WO2016101881A1 - 通过计算机实现的信息推荐方法、装置和计算机设备 - Google Patents

通过计算机实现的信息推荐方法、装置和计算机设备 Download PDF

Info

Publication number
WO2016101881A1
WO2016101881A1 PCT/CN2015/098344 CN2015098344W WO2016101881A1 WO 2016101881 A1 WO2016101881 A1 WO 2016101881A1 CN 2015098344 W CN2015098344 W CN 2015098344W WO 2016101881 A1 WO2016101881 A1 WO 2016101881A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
data
demand
information
contact information
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.)
Ceased
Application number
PCT/CN2015/098344
Other languages
English (en)
French (fr)
Inventor
陈本东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Baidu Online Network Technology Beijing Co Ltd
Original Assignee
Baidu Online Network Technology Beijing Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Baidu Online Network Technology Beijing Co Ltd filed Critical Baidu Online Network Technology Beijing Co Ltd
Publication of WO2016101881A1 publication Critical patent/WO2016101881A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor

Definitions

  • the present invention relates to the field of information processing, and in particular, to a method, an apparatus, and a computer device for information recommendation implemented by a computer.
  • O2O Online to Offline
  • offline business model refers to online marketing online purchase to drive offline operations and offline consumption
  • the news of offline stores is pushed to Internet users, for example, the WeChat public number launched by Tencent and the direct number launched by Baidu.
  • O2O does not have a good entrance, so that its online promotion is difficult.
  • users have to go to various scattered websites or APPs to search for various services, or need users to remember the name of the business. Search (such as WeChat public number "@" search); for merchants, O2O's offline operating costs are high, the cost of updating business information is higher and the accuracy is lower.
  • the object of the present invention is to provide a method, a device and a computer device for information recommendation by using a computer, so as to better match the needs of the user and the merchant providing the service, and timely provide the merchant with valuable user demand information.
  • a method for recommending information implemented by a computer includes: receiving demand data of a first user; acquiring a predetermined number of recommended object data according to the demand data and a pre-trained demand matching sorting model,
  • the recommendation object data includes information of a second user that satisfies a requirement of the first user; and extracts information according to the second user
  • the contact information of the second user is sent to the corresponding second user according to the contact information.
  • an information recommendation apparatus comprising: a data receiving unit, configured to receive demand data of a first user; and an object data obtaining unit, configured to sort according to the demand data and a pre-trained demand match
  • the model obtains a predetermined number of recommended object data, the recommended object data includes information of a second user that satisfies the needs of the first user, and the contact information extracting unit is configured to extract the second user according to the information of the second user
  • the contact information sending unit is configured to separately send the demand data of the first user to the corresponding second user according to the contact information.
  • a computer apparatus comprising: one or more processors; a memory; one or more programs, the one or more programs being stored in the memory and configured to be
  • the one or more processors executing instructions included in the one or more programs for executing an information recommendation method: receiving demand data of a first user; acquiring a predetermined order according to the demand data and a pre-trained demand matching ranking model a number of recommended object data, the recommended object data includes information of the second user that meets the needs of the first user; and extracts contact information of the second user according to the information of the second user; Sending the first user's demand data to the corresponding second user.
  • the computer-implemented information recommendation method, device and computer device provided by the embodiments of the present invention are used to increase the entrance of the O2O, to better match the needs of the user and the merchant providing the service, and timely provide the merchant with valuable user demand information. Reduce the technical requirements and operating costs of the merchant's use of the system, improve the convenience of users to find services, and improve the efficiency of the merchants to find customers.
  • FIG. 1 is a diagram showing an example of information transfer and application of an exemplary embodiment O2O of the present invention
  • FIG. 2 is a view showing an example of an application structure of the overall technical solution of the present invention shown in FIG. 1;
  • FIG. 3 is a flow chart showing a method of recommending information implemented by a computer according to an exemplary embodiment of the present invention
  • FIG. 4 is a block diagram showing the structure of an information recommending apparatus of an exemplary embodiment of the present invention.
  • FIG. 5 is a logic block diagram showing a computer device according to an embodiment of the present invention.
  • the basic idea of the present invention is to establish an O2O service platform, which can push user demand data to the merchant, and push the merchant's information to the user with corresponding requirements, thereby improving the matching between the user demand and the merchant.
  • FIG. 1 is a diagram showing an example of information transfer and application based on an O2O service platform of an exemplary embodiment of the present invention.
  • the user can pass his or her own needs to the O2O service platform.
  • the O2O service platform extracts the user's demand information and transmits the value information (for example, the user's contact information) to the offline merchant.
  • the merchant can register the merchant information through the merchant registration system to the merchant recommendation system; on the other hand, after the user inputs the search term (ie, the query), the O2O service platform performs user demand analysis on the search term, and the user is The demand is transmitted to the merchant recommendation system; on the other hand, the merchant recommendation system matches the business registration information with the user requirements according to the pre-trained demand matching ranking model, and pushes the user requirements to the matching through the merchant push system.
  • the search term ie, the query
  • the O2O service platform performs user demand analysis on the search term, and the user is The demand is transmitted to the merchant recommendation system; on the other hand, the merchant recommendation system matches the business registration information with the user requirements according to the pre-trained demand matching ranking model, and pushes the user requirements to the matching through the merchant push system.
