be obtained respectively. Then, feature embedding can be performed on the user-side features and product-side features respectively, and the vectors obtained by feature embedding can be input into the DNN (Deep Neural Network) model respectively. Then, corresponding user embedding or product embedding can be performed to obtain user feature vectors and product feature vectors. Finally, similarity calculation can be performed using user feature vectors and product feature vectors.

[0113] For example, a large model calling schematic diagram provided by an embodiment of the present disclosure can be as follows: Figure 4 shown.

[0114] like Figure 4 As shown, the disclosed embodiment may include a portrait module, an online service module, an effect monitoring and material rollback module, and a manual verification module. Among them, the portrait module is used to determine the user portrait and the product portrait, the user portrait includes various user-side features, and the product portrait includes various product-side features. In addition, TGI scoring can be performed on each user-side feature and each product-side feature to screen the features. The online service module is used to generate offline materials and call online materials. Among them, offline material generation can be implemented based on a large model call, and online material call can be implemented based on a search engine. The effect monitoring and material rollback module can be used to perform effect monitoring and material rollback on the target material obtained by offline material generation or online material call. Among them, effect monitoring can be implemented through network address indication information, and material rollback can be implemented by performing a significance Z test on each material. The manual verification module can be used for front-end manual verification.

[0115] For example, a schematic diagram of a material delivery architecture provided by an embodiment of the present disclosure may be as follows: Figure 5 shown.

[0116] exist Figure 5In the present invention, the architecture may include a front-end application layer, an API access layer, a business application layer, a service center, and basic resources. Among them, the front-end application layer may include network applications and mobile applications. The API (Application Programming Interface) access layer can be used for API gateway authentication, routing, and flow control. The business application layer can be used for automatic text generation and automatic image generation. The service center may include microservices and middleware, and the microservices may include a large model service A for automatic text generation and a large model service B for automatic image generation. The middleware can be used for service governance and service monitoring. Basic resources may include DOCKER (an application container engine), DB (Database), and DFS (Distributed File System).

[0117] In addition, the front-end application layer, API access layer, and business application layer can achieve load balancing based on NGINX (an application server). The API access layer, business application layer, and service center can perform call link and log analysis.

[0118] For example, a flowchart for determining a target material provided by an embodiment of the present disclosure may be as follows: Figure 6 shown.

[0119] Among them, after starting to execute the material delivery, it can be determined whether the current situation is real-time or offline. If it is real-time, the input parameters are accepted, and the input parameters can include user-side features. After the user-side features of the target user, the user portrait can be matched through the products in the cache. Through the step of matching the user portrait, the product category that the target user is interested in can be determined. Then, the target corresponding to the product category that the target user is interested in is obtained from the material library.

[0120] Alternatively, if it is offline, the input parameters are accepted, and the input parameters accepted at this time can be batches of user-side features and product-side features. After that, the user-side features and product-side features can be processed and the corresponding similarities can be calculated. And the user-side features of the target user and the product-side features of the target product can be TGI scored to screen the target features. Then, the material prompt words can be generated, and the material prompt words can support configuration. Then, the big model API can be called to generate pictures and / or texts. The big model API can be, for example, the big model API of each company. Then the generated target material can be filtered by named entity recognition, and the temporary material can be saved. After that, the target material is manually reviewed and filtered to obtain the target material, and the material is marked, wherein the mark of the target material can include material type, whether AI review, whether manual review, etc. In addition, the target material can be saved in the database of MySQL (My Structured Query Language, a structured query language) and OSS (Object Storage Service, object storage service).

[0121] For example, a schematic diagram of target material delivery provided by an embodiment of the present disclosure may be as follows: Figure 7 shown.

[0122] in, Figure 7 (a) shows a schematic diagram of the target material being printed on a logistics bill, which may include the recipient's name, phone number, contact information, delivery time, warehouse information, order number, etc. The target material may include text, pictures, and may also include a product QR code. In another embodiment, Figure 7 The target material in (a) can also be replaced by Figure 7 (b) Alternatively, you may Figure 7 (a) Add the following to the logistics bill shown: Figure 7 (b) shows the target material. Figure 7 The target material shown in (b) may also be composed of text, pictures, or product QR codes.

