WO2021189949A1 - 信息推荐方法、装置、电子设备及介质 - Google Patents

信息推荐方法、装置、电子设备及介质 Download PDF

Info

Publication number
WO2021189949A1
WO2021189949A1 PCT/CN2020/134886 CN2020134886W WO2021189949A1 WO 2021189949 A1 WO2021189949 A1 WO 2021189949A1 CN 2020134886 W CN2020134886 W CN 2020134886W WO 2021189949 A1 WO2021189949 A1 WO 2021189949A1
Authority
WO
WIPO (PCT)
Prior art keywords
recommendation
recommendation information
user
decision tree
feature
Prior art date
Application number
PCT/CN2020/134886
Other languages
English (en)
French (fr)
Inventor
徐杰
Original Assignee
平安科技(深圳)有限公司
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 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2021189949A1 publication Critical patent/WO2021189949A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/045Combinations of networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0631Item recommendations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/08Insurance

Definitions

  • This application relates to the field of artificial intelligence technology, and in particular to an information recommendation method, device, electronic equipment and medium.
  • the recommendation system is a product of the era of big data and already exists in all aspects of people's lives.
  • the most representative one is similar recommendation, which recommends similar information to users based on information selected in the user's history.
  • the inventor realizes that this kind of recommendation is not personalized enough, and the diversity of recommendation information is not enough, and it is impossible to discover the potential needs of users for other information. Therefore, accurately recommending suitable information for users is an urgent problem to be solved.
  • the purpose of this application is to provide an information recommendation method, device, electronic equipment, and medium that can accurately recommend appropriate information to users.
  • an information recommendation method including: acquiring a user database, acquiring the first characteristic of each user in the user database, and according to the information of each user in the user database.
  • the first feature is to determine the pre-recommended users in the user database through a neural network model, and the pre-recommended users are one or more; for each pre-recommended user, the second feature of the pre-recommended user is acquired, and the The second feature of the pre-recommended user is input into one or more decision tree models, and the pre-recommendation result output by each decision tree model is obtained, wherein each of the decision tree models corresponds to a kind of pre-recommendation information; The recommendation result and the second characteristic of the pre-recommended user select the pre-recommendation information and combine to obtain recommendation information.
  • an information recommendation device including: an acquisition module for acquiring a user database, acquiring the first characteristic of each user in the user database, and according to the user database
  • the first feature of each user in the user database is determined by the neural network model to determine the pre-recommended users in the user database, and the pre-recommended users are one or more;
  • the pre-recommendation module is used to obtain for each pre-recommended user
  • the second feature of the pre-recommended user is input into one or more decision tree models to obtain the pre-recommendation result output by each decision tree model, wherein each of the decision tree models Corresponding to a kind of pre-recommendation information;
  • the recommendation module selects and combines the pre-recommendation information according to the pre-recommendation result and the second characteristic of the pre-recommended user to obtain recommendation information.
  • a computer (readable) program medium which stores computer program instructions (or referred to as computer readable instructions).
  • the computer Perform the following method: obtain a user database, obtain the first characteristic of each user in the user database, and determine the user database in the user database according to the first characteristic of each user in the user database through a neural network model
  • For each pre-recommended user obtain the second characteristic of the pre-recommended user, and input the second characteristic of the pre-recommended user into one or more decision trees
  • the pre-recommendation results output by each decision tree model are obtained, where each of the decision tree models corresponds to a kind of pre-recommendation information;
  • the pre-recommended information is combined to obtain the recommended information.
  • an electronic device including: a processor; a memory, where computer-readable instructions are stored, and when the computer-readable instructions are executed by the processor, the following is achieved Method: Obtain a user database, obtain the first feature of each user in the user database, and determine the preset in the user database through a neural network model according to the first feature of each user in the user database.
  • the number of the pre-recommended users is one or more; for each pre-recommended user, the second feature of the pre-recommended user is acquired, and the second feature of the pre-recommended user is input into one or more decision tree models To obtain the pre-recommendation results output by each decision tree model, where each of the decision tree models corresponds to a kind of pre-recommendation information; the pre-recommendation is selected according to the pre-recommendation result and the second characteristic of the pre-recommended user Information is combined to obtain recommended information.
  • the technical solutions provided by the embodiments of the present application can accurately recommend suitable information for users.
  • FIG. 1 shows a schematic diagram of an exemplary system architecture to which the technical solutions of the embodiments of the present application can be applied;
  • Fig. 2 schematically shows a flowchart of an information recommendation method according to an embodiment of the present application
  • Fig. 3 schematically shows a block diagram of an information recommendation device according to an embodiment of the present application
  • Fig. 4 is a hardware diagram of an electronic device according to an exemplary embodiment
  • Fig. 5 shows a computer-readable storage medium for implementing the foregoing information recommendation method according to an exemplary embodiment.