  • user evaluation, feedback, and merchant's transaction information, favorable information, and the like may be transmitted to the merchant service feedback system for providing user evaluation, feedback information, and transaction information about the merchant.
  • FIG. 3 is a flow chart showing a computer-implemented information recommendation method according to an exemplary embodiment of the present invention.
  • step S110 demand data of the first user is received.
  • the demand data may be a search term of a user.
  • step S120 a predetermined number of recommended object data is acquired according to the demand data and the pre-trained demand matching ordering model, and the recommended object data includes information of the second user that satisfies the needs of the first user.
  • the first user is a consumer user and the second user is a merchant user.
  • step S120 includes the following sub-steps:
  • the natural data analysis is performed on the demand data to obtain primary demand category data and sub-requirement category data of the first user.
  • the user inputs "I want to book western food at 18:00 on April 26", and by performing natural language analysis on the demand data, the main user's main demand category data can be obtained as "food", sub-requirement category.
  • the data is "Western food.”
  • the main requirement category may be a category to which the sub-requirement category belongs, and the sub-requirement category may be a more specific and more detailed requirement of the demand user.
  • the candidate recommendation data may be obtained from the merchant registration information database according to the main demand category data and the sub-requirement category data.
  • the correlation index of the candidate recommendation object data is obtained by using each of the candidate recommendation data and the acquired main requirement category data and the sub-requirement category data as input, wherein the correlation is obtained by using a pre-trained demand matching ranking model.
  • the index may be the probability of selection of candidate recommendation object data.
  • a training method for a demand matching ranking model is given below.
  • Data eg, obtained by natural language analysis of historical demand data
  • the matching ranking model is established based on the primary demand category data, the sub-requirement category data, and the annotated user data and the demand matching ranking model is trained , thereby learning the influence of the values of the main demand category and the sub-requirement category on the relevance index of the candidate recommended object data.
  • a large number of consumer user IDs can be expressed at a certain time (segment time slice) at a certain location (for example, the need for stratified demand analysis)
  • historical demand data such as rejection or acceptance of the recommendation result as the basis for extracting the training characteristics, and using a content management system (for example, DNN), a common machine learning model, and a manual rule to train and fit.
  • DNN content management system
  • the general features can be personalized, thereby evaluating the relevance index of the candidate object data, and providing consumers with more personalized requirements customization.
  • the training of the demand matching ranking model is not limited to using the method described above, and may also Training of the demand matching ranking model is performed using other feature parameters and training methods. Since the model training is not a core improvement point of the present invention, only the above exemplary description is given to the training of the demand matching ranking model.
  • a predetermined number of the candidate recommendation object data is selected as the recommended object data according to the correlation index.
  • the candidate recommended object data is sorted based on the aforementioned correlation index, and a predetermined number of recommended object data is selected therefrom.
  • the contact information of the second user is extracted according to the information of the second user.
  • the information of the second user may be, but is not limited to, an address, a contact information, a user type, a service type, a service range, and the like provided when the second user registers through the registration platform.
  • step S140 the first user's demand data is respectively sent to the corresponding second user according to the contact information.
  • the contact information of the first user is obtained, and the demand data of the first user and the contact information thereof are respectively sent to the corresponding second user according to the contact information of the second user.
  • the second user may also actively contact the first user, for example, by using a phone, a social APP, etc., thereby increasing the offline user of the second user and increasing the merchant (the second user) The volume of the transaction.
  • the computer-implemented information recommendation method provided by the embodiment of the present invention matches the needs of the user and the merchant providing the service, and provides valuable user demand information to the merchant in time to reduce the technical requirements and operating costs of the merchant using the system. Improve the convenience of users to find services, and improve the efficiency of merchants to find customers.
  • Fig. 4 is a block diagram showing the structure of an information recommending apparatus of an exemplary embodiment of the present invention.
  • the information recommendation means includes a data receiving unit 310, an object data acquiring unit 320, a contact information extracting unit 330, and a demand data transmitting unit 340.
  • the data receiving unit 310 is configured to receive the demand data of the first user.
  • the object data obtaining unit 320 is configured to acquire a predetermined number of recommended object data according to the demand data acquired by the data receiving unit 310 and the pre-trained demand matching sorting model, where the recommended object data includes a first meeting that meets the needs of the first user. Two user information.
  • the first user is a consumer user and the second user is a merchant user.
  • the object data obtaining unit 320 is configured to acquire the main demand category data and the sub-requirement category data of the first user according to the demand data acquired by the data receiving unit 310, according to the main demand category data, according to an exemplary embodiment of the present invention. And sub-requirement category data acquisition a plurality of candidate recommendation data, with each of the candidate recommendation data and the acquired main requirement category data and sub-requirement category data as input, respectively acquiring a correlation index of the candidate recommendation object data through a pre-trained demand matching ranking model, according to The correlation index selects a predetermined number of the candidate recommendation object data as the recommended object data.