[0123] It should be noted that the current online personalized advertising delivery in the prior art is only for scenarios with closed-loop data chains, and is not universal for the special scenarios of logistics advertising. In addition, the delivery of general physical advertising ends as soon as it is delivered, and it is difficult to review the effect periodically to form a closed loop, and it is impossible to perform long-term iterative optimization. In addition, common physical advertising delivery can only be widely defined according to the delivery population, such as leaflets in the community, large screens at the airport, offline trials, etc., and it is impossible to make refined and customized advertising recommendations for each customer. The material delivery method provided in the embodiment of the present disclosure can determine the user-side characteristics of the target user to be delivered material in the logistics scenario according to the category information of the goods purchased by the user and the browsing / add-to-cart / purchase information of the user, so as to determine the goods that the target user is most interested in and generate personalized target materials. Therefore, this application can open up the upstream and downstream data links of logistics advertising delivery. In addition, a product QR code can be added to the delivered logistics bill, and the customer can leave a trace after scanning the code, thereby realizing real-time acquisition of browsing / add-to-cart / purchase data, and the generated material can be optimized for a long time.

[0124] Based on the same inventive concept, the present disclosure also provides a material delivery device, as described in the following embodiments. Since the principle of solving the problem in the device embodiment is similar to that in the above method embodiment, the implementation of the device embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0125] Figure 8 A schematic diagram of a material delivery device in an embodiment of the present disclosure is shown. Figure 8 As shown, the device comprises:

[0126] The user-side feature determination module 801 is used to determine the user-side features of the target user to whom the material is to be delivered;

[0127] The target product determination module 802 is used to determine the target product according to the user-side characteristics of the target user;

[0128] The target material determination module 803 is used to determine the target material according to the user-side characteristics of the target user and the product-side characteristics of the target product, wherein the target material includes information corresponding to at least one of the user-side characteristics and the product-side characteristics;

[0129] The delivery module 804 is used to deliver target materials to target users.

[0130] In some embodiments, the target material determination module 803 is used to generate a material prompt based on the user-side characteristics of the target user and the product-side characteristics of the target product, where the material prompt is used to describe the information included in the target material; match the material prompt with the material in the current material library; when the match is successful, obtain the target material from the current material library; when the match fails, generate the target material based on the material prompt.

[0131] In some embodiments, the target material determination module 803 is used to respectively determine the target group index of each user-side feature and each product-side feature, where the target group index is used to indicate the target user's attention to the corresponding feature; the user-side features and the product-side features are screened according to the target group index to obtain at least one target feature; and a material prompt is generated according to at least one target feature.

[0132] In some embodiments, the material delivery device provided by the embodiments of the present disclosure further includes:

[0133] The named entity recognition module is used to identify the named entities in the target material; perform relationship extraction processing on each named entity to obtain the association relationship between each named entity; and determine whether the association relationship between each named entity meets the preset logic;

[0134] The target material storage module is used to store the target material into the current material library when the association relationships between the named entities meet the preset logic.

[0135] In some embodiments, a matching degree determination module is used to determine the matching degree between the target material and the target user when the association relationship between each named entity satisfies a preset logic;

[0136] The target material storage module is used to store the target material into the current material library when the matching degree meets the preset matching condition.

[0137] In some embodiments, the target product determination module 802 is used to determine the product-side features of multiple products to be placed; determine a user feature vector based on the user-side features of the target user; determine corresponding product feature vectors based on the product-side features of the multiple products to be placed; calculate multiple similarities between the user feature vector and the multiple product feature vectors, wherein a similarity is determined between the user feature vector and a product feature vector of a product to be placed, and each similarity is used to indicate the possibility of the corresponding product to be placed being placed to the target user; and screen the multiple products to be placed based on the multiple similarities to obtain the target product.

