CN113837792A - Promotion cooperation object recommendation method, device, equipment and storage medium - Google Patents
Promotion cooperation object recommendation method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN113837792A CN113837792A CN202110978052.6A CN202110978052A CN113837792A CN 113837792 A CN113837792 A CN 113837792A CN 202110978052 A CN202110978052 A CN 202110978052A CN 113837792 A CN113837792 A CN 113837792A
- Authority
- CN
- China
- Prior art keywords
- promotion
- historical
- partner
- ranking
- product
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0242—Determining effectiveness of advertisements
- G06Q30/0245—Surveys
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Molecular Biology (AREA)
- Game Theory and Decision Science (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
本申请涉及一种推广合作对象推荐方法、装置、计算机设备和存储介质。本申请的方法通过获取推广合作对象的基本数据及其历史推广数据，基本数据包括粉丝数目以及推广产品类别；根据推广合作对象的粉丝数目及其历史推广数据，确定推广合作对象对应的推广得分；根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；当接收到推荐请求时，根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。本申请通过历史推广数据以及推广产品类别来对推广合作对象进行排名，依照排名选择推广合作对象进行推荐，从而选择更有效的推广合作对象，来提高推广产品的转化率。
The present application relates to a method, device, computer equipment and storage medium for recommending cooperation objects for promotion. The method of the present application obtains the basic data of the promotion partner and its historical promotion data, the basic data includes the number of fans and the category of the promotion product; according to the number of fans of the promotion partner and its historical promotion data, the promotion score corresponding to the promotion partner is determined; According to the promotion score and the promotion product category of the promotion cooperation object, determine the object ranking of the promotion cooperation object under each promotion product category; when a recommendation request is received, according to the object ranking corresponding to the promotion product category in the recommendation request, push the corresponding promotion cooperation object. This application ranks the promotion partners through historical promotion data and promotion product categories, and selects the promotion partners for recommendation according to the ranking, so as to select more effective promotion partners and improve the conversion rate of the promoted products.
Description
技术领域technical field
本申请涉及计算机领域，特别是涉及一种推广合作对象推荐方法、装置、计算机设备和存储介质。The present application relates to the field of computers, and in particular, to a method, apparatus, computer equipment and storage medium for recommending cooperation objects for promotion.
背景技术Background technique
广告，顾名思义，就是广而告之，向社会广大公众告知某件事物。品牌方一般可以通过广告的形式来向消费者或用户传播产品或服务信息。而目前的线上推广活动中，品牌方一般可以在线上领域选择合适的推广合作对象来帮助进行品牌或者产品的推广。而推广合作对象可以凭借其自身所拥有的粉丝数来帮忙进行产品推广。目前的品牌方选择推广合作对象主要参考推广合作对象的粉丝数，但市场上存在虚假流量的现象。Advertising, as the name suggests, is to advertise, to inform the general public of something. Brands can generally communicate product or service information to consumers or users in the form of advertisements. In the current online promotion activities, the brand side can generally choose suitable promotion partners in the online field to help promote the brand or product. The promotion partners can help to promote products by virtue of the number of fans they have. At present, the brand owner chooses the promotion partner mainly by referring to the number of fans of the promotion partner, but there is a phenomenon of false traffic in the market.
目前，针对该现象，一般可以通过提取粉丝对推广合作对象的直播观看历史记录，而后基于历史记录来识别粉丝对推广合作对象的喜爱程度，从而区分真假流量，来进行有效地推广合作对象推荐，但是这种方法只能区分出真假粉丝，而无法提高广告推广过程的转化率。At present, in response to this phenomenon, it is generally possible to effectively recommend promotion partners by extracting the live viewing history of fans to the promotion partners, and then identifying the fans' liking for the promotion partners based on the historical records, so as to distinguish the true and false traffic. , but this method can only distinguish between true and false fans, and cannot improve the conversion rate of the advertising promotion process.
发明内容SUMMARY OF THE INVENTION
基于此，有必要针对上述技术问题，提供一种能够有效提高推广产品转化率的推广合作对象推荐方法、装置、计算机设备和存储介质。Based on this, it is necessary to provide a method, device, computer equipment and storage medium for recommending promotion cooperation objects that can effectively improve the conversion rate of promotion products, aiming at the above technical problems.
一种推广合作对象推荐方法，所述方法包括：A method for recommending a promotion partner, the method comprising:
获取推广合作对象的基本数据及其历史推广数据，所述基本数据包括粉丝数目以及推广产品类别；Obtain the basic data of the promotion partner and its historical promotion data, the basic data includes the number of fans and the category of the promoted product;
根据所述推广合作对象的粉丝数目以及所述历史推广数据，确定所述推广合作对象对应的推广得分；According to the number of fans of the promotion partner and the historical promotion data, determine the promotion score corresponding to the promotion partner;
根据所述推广得分以及所述推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；According to the promotion score and the promotion product category of the promotion cooperation object, determine the object ranking of the promotion cooperation object under each promotion product category;
当接收到推荐请求时，根据所述推荐请求中的推广产品类别对应的所述对象排名，推送对应的推广合作对象。When a recommendation request is received, the corresponding promotion cooperation object is pushed according to the object ranking corresponding to the promotion product category in the recommendation request.
在其中一个实施例中，所述推广合作对象的历史推广数据包括所述推广合作对象所推广产品的历史转化率数据；In one embodiment, the historical promotion data of the promotion partner includes historical conversion rate data of the product promoted by the promotion partner;
所述根据所述推广合作对象的粉丝数目以及所述历史推广数据，确定所述推广合作对象对应的推广得分包括：According to the number of fans of the promotion partner and the historical promotion data, determining the promotion score corresponding to the promotion partner includes:
根据所述推广合作对象所推广产品的历史转化率数据，生成多个所述推广产品对应的历史转化率排名；generating a plurality of historical conversion rate rankings corresponding to the promotion products according to the historical conversion rate data of the products promoted by the promotion partners;
获取所述历史转化率排名中预设名次内的推广产品，根据所述预设名次内的推广产品确定所述推广合作对象的历史推广得分；Obtaining the promotion products in a preset ranking in the historical conversion rate ranking, and determining the historical promotion score of the promotion partner according to the promotion products in the preset ranking;
根据所述推广合作对象的粉丝数目以及所述历史推广得分，确定所述推广合作对象对应的推广得分。The promotion score corresponding to the promotion cooperation object is determined according to the number of fans of the promotion cooperation object and the historical promotion score.
在其中一个实施例中，所述获取所述历史转化率排名中预设名次内的推广产品，根据所述预设名次内的推广产品确定所述推广合作对象的历史推广得分包括：In one embodiment, the obtaining of the promotion products in a preset ranking in the historical conversion rate ranking, and determining the historical promotion score of the promotion partner according to the promotion products in the preset ranking includes:
获取所述历史转化率排名中预设名次内的推广产品对应的评论文本信息；Obtain the comment text information corresponding to the promoted product in the preset ranking in the historical conversion rate ranking;
基于文本情感分析，提取所述评论文本信息对应的情感特征；Based on text sentiment analysis, the sentiment features corresponding to the comment text information are extracted;
根据所述情感特征获取所述评论文本信息对应的情感分类结果；Obtain the sentiment classification result corresponding to the comment text information according to the sentiment feature;
根据所述情感分类结果，获取所述历史转化率排名中预设名次内的推广产品对应的历史推广得分；According to the sentiment classification result, obtain the historical promotion score corresponding to the promotion product in the preset ranking in the historical conversion rate ranking;
根据所述推广产品对应的历史推广得分的平均值，确定所述推广合作对象的历史推广得分。The historical promotion score of the promotion partner is determined according to the average value of the historical promotion score corresponding to the promotion product.
在其中一个实施例中，所述根据所述情感特征获取所述评论文本信息对应的情感分类结果包括：In one embodiment, the obtaining the sentiment classification result corresponding to the comment text information according to the sentiment feature includes:
将所述情感特征按照特征类型进行组合，获取情感特征组合数据；combining the emotional features according to feature types to obtain emotional feature combination data;
将所述情感特征组合数据输入预设多通道卷积神经网络，通过所述预设多通道卷积神经网络的Sigmoid函数输出评论文本信息对应的情感分类结果。The emotion feature combination data is input into a preset multi-channel convolutional neural network, and the emotion classification result corresponding to the comment text information is output through the sigmoid function of the preset multi-channel convolutional neural network.
