Commodity selection method and device based on commodity portrait, terminal equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a commodity selection method and device based on a commodity portrait, terminal equipment and a storage medium, and the method comprises the steps: constructing the commodity portrait corresponding to any commodity of a certain target category based on the multi-dimensional information corresponding to the commodity; if a certain demand object has the commodity demand for the target category, constructing a live broadcast portrait corresponding to a live broadcast account used by the demand object based on the multi-dimensional information corresponding to the demand object; and calculating a portrait matching degree between the live broadcast portrait and a commodity portrait corresponding to each to-be-selected commodity in the target category, and determining at least one target commodity from all the to-be-selected commodities as a suggested selected commodity of the demand object according to the portrait matching degree. Therefore, goods selection can be carried out based on the commodity portraits, the commodity selection efficiency can be improved, and the matching degree between the goods selection result and the live broadcast room sales demand can be improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a product selection method, device, terminal equipment and storage medium based on product portraits. Background Art

[0002] As live streaming platforms become more and more diverse, live streaming becomes more and more popular, and the audience range of live streaming becomes wider and wider, more and more people or companies are starting to develop live streaming businesses. Live streaming bloggers use live streaming accounts to broadcast live, so as to move the products that were originally sold offline to online sales. This not only expands sales channels, but also increases the exposure of products, thereby attracting more people to buy products.

[0003] In order to attract more live broadcast traffic and increase sales, more and more live broadcast rooms have begun to expand the categories of goods sold in live broadcasts, and one of the important links in expanding the categories of goods is product selection. Currently, most live broadcast room product selection is done by corresponding product selection personnel based on basic product information such as product category, product price, and product ingredient list. This is not only not conducive to improving product selection efficiency, but also not conducive to improving the matching degree between product selection results and live broadcast room sales needs, and thus not conducive to improving the live broadcast sales effect of goods. Summary of the invention

[0004] The present invention provides a product selection method, device, terminal device and storage medium based on product portraits, which can select products based on product portraits, which is not only conducive to improving the efficiency of product selection, but also can improve the matching degree between product selection results and sales needs of live broadcast rooms.

[0005] The first aspect of the present invention discloses a product selection method based on product portraits, the method comprising: For any product of a certain target category, a product portrait corresponding to the product is constructed based on the multi-dimensional information corresponding to the product; wherein the multi-dimensional information corresponding to the product includes product attribute information corresponding to the product, positioning user information corresponding to the product, historical sales records corresponding to the product, historical evaluation information corresponding to the product, and after-sales emotional information corresponding to the product; If a certain demand object has a demand for the product of the target category, a live broadcast portrait corresponding to the live broadcast account used by the demand object is constructed based on the multi-dimensional information corresponding to the demand object; the multi-dimensional information corresponding to the demand object includes live broadcast platform information, live broadcast account information, historical live broadcast records, live broadcast blogger information, expected user traffic and expected user positioning; The portrait matching degree between the live broadcast portrait and the product portrait corresponding to each to-be-selected product in the target category is calculated; and according to the portrait matching degree between the live broadcast portrait and the product portrait corresponding to each to-be-selected product in the target category, at least one target product is determined from all the to-be-selected products as the recommended product for the demand object.

[0006] As an optional implementation, in the first aspect of the present invention, after calculating the portrait matching degree between the live broadcast portrait and the product portrait corresponding to each selected product in the target category, the method further includes: For each product to be selected in the target category, based on the live broadcast dimension information in the multi-dimensional information corresponding to the demand object, the optimal live broadcast discount mechanism of the product to be selected for the demand object is analyzed, and according to the optimal live broadcast discount mechanism of the product to be selected for the demand object and the live broadcast discount demand mechanism corresponding to the demand object, a portrait matching calibration coefficient corresponding to the product to be selected is generated, and based on the portrait matching calibration coefficient corresponding to the product to be selected, the portrait matching degree between the live broadcast portrait and the product portrait corresponding to the product to be selected is calibrated to update the portrait matching degree between the live broadcast portrait and the product portrait corresponding to the product to be selected.

[0007] As an optional implementation, in the first aspect of the present invention, for the demand object, according to the multi-dimensional information corresponding to each live broadcast in the historical live broadcast record corresponding to the demand object, a new product selection suggestion for the demand object is generated, and the new product selection suggestion at least includes a product category suggestion corresponding to the new product required by the demand object; If the commodity category suggestion includes the target category, determining that the demand object has a commodity demand for the target category; Among them, the multi-dimensional information corresponding to each live broadcast in the historical live broadcast record includes the user traffic of each live broadcast, the product information sold in each live broadcast, the sales volume of the products sold in each live broadcast, the interaction information between the live broadcast blogger and the users of the live broadcast room in each live broadcast, the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the product, and the authenticity of the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the product.

[0008] As an optional embodiment, in the first aspect of the present invention, the method further comprises: After determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, then collect existing sales information about the recommended product on the live broadcast platform corresponding to the demand object and advertising information about the recommended product, wherein the existing sales information includes the sales situation of the recommended product in the live broadcast room that sells the recommended product in the form of live broadcast on the live broadcast platform and the sales situation of the recommended product in the video that sells the recommended product by carrying a sales link on the video on the live broadcast platform; Based on the live broadcast room situation of the demand object on the live broadcast platform, the existing sales information and the advertising information, a sales reference strategy is generated for the demand object to sell the recommended selected products on the live broadcast platform through the live broadcast account of the demand object; the sales reference strategy includes sales form, sales time, sales control amount and sales point keywords.