  • the technical solution of this application can be applied to the fields of artificial intelligence, smart city, blockchain and/or big data technology to realize intelligent information recommendation.
  • the data involved in this application can be stored in a database, or can be stored in a blockchain, such as distributed storage through a blockchain, which is not limited in this application.
  • FIG. 1 shows a schematic diagram of an exemplary system architecture 100 to which the technical solutions of the embodiments of the present application can be applied.
  • the system architecture 100 may include a terminal device 101 (the terminal device may be one or more of a smart phone, a tablet computer, a portable computer, and a desktop computer), a network 102 and a server 103.
  • the network 102 is used to provide a medium of a communication link between the terminal device 101 and the server 103.
  • the network 102 may include various connection types, such as wired communication links, wireless communication links, and so on.
  • the numbers of the terminal device 101, the network 102, and the server 103 in FIG. 1 are merely illustrative. According to implementation needs, there may be any number of terminal devices 101, networks 102, and servers 103.
  • the server 103 may be a server cluster composed of multiple servers.
  • the server 103 obtains the first feature of each user in the user database by obtaining the user database, and determines the user information through a neural network model according to the first feature of each user in the user database.
  • the pre-recommended users in the library there are one or more pre-recommended users; for each pre-recommended user, obtain the second characteristic of the pre-recommended user, and enter the second characteristic of the pre-recommended user into one or more decision trees
  • the pre-recommendation results output by each decision tree model are obtained, where each decision tree model corresponds to a kind of pre-recommendation information; the pre-recommendation information is selected and combined according to the pre-recommendation result and the second characteristic of the pre-recommended user to obtain the recommendation information , So as to accurately recommend suitable information for users.
  • the information recommendation method provided by the embodiment of the present application is generally executed by the server 103, and accordingly, the information recommendation device is generally set in the server 103.
  • the terminal device 101 may also have similar functions to the server 103, so as to execute the information recommendation method provided by the embodiments of the present application.
  • FIG. 2 schematically shows a flowchart of an information recommendation method according to an embodiment of the present application.
  • the execution subject of the information recommendation method may be a server, for example, the server 103 shown in FIG. 1.
  • the information recommendation method at least includes steps S210 to S230, which are described in detail as follows:
  • step S210 the user database is acquired, the first characteristic of each user in the user database is acquired, and the pre-recommended user in the user database is determined through the neural network model according to the first characteristic of each user in the user database. There are one or more pre-recommended users.
  • the first feature of each user in the user database can be input into the neural network model to obtain the output result of the neural network model, and the pre-recommendation in the user database can be determined according to the output result of the neural network model user.
  • the output result of the neural network model may be a list of pre-recommended users.
  • the neural network model can be trained by the following method: obtain a first feature sample set, the output result corresponding to each first feature sample in the first feature sample set is known, and the first feature sample corresponds to The output result may be whether the user corresponding to each first feature sample is a pre-recommended user. Input the first feature of each user in the user database into the neural network model, obtain the result of whether the user is a pre-recommended user output by the neural network model, and the output result corresponding to the known same first feature Compare, if they are inconsistent, adjust the neural network model so that the output result is consistent with the known output result corresponding to the same first feature.
  • the first feature may include demographic attributes, business attributes, geographic attributes, insurance awareness, life cycle, underwriting claims records, etc. It can be determined based on the first feature whether the pre-recommended user is a user who may purchase insurance , Only marketing to target users.
  • the first feature may include the user's memory capacity, data occupancy space, etc., and the pre-recommended user is judged to be a user who may need to be cached according to the first feature.
  • step S220 for each pre-recommended user, obtain the second characteristic of the pre-recommended user, and input the second characteristic of the pre-recommended user into one or more decision tree models to obtain each The pre-recommendation result output by the decision tree model, where each decision tree model corresponds to a kind of pre-recommendation information.
  • the second characteristic of the pre-recommended user may be household income level, whether to marry, whether to have a car, whether to have children, etc.
  • Each decision tree model corresponds to a type of insurance.
  • the second feature of the pre-recommended user may be the type of data that the user frequently accesses, the size of the data that the user frequently accesses, the user's online duration, etc., and each decision tree model corresponds to a type of cache information.
  • the second feature of the pre-recommendation information can be obtained, and the second feature of the pre-recommendation information includes one or more factors; according to the one or more factors of the second feature of the pre-recommendation information, The weight of the information is used to establish a decision tree model corresponding to the pre-recommendation information, so that the decision tree model corresponding to each pre-recommendation information conforms to the characteristics of the pre-recommendation information.
  • the bottom-up order of factors in the decision tree may be family income level, whether to marry, whether to have children, and whether to have a car.
  • the bottom-up factor arrangement of the decision tree may be the length of time the user spends online, the type of data that the user frequently accesses, and the size of the data that the user frequently accesses.