在其中一个实施例中，所述根据所述情感分类结果获取所述推广产品对应的历史推广得分包括：In one embodiment, the obtaining the historical promotion score corresponding to the promoted product according to the emotion classification result includes:
根据所述推广产品对应的所有情感分类结果，确定所述推广产品对应的评论情感极性值；According to all sentiment classification results corresponding to the promoted product, determine the comment sentiment polarity value corresponding to the promoted product;
获取所述推广产品对应的浏览参数；Obtain the browsing parameters corresponding to the promoted product;
根据所述推广产品对应的浏览参数以及所述评论情感极性值，获取所述推广产品对应的历史推广得分。According to the browsing parameter corresponding to the promoted product and the sentiment polarity value of the comment, the historical promotion score corresponding to the promoted product is obtained.
在其中一个实施例中，所述当接收到推荐请求时，根据所述推荐请求中的推广产品类别对应的所述对象排名，推送对应的推广合作对象包括：In one embodiment, when a recommendation request is received, according to the object ranking corresponding to the promotion product category in the recommendation request, pushing the corresponding promotion cooperation object includes:
当接收到推荐请求时，获取所述推荐请求所请求的当前推广产品类别以及请求对象数；When a recommendation request is received, obtain the currently promoted product category and the number of requested objects requested by the recommendation request;
确定所述当前推广产品类别对应的对象排名；determining the object ranking corresponding to the currently promoted product category;
根据所述请求对象数以及所述对象排名，推送对应数量的推广合作对象。According to the number of requested objects and the object ranking, a corresponding number of promotion cooperation objects are pushed.
一种推广合作对象推荐装置，所述装置包括：A device for recommending a promotion partner, the device comprising:
数据接收单元，用于获取推广合作对象的基本数据及其历史推广数据，所述基本数据包括粉丝数目以及推广产品类别；a data receiving unit, used to obtain basic data of the promotion partner and historical promotion data, the basic data includes the number of fans and the category of the promotion product;
所述数据接收单元还用于接收推荐请求；The data receiving unit is further configured to receive a recommendation request;
数据处理单元，用于根据所述推广合作对象的粉丝数目以及所述历史推广数据，确定所述推广合作对象对应的推广得分；根据所述推广得分以及所述推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；当接收到推荐请求时，根据所述推荐请求中的推广产品类别对应的所述对象排名，推送对应的推广合作对象。a data processing unit, configured to determine the promotion score corresponding to the promotion cooperation object according to the number of fans of the promotion cooperation object and the historical promotion data; determine the promotion score according to the promotion score and the promotion product category of the promotion cooperation object The object ranking of promotion cooperation objects under each promotion product category; when a recommendation request is received, the corresponding promotion cooperation object is pushed according to the object ranking corresponding to the promotion product category in the recommendation request.
在其中一个实施例中，所述历史推广数据包括所述推广合作对象推广产品的历史转化率数据；所述数据处理单元还用于：根据所述推广合作对象所推广产品的历史转化率数据，生成多个所述推广产品对应的历史转化率排名；获取所述历史转化率排名中预设名次内的推广产品，根据所述预设名次内的推广产品确定所述推广合作对象的历史推广得分；根据所述推广合作对象的粉丝数目以及所述历史推广得分，确定所述推广合作对象对应的推广得分。In one embodiment, the historical promotion data includes historical conversion rate data of the product promoted by the promotion partner; the data processing unit is further configured to: according to the historical conversion rate data of the product promoted by the promotion partner, generating a plurality of historical conversion rate rankings corresponding to the promotion products; obtaining promotion products within a preset ranking in the historical conversion rate ranking, and determining the historical promotion score of the promotion partner according to the promotion products in the preset ranking ; According to the number of fans of the promotion cooperation object and the historical promotion score, determine the promotion score corresponding to the promotion cooperation object.
一种计算机设备，包括存储器和处理器，所述存储器存储有计算机程序，所述处理器执行所述计算机程序时实现以下步骤：A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:
获取推广合作对象的基本数据及其历史推广数据，所述基本数据包括粉丝数目以及推广产品类别；Obtain the basic data of the promotion partner and its historical promotion data, the basic data includes the number of fans and the category of the promoted product;
根据所述推广合作对象的粉丝数目以及所述历史推广数据，确定所述推广合作对象对应的推广得分；According to the number of fans of the promotion partner and the historical promotion data, determine the promotion score corresponding to the promotion partner;
根据所述推广得分以及所述推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；According to the promotion score and the promotion product category of the promotion cooperation object, determine the object ranking of the promotion cooperation object under each promotion product category;
当接收到推荐请求时，根据所述推荐请求中的推广产品类别对应的所述对象排名，推送对应的推广合作对象。When a recommendation request is received, the corresponding promotion cooperation object is pushed according to the object ranking corresponding to the promotion product category in the recommendation request.
一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现以下步骤：A computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:
获取推广合作对象的基本数据及其历史推广数据，所述基本数据包括粉丝数目以及推广产品类别；Obtain the basic data of the promotion partner and its historical promotion data, the basic data includes the number of fans and the category of the promoted product;
根据所述推广合作对象的粉丝数目以及所述历史推广数据，确定所述推广合作对象对应的推广得分；According to the number of fans of the promotion partner and the historical promotion data, determine the promotion score corresponding to the promotion partner;
根据所述推广得分以及所述推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；According to the promotion score and the promotion product category of the promotion cooperation object, determine the object ranking of the promotion cooperation object under each promotion product category;
当接收到推荐请求时，根据所述推荐请求中的推广产品类别对应的所述对象排名，推送对应的推广合作对象。When a recommendation request is received, the corresponding promotion cooperation object is pushed according to the object ranking corresponding to the promotion product category in the recommendation request.
上述推广合作对象推荐方法、装置、计算机设备和存储介质，方法通过获取推广合作对象的基本数据及其历史推广数据，基本数据包括粉丝数目以及推广产品类别；根据推广合作对象的粉丝数目及其历史推广数据，确定推广合作对象对应的推广得分；根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；当接收到推荐请求时，根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。本申请在进行推广合作对象推荐时，先根据推广合作对象的基本数据及其历史推广数据来确定推广合作对象对应的推广得分，而后根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名，在接收到推荐请求，需要推荐推广合作对象时，则可直接根据推广产品类别以及对象排名，推送对应的推广合作对象，本申请通过历史推广数据以及推广产品类别来对推广合作对象进行排名，依照排名选择推广合作对象进行推荐，从而选择更有效的推广合作对象，来提高推广产品转化率。The above recommendation method, device, computer equipment and storage medium for the promotion partner, the method obtains the basic data of the promotion partner and historical promotion data, the basic data includes the number of fans and the category of the promoted product; according to the number of fans of the promotion partner and their history Promotion data, determine the promotion score corresponding to the promotion partner; according to the promotion score and the promotion product category of the promotion partner, determine the object ranking of the promotion partner under each promotion product category; when a recommendation request is received, according to the promotion in the recommendation request. The object ranking corresponding to the product category will push the corresponding promotion cooperation object. When recommending promotion partners in this application, firstly, the promotion score corresponding to the promotion partner is determined according to the basic data of the promotion partner and its historical promotion data, and then each promotion product is determined according to the promotion score and the promotion product category of the promotion partner. The object ranking of the promotion partners under the category. When a recommendation request is received and a promotion partner needs to be recommended, the corresponding promotion partner can be pushed directly according to the promotion product category and object ranking. This application uses historical promotion data and promotion product categories. To rank the promotion partners, select the promotion partners to recommend according to the ranking, so as to select more effective promotion partners to improve the conversion rate of the promoted products.