[0009] As an optional embodiment, in the first aspect of the present invention, the method further comprises: After determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, then based on the positioning user information corresponding to the recommended product, the user of the recommended product, the user stickiness of the live broadcast account of the demand object corresponding to the live broadcast room, and the historical purchase record of high-stickiness users of the live broadcast account of the demand object in the live broadcast room, the related product corresponding to the recommended product is determined for the demand object and recommended to the demand object.

[0010] As an optional embodiment, in the first aspect of the present invention, the method further comprises: After determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, based on the currently sold products corresponding to the live broadcast account of the demand object, determine whether the recommended product is an upgraded product of the existing product of the same category; If the recommended product is an upgraded product of an existing product of the same category, a live broadcast transition suggestion is generated for the live broadcast room corresponding to the live broadcast account of the demand object to transition from the existing product of the same category to the recommended product based on the historical live broadcast sales of the existing product of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of the consumer users who purchased the existing product of the same category in the live broadcast room corresponding to the live broadcast account of the demand object towards the live broadcast room corresponding to the live broadcast account of the demand object.

[0011] As an optional implementation, in the first aspect of the present invention, if the recommended product is an upgraded product of an existing product of the same category, the method further includes: According to the commodity category information corresponding to the existing commodities of the same category and the commodity level information corresponding to the existing commodities of the same category, the user replenishment period corresponding to the existing commodities of the same category is calculated to determine whether the user replenishment period is less than or equal to the preset replenishment period; if so, the step of generating a live broadcast transition suggestion for the live broadcast room corresponding to the live broadcast account of the demand object to transition from the existing commodities of the same category to the recommended selected commodities is performed based on the historical live broadcast sales of the existing commodities of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of the consumer users who purchased the existing commodities of the same category in the live broadcast room corresponding to the live broadcast account of the demand object towards the live broadcast room corresponding to the live broadcast account of the demand object.

[0012] A second aspect of the present invention discloses a product selection device based on product portraits, the device comprising: The first construction module is used to construct a product portrait corresponding to any product of a certain target category based on the multi-dimensional information corresponding to the product; wherein the multi-dimensional information corresponding to the product includes product attribute information corresponding to the product, positioning user information corresponding to the product, historical sales records corresponding to the product, historical evaluation information corresponding to the product, and after-sales emotional information corresponding to the product; The second construction module is used to construct a live broadcast portrait corresponding to the live broadcast account used by a certain demand object based on the multi-dimensional information corresponding to the demand object if there is a demand for the product of the target category; the multi-dimensional information corresponding to the demand object includes live broadcast platform information, live broadcast account information, historical live broadcast records, live broadcast blogger information, expected user traffic and expected user positioning; A matching degree calculation module, used to calculate the image matching degree between the live broadcast image and the product image corresponding to each selected product in the target category; The product selection suggestion module is used to determine at least one target product from all the products to be selected as the recommended product for the demand object based on the image matching degree between the live broadcast image and the product image corresponding to each product to be selected in the target category.

[0013] As an optional implementation, in the second aspect of the present invention, the device further includes: A matching calibration module is used to analyze, for each candidate product in the target category, the optimal live broadcast preferential mechanism of the candidate product for the demand object based on the live broadcast dimension information in the multi-dimensional information corresponding to the demand object, generate a portrait matching calibration coefficient corresponding to the candidate product according to the optimal live broadcast preferential mechanism of the candidate product for the demand object and the live broadcast preferential demand mechanism corresponding to the demand object, and calibrate the portrait matching degree between the live broadcast portrait and the product portrait corresponding to the candidate product based on the portrait matching calibration coefficient corresponding to the candidate product, so as to update the portrait matching degree between the live broadcast portrait and the product portrait corresponding to the candidate product.

[0014] As an optional implementation, in the second aspect of the present invention, the device further includes: A demand analysis module is used to generate, for the demand object, a new product selection suggestion for the demand object based on the multi-dimensional information corresponding to each live broadcast in the historical live broadcast record corresponding to the demand object, wherein the new product selection suggestion at least includes a product category suggestion corresponding to the new product required by the demand object; if the product category suggestion includes the target category, it is determined that the demand object has a product demand for the target category; Among them, the multi-dimensional information corresponding to each live broadcast in the historical live broadcast record includes the user traffic of each live broadcast, the product information sold in each live broadcast, the sales volume of the products sold in each live broadcast, the interaction information between the live broadcast blogger and the users of the live broadcast room in each live broadcast, the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the product, and the authenticity of the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the product.

[0015] As an optional implementation, in the second aspect of the present invention, the device further includes: A strategy generation module is used for, after determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, then collecting the existing sales information about the recommended product on the live broadcast platform corresponding to the demand object and the advertising information about the recommended product, the existing sales information including the sales situation of the recommended product in the live broadcast room that sells the recommended product through live broadcast on the live broadcast platform and the sales situation of the recommended product in the video that sells the recommended product on the live broadcast platform by carrying a sales link on the video; and, based on the live broadcast room situation of the demand object on the live broadcast platform, the existing sales information and the advertising information, generating a sales reference strategy for the demand object to sell the recommended product on the live broadcast platform through the live broadcast account of the demand object; the sales reference strategy includes sales form, sales time, sales control amount and selling point keywords.