  • the second characteristic of the pre-recommended user includes one or more factors, and one or more factors in the second characteristic of the pre-recommended user can be input into each decision tree model, based on The second feature of the pre-recommended user and the second feature of the pre-recommended information in each decision tree model are used to obtain the matching probability between the pre-recommended user output by each decision tree model and the pre-recommended information corresponding to the decision tree model.
  • step S230 the pre-recommendation information is selected and combined according to the pre-recommendation result and the second characteristic of the pre-recommended user to obtain recommendation information.
  • the pre-recommendation information can be used as the recommendation information.
  • the second feature of the pre-recommended user may include the total resource W; N pre-recommendation information is selected from the pre-recommendation information corresponding to the M model trees, and the selected N-th pre-recommendation information corresponds to
  • the total resource W may be the user’s annual household income.
  • the type of insurance suitable for the family is determined.
  • Select N types of insurance from M types of insurance that are suitable for recommendation to the user, and match the user’s second feature with these N types of insurance, and obtain the matching probability of each insurance in the N types of insurance that is suitable for the user, denoted as P MN (N 1, 2, 3).
  • the N types of insurances themselves have their corresponding weights.
  • the ⁇ MN for critical illness insurance can be 50%
  • the ⁇ MN for cancer prevention insurance can be 20%...
  • the total resource W may be the total number of cached information.
  • the user can find suitable N types of cache modes from the M types of cache modes. These N types of cache modes and the number of cached information
  • the matching probability is P MN
  • the weight of the N-th cache method itself is ⁇ MN
  • the value of ⁇ may be 10%.
  • the amount used to purchase insurance in a family can be 10% of the annual income, which will not affect family life while protecting family health.
  • Fig. 3 schematically shows a block diagram of an information recommendation apparatus according to an embodiment of the present application.
  • the information recommendation device 300 includes an acquisition module 301, a pre-recommendation module 302, and a recommendation module 303.
  • the obtaining module 301 is used to obtain a user database, obtain the first feature of each user in the user database, and pass The neural network model determines the pre-recommended users in the user database, and there are one or more pre-recommended users; the pre-recommendation module 302 is used for each pre-recommended user to obtain the second characteristic of the pre-recommended user, and the pre-recommended user
  • the second feature of is input into one or more decision tree models to obtain the pre-recommendation results output by each decision tree model, where each decision tree model corresponds to a kind of pre-recommendation information;
  • the recommendation module 303 is used to obtain pre-recommendation results according to the pre-recommendation results
  • the pre-recommendation information is selected and combined with the second characteristic of the pre-recommended user to obtain the recommendation information.
  • the pre-recommendation module 302 is configured to: obtain a second feature of the pre-recommendation information, the second feature of the pre-recommendation information includes one or more factors; according to the second feature of the pre-recommendation information The weight of one or more factors of the characteristic in the pre-recommendation information is used to establish a decision tree model corresponding to the pre-recommendation information.
  • the second characteristic of the pre-recommended user includes one or more factors; the pre-recommendation module 302 is configured to: one or more factors in the second characteristic of the pre-recommended user Input into each decision tree model; based on the second feature of the pre-recommended user and the second feature of the pre-recommended information in each decision tree model, the pre-recommended user and the pre-recommended user output by each decision tree model are obtained.
  • the matching probability between the pre-recommendation information corresponding to the decision tree model is obtained.
  • the recommendation module 303 is configured to use the pre-recommendation information as the recommendation information.
  • the recommendation module 303 is configured such that the value of ⁇ is 10%.
  • An embodiment of the present application also provides an electronic device, which may include one or more processors and storage devices.
  • the storage device is used to store one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors are allowed to implement part or All the steps are not repeated here.
  • the processor may also be referred to as a processing unit, and the memory may be referred to as a storage unit.
  • the electronic device 40 according to this embodiment of the present application will be described below with reference to FIG. 4.
  • the electronic device 40 shown in FIG. 4 is only an example, and should not bring any limitation to the function and scope of use of the embodiments of the present application.
  • the electronic device 40 is represented in the form of a general-purpose computing device.
  • the components of the electronic device 40 may include, but are not limited to: the aforementioned at least one processing unit 41, the aforementioned at least one storage unit 42, a bus 43 connecting different system components (including the storage unit 42 and the processing unit 41), and a display unit 44.
  • the storage unit stores program code, and the program code can be executed by the processing unit 41, so that the processing unit 41 executes the various exemplary methods described in the "Methods of Embodiments" section of this specification. Steps of implementation.
  • the storage unit 42 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 421 and/or a cache storage unit 422, and may further include a read-only storage unit (ROM) 423.
  • RAM random access storage unit
  • ROM read-only storage unit
  • the storage unit 42 may also include a program/utility tool 424 having a set of (at least one) program modules 425.
  • program modules 425 include but are not limited to: an operating system, one or more application programs, other program modules, and program data, Each of these examples or some combination may include the implementation of a network environment.