附图说明Description of drawings
图1为一个实施例中推广合作对象推荐方法的应用环境图；Fig. 1 is the application environment diagram of promoting the method for recommending cooperation objects in one embodiment;
图2为一个实施例中推广合作对象推荐方法的流程示意图；2 is a schematic flowchart of a method for promoting a partner recommendation in an embodiment;
图3为一个实施例中图2中步骤203的子流程示意图；Fig. 3 is a sub-flow schematic diagram of step 203 in Fig. 2 in one embodiment;
图4为一个实施例中图3中步骤304的子流程示意图；FIG. 4 is a schematic sub-flow diagram of step 304 in FIG. 3 in one embodiment;
图5为一个实施例中图4中步骤407的子流程示意图；Fig. 5 is a sub-flow schematic diagram of step 407 in Fig. 4 in one embodiment;
图6为一个实施例中图2中步骤207的子流程示意图；Fig. 6 is a sub-flow schematic diagram of step 207 in Fig. 2 in one embodiment;
图7为一个实施例中推广合作对象推荐装置的结构框图；Fig. 7 is a structural block diagram of an apparatus for promoting cooperation object recommendation in one embodiment;
图8为一个实施例中计算机设备的内部结构图。FIG. 8 is a diagram of the internal structure of a computer device in one embodiment.
具体实施方式Detailed ways
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.
本申请提供的推广合作对象推荐方法，可以应用于如图1所示的应用环境中。其中，终端102通过网络与服务器104通过网络进行通信。终端102可以通过网络向服务器104发送推广合作对象的基本数据及其历史推广数据，同时，终端102还可以在需要取得推广合作对象的推荐时，向服务器104发送推荐请求。而后服务器104获取推广合作对象的基本数据及其历史推广数据，基本数据包括粉丝数目以及推广产品类别；根据推广合作对象的粉丝数目及其历史推广数据，确定推广合作对象对应的推广得分；根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；当接收到推荐请求时，根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。其中，终端102可以但不限于是各种个人计算机、笔记本电脑、智能手机、平板电脑和便携式可穿戴设备，服务器104可以用独立的服务器或者是多个服务器组成的服务器集群来实现。The method for recommending cooperation objects for promotion provided in this application can be applied to the application environment shown in FIG. 1 . The terminal 102 communicates with the
在一个实施例中，如图2所示，提供了一种推广合作对象推荐方法，以该方法应用于图1中的终端102为例进行说明，包括以下步骤：In one embodiment, as shown in FIG. 2 , a method for recommending a promotion partner is provided, and the method is applied to the terminal 102 in FIG. 1 as an example for description, including the following steps:
步骤201，获取推广合作对象的基本数据及其历史推广数据，基本数据包括粉丝数目以及推广产品类别。Step 201: Acquire basic data of the promotion partner and historical promotion data, where the basic data includes the number of fans and the category of the promoted product.
其中，推广合作对象是指可以帮助进行产品推广的对象，可以是团体也可以是个人。一般地，推广合作对象具体为可以在互联网上帮助进行产品推广的对象，如某些“带货”的网络主播、网红或者网络达人。这些推广合作对象可以以直播等形式向其粉丝进行产品推广。推广合作对象的基本数据包括但不限于推广合作对象的名字、粉丝数以及推广产品类别等。其中，推广产品类别具体是指推广合作对象为自己设定的类别，包括了当前推广合作对象可以推广哪些类别下的产品，包括但不限于食品、美妆、健身、数码等。而历史推广数据即推广合作对象以前推广过的产品信息，包括但不限于产品名、详细类别，价格，转化率信息等。Among them, the object of promotion cooperation refers to the object who can help to promote the product, which can be a group or an individual. Generally, the target of promotion cooperation is the target who can help to promote the product on the Internet, such as some network anchors, Internet celebrities or Internet experts who "carry goods". These promotion partners can promote products to their fans in the form of live broadcasts. The basic data of the promotion partner includes but is not limited to the name of the promotion partner, the number of fans, and the category of the promoted product. Among them, the promotion product category specifically refers to the category set by the promotion partner for themselves, including which categories of products the current promotion partner can promote, including but not limited to food, beauty, fitness, digital, etc. The historical promotion data refers to the product information previously promoted by the promotion partner, including but not limited to product name, detailed category, price, conversion rate information, etc.
具体地，本申请主要用于在产品需要进行推广时，为需要推广的产品来推荐对应的推广合作对象。因此，可以在进行产品推广前，先根据推广合作对象的基本数据及其历史推广数据等信息，来确定各个预设的推广产品类别下，推广合作对象的排名，从而可以在某个产品需要进行推广合作对象推荐时，可以根据产品所属的推广产品类别，来为其推荐对应的推广合作对象。因此，在进行推广合作对象推荐时，可以先获取推广合作对象的基本数据及其历史推广数据，根据这些数据来对各个现有的推广合作对象进行后续分析。Specifically, this application is mainly used to recommend corresponding promotion partners for the products to be promoted when the products need to be promoted. Therefore, before product promotion, you can first determine the ranking of the promotion partners under each preset promotion product category based on the basic data of the promotion partners and their historical promotion data, so that the promotion partners can be promoted when a product needs to be promoted. When recommending a promotion partner, you can recommend the corresponding promotion partner according to the promotion product category to which the product belongs. Therefore, when recommending promotion partners, the basic data and historical promotion data of the promotion partners can be obtained first, and subsequent analysis of each existing promotion partner can be performed according to these data.
步骤203，根据推广合作对象的粉丝数目及其历史推广数据，确定推广合作对象对应的推广得分。Step 203: Determine the promotion score corresponding to the promotion cooperation object according to the number of fans of the promotion cooperation object and its historical promotion data.
其中，推广合作对象对应的推广得分，具体是根据推广合作对象的粉丝数以及推广合作对象对各种类型的产品推广的推广成绩，来综合分析所得到的一个数据，具体用于展示当前的推广合作对象在产品推广领域所得的分数。Among them, the promotion score corresponding to the promotion partner is a piece of data obtained by comprehensive analysis based on the number of fans of the promotion partner and the promotion results of the promotion partner on various types of product promotion, and is specifically used to display the current promotion The score obtained by the partner in the field of product promotion.
具体地，为了能够将推广合作对象的推广效果进行量化，可以在推广合作对象推荐前，基于推广合作对象的粉丝数目及其历史推广数据，来确定推广合作对象的推广得分。具体地，可以选取历史推广数据中转化率较高的几个产品，来作为确定推广合作对象推广得分的依据。而粉丝的数量显然可以影响最终的推广效果，因此，可以综合粉丝数目及其历史推广数据，来确定推广合作对象对应的推广得分。Specifically, in order to be able to quantify the promotion effect of the promotion partner, the promotion score of the promotion partner may be determined based on the number of fans of the promotion partner and historical promotion data before the promotion partner is recommended. Specifically, several products with higher conversion rates in the historical promotion data may be selected as the basis for determining the promotion score of the promotion partner. The number of fans can obviously affect the final promotion effect. Therefore, the number of fans and their historical promotion data can be combined to determine the promotion score corresponding to the promotion partner.
步骤205，根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名。Step 205 , according to the promotion score and the promotion product category of the promotion cooperation object, determine the object ranking of the promotion cooperation object under each promotion product category.
具体地，推广合作对象的基本数据中包括了其对应的推广产品类别。因此，在得到所有推广合作对象对应的推广得分后，可以针对每个推广产品类别，来进行推广合作对象的推广得分排名，确定各个推广产品类别下推广合作对象的对象排名，分数越高的推广合作对象在对象排名中越靠前。Specifically, the basic data of the promotion partner includes its corresponding promotion product category. Therefore, after obtaining the promotion scores corresponding to all the promotion partners, the promotion score ranking of the promotion partners can be carried out for each promotion product category, and the object ranking of the promotion partners under each promotion product category can be determined. The higher the partner is in the object ranking.
步骤207，当接收到推荐请求时，根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。Step 207: When the recommendation request is received, push the corresponding promotion cooperation object according to the object ranking corresponding to the promotion product category in the recommendation request.