  • the object data obtaining unit 320 performs natural language analysis on the demand data, and acquires main demand category data and sub-requirement category data of the first user.
  • the contact information extracting unit 330 is configured to extract contact information of the second user according to the information of the second user.
  • the requirement data sending unit 340 is configured to separately send the demand data of the first user to the corresponding second user according to the contact information.
  • the demand data sending unit 340 is configured to acquire contact information of the first user, and send the first user's demand data and contact information to the corresponding first according to the contact information of the second user. Two users.
  • the computer-implemented information recommendation device provided by the embodiment of the present invention matches the needs of the user and the merchant providing the service, provides valuable user demand information to the merchant in time, reduces the technical requirements of the merchant to use the system, and operates Cost, improve the convenience of users to find services, and improve the efficiency of merchants to find customers.
  • FIG. 5 is a logic block diagram showing a computer device according to an embodiment of the present invention.
  • a computer device can be used to implement the information recommendation method provided in the above embodiments. Specifically:
  • Computer devices may vary considerably depending on configuration or performance, and may include one or more processors (such as Central Processing Units, CPU) 510 and memory 520.
  • the memory 520 can be short-term storage or persistent storage.
  • One or more programs may be stored in memory 520, each of which may include a series of instruction operations in a computer device.
  • the processor 510 can be in communication with the memory 520 to perform a series of instruction operations in the memory 520 on the computer device.
  • the memory 520 also stores data of one or more operating systems, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, and the like.
  • the computer device may also include one or more power sources 530, one or more wired or wireless network interfaces 540, one or more input and output interfaces 550, and the like.
  • the computer device includes one or more processors 510 and stores A storage 520, and one or more programs, one or more programs are stored in the memory 520, and configured to execute, by the one or more processors 510, instructions for executing the information recommendation method included in one or more programs
  • Receiving the demand data of the first user ; acquiring a predetermined number of recommended object data according to the demand data and the pre-trained demand matching sorting model, where the recommended object data includes information of the second user that meets the requirement of the first user; Extracting contact information of the second user according to the information of the second user; and sending the demand data of the first user to the corresponding second user according to the contact information.
  • the processing instruction for respectively transmitting the first user's demand data to the corresponding second user according to the contact information may include: acquiring the first user's contact information, according to The contact information of the second user respectively sends the demand data of the first user and the contact information thereof to the corresponding second user.
  • the processing instruction of acquiring the predetermined number of recommended object data according to the demand data and the pre-trained demand matching sorting model may include: acquiring the main user category data and the sub-requirement category data of the first user according to the demand data, according to The main demand category data and the sub-requirement category data acquire a plurality of candidate recommendation data, and each of the candidate recommendation data and the acquired main demand category data and sub-requirement category data are input as inputs, respectively, and are obtained by a pre-trained demand matching ranking model.
  • a correlation index of each candidate recommended object data is selected as a recommended target data by a predetermined number of the candidate recommended object data based on the correlation index.
  • the processing instruction of acquiring the main demand category data and the sub-requirement category data of the first user according to the demand data may include: performing natural language analysis on the demand data, and acquiring the main demand category data and the sub-requirement category of the first user. data.
  • the first user may be a consumer user and the second user may be a merchant user.
  • the computer equipment provided by the invention better matches the needs of the users and the merchants providing the services, timely provides the merchants with valuable user demand information, reduces the technical requirements of the merchants using the system, and operates the costs, and improves the users to find services. Convenience and increase the efficiency of merchants finding customers.
  • the above method according to the present invention can be implemented in hardware or firmware, or can be implemented as Software or computer code stored in a recording medium such as a CD ROM, RAM, floppy disk, hard disk or magneto-optical disk, or originally stored on a network, stored in a remote recording medium or a non-transitory machine readable medium and stored
  • the computer code in the medium is recorded locally such that the methods described herein can be stored in such software processing on a recording medium using a general purpose computer, a special purpose processor, or programmable or special purpose hardware such as an ASIC or an FPGA.
  • a computer, processor, microprocessor controller or programmable hardware includes storage components (eg, RAM, ROM, flash memory, etc.) that can store or receive software or computer code, when the software or computer code is The processing methods described herein are implemented when the processor or hardware is accessed and executed. Moreover, when a general purpose computer accesses code for implementing the processing shown herein, the execution of the code converts the general purpose computer into a special purpose computer for performing the processing shown herein.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • General Engineering & Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Databases & Information Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种通过计算机实现的信息推荐方法、装置和计算机设备，所述方法包括：接收第一用户的需求数据（S110）；根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据（S120），所述推荐对象数据包括满足第一用户的需求的第二用户的信息；根据所述第二用户的信息提取所述第二用户的联系信息（S130）；根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户（S140）。通过所述方法、装置和计算机设备，可提高用户需求和提供服务的商户的匹配度，降低商户使用系统的技术要求及运营成本，提高用户找到服务的便捷性，且提高商户找到客户的效率。