  • the bus 43 may represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local area using any bus structure among multiple bus structures. bus.
  • the electronic device 40 may also communicate with one or more external devices (such as keyboards, pointing devices, Bluetooth devices, etc.), and may also communicate with one or more devices that enable a user to interact with the electronic device 40, and/or communicate with
  • the electronic device 40 can communicate with any device (such as a router, modem, etc.) that communicates with one or more other computing devices. This communication can be performed through an input/output (I/O) interface 45.
  • the electronic device 40 may also communicate with one or more networks (for example, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 46. As shown in the figure, the network adapter 46 communicates with other modules of the electronic device 40 through the bus 43.
  • LAN local area network
  • WAN wide area network
  • public network such as the Internet
  • the example embodiments described here can be implemented by software, or can be implemented by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, U disk, mobile hard disk, etc.) or on the network , Including several instructions to make a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) execute the method according to the embodiment of the present application.
  • a computing device which can be a personal computer, a server, a terminal device, or a network device, etc.
  • a computer-readable storage medium (or referred to as a computer program medium) is also provided, on which a program product (such as a computer-readable instruction) capable of implementing the above method of this specification is stored.
  • a program product such as a computer-readable instruction
  • various aspects of the present application can also be implemented in the form of a program product, which includes program code.
  • the program product runs on a terminal device, the program code is used to make the The terminal device executes the steps according to various exemplary embodiments of the present application described in the above-mentioned "Exemplary Method" section of this specification.
  • the medium involved in this application such as a computer-readable storage medium, may be non-volatile or volatile.
  • a program product 50 for implementing the above method according to an embodiment of the present application is described. It can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be installed in a terminal device, For example, running on a personal computer.
  • CD-ROM compact disk read-only memory
  • the program product of this application is not limited to this.
  • the readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, device, or device.
  • the program product can use any combination of one or more readable media.
  • the readable medium may be a readable signal medium or a readable storage medium.