其中，推荐请求可以由终端102发送，用于请求服务器104来进行对应推广产品类别的推广合作对象推荐。The recommendation request may be sent by the terminal 102 for requesting the
具体地，在建立完各个推广产品类别下推广合作对象的对象排名后，当终端102方的工作需求为指定的产品推荐推广合作对象时，可以发送包括该指定的产品所属推广产品类别的推荐请求至服务器104。服务器104根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。Specifically, after establishing the object ranking of the promotion cooperation objects under each promotion product category, when the work requirement of the terminal 102 is the specified product recommendation promotion cooperation object, a recommendation request including the promotion product category to which the specified product belongs can be sent. to
上述推广合作对象推荐方法，通过获取推广合作对象的基本数据及其历史推广数据，基本数据包括粉丝数目以及推广产品类别；根据推广合作对象的粉丝数目及其历史推广数据，确定推广合作对象对应的推广得分；根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；当接收到推荐请求时，根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。本申请在进行推广合作对象推荐时，先根据推广合作对象的基本数据及其历史推广数据来确定推广合作对象对应的推广得分，而后根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名，在接收到推荐请求，需要推荐推广合作对象时，则可直接根据推广产品类别以及对象排名，推送对应的推广合作对象，本申请通过历史推广数据以及推广产品类别来对推广合作对象进行排名，依照排名选择推广合作对象进行推荐，从而选择更有效的推广合作对象，来提高推广产品的转化率。The above promotion partner recommendation method obtains the basic data of the promotion partner and its historical promotion data. The basic data includes the number of fans and the category of the promotion product; according to the number of fans of the promotion partner and their historical promotion data, the corresponding promotion partner is determined. Promotion score; according to the promotion score and the promotion product category of the promotion partner, determine the object ranking of the promotion partner under each promotion product category; when a recommendation request is received, according to the object ranking corresponding to the promotion product category in the recommendation request, push the corresponding promotion partners. When recommending promotion partners in this application, firstly, the promotion score corresponding to the promotion partner is determined according to the basic data of the promotion partner and its historical promotion data, and then each promotion product is determined according to the promotion score and the promotion product category of the promotion partner. The object ranking of the promotion partners under the category. When a recommendation request is received and a promotion partner needs to be recommended, the corresponding promotion partner can be pushed directly according to the promotion product category and object ranking. This application uses historical promotion data and promotion product categories. To rank the promotion partners, select the promotion partners to recommend according to the ranking, so as to select more effective promotion partners to improve the conversion rate of the promoted products.
在一个实施例中，推广合作对象的历史推广数据包括推广合作对象所推广产品的历史转化率数据。如图3所示，步骤203包括：In one embodiment, the historical promotion data of the promotion partner includes historical conversion rate data of the products promoted by the promotion partner. As shown in Figure 3, step 203 includes:
步骤302，根据推广合作对象所推广产品的历史转化率数据，生成多个推广产品对应的历史转化率排名。Step 302 , according to the historical conversion rate data of the products promoted by the promotion partner, generate historical conversion rate rankings corresponding to the plurality of promoted products.
步骤304，获取历史转化率排名中预设名次内的推广产品，根据预设名次内的推广产品确定推广合作对象的历史推广得分。Step 304: Acquire the promotion products in the preset ranking in the historical conversion rate ranking, and determine the historical promotion score of the promotion partner according to the promotion products in the preset ranking.
步骤306，根据推广合作对象的粉丝数目以及历史推广得分，确定推广合作对象对应的推广得分。Step 306: Determine the promotion score corresponding to the promotion cooperation object according to the number of fans of the promotion cooperation object and the historical promotion score.
其中，转化率指在一个统计周期内，完成转化行为的次数占推广信息总点击次数的比率。计算公式为：转化率＝(转化次数/点击量)×100％。例如：10名用户看到某个搜索推广的结果，其中5名用户点击了某一推广结果并被跳转到目标URL上，之后，其中2名用户有了后续转化的行为。那么，这条推广结果的转化率就是(2/5)×100％＝40％。在其中一个实施例中，预设名次具体可以推广合作对象所推广产品的数目来确定。而此处的推广合作对象的历史推广得分，则用于体现推广合作对象在推广预设名次内的推广产品时的推广成绩体现。通过历史推广得分综合推广合作对象的粉丝数目，可以得出最终的推广合作对象对应的推广得分。The conversion rate refers to the ratio of the number of completed conversion behaviors to the total number of clicks on the promotion information in a statistical period. The calculation formula is: conversion rate=(number of conversions/clicks)×100%. For example: 10 users saw the results of a certain search promotion, 5 of them clicked on a certain promotion result and were redirected to the target URL, and then 2 of them had subsequent conversion behaviors. Then, the conversion rate of this promotion result is (2/5)×100%=40%. In one embodiment, the preset ranking can be specifically determined by the number of products promoted by the promotion partner. The historical promotion score of the promotion partner here is used to reflect the promotion performance of the promotion partner when promoting the promotion product within the preset ranking. Through the historical promotion score and the number of fans of the promotion partner, the promotion score corresponding to the final promotion partner can be obtained.
具体地，在确定推广合作对象对应的推广得分时，可以先参考历史推广数据中的转化率数据，针对每个推广合作对象，生成其推广产品对应的历史转化率排名，而后选取转化率较高的推广产品，依据这些推广产品的推广效果来确定推广合作对象的历史推广得分。具体地，可以计算出转化率较高的推广产品中，每个推广产品对应的历史推广得分，而后通过加权平均得到推广合作对象的最终历史推广得分。而综合历史推广得分History_Score(U)与推广合作对象的粉丝数Fans，得到推广合作对象对应的推广得分Score(U)。Specifically, when determining the promotion score corresponding to the promotion partner, you can first refer to the conversion rate data in the historical promotion data, and for each promotion partner, generate the historical conversion rate ranking corresponding to the promoted product, and then select the higher conversion rate. According to the promotion effect of these promotion products, the historical promotion score of the promotion partner is determined. Specifically, the historical promotion score corresponding to each promotion product among the promotion products with a high conversion rate can be calculated, and then the final historical promotion score of the promotion partner can be obtained through a weighted average. And the comprehensive historical promotion score History_Score(U) and the number of fans Fans of the promotion partner get the promotion score Score(U) corresponding to the promotion partner.
其中，wh为历史推广得分的权重，wf为粉丝数的权重，一般认为历史推广得分更具参考性，故wh＞wf。本实施例中，通过获取历史转化率排名中预设名次内的推广产品，来计算推广合作对象的历史推广得分，可以有效保证历史推广得分计算的准确率，从而保证推广合作对象推荐的准确率。Among them, wh is the weight of the historical promotion score, and wf is the weight of the number of fans. It is generally considered that the historical promotion score is more informative, so wh>wf. In this embodiment, the historical promotion score of the promotion partner is calculated by obtaining the promotion products within the preset ranks in the historical conversion rate ranking, which can effectively ensure the accuracy of the calculation of the historical promotion score, thereby ensuring the accuracy of the promotion partner recommendation. .
在一个实施例中，如图4所示，步骤304包括：In one embodiment, as shown in FIG. 4, step 304 includes:
步骤401，获取历史转化率排名中预设名次内的推广产品对应的评论文本信息。Step 401: Obtain comment text information corresponding to the promoted product in the preset ranking in the historical conversion rate ranking.
步骤403，基于文本情感分析，提取评论文本信息对应的情感特征。Step 403 , based on the text sentiment analysis, extract sentiment features corresponding to the comment text information.
步骤405，根据情感特征获取评论文本信息对应的情感分类结果。Step 405: Obtain the sentiment classification result corresponding to the comment text information according to the sentiment feature.
步骤407，根据情感分类结果，获取历史转化率排名中预设名次内的推广产品对应的历史推广得分。Step 407: According to the sentiment classification result, obtain the historical promotion score corresponding to the promotion product in the preset ranking in the historical conversion rate ranking.
步骤409，根据推广产品对应的历史推广得分的平均值，确定推广合作对象的历史推广得分。Step 409: Determine the historical promotion score of the promotion partner according to the average value of the historical promotion score corresponding to the promotion product.
其中，文本情感分析又称意见挖掘、倾向性分析等。简单而言,是对带有情感色彩的主观性文本进行分析、处理、归纳和推理的过程。互联网(如博客和论坛以及社会服务网络如大众点评)上产生了大量的用户参与的、对于诸如人物、事件、产品等有价值的评论信息。这些评论信息表达了人们的各种情感色彩和情感倾向性,如喜、怒、哀、乐和批评、赞扬等。基于此，潜在的用户就可以通过浏览这些主观色彩的评论来了解大众舆论对于某一事件或产品的看法。Among them, text sentiment analysis is also called opinion mining, tendency analysis, etc. In short, it is the process of analyzing, processing, summarizing and reasoning about subjective texts with emotional colors. The Internet (such as blogs and forums, as well as social service networks such as Dianping) has produced a large number of user-participated, valuable comments about people, events, and products. These comments express people's various emotional colors and emotional tendencies, such as joy, anger, sadness, joy, criticism and praise. Based on this, potential users can browse these subjective comments to understand the public opinion on an event or product.