[0016] As an optional implementation, in the second aspect of the present invention, the product selection suggestion module is further used to: After determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, then based on the positioning user information corresponding to the recommended product, the user of the recommended product, the user stickiness of the live broadcast account of the demand object corresponding to the live broadcast room, and the historical purchase record of high-stickiness users of the live broadcast account of the demand object in the live broadcast room, the related product corresponding to the recommended product is determined for the demand object and recommended to the demand object.

[0017] As an optional implementation, in the second aspect of the present invention, the device further includes: A transition suggestion module is used to determine the recommended product for the demand object. If the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, based on the currently sold products corresponding to the live broadcast account of the demand object, determine whether the recommended product is an upgraded product for the existing products of the same category; if the recommended product is an upgraded product for the existing products of the same category, then based on the historical live broadcast sales of the existing products of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of the consumer users who purchased the existing products of the same category in the live broadcast room corresponding to the live broadcast account of the demand object to the live broadcast room corresponding to the live broadcast account of the demand object, generate a live broadcast transition suggestion for the live broadcast room corresponding to the live broadcast account of the demand object to transition from the existing products of the same category to the recommended product.

[0018] As an optional implementation, in the second aspect of the present invention, the transition suggestion module is further used for: If the recommended product is an upgraded product of an existing product of the same category, the user replenishment period corresponding to the existing product of the same category is calculated according to the product category information corresponding to the existing product of the same category and the product level information corresponding to the existing product of the same category, and a judgment is made as to whether the user replenishment period is less than or equal to a preset replenishment period. If so, the operation of generating a live transition suggestion for the live broadcast room corresponding to the live broadcast account of the demand object to transition from the existing product of the same category to the recommended product is performed based on the historical live broadcast sales of the existing product of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of the consumer users who purchased the existing product of the same category in the live broadcast room corresponding to the live broadcast account of the demand object towards the live broadcast room corresponding to the live broadcast account of the demand object.

[0019] A third aspect of the present invention discloses a terminal device, the terminal device comprising: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute part or all of the steps in the product selection method based on product portraits disclosed in the first aspect of the present invention.

[0020] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute part or all of the steps in the product selection method based on product portraits disclosed in the first aspect of the present invention.

[0021] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: In the embodiments of the present invention, a product portrait can be constructed based on the multi-dimensional information corresponding to the product, thereby improving the accuracy of constructing the product portrait. When a certain demand object has a demand for a product of a corresponding category, a live broadcast portrait can also be constructed based on the multi-dimensional information corresponding to the demand object (including multiple information of the live broadcast dimension), thereby improving the accuracy of constructing the live broadcast portrait corresponding to the demand object. Then, based on the matching degree between the product portrait and the live broadcast portrait, recommended selected products are determined for the demand object, thereby providing an intelligent method for determining selected products, which is not only conducive to improving the efficiency of product selection, but also can improve the matching degree between the selection results and the sales demand of the live broadcast room. In addition, after calculating the matching degree between the live broadcast portrait and the product portrait, It is able to analyze the optimal live broadcast preferential mechanism of the selected products for the demand object and the live broadcast preferential demand mechanism of the demand object itself to generate a matching calibration system, and calibrate the matching degree, which is conducive to improving the accuracy of the determined portrait matching degree, and thus is conducive to further improving the accuracy of product selection; in addition, it is also able to intelligently generate product category suggestions corresponding to the products that need to be added for the demand object based on the multi-dimensional information involved in the historical live broadcast of the corresponding live broadcast room of the demand object, thereby improving the accuracy of product category suggestions; in addition, after determining the recommended product selection products for the demand object, if the demand object has a greater degree of product selection intention, further collect the existing sales information on the relevant live broadcast platform about the aforementioned recommended product selection products and The advertising placement information can be used to determine the recommended product selection for the live broadcast room of the target audience, and based on the existing sales information and the advertising placement information, a sales reference strategy for selling the aforementioned recommended product selection for the live broadcast room of the target audience is provided, which is not only conducive to improving the accuracy and reliability of the sales reference strategy, but also can be further conducive to improving the live broadcast sales effect of the live broadcast room of the target audience for the aforementioned recommended product selection (such as live broadcast sales volume and / or user flow brought to the live broadcast room and / or increased user stickiness for the live broadcast room); in addition, when the recommended product selection for the target audience is determined and the target audience has a greater intention to select the aforementioned recommended product selection, it is also possible to further determine the related product selection for the target audience based on the diversified information, which not only improves the accuracy of the related product selection The accuracy of the product selection is not only conducive to further improving the comprehensiveness and accuracy of the recommended product selection, but also helps to further improve the comprehensiveness and accuracy of the recommended product selection. In addition, when the recommended product selection products for the demand object are determined and the demand object has a strong intention to select the aforementioned recommended product selection products, the new types of recommended product selection products can be further analyzed. If the existing product is upgraded, the live broadcast transition suggestion is determined intelligently to reduce the situation where user stickiness is reduced due to direct product upgrades. In addition, if the recommended product selection product is an upgrade of the existing product, the user replenishment cycle of the existing product can also be calculated based on the category information and level information of the existing product. If the user replenishment cycle is short, when executing the live broadcast transition suggestion generation operation, it is conducive to improving the execution reliability of the live broadcast transition suggestion generation operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 It is a flowchart of a product selection method based on product portraits disclosed in an embodiment of the present invention; Figure 2 It is a structural schematic diagram of a product selection device based on product portraits disclosed in an embodiment of the present invention; Figure 3 It is a structural schematic diagram of another product selection device based on product portraits disclosed in an embodiment of the present invention; Figure 4 It is a structural schematic diagram of a terminal device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0026] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0027] The present invention discloses a product selection method, device, terminal device and storage medium based on product portrait, which can construct product portrait according to the multi-dimensional information corresponding to the product, thereby improving the accuracy of constructing product portrait. When a demand object has a demand for a product of a corresponding category, a live broadcast portrait can also be constructed according to the multi-dimensional information corresponding to the demand object (including multiple information of the live broadcast dimension), thereby improving the accuracy of constructing the live broadcast portrait corresponding to the demand object. Then, according to the matching degree between the product portrait and the live broadcast portrait, a recommended product selection product is determined for the demand object, thereby providing an intelligent determination method for selecting product selection products, which is not only conducive to improving the efficiency of product selection, but also can improve the matching degree between the selection results and the sales demand of the live broadcast room. The following are detailed descriptions.