  • the readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or a combination of any of the above. More specific examples (non-exhaustive list) of readable storage media include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable Type programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • the computer-readable signal medium may include a data signal propagated in baseband or as a part of a carrier wave, and readable program code is carried therein. This propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • the readable signal medium may also be any readable medium other than a readable storage medium, and the readable medium may send, propagate, or transmit a program for use by or in combination with the instruction execution system, apparatus, or device.
  • the program code contained on the readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.
  • the program code used to perform the operations of this application can be written in any combination of one or more programming languages.
  • the programming languages include object-oriented programming languages—such as Java, C++, etc., as well as conventional procedural programming languages. Programming language-such as "C" language or similar programming language.
  • the program code can be executed entirely on the user's computing device, partly on the user's device, executed as an independent software package, partly on the user's computing device and partly executed on the remote computing device, or entirely on the remote computing device or server Executed on.
  • the remote computing device can be connected to a user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (for example, using Internet service providers). Business to connect via the Internet).
  • LAN local area network
  • WAN wide area network
  • Internet service providers for example, using Internet service providers.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Health & Medical Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Technology Law (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种信息推荐方法、装置。该方法包括获取用户资料库,获取用户资料库中每个用户的第一特征,根据用户资料库中每个用户的第一特征,通过神经网络模型确定用户资料库中的预推荐用户,预推荐用户为一个或多个(S210)。对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个决策树模型对应一种预推荐信息(S220);根据预推荐结果和预推荐用户的第二特征选取预推荐信息进行组合得到推荐信息(S230),能够准确的为用户推荐合适的信息。

Description

信息推荐方法、装置、电子设备及介质
本申请要求于2020年3月26日提交中国专利局、申请号为202010223996.8,发明名称为“信息推荐方法、装置、电子设备及介质”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
本申请涉及人工智能技术领域,特别涉及一种信息推荐方法、装置、电子设备及介质。
背景技术
随着生活水平的提高,人们对各种应用的要求越来越高,推荐系统应运而生。推荐系统是大数据时代的产物,已经存在于人们生活的方方面面。
现有的推荐系统中,最具有代表性的一种是同类推荐,根据用户历史选择过的信息,向用户推荐同类信息。发明人意识到,这种推荐个性化不足,推荐信息的多样性不够,不能发现用户对其他信息的潜在需求。因此,准确的为用户推荐合适的信息是亟待解决的问题。
发明内容
本申请旨在提供一种信息推荐方法、装置、电子设备及介质,能够准确的为用户推荐合适的信息。
根据本申请实施例的一个方面,提供了一种信息推荐方法,包括:获取用户资料库,获取所述用户资料库中每个用户的第一特征,根据所述用户资料库中每个用户的第一特征,通过神经网络模型确定所述用户资料库中的预推荐用户,所述预推荐用户为一个或多个;对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个所述决策树模型对应一种预推荐信息;根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息。