具体地，本申请中可以结合推广产品对应的评论文本信息来进行历史推广得分的计算，综合用户在推广产品的评论，对推广的效果可以进行有效地识别。在历史推广得分计算时，通过文本情感分析，提取评论文本信息对应的情感特征，并进行相应的情感分类，而后基于情感分类结果来计算推广合作对象对应各推广产品的历史推广得分。最终根据历史转化率排名中预设名次内的推广产品对应的历史推广得分的平均值，确定推广合作对象的历史推广得分本实施例中，通过文本情感分析，来提取评论文本信息对应的情感特征，从而进行推广效果地识别，可以有效保证推广合作对象的历史推广得分计算的准确率，保证推荐效果。Specifically, in this application, the historical promotion score can be calculated in combination with the comment text information corresponding to the promoted product, and the effect of the promotion can be effectively identified by synthesizing user comments on the promoted product. When calculating the historical promotion score, the emotional features corresponding to the comment text information are extracted through text sentiment analysis, and the corresponding sentiment classification is performed, and then the historical promotion score of the promotion partner corresponding to each promotion product is calculated based on the sentiment classification result. Finally, according to the average value of the historical promotion scores corresponding to the promotion products in the preset ranks in the historical conversion rate ranking, the historical promotion scores of the promotion partners are determined. In this embodiment, the emotional features corresponding to the comment text information are extracted through text emotion analysis. , so as to identify the promotion effect, which can effectively ensure the accuracy of the historical promotion score calculation of the promotion partner and ensure the recommendation effect.
在其中一个实施例中，步骤405包括：将情感特征按照特征类型进行组合，获取情感特征组合数据；将情感特征组合数据输入预设多通道卷积神经网络，通过预设多通道卷积神经网络的Sigmoid函数输出评论文本信息对应的情感分类结果。In one embodiment, step 405 includes: combining the emotional features according to feature types to obtain emotional feature combination data; inputting the emotional feature combination data into a preset multi-channel convolutional neural network, through the preset multi-channel convolutional neural network The Sigmoid function outputs the sentiment classification results corresponding to the comment text information.
其中，情感特征具体可以包括词特征、词性特征、情感符号特征和情感标签特征这四种。Among them, the emotional features may specifically include four types: word features, part-of-speech features, emotional symbol features, and emotional tag features.
词特征：以评论文本句子的词为单位，每个词映射成分布式词向量，在词向量词典中有两列，一列是词语，一列是对应的分布式词向量，将每个句子序列中每个词语对应的词向量依次拼接起来，得到整个句子序列的词向量矩阵。Word feature: Take the word of the comment text sentence as the unit, and each word is mapped to a distributed word vector. There are two columns in the word vector dictionary, one column is the word, and the other is the corresponding distributed word vector. The word vectors corresponding to each word are spliced in turn to obtain the word vector matrix of the entire sentence sequence.
词性特征：为了获取词语的情感的分，采用HowNet情感词典作为基准词典。此外针对不同平台，采集样本数据集，对常用的词语、情感表情符及情感符号进行人工标注，合并不同平台获得的标注情感词与HowNet情感词典，最终得到目标情感词典。不同的词有不同的情感倾向程度，量化词典中各个词的情感倾向程度的计算方法为：Part-of-speech features: In order to obtain the sentiment scores of words, the HowNet sentiment dictionary is used as the benchmark dictionary. In addition, for different platforms, sample data sets are collected, commonly used words, emotional emoticons and emotional symbols are manually labeled, and the labeled emotional words obtained from different platforms are combined with the HowNet emotional dictionary, and finally the target emotional dictionary is obtained. Different words have different degrees of emotional inclination. The calculation method to quantify the degree of emotional inclination of each word in the dictionary is as follows:
Freq(ti)＝|p*Pos(ti)-n*Neg(ti)|Freq(ti)=|p*Pos(ti)-n*Neg(ti)|
其中，Pos(ti)表征第i个情感词ti在积极的文档中出现频率，Neg(ti)表征第i个情感词ti在消极的文档中出现频率。||为取绝对值，[]表示取整。Freq(ti)表征第i个情感词ti在数据集的文档频数。Sen(ti)表示ti的情感得分，p表示积极文档频数的重要程度权重，n表示消极文档频数的重要程度权重。s用于控制情感词得分阈值。Among them, Pos(ti) represents the frequency of the i-th emotional word ti in positive documents, and Neg(ti) represents the frequency of the i-th emotional word ti in negative documents. || is the absolute value, [] means rounding. Freq(ti) represents the document frequency of the i-th sentiment word ti in the dataset. Sen(ti) represents the sentiment score of ti, p represents the importance weight of the frequency of positive documents, and n represents the weight of the importance of the frequency of negative documents. s is used to control the sentiment word score threshold.
情感符号特征：对于每一个句子，如果存在表情符号，将表情符号转化为分布式向量，并在Hownet中找到其对应的情感极性：积极或消极。如果属于积极情感符号，将其词性重新标注为pos，如果属于消极情感符号，将其词性重新标注为neg。若在Hownet中未找到对应的情感极性，则无需重新标注。Sentiment Symbol Features: For each sentence, if there is an emoji, convert the emoji into a distributed vector and find its corresponding sentiment polarity in Hownet: positive or negative. If it is a positive sentiment symbol, re-mark its part of speech as pos, and if it is a negative sentiment symbol, re-mark its part of speech as neg. If the corresponding sentiment polarity is not found in Hownet, there is no need to relabel.
其中，
情感标签特征：为了充分利用标签表达的情感信息，将积极标签、消极标签和无情感标签分别转化为对应的分布式向量。向量化的标签用L表示，L＝{l1，l2，……lm}，其中m是标签的类别数，li为标签yi的向量。Sentiment label features: In order to make full use of the emotional information expressed by the labels, positive labels, negative labels and no-sentiment labels are converted into corresponding distributed vectors respectively. The vectorized labels are denoted by L, L={l1, l2,...lm}, where m is the number of categories of labels, and li is the vector of labels yi.
预设多通道卷积神经网络具体可以为多通道卷积神经网络MF-MCNN。The preset multi-channel convolutional neural network may specifically be a multi-channel convolutional neural network MF-MCNN.
具体地，在进行情感分类的过程中，可以使用多通道卷积神经网络MF-MCNN的输入层使用三个通道来接收句子的不同特征组合：①词特征+情感符号特征+词性特征；②词特征+情感符号特征；③情感标签特征。并将不同特征映射为分布式向量的形式，不同的通道使得模型获取更加丰富的特征信息。三个通道的特征(X1，X2,Yj)经过卷积和池化后进行拼接合并，组成最终的向量输入到全连接层。而后通过预设多通道卷积神经网络的Sigmoid函数输出评论文本信息对应的情感分类结果。采用Sigmoid函数输出待分类句子的情感分类结果，非线性函数变换后，文本和每一个情感标签产生一个匹配分数，在0-1之间，值越大，表明文本的情感类别越接近该情感标签。训练过程中，模型通过最小交叉熵损失函数Loss来调整模型参数，提升分类性能。Specifically, in the process of sentiment classification, the input layer of the multi-channel convolutional neural network MF-MCNN can use three channels to receive different feature combinations of sentences: ① word feature + emotion symbol feature + part of speech feature; ② word feature Feature + emotion symbol feature; ③ emotion label feature. Different features are mapped into the form of distributed vectors, and different channels enable the model to obtain richer feature information. The features of the three channels (X1, X2, Yj) are convolutional and pooled and then spliced and merged to form the final vector input to the fully connected layer. Then, the Sigmoid function of the preset multi-channel convolutional neural network is used to output the sentiment classification result corresponding to the comment text information. The sigmoid function is used to output the sentiment classification result of the sentence to be classified. After the nonlinear function transformation, the text and each sentiment label generate a matching score. Between 0 and 1, the larger the value, the closer the sentiment class of the text is to the sentiment label. . During the training process, the model uses the minimum cross entropy loss function Loss to adjust the model parameters to improve the classification performance.