Description

通过计算机实现的信息推荐方法、装置和计算机设备
本申请要求于2014年12月23日提交中国专利局、申请号为201410815315.1、发明名称为“通过计算机实现的信息推荐方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及信息处理领域，尤其涉及一种通过计算机实现的信息推荐方法、装置和计算机设备。
背景技术
普通用户需要获取各种服务时，不能非常方便的联系到商户，例如，订餐、订鲜花或者英语培训，用户不得不去网站(或者APP)上搜索的商家，而且网站搜索的商家可能离用户较远，不便于交易。
目前，O2O(Online to Offline)营销模式(又称离线商务模式，是指线上营销线上购买带动线下经营和线下消费)最为热门，其通过打折、提供信息、服务预订等方式，把线下商店的消息推送给互联网用户，例如，腾讯推出的微信公众号、百度推出的直达号等。
但是，就现目前来看，O2O尚无一个好的入口，以致其线上推广困难，对于用户来讲，用户不得不去各个分散的网站或者APP去搜索各种服务，或者需要用户记忆商家名称进行搜索(如微信公众号的“@”搜索)；而对于商户来讲，O2O的线下运营成本高昂，更新商家信息成本较高且准确率较低。
发明内容
本发明的目的在于，提供一种通过计算机实现的信息推荐方法、装置和计算机设备，以较好地对用户的需求以及提供服务的商户进行匹配，及时向商户提供有价值的用户需求信息。
根据本发明的一方面，提供一种通过计算机实现的信息推荐方法，包括：接收第一用户的需求数据；根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息；根据所述第二用户的信息提取 所述第二用户的联系信息；根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。
根据本发明的另一方面，提供一种信息推荐装置，包括：数据接收单元，用于接收第一用户的需求数据；对象数据获取单元，用于根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息；联系信息提取单元，用于根据所述第二用户的信息提取所述第二用户的联系信息；需求数据发送单元，用于根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。
根据本发明的另一方面，提供一种计算机设备，包括：一个或多个处理器；存储器；一个或多个程序，所述一个或多个程序存储在所述存储器中，且经配置以由所述一个或者多个处理器执行所述一个或者多个程序包含的用于执行信息推荐方法的指令：接收第一用户的需求数据；根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息；根据所述第二用户的信息提取所述第二用户的联系信息；根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。
本发明实施例提供的计算机实现的信息推荐方法、装置和计算机设备，以增加O2O的入口，以较好地对用户的需求以及提供服务的商户进行匹配，及时向商户提供有价值的用户需求信息，降低商户使用系统的技术要求及运营成本，提高用户找到服务的便捷性，且提高商户找到客户的效率。
附图说明
图1是示出本发明示例性实施例O2O的信息传递及应用示例图；
图2是示出图1所示的本发明的整体技术方案的应用结构示例图；
图3是示出本发明示例性实施例的通过计算机实现的信息推荐方法的流程示意图；
图4是示出本发明示例性实施例的信息推荐装置的结构框图；
图5是示出本发明实施例提供的计算机设备的逻辑框图。
具体实施方式
本发明的基本构思是，建立一种O2O服务平台，可将用户需求数据推送给商户，且将商户的信息推送给具有相应需求的用户，从而提高用户需求与商户的匹配。
下面结合附图对本发明示例性实施例的一种通过计算机实现的信息推荐方法、装置和计算机设备进行详细描述。
图1是示出本发明示例性实施例基于O2O服务平台的信息传递及应用示例图。如图1所示，用户可以将自己的需求传递至O2O服务平台，O2O服务平台对用户的需求信息进行提取，将价值信息(例如，用户的联系信息)传递给线下商家。
图2是示出本发明示例性实施例的O2O服务平台整体架构的示意图。参照图2，商户可以通过其商户登记系统注册商户信息，传至商户推荐系统；另一方面，用户输入搜索词(即query)后，O2O服务平台对所述搜索词进行用户需求分析，将用户需求传至商户推荐系统；另一方面，商户推荐系统根据预先训练的需求匹配排序模型将商户登记的信息与所述用户需求进行匹配排序，并且通过商户推送系统将所述用户需求推送给匹配的商户。
此外，还可将用户评价、反馈、以及商户的成交信息、好评信息等传至商户服务反馈系统，以用于向用户提供关于商家的用户评价、反馈信息以及成交信息。
下面将参照图3～图4详细描述本发明的示例性实施例。