根据本申请实施例的一个方面,提供了一种信息推荐装置,包括:获取模块,用于获取用户资料库,获取所述用户资料库中每个用户的第一特征,根据所述用户资料库中每个用户的第一特征,通过神经网络模型确定所述用户资料库中的预推荐用户,所述预推荐用户为一个或多个;预推荐模块,用于对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个所述决策树模型对应一种预推荐信息;推荐模块,根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息。
根据本申请实施例的一个方面,提供了一种计算机(可读)程序介质,其存储有计算机程序指令(或者称为计算机可读指令),当所述计算机程序指令被计算机执行时,使计算机执行以下方法:获取用户资料库,获取所述用户资料库中每个用户的第一特征,根据所述用户资料库中每个用户的第一特征,通过神经网络模型确定所述用户资料库中的预推荐用户,所述预推荐用户为一个或多个;对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个所述决策树模型对应一种预推荐信息;根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息。
根据本申请实施例的一个方面,提供了一种电子装置,包括:处理器;存储器,所述存储器上存储有计算机可读指令,所述计算机可读指令被所述处理器执行时,实现以下方法:获取用户资料库,获取所述用户资料库中每个用户的第一特征,根据所述用户资料库中每个用户的第一特征,通过神经网络模型确定所述用户资料库中的预推荐用户,所述预推荐用户为一个或多个;对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结 果,其中,每个所述决策树模型对应一种预推荐信息;根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息。
本申请的实施例提供的技术方案可以准确的为用户推荐合适的信息。
应当理解的是,以上的一般描述和后文的细节描述仅是示例性的,并不能限制本申请。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分,示出了符合本申请的实施例,并于说明书一起用于解释本申请的原理。
图1示出了可以应用本申请实施例的技术方案的示例性系统架构的示意图;
图2示意性示出了根据本申请的一个实施例的信息推荐方法的流程图;
图3示意性示出了根据本申请的一个实施例的信息推荐装置的框图;
图4是根据一示例性实施例示出的一种电子装置的硬件图;
图5是根据一示例性实施例示出的一种用于实现上述信息推荐方法的计算机可读存储介质。
具体实施方式
现在将参考附图更全面地描述示例实施方式。然而,示例实施方式能够以多种形式实施,且不应被理解为限于在此阐述的范例;相反,提供这些实施方式使得本申请将更加全面和完整,并将示例实施方式的构思全面地传达给本领域的技术人员。
此外,所描述的特征、结构或特性可以以任何合适的方式结合在一个或更多实施例中。在下面的描述中,提供许多具体细节从而给出对本申请的实施例的充分理解。然而,本领域技术人员将意识到,可以实践本申请的技术方案而没有特定细节中的一个或更多,或者可以采用其它的方法、组元、装置、步骤等。在其它情况下,不详细示出或描述公知方法、装置、实现或者操作以避免模糊本申请的各方面。
附图中所示的方框图仅仅是功能实体,不一定必须与物理上独立的实体相对应。即,可以采用软件形式来实现这些功能实体,或在一个或多个硬件模块或集成电路中实现这些功能实体,或在不同网络和/或处理器装置和/或微控制器装置中实现这些功能实体。
附图中所示的流程图仅是示例性说明,不是必须包括所有的内容和操作/步骤,也不是必须按所描述的顺序执行。例如,有的操作/步骤还可以分解,而有的操作/步骤可以合并或部分合并,因此实际执行的顺序有可能根据实际情况改变。
本申请的技术方案可应用于人工智能、智慧城市、区块链和/或大数据技术领域,以实现智能化信息推荐。可选的,本申请涉及的数据如特征和/或推荐信息等可存储于数据库中,或者可以存储于区块链中,比如通过区块链分布式存储,本申请不做限定。
图1示出了可以应用本申请实施例的技术方案的示例性系统架构100的示意图。
如图1所示,系统架构100可以包括终端设备101(终端设备可以为智能手机、平板电脑、便携式计算机、台式计算机中的一种或多种)、网络102和服务器103。网络102用以在终端设备101和服务器103之间提供通信链路的介质。网络102可以包括各种连接类型,例如有线通信链路、无线通信链路等等。
应该理解,图1中的终端设备101、网络102和服务器103的数目仅仅是示意性的。根据实现需要,可以具有任意数目的终端设备101、网络102和服务器103。比如服务器103可以是多个服务器组成的服务器集群等。
在本申请的一个实施例中,服务器103通过获取用户资料库,获取用户资料库中每个用户的第一特征,根据用户资料库中每个用户的第一特征,通过神经网络模型确定用户资料库中的预推荐用户,预推荐用户为一个或多个;对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个决策树模型对应一种预推荐信息;根 据预推荐结果和预推荐用户的第二特征选取预推荐信息进行组合得到推荐信息,从而准确的为用户推荐合适的信息。
需要说明的是,本申请实施例所提供的信息推荐方法一般由服务器103执行,相应地,信息推荐装置一般设置于服务器103中。但是,在本申请的其它实施例中,终端设备101也可以与服务器103具有相似的功能,从而执行本申请实施例所提供的信息推荐方法。
以下对本申请实施例的技术方案的实现细节进行详细阐述:
图2示意性示出了根据本申请的一个实施例的信息推荐方法的流程图,该信息推荐方法的执行主体可以是服务器,比如可以是图1中所示的服务器103。
参照图2所示,该信息推荐方法至少包括步骤S210至步骤S230,详细介绍如下:
在步骤S210中,获取用户资料库,获取用户资料库中每个用户的第一特征,根据用户资料库中每个用户的第一特征,通过神经网络模型确定用户资料库中的预推荐用户,预推荐用户为一个或多个。
在本申请的一个实施例中,可以将用户资料库中每个用户的第一特征输入神经网络模型,得到神经网络模型的输出结果,根据神经网络模型的输出结果确定用户资料库中的预推荐用户。
在本申请的一个实施例中,神经网络模型的输出结果可以为预推荐用户名单。
在本申请的一个实施例中,可以通过以下方法训练神经网络模型:获取第一特征样本集合,第一特征样本集合中每个第一特征样本对应的输出结果已知,第一特征样本对应的输出结果可以是每个第一特征样本对应的用户是否为预推荐用户。将用户资料库中每个用户的第一特征输入神经网络模型,获取神经网络模型输出的该用户是否为预推荐用户的结果,将输出的结果和已知的相同的第一特征对应的输出结果比较,如果不一致,调整神经网络模型,使得输出的结果和已知的相同的第一特征对应的输出结果一致。