其中，σ表示sigmoid函数，y表示句子序列对应的真实情感标签，使用0-1向量表示，Yj表示第j个标签提取的最终特征表示。本实施例中，通过预设卷积神经网络来将情感特征按照特征类型进行组合，从而进行评论文本信息对应的情感分类。可以有效保证情感分类处理的准确性。Among them, σ represents the sigmoid function, y represents the real sentiment label corresponding to the sentence sequence, which is represented by a 0-1 vector, and Yj represents the final feature representation extracted from the jth label. In this embodiment, the emotion features are combined according to feature types by using a preset convolutional neural network, so as to perform emotion classification corresponding to the comment text information. It can effectively guarantee the accuracy of emotion classification processing.
在其中一个实施例中，如图5所示，步骤407包括：In one embodiment, as shown in FIG. 5 , step 407 includes:
步骤502，根据推广产品对应的所有情感分类结果，确定推广产品对应的评论情感极性值。Step 502 , according to all the sentiment classification results corresponding to the promoted product, determine the comment sentiment polarity value corresponding to the promoted product.
步骤504，获取推广产品对应的浏览参数。Step 504: Obtain browsing parameters corresponding to the promoted product.
步骤506，根据推广产品对应的浏览参数以及评论情感极性值，获取推广产品对应的历史推广得分。Step 506: Obtain historical promotion scores corresponding to the promoted product according to the browsing parameters corresponding to the promoted product and the comment sentiment polarity value.
其中，浏览参数具体包括浏览量、点击量以及跳转购买量等数据。具体地，可以通过情感分析，来为对应的评论文本添加积极标签、消极标签和无情感标签。而后在情感分类过程中，将文本信息分为不同类型非线性函数变换后，文本和每一个情感标签产生一个匹配分数，在0-1之间，值越大，表明文本的情感类别越接近该情感标签。最终确定评论文本信息对应的情感分类结果为消极还是积极。在分类完成后，可以以不对称二进制来统计推广合作对象U历史推广的产品gi的评论的情感极性数量，用1表示积极评论，0表示消极评论，并归一化，最终推广合作对象U历史推广的产品gi的评论情感极性SenScore(Ugi)值越靠近1，评论越积极。而后在计算历史推广得分时，则可以将商品的评论情感极性SenScore，点击量D，浏览量V，跳转购买量B结合计算出达人U推广该商品gi的历史推广得分(百分制)。The browsing parameters specifically include data such as pageviews, clicks, and jump purchases. Specifically, sentiment analysis can be used to add positive labels, negative labels and no sentiment labels to the corresponding comment text. Then in the sentiment classification process, after dividing the text information into different types of nonlinear functions, a matching score is generated between the text and each sentiment label. Between 0 and 1, the larger the value, the closer the sentiment category of the text is. Emotional labels. Finally, determine whether the sentiment classification result corresponding to the comment text information is negative or positive. After the classification is completed, the number of sentiment polarities of the comments of the product gi that the promotion partner U has historically promoted can be counted by asymmetric binary, with 1 representing positive comments and 0 representing negative comments, and normalized, and finally the partner U is promoted. The comment sentiment polarity SenScore(U gi ) value of the historically promoted product gi is closer to 1, the more positive the comment is. Then, when calculating the historical promotion score, you can combine the product's comment sentiment polarity SenScore, click volume D, pageview volume V, and jump purchase volume B to calculate the historical promotion score (percentage system) of the product gi promoted by Daren U.
其中，ws、wd、wv以及wb分别为评论情感极性、点击量、浏览量以及跳转购买量所对应的权重。该权重可以根据实际推广效果来进行调整。本实施例中，通过综合推广产品对应的浏览参数Among them, ws, wd, wv and wb are the weights corresponding to comment sentiment polarity, click volume, pageview volume and jump purchase volume, respectively. The weight can be adjusted according to the actual promotion effect. In this embodiment, the browsing parameters corresponding to the products are comprehensively promoted
在其中一个实施例中，如图6所示，步骤207包括：In one embodiment, as shown in FIG. 6 , step 207 includes:
步骤601，当接收到推荐请求时，获取推荐请求所请求的当前推广产品类别以及请求对象数。Step 601, when a recommendation request is received, obtain the currently promoted product category and the number of requested objects requested by the recommendation request.
步骤603，确定当前推广产品类别对应的对象排名。Step 603: Determine the object ranking corresponding to the currently promoted product category.
步骤605，根据请求对象数以及对象排名，推送对应数量的推广合作对象。Step 605: Push a corresponding number of promotion cooperation objects according to the number of requested objects and the object ranking.
具体地，推荐请求包括所请求的当前推广产品类别，从而确定需要推荐的产品所对应的对象排名。推荐请求中还可以指定所需推广合作对象的数量。而后服务器104可以根据对象排名，选取数量对应排名范围内的推广合作对象推送至终端102。本实施例中，服务器104根据当前推广产品类别以及请求对象数可以有效确定终端102所需要的推荐信息，从而实现对应数量的推广合作对象的推送，保证推送范围的有效性。Specifically, the recommendation request includes the requested currently promoted product category, so as to determine the object ranking corresponding to the product to be recommended. The referral request can also specify the number of partners you want to promote. Then, the
应该理解的是，虽然图2-6的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图2-6中的至少一部分步骤可以包括多个步骤或者多个阶段，这些步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤中的步骤或者阶段的至少一部分轮流或者交替地执行。It should be understood that although the steps in the flowcharts of FIGS. 2-6 are shown in sequence according to the arrows, these steps are not necessarily executed in the sequence shown by the arrows. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and these steps may be performed in other orders. Moreover, at least a part of the steps in FIGS. 2-6 may include multiple steps or multiple stages. These steps or stages are not necessarily executed and completed at the same time, but may be executed at different times. The execution of these steps or stages The order is also not necessarily sequential, but may be performed alternately or alternately with other steps or at least a portion of the steps or phases within the other steps.
在一个实施例中，如图7所示，提供了一种推广合作对象推荐装置，包括：In one embodiment, as shown in FIG. 7 , a device for recommending cooperation objects for promotion is provided, including:
数据接收单元702，用于获取推广合作对象的基本数据及其历史推广数据，基本数据包括粉丝数目以及推广产品类别；The
数据接收单元702还用于接收推荐请求；The
数据处理单元704，用于根据推广合作对象的粉丝数目及其历史推广数据，确定推广合作对象对应的推广得分；根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；当接收到推荐请求时，根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。The
在其中一个实施例中，历史推广数据包括推广合作对象推广产品的历史转化率数据；数据处理单元704还用于：根据推广合作对象所推广产品的历史转化率数据，生成多个推广产品对应的历史转化率排名；获取历史转化率排名中预设名次内的推广产品，根据预设名次内的推广产品确定推广合作对象的历史推广得分；根据推广合作对象的粉丝数目以及历史推广得分，确定推广合作对象对应的推广得分。In one embodiment, the historical promotion data includes historical conversion rate data of the products promoted by the promotion partner; the
在其中一个实施例中，获取历史转化率排名中预设名次内的推广产品，数据处理单元704还用于：获取历史转化率排名中预设名次内的推广产品对应的评论文本信息；基于文本情感分析，提取评论文本信息对应的情感特征；根据情感特征获取评论文本信息对应的情感分类结果；根据情感分类结果，获取历史转化率排名中预设名次内的推广产品对应的历史推广得分；根据推广产品对应的历史推广得分的平均值，确定推广合作对象的历史推广得分。In one embodiment, to obtain the promoted products in the preset ranking in the historical conversion rate ranking, the
在其中一个实施例中，数据处理单元704还用于：将情感特征按照特征类型进行组合，获取情感特征组合数据；将情感特征组合数据输入预设多通道卷积神经网络，通过预设多通道卷积神经网络的Sigmoid函数输出评论文本信息对应的情感分类结果。In one embodiment, the
在其中一个实施例中，数据处理单元704还用于：根据推广产品对应的所有情感分类结果，确定推广产品对应的评论情感极性值；获取推广产品对应的浏览参数；根据推广产品对应的浏览参数以及评论情感极性值，获取推广产品对应的历史推广得分。In one embodiment, the
在其中一个实施例中，数据处理单元704还用于：当接收到推荐请求时，获取推荐请求所请求的当前推广产品类别以及请求对象数；确定当前推广产品类别对应的对象排名；根据请求对象数以及对象排名，推送对应数量的推广合作对象。In one embodiment, the
关于推广合作对象推荐装置的具体限定可以参见上文中对于推广合作对象推荐方法的限定，在此不再赘述。上述推广合作对象推荐装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。For the specific limitation of the apparatus for recommending a promotion partner, please refer to the above limitation on the method for recommending a partner to be promoted, which will not be repeated here. All or part of the modules in the above-mentioned device for recommending objects for promotion and cooperation can be implemented by software, hardware and combinations thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
在一个实施例中，提供了一种计算机设备，该计算机设备可以是服务器，其内部结构图可以如图8所示。该计算机设备包括通过系统总线连接的处理器、存储器和网络接口。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统、计算机程序和数据库。该内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该计算机设备的数据库用于存储推广合作对象推荐数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机程序被处理器执行时以实现一种推广合作对象推荐方法。In one embodiment, a computer device is provided, and the computer device may be a server, and its internal structure diagram may be as shown in FIG. 8 . The computer device includes a processor, memory, and a network interface connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The nonvolatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing recommendation data of promotion cooperation objects. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for recommending cooperative objects for promotion is realized.