图3是示出本发明示例性实施例的通过计算机实现的信息推荐方法的流程示意图。
参照图3，在步骤S110，接收第一用户的需求数据。所述需求数据可以是用户的搜索词。
在步骤S120，根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息。
优选地，所述第一用户是消费者用户，所述第二用户是商户用户。
具体地，根据本发明的示例性实施例，步骤S120包括如下子步骤：
首先，根据所述需求数据获取第一用户的主需求类别数据和子需求 类别数据。优选地，对所述需求数据进行自然语言分析，获取第一用户的主需求类别数据和子需求类别数据。
例如，用户输入“我要预订4月26日晚上18:00点的西餐”，通过对所述需求数据进行自然语言分析，可以获取第一用户的主需求类别数据为“餐饮”，子需求类别数据是“西餐”。
其次，根据所述主需求类别数据和子需求类别数据获取多个候选推荐数据。其中，所述主需求类别可以是子需求类别所属的类目，子需求类别可以是需求用户的更具体化、更细化的需求。
具体地，可根据所述主需求类别数据和子需求类别数据从商户注册信息数据库获取所述候选推荐数据。
再次，以各个所述候选推荐数据以及获取的主需求类别数据和子需求类别数据作为输入，分别通过预先训练的需求匹配排序模型获取所述各候选推荐对象数据的相关性指数，其中，该相关性指数可以是候选推荐对象数据的选取概率。目前已有各种根据预定的特征指标通过预先选好的标注数据进行排序模型训练的技术。
下面对需求匹配排序模型的一种训练方法给予示例性说明。首先，从一组历史需求数据以及为其标注的用户数据(例如，商户的成交数量、地理位置、好评数以及近期活跃度)；此后，根据这些历史需求数据分别获取主需求类别数据和子需求类别数据(例如，通过对历史需求数据进行自然语言分析获取)；然后，根据该主需求类别数据、子需求类别数据和所述标注的用户数据建立所述匹配排序模型并且训练所述需求匹配排序模型，从而学习主需求类别和子需求类别的值对候选推荐对象数据的相关性指数的影响。
更具体而言，可以将大量的消费者用户ID(性别、年龄)在某个时间(分片的时间片段)某个地点，表达了怎样的需求(例如，将该需要进行分层需求分析)，以及对推荐结果是拒绝还是接受等历史需求数据作为提取训练特征的依据，并采用内容管理系统(例如，DNN)、普通的机器学习模型及人工规则方式训练拟合得到。通过所述需求匹配排序模型可以将一般特征个性化，从而对候选对象数据的相关性指数进行评价，为消费者用户提供更个性化的需求定制。
所述需求匹配排序模型的训练不限于使用前述描述的方法，还可以 使用其他特征参数和训练方法来进行所述需求匹配排序模型的训练。由于所述模型训练不是本发明的核心改进点，在此仅对需求匹配排序模型的训练给予上述示例性说明。
最后，根据所述相关性指数选取预定个数的所述候选推荐对象数据作为推荐对象数据。具体地，基于前述的相关性指数对候选推荐对象数据进行排序，从中筛选出预定个数推荐对象数据。
在步骤S130，根据所述第二用户的信息提取所述第二用户的联系信息。其中，第二用户的信息可以但不限于是第二用户通过登记平台注册时提供的地址、联系方式、用户类型、服务类型、服务范围等。
在步骤S140，根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。优选地，获取所述第一用户的联系信息，根据所述第二用户的联系信息分别将所述第一用户的需求数据及其联系信息发送给相应的第二用户。
第二用户获取到第一用户的需求数据及其联系信息后，还可以主动联系第一用户，例如，通过电话、社交APP等，从而增加第二用户的线下用户，增加商户(第二用户)的成交量。
本发明实施例提供的计算机实现的信息推荐方法，以较好地对用户的需求以及提供服务的商户进行匹配，及时向商户提供有价值的用户需求信息降低商户使用系统的技术要求及运营成本，提高用户找到服务的便捷性，且提高商户找到客户的效率。
图4示出本发明示例性实施例的信息推荐装置的结构框图。
参照图4，所述信息推荐装置包括数据接收单元310、对象数据获取单元320、联系信息提取单元330以及需求数据发送单元340。
数据接收单元310用于接收第一用户的需求数据。
对象数据获取单元320用于根据所述数据接收单元310获取到的需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息。
例如，所述第一用户是消费者用户，所述第二用户是商户用户。
具体地，根据本发明的示例性实施例，对象数据获取单元320用于根据数据接收单元310获取到的需求数据获取第一用户的主需求类别数据和子需求类别数据，根据所述主需求类别数据和子需求类别数据获取 多个候选推荐数据，以各个所述候选推荐数据以及获取的主需求类别数据和子需求类别数据作为输入，分别通过预先训练的需求匹配排序模型获取所述各候选推荐对象数据的相关性指数，根据所述相关性指数选取预定个数的所述候选推荐对象数据作为推荐对象数据。