在本申请的一个实施例中,第一特征可以包括人口属性、企业属性、地理属性、保险意识、生命周期、承保理赔记录等,可以根据第一特征判断预推荐用户是否为可能购买保险的用户,只对目标用户营销。
在本申请的一个实施例中,第一特征可以包括用户的内存容量、数据占用空间等,根据第一特征判断预推荐用户为可能需要缓存用户。
继续参照图2,在步骤S220中,对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个决策树模型对应一种预推荐信息。
在本申请的一个实施例中,预推荐用户的第二特征可以为家庭收入水平、是否结婚、是否有车、是否有小孩等,每个决策树模型对应一种保险。
在本申请的一个实施例中,预推荐用户的第二特征可以为用户经常访问的数据类型、用户经常访问的数据大小、用户的上网时长等,每个决策树模型对应一种缓存信息。
在本申请的一个实施例中,可以获取预推荐信息的第二特征,预推荐信息的第二特征包括一个或多个因素;根据预推荐信息的第二特征的一个或多个因素在预推荐信息中所占的权重,建立预推荐信息对应的决策树模型,使每个预推荐信息对应的决策树模型都符合该预推荐信息的特点。
在本申请的一个实施例中,由于位于决策树根部的因素对决策树的结果影响最大,可以将影响越大的因素设置得越靠近决策树根部。
在本申请的一个实施例中,决策树由下至上的因素排列可以为家庭收入水平、是否结婚、是否有小孩、是否有车。
在本申请的一个实施例中,决策树由下至上的因素排列可以为用户的上网时长、用户 经常访问的数据类型、用户经常访问的数据大小。
在本申请的一个实施例中,预推荐用户的第二特征包括一个或多个因素,可以将该预推荐用户的第二特征中的一个或多个因素输入至每个决策树模型中,基于预推荐用户的第二特征与每个决策树模型中预推荐信息的第二特征,得到每个决策树模型输出的该预推荐用户与该决策树模型对应的预推荐信息之间的匹配概率。
继续参照图2,在步骤S230中,根据预推荐结果和预推荐用户的第二特征选取预推荐信息进行组合得到推荐信息。
在本申请的一个实施例中,若预推荐信息只有一种,则可以将预推荐信息作为推荐信息。
在本申请的一个实施例中,预推荐用户的第二特征可以包括总资源W;从M个模型树对应的预推荐信息中选取N个预推荐信息,将选取的第N个预推荐信息对应的决策树模型输出的匹配概率记为P MN(N=1,2,3……);将第N个决策树模型对应的预推荐信息预设权重记为θ M(M=1,2,3…M),通过公式:W N=P MNθ MW×β/(P M1+P M2+…+P MN)得到第N个预推荐信息的建议分配资源W M,其中,β为预设系数。
在本申请的一个实施例中,总资源W可以为用户的家庭年收入,根据该用户的第二特征,如是否有小孩、是否有车、是否有老人,来确定适合该家庭的保险种类,从M个种类的保险中选出N种保险适合推荐给该用户,用户的第二特征和这N种保险进行匹配,得到这N种保险中每个保险适合该用户的匹配概率,记为P MN(N=1,2,3……)。其中,这N种保险本身有其对应的权重,如重疾险的θ MN可以为50%,防癌险的θ MN可以为20%……则该用户购买第N种保险花费的合理金额为:W N=P MNθ MNW×β/(P M1+P M2+…+P MN)。
在本申请的一个实施例中,总资源W可以为缓存信息的总数,根据缓存信息的种类可以为用户从M种缓存方式中找到合适的N种缓存方式,这N种缓存方式和缓存信息的匹配概率为P MN,第N种缓存方式自身的权重为θ MN,则用户将缓存信息分配至第N种缓存方式中的信息量为:W N=P MNθ MNW×β/(P M1+P M2+…+P MN)。
在本申请的一个实施例中,β的值可以为10%。
在本申请的一个实施例中,在一个家庭中用于购买保险的金额可以为年收入的10%,在保障家庭健康的同时又不会影响家庭生活。
以下介绍本申请的装置实施例,可以用于执行本申请上述实施例中的信息推荐方法。对于本申请装置实施例中未披露的细节,请参照本申请上述的信息推荐方法的实施例。
图3示意性示出了根据本申请的一个实施例的信息推荐装置的框图。
参照图3所示,根据本申请的一个实施例的信息推荐装置300,包括获取模块301、预推荐模块302和推荐模块303。
在本申请的一些实施例中,基于前述方案,获取模块301用于获取用户资料库,获取用户资料库中每个用户的第一特征,根据用户资料库中每个用户的第一特征,通过神经网络模型确定用户资料库中的预推荐用户,预推荐用户为一个或多个;预推荐模块302用于对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个决策树模 型对应一种预推荐信息;推荐模块303用于根据预推荐结果和预推荐用户的第二特征选取预推荐信息进行组合得到推荐信息。
在本申请的一些实施例中,基于前述方案,预推荐模块302配置为:获取预推荐信息的第二特征,预推荐信息的第二特征包括一个或多个因素;根据预推荐信息的第二特征的一个或多个因素在预推荐信息中所占的权重,建立预推荐信息对应的决策树模型。
在本申请的一些实施例中,基于前述方案,预推荐用户的第二特征包括一个或多个因素;预推荐模块302配置为:将该预推荐用户的第二特征中的一个或多个因素输入至每个决策树模型中;基于所述预推荐用户的第二特征与每个决策树模型中所述预推荐信息的第二特征,得到每个决策树模型输出的该预推荐用户与该决策树模型对应的预推荐信息之间的匹配概率。
在本申请的一些实施例中,基于前述方案,推荐模块303配置为:将预推荐信息作为推荐信息。
在本申请的一些实施例中,基于前述方案,推荐模块303配置为:预推荐用户的第二特征包括总资源W;将第M个决策树模型对应的预推荐信息预设权重记为θ M(M=1,2,3…M),第M个决策树模型输出的匹配概率记为P M;从M个模型树对应的预推荐信息中选取N个预推荐信息,将选取的第N个预推荐信息对应的决策树模型输出的匹配概率记为P MN(N=1,2,3……);通过公式:W N=P MNθ MW×β/(P M1+P M2+…+P MN)得到第N个预推荐信息的建议分配资源W M,其中,β为预设系数。
在本申请的一些实施例中,基于前述方案,推荐模块303配置为:β的值为10%。
所属技术领域的技术人员能够理解,本申请的各个方面可以实现为系统、方法或程序产品。因此,本申请的各个方面可以具体实现为以下形式,即:完全的硬件实施方式、完全的软件实施方式(包括固件、微代码等),或硬件和软件方面结合的实施方式,这里可以统称为“电路”、“模块”或“系统”。
本申请实施例还提供了一种电子设备,该电子设备可包括一个或多个处理器和存储装置。其中,存储装置,用于存储一个或多个程序,当所述一个或多个程序被所述一个或多个处理器执行时,使得所述一个或多个处理器实现上述方法中的部分或全部步骤,此处不赘述。该处理器还可称为处理单元,存储器可以称为存储单元。