[0028] Embodiment 1 See also Figure 1 , Figure 1 : is a flowchart of a product selection method based on product portrait disclosed in an embodiment of the present invention. Figure 1 The described method can be applied to a selection device for determining recommended products for a demand object. The selection device can be integrated in a terminal device, or in a live broadcast platform or live broadcast App, which is not limited in the embodiments of the present invention. Figure 1 As shown, the method may include: 101. For any product of a certain target category, a product portrait corresponding to the product is constructed based on the multi-dimensional information corresponding to the product.

[0029] In an embodiment of the present invention, the multi-dimensional information corresponding to the product includes product attribute information corresponding to the product, positioning user information corresponding to the product, historical sales records corresponding to the product, historical evaluation information corresponding to the product, and after-sales emotional information corresponding to the product (such as a combination of one or more of evaluation information, after-sales information, and repeat purchase information).

[0030] 102. If a certain demand object has a demand for products of the target category, a live broadcast portrait corresponding to the live broadcast account used by the demand object is constructed based on the multi-dimensional information corresponding to the demand object.

[0031] In the embodiment of the present invention, the multi-dimensional information corresponding to the demand object includes live broadcast platform information, live broadcast account information, historical live broadcast records, live broadcast blogger information (including live broadcast style, live broadcast interaction status, etc.), expected user traffic and expected user positioning.

[0032] 103. Calculate the portrait matching degree between the live broadcast portrait and the product portrait corresponding to each selected product in the target category.

[0033] Among them, the portrait matching degree can be calculated by inputting the live broadcast portrait and the product portrait into a matching degree calculation model that has been trained to convergence, or it can be calculated according to a pre-selected calculation formula. For example, the live broadcast portrait includes information in multiple dimensions, and the product portrait includes information in multiple dimensions. From the perspective of the product portrait, according to the information of each dimension of the product portrait, determine the matching score corresponding to the information of the live broadcast portrait in the corresponding dimension, and then comprehensively calculate the portrait matching degree based on the matching scores corresponding to the information of all dimensions.

[0034] 104. According to the image matching degree between the live broadcast image and the product image corresponding to each product to be selected in the target category, at least one target product is determined from all the products to be selected as the recommended product for the demand object.

[0035] In the embodiment of the present invention, the candidate products with a portrait matching degree greater than or equal to a preset matching degree threshold can be selected as the recommended selected products corresponding to the demand.

[0036] It can be seen that the embodiments of the present invention can construct a product portrait based on the multi-dimensional information corresponding to the product, thereby improving the accuracy of constructing the product portrait. When a certain demand object has a demand for a product of a corresponding category, a live broadcast portrait can also be constructed based on the multi-dimensional information corresponding to the demand object (including multiple information of the live broadcast dimension), thereby improving the accuracy of constructing the live broadcast portrait corresponding to the demand object. Then, based on the matching degree between the product portrait and the live broadcast portrait, recommended selected products are determined for the demand object, thereby providing an intelligent method for determining selected products, which is not only conducive to improving the efficiency of product selection, but also can improve the matching degree between the selection results and the sales demand of the live broadcast room.

[0037] In an optional embodiment, after calculating the image matching degree between the live broadcast image and the product image corresponding to each selected product in the target category, the method further includes: For each product to be selected in the target category, based on the live broadcast dimension information in the multi-dimensional information corresponding to the demand object, analyze the optimal live broadcast discount mechanism of the product to be selected for the demand object, and generate a portrait matching calibration coefficient corresponding to the product to be selected according to the optimal live broadcast discount mechanism of the product to be selected for the demand object and the live broadcast discount demand mechanism corresponding to the demand object. Based on the portrait matching calibration coefficient corresponding to the product to be selected, calibrate the portrait matching degree between the live broadcast portrait and the product portrait corresponding to the product to be selected, so as to update the portrait matching degree between the live broadcast portrait and the product portrait corresponding to the product to be selected.

[0038] It can be seen that after calculating the matching degree between the live broadcast portrait and the product portrait, this optional embodiment can also analyze the optimal live broadcast preferential mechanism of the selected products for the demand object and the live broadcast preferential demand mechanism of the demand object itself to generate a matching degree calibration system, and calibrate the matching degree, which is beneficial to improve the accuracy of the determined portrait matching degree, and thus is beneficial to further improve the accuracy of product selection.

[0039] In another optional embodiment, the method may further include: For the demand object, based on the multi-dimensional information corresponding to each live broadcast in the historical live broadcast records corresponding to the demand object, generate new product selection suggestions for the demand object, and the new product selection suggestions at least include product category suggestions corresponding to the new products required by the demand object; If the product category suggestion includes the target category, it is determined that the demand object has a demand for products in the target category.