优选地，对象数据获取单元320对所述需求数据进行自然语言分析，获取第一用户的主需求类别数据和子需求类别数据。
联系信息提取单元330用于根据所述第二用户的信息提取所述第二用户的联系信息。
需求数据发送单元340用于根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。
优选地，所述需求数据发送单元340用于获取所述第一用户的联系信息，根据所述第二用户的联系信息分别将所述第一用户的需求数据及其联系信息发送给相应的第二用户。
本发明实施例提供的计算机实现的信息推荐装置，以较好地对用户的需求以及提供服务的商户进行匹配，及时向商户提供有价值的用户需求信息，降低商户使用系统的技术要求，及运营成本，提高用户找到服务的便捷性，且提高商户找到客户的效率。
图5是示出本发明实施例提供的计算机设备的逻辑框图。
参照图5，计算机设备可用于实施上述实施例中提供的信息推荐方法。具体来讲：
计算机设备可因配置或性能不同而产生比较大的差异，可以包括一个或一个以上处理器(如Central Processing Units，CPU)510和存储器520。其中，存储器520可以是短暂存储或持久存储。存储器520中可存储有一个或一个以上的程序，每个程序可包括对计算机设备中的一系列指令操作。更进一步地，处理器510可与存储器520通信，在计算机设备上执行存储器520中的一系列指令操作。特别地，存储器520中还存储有一个或一个以上的操作系统的数据，例如Windows ServerTM，Mac OS XTM，UnixTM，LinuxTM，FreeBSDTM等等。计算机设备还可包括一个或一个以上电源530，一个或一个以上有线或无线网络接口540，一个或一个以上输入输出接口550等。
具体在本实施例中，计算机设备包括一个或者多个处理器510、存 储器520，以及一个或者多个程序，一个或者多个程序存储于存储器520中，且经配置以由一个或者多个处理器510执行一个或者多个程序包含的用于执行信息推荐方法的指令：接收第一用户的需求数据；根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息；根据所述第二用户的信息提取所述第二用户的联系信息；根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。
这里，根据本发明的示例性实施例，根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户的处理指令可包括：获取所述第一用户的联系信息，根据所述第二用户的联系信息分别将所述第一用户的需求数据及其联系信息发送给相应的第二用户。
进一步地，根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据的处理指令可包括：根据所述需求数据获取第一用户的主需求类别数据和子需求类别数据，根据所述主需求类别数据和子需求类别数据获取多个候选推荐数据，以各个所述候选推荐数据以及获取的主需求类别数据和子需求类别数据作为输入，分别通过预先训练的需求匹配排序模型获取所述各候选推荐对象数据的相关性指数，根据所述相关性指数选取预定个数的所述候选推荐对象数据作为推荐对象数据。
更进一步地，根据所述需求数据获取第一用户的主需求类别数据和子需求类别数据的处理指令可包括：对所述需求数据进行自然语言分析，获取第一用户的主需求类别数据和子需求类别数据。
需要说明的是，第一用户可以是消费者用户，第二用户可以是商户用户。
本发明提供的计算机设备，以较好地对用户的需求以及提供服务的商户进行匹配，及时向商户提供有价值的用户需求信息，降低商户使用系统的技术要求，及运营成本，提高用户找到服务的便捷性，且提高商户找到客户的效率。
需要指出，根据实施的需要，可将本申请中描述的各个步骤拆分为更多步骤，也可将两个或多个步骤或者步骤的部分操作组合成新的步骤，以实现本发明的目的。
上述根据本发明的方法可在硬件、固件中实现，或者被实现为可存 储在记录介质(诸如CD ROM、RAM、软盘、硬盘或磁光盘)中的软件或计算机代码，或者被实现通过网络下载的原始存储在远程记录介质或非暂时机器可读介质中并将被存储在本地记录介质中的计算机代码，从而在此描述的方法可被存储在使用通用计算机、专用处理器或者可编程或专用硬件(诸如ASIC或FPGA)的记录介质上的这样的软件处理。可以理解，计算机、处理器、微处理器控制器或可编程硬件包括可存储或接收软件或计算机代码的存储组件(例如，RAM、ROM、闪存等)，当所述软件或计算机代码被计算机、处理器或硬件访问且执行时，实现在此描述的处理方法。此外，当通用计算机访问用于实现在此示出的处理的代码时，代码的执行将通用计算机转换为用于执行在此示出的处理的专用计算机。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以所述权利要求的保护范围为准。