本领域技术人员可以理解，图8中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。Those skilled in the art can understand that the structure shown in FIG. 8 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.
在一个实施例中，提供了一种计算机设备，包括存储器和处理器，存储器中存储有计算机程序，该处理器执行计算机程序时实现以下步骤：In one embodiment, a computer device is provided, including a memory and a processor, a computer program is stored in the memory, and the processor implements the following steps when executing the computer program:
获取推广合作对象的基本数据及其历史推广数据，基本数据包括粉丝数目以及推广产品类别；Obtain the basic data of promotion partners and their historical promotion data, including the number of fans and the category of products to be promoted;
根据推广合作对象的粉丝数目及其历史推广数据，确定推广合作对象对应的推广得分；According to the number of fans of the promotion partner and their historical promotion data, determine the promotion score corresponding to the promotion partner;
根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；According to the promotion score and the promotion product category of the promotion partner, determine the object ranking of the promotion partner under each promotion product category;
当接收到推荐请求时，根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。When a recommendation request is received, the corresponding promotion cooperation object is pushed according to the object ranking corresponding to the promotion product category in the recommendation request.
在一个实施例中，处理器执行计算机程序时还实现以下步骤：根据推广合作对象所推广产品的历史转化率数据，生成多个推广产品对应的历史转化率排名；获取历史转化率排名中预设名次内的推广产品，根据预设名次内的推广产品确定推广合作对象的历史推广得分；根据推广合作对象的粉丝数目以及历史推广得分，确定推广合作对象对应的推广得分。In one embodiment, the processor further implements the following steps when executing the computer program: generating historical conversion rate rankings corresponding to a plurality of promoted products according to historical conversion rate data of the products promoted by the promotion partner; obtaining presets in the historical conversion rate rankings For the promotion products in the ranking, the historical promotion score of the promotion partner is determined according to the promotion product in the preset ranking; according to the number of fans of the promotion partner and the historical promotion score, the promotion score corresponding to the promotion partner is determined.
在一个实施例中，处理器执行计算机程序时还实现以下步骤：获取历史转化率排名中预设名次内的推广产品对应的评论文本信息；基于文本情感分析，提取评论文本信息对应的情感特征；根据情感特征获取评论文本信息对应的情感分类结果；根据情感分类结果，获取历史转化率排名中预设名次内的推广产品对应的历史推广得分；根据推广产品对应的历史推广得分的平均值，确定推广合作对象的历史推广得分。In one embodiment, the processor also implements the following steps when executing the computer program: obtaining comment text information corresponding to the promotion product in the preset ranking in the historical conversion rate ranking; based on text sentiment analysis, extracting the sentiment feature corresponding to the comment text information; Obtain the sentiment classification result corresponding to the comment text information according to the sentiment feature; obtain the historical promotion score corresponding to the promoted product in the preset ranking in the historical conversion rate ranking according to the sentiment classification result; determine the average value of the historical promotion score corresponding to the promoted product The historical promotion score of the promotion partner.
在一个实施例中，处理器执行计算机程序时还实现以下步骤：将情感特征按照特征类型进行组合，获取情感特征组合数据；将情感特征组合数据输入预设多通道卷积神经网络，通过预设多通道卷积神经网络的Sigmoid函数输出评论文本信息对应的情感分类结果。In one embodiment, when the processor executes the computer program, the following steps are further implemented: combining emotional features according to feature types to obtain emotional feature combination data; inputting the emotional feature combination data into a preset multi-channel convolutional neural network, and by preset The Sigmoid function of the multi-channel convolutional neural network outputs the sentiment classification results corresponding to the comment text information.
在一个实施例中，处理器执行计算机程序时还实现以下步骤：根据推广产品对应的所有情感分类结果，确定推广产品对应的评论情感极性值；获取推广产品对应的浏览参数；根据推广产品对应的浏览参数以及评论情感极性值，获取推广产品对应的历史推广得分。In one embodiment, when the processor executes the computer program, the following steps are further implemented: determining the sentiment polarity value of comments corresponding to the promoted product according to all sentiment classification results corresponding to the promoted product; obtaining browsing parameters corresponding to the promoted product; to obtain the historical promotion score corresponding to the promoted product.
在一个实施例中，处理器执行计算机程序时还实现以下步骤：当接收到推荐请求时，获取推荐请求所请求的当前推广产品类别以及请求对象数；确定当前推广产品类别对应的对象排名；根据请求对象数以及对象排名，推送对应数量的推广合作对象。In one embodiment, when the processor executes the computer program, the following steps are further implemented: when a recommendation request is received, obtain the current promotion product category requested by the recommendation request and the number of requested objects; determine the object ranking corresponding to the current promotion product category; Request the number of objects and object ranking, and push the corresponding number of promotion partners.
在一个实施例中，提供了一种计算机可读存储介质，其上存储有计算机程序，计算机程序被处理器执行时实现以下步骤：In one embodiment, a computer-readable storage medium is provided on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:
获取推广合作对象的基本数据及其历史推广数据，基本数据包括粉丝数目以及推广产品类别；Obtain the basic data of promotion partners and their historical promotion data, including the number of fans and the category of products to be promoted;
根据推广合作对象的粉丝数目及其历史推广数据，确定推广合作对象对应的推广得分；According to the number of fans of the promotion partner and their historical promotion data, determine the promotion score corresponding to the promotion partner;
根据推广得分以及推广合作对象的推广产品类别，确定各个推广产品类别下推广合作对象的对象排名；According to the promotion score and the promotion product category of the promotion partner, determine the object ranking of the promotion partner under each promotion product category;
当接收到推荐请求时，根据推荐请求中的推广产品类别对应的对象排名，推送对应的推广合作对象。When a recommendation request is received, the corresponding promotion cooperation object is pushed according to the object ranking corresponding to the promotion product category in the recommendation request.
在一个实施例中，计算机程序被处理器执行时还实现以下步骤：根据推广合作对象所推广产品的历史转化率数据，生成多个推广产品对应的历史转化率排名；获取历史转化率排名中预设名次内的推广产品，根据预设名次内的推广产品确定推广合作对象的历史推广得分；根据推广合作对象的粉丝数目以及历史推广得分，确定推广合作对象对应的推广得分。In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: generating historical conversion rate rankings corresponding to a plurality of promoted products according to historical conversion rate data of the products promoted by the promotion partner; Set the promotion products in the ranking, and determine the historical promotion score of the promotion partner according to the promotion product in the preset ranking; according to the number of fans of the promotion partner and the historical promotion score, determine the promotion score corresponding to the promotion partner.