[0040] Among them, the multi-dimensional information corresponding to each live broadcast in the historical live broadcast records includes the user traffic of each live broadcast, the product information sold in each live broadcast, the sales volume of the products sold in each live broadcast, the interaction information between the live blogger and the users of the live broadcast room in each live broadcast, the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the products, and the authenticity of the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the products.

[0041] It can be seen that this optional embodiment can also intelligently generate product category suggestions corresponding to the products that need to be added for the demand object based on the multi-dimensional information involved in the historical live broadcasts of the live broadcast room corresponding to the demand object, thereby improving the accuracy of the product category suggestions.

[0042] In yet another optional embodiment, the method may further include: After determining the recommended products for the demand object, if the demand object's selection intention for the recommended products is greater than or equal to a preset degree threshold, then collect the existing sales information about the recommended products and the advertising information about the recommended products on the live broadcast platform corresponding to the demand object, the existing sales information includes the sales situation of the recommended products in the live broadcast room that sells the recommended products through live broadcast on the live broadcast platform, and the sales situation of the recommended products in the video that sells the recommended products by carrying a sales link on the video on the live broadcast platform; Based on the live broadcast room situation, existing sales information and advertising information of the demand object on the live broadcast platform, a sales reference strategy is generated for selling recommended selected products through the demand object's live broadcast account on the live broadcast platform; the sales reference strategy includes sales form, sales time, sales control volume and sales point keywords.

[0043] It can be seen that after this optional embodiment determines the recommended products for the demand object, if the demand object has a greater degree of product selection intention, it further collects the existing sales information and advertising information about the aforementioned recommended products on the relevant live broadcast platform, and provides a sales reference strategy for selling the aforementioned recommended products in the live broadcast room of the demand object based on the existing sales information and advertising information. This is not only conducive to improving the accuracy and reliability of the sales reference strategy, but also can further help improve the live broadcast sales effect of the live broadcast room of the demand object for the aforementioned recommended products (such as live broadcast sales volume and / or user flow brought to the live broadcast room and / or increased user stickiness for the live broadcast room).

[0044] In yet another optional embodiment, the method may further include: After determining the recommended product selection for the demand object, if the demand object's selection intention for the recommended product selection is greater than or equal to the preset degree threshold, then based on the positioning user information corresponding to the recommended product selection, the user of the recommended product selection, the user stickiness of the live broadcast account of the demand object corresponding to the live broadcast room, and the historical purchase record of high-stickiness users of the live broadcast account of the demand object in the live broadcast room, the related product selection corresponding to the recommended product selection is determined for the demand object and recommended to the demand object.

[0045] For example, if the recommended product is toothpaste, then its related products can be toothbrushes, water flossers, etc.

[0046] It can be seen that when this optional embodiment determines the recommended product selection for the demand object and the demand object has a strong selection intention for the aforementioned recommended product selection, it can further determine the related product selection corresponding to the recommended product selection for the demand object based on diversified information, which not only improves the accuracy of the related product selection, but also helps to further improve the comprehensiveness and accuracy of the recommended product selection.

[0047] In yet another optional embodiment, the method may further include: After determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, based on the currently sold products corresponding to the live broadcast account of the demand object, determine whether the recommended product is an upgraded product of the existing products of the same category; If the recommended product is an upgraded product of an existing product of the same category, a live broadcast transition suggestion is generated for the live broadcast room corresponding to the live broadcast account of the demand object to transition from existing products of the same category to the recommended product based on the historical live broadcast sales of existing products of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of consumers who purchased existing products of the same category in the live broadcast room corresponding to the live broadcast account of the demand object towards the live broadcast room corresponding to the live broadcast account of the demand object.

[0048] It can be seen that this optional embodiment can further analyze the new types of recommended products when the recommended products for the demand object are determined and the demand object has a strong intention to select the aforementioned recommended products. If the existing products are upgraded, the live broadcast transition suggestions are intelligently determined to reduce the situation where user stickiness is reduced due to direct product upgrades.

[0049] In yet another optional embodiment, if the recommended selected product is an upgraded product of an existing product of the same category, the method further includes: Based on the product category information corresponding to the existing products of the same category and the product level information corresponding to the existing products of the same category, the user replenishment cycle corresponding to the existing products of the same category is calculated to determine whether the user replenishment cycle is less than or equal to the preset replenishment cycle. If so, the steps of generating a live broadcast transition suggestion for the live broadcast room corresponding to the live broadcast account of the demand object to transition from existing products of the same category to recommended products are executed based on the historical live broadcast sales of existing products of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of consumer users who purchase existing products of the same category in the live broadcast room corresponding to the live broadcast account of the demand object towards the live broadcast room corresponding to the demand object.

[0050] It can be seen that in this optional embodiment, if the recommended product is an upgrade of an existing product, the user replenishment cycle of the existing product can also be calculated based on the category information and level information of the existing product. If the user replenishment cycle is short, when executing the live transition suggestion generation operation, it is beneficial to improve the execution reliability of the live transition suggestion generation operation.