Claims (11)

  1. 一种通过计算机实现的信息推荐方法，其特征在于，所述方法包括：
    接收第一用户的需求数据；
    根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息；
    根据所述第二用户的信息提取所述第二用户的联系信息；
    根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户的处理包括：
    获取所述第一用户的联系信息，
    根据所述第二用户的联系信息分别将所述第一用户的需求数据及其联系信息发送给相应的第二用户。
  3. 根据权利要求1或2所述的方法，其特征在于，所述根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据的处理包括：
    根据所述需求数据获取第一用户的主需求类别数据和子需求类别数据，
    根据所述主需求类别数据和子需求类别数据获取多个候选推荐数据，
    以各个所述候选推荐数据以及获取的主需求类别数据和子需求类别数据作为输入，分别通过预先训练的需求匹配排序模型获取所述各候选推荐对象数据的相关性指数，
    根据所述相关性指数选取预定个数的所述候选推荐对象数据作为推荐对象数据。
  4. 根据权利要求1～3中任一项所述的方法，其特征在于，所述根据所述需求数据获取第一用户的主需求类别数据和子需求类别数据的处理包括：
    对所述需求数据进行自然语言分析，获取第一用户的主需求类别数据和子需求类别数据。
  5. 根据权利要求4所述的方法，其特征在于，所述第一用户是消费者用户，所述第二用户是商户用户。
  6. 一种信息推荐装置，其特征在于，所述装置包括：
    数据接收单元，用于接收第一用户的需求数据；
    对象数据获取单元，用于根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息；
    联系信息提取单元，用于根据所述第二用户的信息提取所述第二用户的联系信息；
    需求数据发送单元，用于根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。
  7. 根据权利要求6所述的装置，其特征在于，所述需求数据发送单元用于获取所述第一用户的联系信息，根据所述第二用户的联系信息分别将所述第一用户的需求数据及其联系信息发送给相应的第二用户。
  8. 根据权利要求6或7所述的装置，其特征在于，所述对象数据获取单元用于根据所述需求数据获取第一用户的主需求类别数据和子需求类别数据，
    根据所述主需求类别数据和子需求类别数据获取多个候选推荐数据，
    以各个所述候选推荐数据以及获取的主需求类别数据和子需求类别数据作为输入，分别通过预先训练的需求匹配排序模型获取所述各候选推荐对象数据的相关性指数，
    根据所述相关性指数选取预定个数的所述候选推荐对象数据作为推荐对象数据。
  9. 根据权利要求6～8中任一项所述的装置，其特征在于，对象数据获取单元对所述需求数据进行自然语言分析，获取第一用户的主需求类别数据和子需求类别数据。
  10. 根据权利要求9所述的装置，其特征在于，所述第一用户是消费者用户，所述第二用户是商户用户。
  11. 一种计算机设备，其特征在于，包括：
    一个或多个处理器；
    存储器；
    一个或多个程序，所述一个或多个程序存储在所述存储器中，且经 配置以由所述一个或者多个处理器执行所述一个或者多个程序包含的用于执行信息推荐方法的指令：
    接收第一用户的需求数据；
    根据所述需求数据以及预先训练的需求匹配排序模型获取预定个数的推荐对象数据，所述推荐对象数据包括满足第一用户的需求的第二用户的信息；
    根据所述第二用户的信息提取所述第二用户的联系信息；
    根据所述联系信息分别将所述第一用户的需求数据发送给相应的第二用户。
PCT/CN2015/098344 2014-12-23 2015-12-22 通过计算机实现的信息推荐方法、装置和计算机设备 Ceased WO2016101881A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410815315.1A CN104537552B (zh) 2014-12-23 2014-12-23 通过计算机实现的信息推荐方法及装置
CN201410815315.1 2014-12-23

Publications (1)

Publication Number Publication Date
WO2016101881A1 true WO2016101881A1 (zh) 2016-06-30

Family

ID=52853071

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/098344 Ceased WO2016101881A1 (zh) 2014-12-23 2015-12-22 通过计算机实现的信息推荐方法、装置和计算机设备

Country Status (2)

Country Link
CN (1) CN104537552B (zh)
WO (1) WO2016101881A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114500430A (zh) * 2022-02-09 2022-05-13 携程计算机技术（上海）有限公司 对话方法、系统、设备及存储介质
CN115659033A (zh) * 2022-11-01 2023-01-31 中国联合网络通信集团有限公司 商户推荐方法、装置及计算机可读存储介质

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104537552B (zh) * 2014-12-23 2018-01-05 百度在线网络技术（北京）有限公司 通过计算机实现的信息推荐方法及装置
CN106997357B (zh) * 2016-01-22 2020-10-09 腾讯科技（深圳）有限公司 一种消息处理方法、装置及系统
CN108022142A (zh) * 2016-11-03 2018-05-11 阿里巴巴集团控股有限公司 一种导购用户的确定方法、装置及电子设备
CN107977860A (zh) * 2017-11-15 2018-05-01 广东原始源正电子商务股份有限公司 一种基于移动网络的信息处理方法及系统
CN109919704A (zh) * 2019-01-21 2019-06-21 浙江口碑网络技术有限公司 一种资源数据的分发方法、装置及系统
CN113688311A (zh) * 2021-06-18 2021-11-23 诺正集团股份有限公司 基于数据交互的信息推荐方法、装置、设备及存储介质
CN113590977B (zh) * 2021-07-26 2025-03-25 北京沃东天骏信息技术有限公司 信息推荐方法、装置、电子设备和存储介质
CN115964397B (zh) * 2022-09-20 2023-09-19 成都比特信安科技有限公司 一种数据种子植入及溯源追踪方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080091448A1 (en) * 2006-10-16 2008-04-17 Niheu Eric K System and method of integrating enterprise applications
CN101488968A (zh) * 2008-01-15 2009-07-22 北京标比传媒科技有限公司 一种移动通信网络中信息推送的方法和系统
CN103927324A (zh) * 2014-03-13 2014-07-16 中国联合网络通信集团有限公司 一种信息推送的方法及装置
CN103942255A (zh) * 2014-03-19 2014-07-23 华存数据信息技术有限公司 个性化信息推荐系统及其推荐方法
CN104537552A (zh) * 2014-12-23 2015-04-22 百度在线网络技术（北京）有限公司 通过计算机实现的信息推荐方法及装置