[0138] In some embodiments, the target material also includes network address indication information, and the network address indication information is used for the target user to view the network details page of the target product. The material delivery device provided by the embodiment of the present disclosure also includes:

[0139] The delivery effect data module is used to determine the delivery effect data corresponding to each material in the current material library. The delivery effect data corresponding to any material is used to describe the possibility of viewing the corresponding network details page through the network address indication information in the material;

[0140] A material deletion module is used to delete each material whose delivery effect data meets the preset low delivery effect condition;

[0141] The target material determination module 803 is used to screen the multiple materials according to the delivery effect data to obtain the target material if the material prompt successfully matches the multiple materials in the current material library.

[0142] In some embodiments, the delivery effect data module is used to determine the explosive coefficient of each material in the current material library. The explosive coefficient is used to describe the possibility that the corresponding material is a explosive material; based on the explosive coefficient, the delivery effect data corresponding to each material in the current material library is determined.

[0143] The device provided by the embodiment of the present disclosure can determine the target product based on the user-side characteristics of the target user to be delivered the material, and can determine the personalized target material based on the user-side characteristics of the target user and the product-side characteristics of the target product. Therefore, the present disclosure can determine personalized materials for different users and achieve accurate delivery, improve the accuracy of material delivery, and ensure content conversion effect and user experience.

[0144] It should be noted that the user-side feature determination module 801, the target product determination module 802, the target material determination module 803, and the delivery module 804 correspond to S202 to S208 in the method embodiment, and the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above method embodiment. It should be noted that the above modules, as part of the device, can be executed in a computer system such as a set of computer executable instructions.

[0145] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0146] The embodiment of the present disclosure provides an electronic device, which exemplarily includes: a processor and a memory. The memory can be used to store executable instructions of the processor. The processor is configured to implement the material delivery method provided by the embodiment of the present disclosure by executing the above executable instructions.

[0147] Refer to the following Fig. 9 The electronic device 900 according to this embodiment of the present disclosure is described. Fig. 9 The electronic device 900 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0148] like Fig. 9 As shown, the electronic device 900 is in the form of a general computing device. The components of the electronic device 900 may include, but are not limited to: at least one processing unit 910, at least one storage unit 920, and a bus 930 connecting different system components (including the storage unit 920 and the processing unit 910).

[0149] The storage unit stores a program code, which can be executed by the processing unit 910, so that the processing unit 910 performs the steps of various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 910 can perform the following steps of the above method embodiment:

[0150] Determine the user-side characteristics of the target user to whom the material is to be delivered; determine the target product based on the user-side characteristics of the target user; determine the target material based on the user-side characteristics of the target user and the product-side characteristics of the target product, the target material including information corresponding to at least one of the user-side characteristics and the product-side characteristics; deliver the target material to the target user.

[0151] The storage unit 920 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 9201 and / or a cache storage unit 9202 , and may further include a read-only storage unit (ROM) 9203 .

[0152] The storage unit 920 may also include a program / utility 9204 having a set (at least one) of program modules 9205, such program modules 9205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0153] Bus 930 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0154] The electronic device 900 may also communicate with one or more external devices 940 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 900, and / or may communicate with any device that enables the electronic device 900 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 950. Furthermore, the electronic device 900 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 960. As shown, the network adapter 960 communicates with other modules of the electronic device 900 via a bus 930. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 900, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0155] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0156] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer program product, which includes: a computer program, which implements the above-mentioned material delivery method when executed by a processor.

[0157] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, and when the computer program is executed by a processor, the material delivery method provided in the embodiment of the present disclosure can be implemented. The computer-readable storage medium can be a readable signal medium or a readable storage medium.

[0158] Fig.10 A schematic diagram of a computer-readable storage medium in an embodiment of the present disclosure is shown. Fig.10As shown, a program product capable of implementing the above method of the present disclosure is stored on the computer-readable storage medium 1000. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above “Exemplary Method” section of this specification.

[0159] More specific examples of computer-readable storage media in the present disclosure may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0160] In the present disclosure, a computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein a readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A readable signal medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0161] Alternatively, the program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0162] In a specific implementation, the program code for performing the operation of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0163] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0164] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0165] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0166] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The description and examples are to be regarded as exemplary only, and the true scope of the present disclosure is indicated by the appended claims.