[0051] Embodiment 2 See also Figure 2 , Figure 2 : is a schematic diagram of a product selection device based on product portraits disclosed in an embodiment of the present invention. Figure 2 The described device is used to determine recommended products for the demand object, and the selection device can be integrated in the terminal device, or integrated in the live broadcast platform or live broadcast App, which is not limited in the embodiment of the present invention. Figure 2 As shown, the device may include: The first construction module 201 is used to construct a product portrait corresponding to any product of a certain target category based on the multi-dimensional information corresponding to the product; wherein the multi-dimensional information corresponding to the product includes product attribute information corresponding to the product, positioning user information corresponding to the product, historical sales records corresponding to the product, historical evaluation information corresponding to the product, and after-sales emotional information corresponding to the product; The second construction module 202 is used to construct a live broadcast portrait corresponding to the live broadcast account used by a certain demand object based on the multi-dimensional information corresponding to the demand object if there is a demand for the product of the target category; the multi-dimensional information corresponding to the demand object includes live broadcast platform information, live broadcast account information, historical live broadcast records, live broadcast blogger information, expected user traffic and expected user location; A matching degree calculation module 203 is used to calculate the image matching degree between the live broadcast image and the product image corresponding to each selected product in the target category; The product selection suggestion module 204 is used to determine at least one target product from all the products to be selected as a recommended product for the demand object based on the image matching degree between the live broadcast image and the product image corresponding to each product to be selected in the target category.

[0052] It can be seen that the embodiments of the present invention can construct a product portrait based on the multi-dimensional information corresponding to the product, thereby improving the accuracy of constructing the product portrait. When a certain demand object has a demand for a product of a corresponding category, a live broadcast portrait can also be constructed based on the multi-dimensional information corresponding to the demand object (including multiple information of the live broadcast dimension), thereby improving the accuracy of constructing the live broadcast portrait corresponding to the demand object. Then, based on the matching degree between the product portrait and the live broadcast portrait, recommended selected products are determined for the demand object, thereby providing an intelligent method for determining selected products, which is not only conducive to improving the efficiency of product selection, but also can improve the matching degree between the selection results and the sales demand of the live broadcast room.

[0053] In an optional embodiment, if Figure 3 As shown, the device also includes: The matching calibration module 205 is used to analyze the optimal live broadcast preferential mechanism of the selected commodity for the demand object for each selected commodity in the target category based on the live broadcast dimension information in the multi-dimensional information corresponding to the demand object, generate a portrait matching calibration coefficient corresponding to the selected commodity according to the optimal live broadcast preferential mechanism of the selected commodity for the demand object and the live broadcast preferential demand mechanism corresponding to the demand object, and calibrate the portrait matching degree between the live broadcast portrait and the commodity portrait corresponding to the selected commodity based on the portrait matching calibration coefficient corresponding to the selected commodity, so as to update the portrait matching degree between the live broadcast portrait and the commodity portrait corresponding to the selected commodity.

[0054] It can be seen that after calculating the matching degree between the live broadcast portrait and the product portrait, this optional embodiment can also analyze the optimal live broadcast preferential mechanism of the selected products for the demand object and the live broadcast preferential demand mechanism of the demand object itself to generate a matching degree calibration system, and calibrate the matching degree, which is beneficial to improve the accuracy of the determined portrait matching degree, and thus is beneficial to further improve the accuracy of product selection.

[0055] In another optional embodiment, Figure 3 As shown, the device also includes: The demand analysis module 206 is used to generate a new product selection suggestion for the demand object according to the multi-dimensional information corresponding to each live broadcast in the historical live broadcast record corresponding to the demand object, wherein the new product selection suggestion at least includes a product category suggestion corresponding to the new product required by the demand object; if the product category suggestion includes the target category, it is determined that the demand object has a product demand for the target category; Among them, the multi-dimensional information corresponding to each live broadcast in the historical live broadcast record includes the user traffic of each live broadcast, the product information sold in each live broadcast, the sales volume of the products sold in each live broadcast, the interaction information between the live broadcast blogger and the users of the live broadcast room in each live broadcast, the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the product, and the authenticity of the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the product.

[0056] It can be seen that this optional embodiment can also intelligently generate product category suggestions corresponding to the products that need to be added for the demand object based on the multi-dimensional information involved in the historical live broadcasts of the live broadcast room corresponding to the demand object, thereby improving the accuracy of the product category suggestions.

[0057] In yet another optional embodiment, Figure 3 As shown, the device also includes: The strategy generation module 207 is used to collect the existing sales information and advertising information about the recommended products on the live broadcast platform corresponding to the demand object after determining the recommended products for the demand object, if the demand object's selection intention for the recommended products is greater than or equal to a preset degree threshold, the existing sales information including the sales situation of the recommended products in the live broadcast room that sells the recommended products through live broadcast on the live broadcast platform and the sales situation of the recommended products in the video that sells the recommended products by carrying a sales link on the video on the live broadcast platform; and, based on the live broadcast room situation, existing sales information and advertising information of the demand object on the live broadcast platform, generate a sales reference strategy for the demand object to sell the recommended products on the live broadcast platform through the demand object's live broadcast account; the sales reference strategy includes sales form, sales time, sales control amount and sales point keywords.

[0058] It can be seen that after this optional embodiment determines the recommended products for the demand object, if the demand object has a greater degree of product selection intention, it further collects the existing sales information and advertising information about the aforementioned recommended products on the relevant live broadcast platform, and provides a sales reference strategy for selling the aforementioned recommended products in the live broadcast room of the demand object based on the existing sales information and advertising information. This is not only conducive to improving the accuracy and reliability of the sales reference strategy, but also can further help improve the live broadcast sales effect of the live broadcast room of the demand object for the aforementioned recommended products (such as live broadcast sales volume and / or user flow brought to the live broadcast room and / or increased user stickiness for the live broadcast room).