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000132567A (ja) * 1998-10-27 2000-05-12 Nippon Telegr & Teleph Corp <Ntt> メッセージマッチング方法及び装置及びメッセージマッチングプログラムを格納した記憶媒体
CN101071424B (zh) * 2006-06-23 2010-08-25 腾讯科技（深圳）有限公司 一种个性化信息推送系统和方法
CN101383032A (zh) * 2008-09-12 2009-03-11 金斌斌 一种能接收外部数据并进行处理的商务交易系统及方法
CA2889279A1 (en) * 2012-10-10 2014-04-17 Ebay Inc. Connecting buyers and sellers
CN102915358B (zh) * 2012-10-16 2015-11-25 北京奇虎科技有限公司 导航网站实现方法和装置
CN103514299B (zh) * 2013-10-18 2018-04-17 北京奇虎科技有限公司 信息搜索方法和装置
CN104036045B (zh) * 2014-07-01 2018-04-03 彩带网络科技（北京）有限公司 一种信息分析方法及服务平台

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080091448A1 (en) * 2006-10-16 2008-04-17 Niheu Eric K System and method of integrating enterprise applications
CN101488968A (zh) * 2008-01-15 2009-07-22 北京标比传媒科技有限公司 一种移动通信网络中信息推送的方法和系统
CN103927324A (zh) * 2014-03-13 2014-07-16 中国联合网络通信集团有限公司 一种信息推送的方法及装置
CN103942255A (zh) * 2014-03-19 2014-07-23 华存数据信息技术有限公司 个性化信息推荐系统及其推荐方法
CN104537552A (zh) * 2014-12-23 2015-04-22 百度在线网络技术（北京）有限公司 通过计算机实现的信息推荐方法及装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114500430A (zh) * 2022-02-09 2022-05-13 携程计算机技术（上海）有限公司 对话方法、系统、设备及存储介质
CN115659033A (zh) * 2022-11-01 2023-01-31 中国联合网络通信集团有限公司 商户推荐方法、装置及计算机可读存储介质

Also Published As

Publication number Publication date
CN104537552A (zh) 2015-04-22
CN104537552B (zh) 2018-01-05

Similar Documents

Publication Publication Date Title
US11804035B2 (en) Intelligent online personal assistant with offline visual search database
CN111400507B (zh) 实体匹配方法及其装置
CN104281622B (zh) 一种社交媒体中的信息推荐方法和装置
JP2020042779A (ja) インターネットテキストマイニングに基づく関心地点の有効性の判断方法および装置
CN114528474A (zh) 推荐对象确定方法、装置、电子设备及存储介质
WO2018071525A1 (en) Personal assistant with visual multi-turn dialog
JP6846390B2 (ja) ドメイン間の文法スロットによる分類
WO2016206196A1 (zh) 用于获取用户属性信息的方法、装置及服务器
CN104537552B (zh) 通过计算机实现的信息推荐方法及装置
CN107729453B (zh) 一种提取中心产品词的方法和装置
CN105354199B (zh) 一种基于场景信息的实体含义识别方法和系统
WO2018126740A1 (zh) 用于推送信息的方法和装置
CN107730346A (zh) 物品聚类的方法和装置
CN106993030A (zh) 基于人工智能的信息推送方法和装置
JP2016038822A (ja) 抽出装置、抽出方法及び抽出プログラム
CN106776707A (zh) 信息推送的方法和装置
WO2019141109A1 (zh) 内容推荐方法及装置
KR102322668B1 (ko) 콘텐츠 크리에이터의 창작활동 활성화를 위한 다중 플랫폼 서비스 제공 시스템
CN106445971A (zh) 一种应用推荐方法和系统
CN107862004A (zh) 智能排序方法及装置、存储介质、电子设备
KR20210052746A (ko) 제품화 대상 콘텐츠의 예상 판매량 추정 방법, 장치 및 컴퓨터프로그램
CN111882224B (zh) 对消费场景进行分类的方法和装置
CN109739965A (zh) 跨领域对话策略的迁移方法及装置、设备、可读存储介质
CN111787042B (zh) 用于推送信息的方法和装置
KR101754124B1 (ko) 레스토랑 추천 시스템 및 추천 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15871949

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15871949

Country of ref document: EP

Kind code of ref document: A1