Claims

1. A material delivery method, characterized in that: include: Determine the user-side characteristics of the target users to whom the material is to be delivered; Determining target products according to user-side characteristics of the target user; Determine a target material according to the user-side feature of the target user and the product-side feature of the target product, wherein the target material includes information corresponding to at least one of the user-side feature and the product-side feature; The target material is delivered to the target user.

2. The material delivery method according to claim 1, characterized in that: The determining the target material according to the user-side characteristics of the target user and the product-side characteristics of the target product includes: Generate a material prompt according to the user-side characteristics of the target user and the product-side characteristics of the target product, wherein the material prompt is used to describe information included in the target material; Matching the material prompt with the material in the current material library; When the match is successful, the target material is obtained from the current material library; when the match fails, the target material is generated according to the material prompt.

3. The material delivery method according to claim 2, characterized in that: The generating of the material prompt according to the user-side characteristics of the target user and the product-side characteristics of the target product includes: Determine the target group index of each user-side feature and each product-side feature respectively, wherein the target group index is used to indicate the degree of attention paid by the target user to the corresponding feature; The user-side features and the product-side features are screened according to the target group index to obtain at least one target feature; The material prompt is generated according to at least one target feature.

4. The material delivery method according to any one of claims 1 to 3, characterized in that: After determining the target material, the method further includes: Identifying named entities in the target material; Perform relationship extraction on each named entity to obtain the association relationship between each named entity; Determine whether the association relationship between each named entity meets the preset logic; When the association relationships between the named entities all satisfy the preset logic, the target material is stored in the current material library.

5. The material delivery method according to claim 4, characterized in that: When the association relationships between the named entities all satisfy the preset logic, storing the target material in the current material library includes: When the association relationships between the named entities all satisfy the preset logic, determining the matching degree between the target material and the target user; When the matching degree meets the preset matching condition, the target material is stored in the current material library.

6. The material delivery method according to any one of claims 1 to 3, characterized in that: The determining the target product according to the user-side characteristics of the target user includes: Determine product-side features of multiple products to be launched; Determine a user feature vector according to the user-side features of the target user; determine corresponding product feature vectors according to the product-side features of multiple products to be placed; Calculating multiple similarities between the user feature vector and multiple product feature vectors, wherein the user feature vector and a product feature vector of a product to be placed determine a similarity, and each similarity is used to indicate the possibility that the corresponding product to be placed is placed to the target user; A plurality of commodities to be put on the market are screened according to a plurality of similarities to obtain the target commodity.

7. The material delivery method according to claim 2 or 3, characterized in that: The target material also includes network address indication information, and the network address indication information is used by the target user to view the network details page of the target product. The method also includes: Determine the delivery effect data corresponding to each material in the current material library, wherein the delivery effect data corresponding to any material is used to describe the possibility of viewing the corresponding network details page through the network address indication information in the material; Deleting each material whose delivery effect data meets a preset low delivery effect condition; Wherein, when the match is successful, obtaining the target material from the current material library includes: If the material prompt successfully matches multiple materials in the current material library, the multiple materials are screened according to the delivery effect data to obtain the target material.

8. The material delivery method according to claim 7, characterized in that: The determining of the delivery effect data corresponding to each material in the current material library includes: Determine the explosive product coefficient of each material in the current material library, where the explosive product coefficient is used to describe the possibility that the corresponding material is a explosive material; According to the explosive product coefficient, the delivery effect data corresponding to each material in the current material library is determined.

9. A material delivery device, characterized in that: include: A user-side feature determination module, used to determine the user-side features of the target user to whom the material is to be delivered; A target product determination module, used to determine the target product according to the user-side characteristics of the target user; A target material determination module, configured to determine a target material according to a user-side feature of the target user and a product-side feature of the target product, wherein the target material includes information corresponding to at least one of the user-side feature and the product-side feature; A delivery module is used to deliver the target material to the target user.

10. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the material delivery method according to any one of claims 1 to 8 by executing the executable instructions.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the material delivery method described in any one of claims 1 to 8 is implemented.

12. A computer program product, characterized in that The computer program product includes computer instructions, which are stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the material delivery method as described in any one of claims 1 to 8.

Images