[0059] In yet another optional embodiment, the product selection suggestion module 204 is further configured to: After determining the recommended product selection for the demand object, if the demand object's selection intention for the recommended product selection is greater than or equal to the preset degree threshold, then based on the positioning user information corresponding to the recommended product selection, the user of the recommended product selection, the user stickiness of the live broadcast account of the demand object corresponding to the live broadcast room, and the historical purchase record of high-stickiness users of the live broadcast account of the demand object in the live broadcast room, the related product selection corresponding to the recommended product selection is determined for the demand object and recommended to the demand object.

[0060] It can be seen that when this optional embodiment determines the recommended product selection for the demand object and the demand object has a strong selection intention for the aforementioned recommended product selection, it can further determine the related product selection corresponding to the recommended product selection for the demand object based on diversified information, which not only improves the accuracy of the related product selection, but also helps to further improve the comprehensiveness and accuracy of the recommended product selection.

[0061] In yet another optional embodiment, Figure 3 As shown, the device also includes: The transition suggestion module 208 is used to determine the recommended product for the demand object. If the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, based on the currently sold products corresponding to the live broadcast account of the demand object, determine whether the recommended product is an upgraded product for the existing products of the same category; if the recommended product is an upgraded product for the existing products of the same category, then based on the historical live broadcast sales of the existing products of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of the consumer users who purchased the existing products of the same category in the live broadcast room corresponding to the live broadcast account of the demand object to the live broadcast room corresponding to the live broadcast account of the demand object, generate a live broadcast transition suggestion for the live broadcast room corresponding to the live broadcast account of the demand object to transition from the existing products of the same category to the recommended product.

[0062] It can be seen that this optional embodiment can further analyze the new types of recommended products when the recommended products for the demand object are determined and the demand object has a strong intention to select the aforementioned recommended products. If the existing products are upgraded, the live broadcast transition suggestions are intelligently determined to reduce the situation where user stickiness is reduced due to direct product upgrades.

[0063] Further optionally, the transition suggestion module 208 is further configured to: If the recommended product is an upgraded product for an existing product of the same category, the user replenishment cycle for the existing products of the same category is calculated based on the product category information and the product level information of the existing products of the same category, and a determination is made as to whether the user replenishment cycle is less than or equal to the preset replenishment cycle. If so, the step of generating a live broadcast transition suggestion for the live broadcast room corresponding to the live broadcast account of the demand object from existing products of the same category to the recommended product is performed based on the historical live broadcast sales of the existing products of the same category in the live broadcast room corresponding to the demand object's live broadcast account and the user stickiness of the consumer users who purchased the existing products of the same category in the live broadcast room corresponding to the demand object's live broadcast account towards the live broadcast room corresponding to the demand object.

[0064] It can be seen that in this optional embodiment, if the recommended product is an upgrade of an existing product, the user replenishment cycle of the existing product can also be calculated based on the category information and level information of the existing product. If the user replenishment cycle is short, when executing the live transition suggestion generation operation, it is beneficial to improve the execution reliability of the live transition suggestion generation operation.

[0065] Embodiment 3 See also Figure 4 , Figure 4 is a schematic diagram of the structure of a terminal device disclosed in an embodiment of the present invention. The terminal device is used to determine recommended products for a demand object. Figure 4 As shown, the terminal device may include: A memory 301 storing executable program codes; a processor 302 coupled to a memory; The processor 302 calls the executable program code stored in the memory 301 to execute the steps in any of the product selection methods based on product portraits described in Example 1.

[0066] Embodiment 4 The embodiment of the present invention discloses another terminal device, which is used to determine recommended selected products for a demand object. The terminal device may include any product selection device based on product portrait described in the second embodiment, and the embodiment of the present invention will not be repeated.

[0067] Embodiment 5 An embodiment of the present invention discloses a computer storage medium that stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute any step of the product selection method based on product portraits described in Embodiment 1.

[0068] Embodiment 6 An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute any step of the product selection method based on product portraits described in Example 1.

[0069] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0070] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0071] Finally, it should be noted that the product selection method, device, terminal device and storage medium based on commodity portraits disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A product selection method based on product portrait, characterized in that: The method comprises: For any product of a certain target category, a product portrait corresponding to the product is constructed based on the multi-dimensional information corresponding to the product; wherein the multi-dimensional information corresponding to the product includes product attribute information corresponding to the product, positioning user information corresponding to the product, historical sales records corresponding to the product, historical evaluation information corresponding to the product, and after-sales emotional information corresponding to the product; If a certain demand object has a demand for the product of the target category, a live broadcast portrait corresponding to the live broadcast account used by the demand object is constructed based on the multi-dimensional information corresponding to the demand object; the multi-dimensional information corresponding to the demand object includes live broadcast platform information, live broadcast account information, historical live broadcast records, live broadcast blogger information, expected user traffic and expected user positioning; The portrait matching degree between the live broadcast portrait and the product portrait corresponding to each to-be-selected product in the target category is calculated; and according to the portrait matching degree between the live broadcast portrait and the product portrait corresponding to each to-be-selected product in the target category, at least one target product is determined from all the to-be-selected products as the recommended product for the demand object.