在一个实施例中，计算机程序被处理器执行时还实现以下步骤：获取历史转化率排名中预设名次内的推广产品对应的评论文本信息；基于文本情感分析，提取评论文本信息对应的情感特征；根据情感特征获取评论文本信息对应的情感分类结果；根据情感分类结果，获取历史转化率排名中预设名次内的推广产品对应的历史推广得分；根据推广产品对应的历史推广得分的平均值，确定推广合作对象的历史推广得分。In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining comment text information corresponding to the promoted products in the preset rankings in the historical conversion rate ranking; extracting sentiment features corresponding to the comment text information based on text sentiment analysis ; Obtain the sentiment classification result corresponding to the comment text information according to the sentiment feature; obtain the historical promotion score corresponding to the promotion product in the preset ranking in the historical conversion rate ranking according to the sentiment classification result; According to the average value of the historical promotion score corresponding to the promotion product, Determine the historical promotion score of the promotion partner.
在一个实施例中，计算机程序被处理器执行时还实现以下步骤：将情感特征按照特征类型进行组合，获取情感特征组合数据；将情感特征组合数据输入预设多通道卷积神经网络，通过预设多通道卷积神经网络的Sigmoid函数输出评论文本信息对应的情感分类结果。In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: combining emotional features according to feature types to obtain emotional feature combination data; inputting the emotional feature combination data into a preset multi-channel convolutional neural network, Let the Sigmoid function of the multi-channel convolutional neural network output the sentiment classification result corresponding to the comment text information.
在一个实施例中，计算机程序被处理器执行时还实现以下步骤：根据推广产品对应的所有情感分类结果，确定推广产品对应的评论情感极性值；获取推广产品对应的浏览参数；根据推广产品对应的浏览参数以及评论情感极性值，获取推广产品对应的历史推广得分。In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: determine the comment sentiment polarity value corresponding to the promoted product according to all the sentiment classification results corresponding to the promoted product; obtain the browsing parameters corresponding to the promoted product; The corresponding browsing parameters and comment sentiment polarity values are used to obtain the historical promotion scores corresponding to the promoted products.
在一个实施例中，计算机程序被处理器执行时还实现以下步骤：当接收到推荐请求时，获取推荐请求所请求的当前推广产品类别以及请求对象数；确定当前推广产品类别对应的对象排名；根据请求对象数以及对象排名，推送对应数量的推广合作对象。In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: when a recommendation request is received, obtain the current promotion product category requested by the recommendation request and the number of requested objects; determine the object ranking corresponding to the current promotion product category; According to the number of requested objects and object ranking, push the corresponding number of promotion partners.
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，的计算机程序可存储于一非易失性计算机可读取存储介质中，该计算机程序在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和易失性存储器中的至少一种。非易失性存储器可包括只读存储器(Read-OnlyMemory，ROM)、磁带、软盘、闪存或光存储器等。易失性存储器可包括随机存取存储器(RandomAccessMemory，RAM)或外部高速缓冲存储器。作为说明而非局限，RAM可以是多种形式，比如静态随机存取存储器(StaticRandomAccessMemory，SRAM)或动态随机存取存储器(DynamicRandomAccessMemory，DRAM)等。Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium , when the computer program is executed, it may include the processes of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other media used in the various embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, or optical memory, and the like. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, the RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM).
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features It is considered to be the range described in this specification.
以上实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。The above examples only represent several embodiments of the present application, and the descriptions thereof are relatively specific and detailed, but should not be construed as a limitation on the scope of the invention patent. It should be noted that, for those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the scope of protection of the patent of the present application shall be subject to the appended claims.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110978052.6A CN113837792A (en) | 2021-08-23 | 2021-08-23 | Promotion cooperation object recommendation method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110978052.6A CN113837792A (en) | 2021-08-23 | 2021-08-23 | Promotion cooperation object recommendation method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113837792A true CN113837792A (en) | 2021-12-24 |
Family
ID=78961174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110978052.6A Pending CN113837792A (en) | 2021-08-23 | 2021-08-23 | Promotion cooperation object recommendation method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113837792A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118427411A (en) * | 2024-06-06 | 2024-08-02 | 北京百度网讯科技有限公司 | Resource recommendation method, device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109951724A (en) * | 2017-12-20 | 2019-06-28 | 阿里巴巴集团控股有限公司 | Recommended method, main broadcaster's recommended models training method and relevant device is broadcast live |
| CN111586448A (en) * | 2020-01-20 | 2020-08-25 | 全息空间(深圳)智能科技有限公司 | Live broadcast platform anchor pre-screening method and device and storage medium |
| CN112672188A (en) * | 2019-10-15 | 2021-04-16 | 阿里巴巴集团控股有限公司 | Anchor recommendation method, device and storage medium |
| CN113205382A (en) * | 2021-04-30 | 2021-08-03 | 北京有竹居网络技术有限公司 | Method, apparatus, device, storage medium and program product for determining object |
| CN113205362A (en) * | 2021-04-30 | 2021-08-03 | 北京有竹居网络技术有限公司 | Method, apparatus, device, storage medium and program product for determining a promoter |
-
2021
- 2021-08-23 CN CN202110978052.6A patent/CN113837792A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109951724A (en) * | 2017-12-20 | 2019-06-28 | 阿里巴巴集团控股有限公司 | Recommended method, main broadcaster's recommended models training method and relevant device is broadcast live |
| CN112672188A (en) * | 2019-10-15 | 2021-04-16 | 阿里巴巴集团控股有限公司 | Anchor recommendation method, device and storage medium |
| CN111586448A (en) * | 2020-01-20 | 2020-08-25 | 全息空间(深圳)智能科技有限公司 | Live broadcast platform anchor pre-screening method and device and storage medium |
| CN113205382A (en) * | 2021-04-30 | 2021-08-03 | 北京有竹居网络技术有限公司 | Method, apparatus, device, storage medium and program product for determining object |
| CN113205362A (en) * | 2021-04-30 | 2021-08-03 | 北京有竹居网络技术有限公司 | Method, apparatus, device, storage medium and program product for determining a promoter |
Non-Patent Citations (2)
| Title |
|---|
| 潘海峡等: "深度学习 工程师认证初级教程", 31 May 2020, 北京航空航天大学出版社, pages: 203 * |
| 陶永才 等: "面向短文本情感分析的多特征融合方法研究", 小型微型计算机系统, 30 June 2020 (2020-06-30), pages 1126 - 1132 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118427411A (en) * | 2024-06-06 | 2024-08-02 | 北京百度网讯科技有限公司 | Resource recommendation method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020125445A1 (en) | Classification model training method, classification method, device and medium | |
| CN112733042B (en) | Recommendation information generation method, related device and computer program product | |
| CN108845986A (en) | A kind of sentiment analysis method, equipment and system, computer readable storage medium | |
| CA3147634A1 (en) | Method and apparatus for analyzing sales conversation based on voice recognition | |
| CN112926308B (en) | Method, device, equipment, storage medium and program product for matching text | |
| CN111104590A (en) | Information recommendation method, device, medium and electronic equipment | |
| CN104063799A (en) | Promotion message pushing method and device | |
| US20180239790A1 (en) | Provision device, provision method and non-transitory computer readable storage medium | |
| CN111695922B (en) | Potential user determination method and device, storage medium and electronic device | |
| CN119739987A (en) | Label system construction, label providing method, device, equipment and storage medium | |
| CN112200602B (en) | Neural network model training method and device for advertisement recommendation | |
| CN113837792A (en) | Promotion cooperation object recommendation method, device, equipment and storage medium | |
| CN113360769A (en) | Information query method and device, electronic equipment and storage medium | |
| WO2020211616A1 (en) | Method and device for processing user interaction information | |
| CN115292489A (en) | Enterprise public opinion analysis method, device, equipment and storage medium | |
| CN114612155A (en) | Method, device, equipment and medium for determining target population | |
| US10937078B1 (en) | Dynamic generation of item description content | |
| CN113516515A (en) | Information pushing method, device and system, electronic equipment and storage medium | |
| CN114595377A (en) | Vocabulary recommendation method and device, terminal and readable storage medium | |
| KR102595097B1 (en) | process and system for product marketing efficiency analysis | |
| TWI795001B (en) | Content generation method, content delivery method, content generation program and content delivery program | |
| CN117474598B (en) | Advertisement searching method, model training method and device | |
| Scientific | Sentiment Analysis Based On Pubgm Player Aspects From App Store Reviews Using Bidirectional Encoder Representation From Transformer (BERT) | |
| CN113781076B (en) | Prompting method, device, equipment and readable storage medium | |
| US20250124264A1 (en) | Generating customized content descriptions using artificial intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211224 |