2. The product selection method based on product portrait according to claim 1 is characterized in that: After calculating the image matching degree between the live broadcast image and the product image corresponding to each selected product in the target category, the method further includes: For each product to be selected in the target category, based on the live broadcast dimension information in the multi-dimensional information corresponding to the demand object, the optimal live broadcast discount mechanism of the product to be selected for the demand object is analyzed, and according to the optimal live broadcast discount mechanism of the product to be selected for the demand object and the live broadcast discount demand mechanism corresponding to the demand object, a portrait matching calibration coefficient corresponding to the product to be selected is generated, and based on the portrait matching calibration coefficient corresponding to the product to be selected, the portrait matching degree between the live broadcast portrait and the product portrait corresponding to the product to be selected is calibrated to update the portrait matching degree between the live broadcast portrait and the product portrait corresponding to the product to be selected.

3. The product selection method based on product portrait according to claim 2 is characterized in that: The method further comprises: For the demand object, based on the multi-dimensional information corresponding to each live broadcast in the historical live broadcast record corresponding to the demand object, a new product selection suggestion for the demand object is generated, and the new product selection suggestion at least includes a product category suggestion corresponding to the new product required by the demand object; If the commodity category suggestion includes a target category, determining that the demand object has a commodity demand for the target category; Among them, the multi-dimensional information corresponding to each live broadcast in the historical live broadcast record includes the user traffic of each live broadcast, the product information sold in each live broadcast, the sales volume of the products sold in each live broadcast, the interaction information between the live broadcast blogger and the users of the live broadcast room in each live broadcast, the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the product, and the authenticity of the barrage information sent by the users of the live broadcast room in each live broadcast to indicate that they have demand for the product.

4. The product selection method based on product portrait according to any one of claims 1 to 3, characterized in that: The method further comprises: After determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, then collect existing sales information about the recommended product on the live broadcast platform corresponding to the demand object and advertising information about the recommended product, wherein the existing sales information includes the sales situation of the recommended product in the live broadcast room that sells the recommended product in the form of live broadcast on the live broadcast platform and the sales situation of the recommended product in the video that sells the recommended product by carrying a sales link on the video on the live broadcast platform; Based on the live broadcast room situation of the demand object on the live broadcast platform, the existing sales information and the advertising information, a sales reference strategy is generated for the demand object to sell the recommended selected products on the live broadcast platform through the live broadcast account of the demand object; the sales reference strategy includes sales form, sales time, sales control amount and sales point keywords.

5. The product selection method based on product portrait according to claim 4 is characterized in that: The method further comprises: After determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, then based on the positioning user information corresponding to the recommended product, the user of the recommended product, the user stickiness of the live broadcast account of the demand object corresponding to the live broadcast room, and the historical purchase record of high-stickiness users of the live broadcast account of the demand object in the live broadcast room, the related product corresponding to the recommended product is determined for the demand object and recommended to the demand object.

6. The product selection method based on product image according to claim 5 is characterized in that: The method further comprises: After determining the recommended product for the demand object, if the demand object's selection intention for the recommended product is greater than or equal to a preset degree threshold, based on the currently sold products corresponding to the live broadcast account of the demand object, determine whether the recommended product is an upgraded product of the existing product of the same category; If the recommended product is an upgraded product of an existing product of the same category, a live broadcast transition suggestion is generated for the live broadcast room corresponding to the live broadcast account of the demand object to transition from the existing product of the same category to the recommended product based on the historical live broadcast sales of the existing product of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of the consumer users who purchased the existing product of the same category in the live broadcast room corresponding to the live broadcast account of the demand object towards the live broadcast room corresponding to the live broadcast account of the demand object.

7. The product selection method based on product portrait according to claim 6, characterized in that: If the recommended product is an upgraded product of an existing product of the same category, the method further includes: According to the commodity category information corresponding to the existing commodities of the same category and the commodity level information corresponding to the existing commodities of the same category, the user replenishment period corresponding to the existing commodities of the same category is calculated to determine whether the user replenishment period is less than or equal to the preset replenishment period; if so, the step of generating a live broadcast transition suggestion for the live broadcast room corresponding to the live broadcast account of the demand object to transition from the existing commodities of the same category to the recommended selected commodities is performed based on the historical live broadcast sales of the existing commodities of the same category in the live broadcast room corresponding to the live broadcast account of the demand object and the user stickiness of the consumer users who purchased the existing commodities of the same category in the live broadcast room corresponding to the live broadcast account of the demand object towards the live broadcast room corresponding to the live broadcast account of the demand object.

8. A product selection device based on product portraits, characterized in that: The device comprises: The first construction module is used to construct a product portrait corresponding to any product of a certain target category based on the multi-dimensional information corresponding to the product; wherein the multi-dimensional information corresponding to the product includes product attribute information corresponding to the product, positioning user information corresponding to the product, historical sales records corresponding to the product, historical evaluation information corresponding to the product, and after-sales emotional information corresponding to the product; The second construction module is used to construct a live broadcast portrait corresponding to the live broadcast account used by a certain demand object based on the multi-dimensional information corresponding to the demand object if there is a demand for the product of the target category; the multi-dimensional information corresponding to the demand object includes live broadcast platform information, live broadcast account information, historical live broadcast records, live broadcast blogger information, expected user traffic and expected user positioning; A matching degree calculation module, used to calculate the image matching degree between the live broadcast image and the product image corresponding to each selected product in the target category; The product selection suggestion module is used to determine at least one target product from all the products to be selected as the recommended product for the demand object based on the image matching degree between the live broadcast image and the product image corresponding to each product to be selected in the target category.

9. A terminal device, characterized in that: The terminal device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the product selection method based on product portrait as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores computer instructions, which, when called, are used to execute the product selection method based on product portraits as described in any one of claims 1 to 7.

Images