例如,下面参照图4来描述根据本申请的这种实施方式的电子设备40。图4显示的电子设备40仅仅是一个示例,不应对本申请实施例的功能和使用范围带来任何限制。
如图4所示,电子设备40以通用计算设备的形式表现。电子设备40的组件可以包括但不限于:上述至少一个处理单元41、上述至少一个存储单元42、连接不同系统组件(包括存储单元42和处理单元41)的总线43、显示单元44。
其中,所述存储单元存储有程序代码,所述程序代码可以被所述处理单元41执行,使得所述处理单元41执行本说明书上述“实施例方法”部分中描述的根据本申请各种示例性实施方式的步骤。
存储单元42可以包括易失性存储单元形式的可读介质,例如随机存取存储单元(RAM)421和/或高速缓存存储单元422,还可以进一步包括只读存储单元(ROM)423。
存储单元42还可以包括具有一组(至少一个)程序模块425的程序/实用工具424,这样的程序模块425包括但不限于:操作系统、一个或者多个应用程序、其它程序模块以及程序数据,这些示例中的每一个或某种组合中可能包括网络环境的实现。
总线43可以为表示几类总线结构中的一种或多种,包括存储单元总线或者存储单元控制器、外围总线、图形加速端口、处理单元或者使用多种总线结构中的任意总线结构的局域总线。
电子设备40也可以与一个或多个外部设备(例如键盘、指向设备、蓝牙设备等)通信,还可与一个或者多个使得用户能与该电子设备40交互的设备通信,和/或与使得该电子设备40能与一个或多个其它计算设备进行通信的任何设备(例如路由器、调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口45进行。并且,电子设备40还可以通过网络适配器46与一个或者多个网络(例如局域网(LAN),广域网(WAN)和/或公共网络,例如因特网)通信。如图所示,网络适配器46通过总线43与电子设备40的其它模块通信。应当明白,尽管图中未示出,可以结合电子设备40使用其它硬件和/或软件模块,包括但不限于:微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、RAID系统、磁带驱动器以及数据备份存储系统等。
通过以上的实施方式的描述,本领域的技术人员易于理解,这里描述的示例实施方式可以通过软件实现,也可以通过软件结合必要的硬件的方式来实现。因此,根据本申请实施方式的技术方案可以以软件产品的形式体现出来,该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM,U盘,移动硬盘等)中或网络上,包括若干指令以使得一台计算设备(可以是个人计算机、服务器、终端装置、或者网络设备等)执行根据本申请实施方式的方法。
根据本申请一个实施例,还提供了一种计算机可读存储介质(或者称为计算机程序介质),其上存储有能够实现本说明书上述方法的程序产品(如计算机可读指令)。在一些可能的实施方式中,本申请的各个方面还可以实现为一种程序产品的形式,其包括程序代码,当所述程序产品在终端设备上运行时,所述程序代码用于使所述终端设备执行本说明书上述“示例性方法”部分中描述的根据本申请各种示例性实施方式的步骤。
可选的,本申请涉及的介质如计算机可读存储介质可以是非易失性的,也可以是易失性的。
参考图5所示,描述了根据本申请的实施方式的用于实现上述方法的程序产品50,其可以采用便携式紧凑盘只读存储器(CD-ROM)并包括程序代码,并可以在终端设备,例如个人电脑上运行。然而,本申请的程序产品不限于此,在本文件中,可读存储介质可以是任何包含或存储程序的有形介质,该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
所述程序产品可以采用一个或多个可读介质的任意组合。可读介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以为但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件,或者任意以上的组合。可读存储介质的更具体的例子(非穷举的列表)包括:具有一个或多个导线的电连接、便携式盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。
计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号,其中承载了可读程序代码。这种传播的数据信号可以采用多种形式,包括但不限于电磁信号、光信号或上述的任意合适的组合。可读信号介质还可以是可读存储介质以外的任何可读介质,该可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。
可读介质上包含的程序代码可以用任何适当的介质传输,包括但不限于无线、有线、光缆、RF等等,或者上述的任意合适的组合。
可以以一种或多种程序设计语言的任意组合来编写用于执行本申请操作的程序代码, 所述程序设计语言包括面向对象的程序设计语言—诸如Java、C++等,还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算设备上执行、部分地在用户设备上执行、作为一个独立的软件包执行、部分在用户计算设备上部分在远程计算设备上执行、或者完全在远程计算设备或服务器上执行。在涉及远程计算设备的情形中,远程计算设备可以通过任意种类的网络,包括局域网(LAN)或广域网(WAN),连接到用户计算设备,或者,可以连接到外部计算设备(例如利用因特网服务提供商来通过因特网连接)。
此外,上述附图仅是根据本申请示例性实施例的方法所包括的处理的示意性说明,而不是限制目的。易于理解,上述附图所示的处理并不表明或限制这些处理的时间顺序。另外,也易于理解,这些处理可以是例如在多个模块中同步或异步执行的。
应当理解的是,本申请并不局限于上面已经描述并在附图中示出的精确结构,并且可以在不脱离其范围执行各种修改和改变。本申请的范围仅由所附的权利要求来限制。

Claims (20)

  1. 一种信息推荐方法,其中,包括:
    获取用户资料库,获取所述用户资料库中每个用户的第一特征,根据所述用户资料库中每个用户的第一特征,通过神经网络模型确定所述用户资料库中的预推荐用户,所述预推荐用户为一个或多个;
    对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个所述决策树模型对应一种预推荐信息;
    根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息。
  2. 根据权利要求1所述的信息推荐方法,其中,在将该预推荐用户的第二特征输入至一个或多个决策树模型中之前,所述方法包括:
    获取所述预推荐信息的第二特征,所述预推荐信息的第二特征包括一个或多个因素;
    根据所述预推荐信息的第二特征的一个或多个因素在所述预推荐信息中所占的权重,建立所述预推荐信息对应的决策树模型。
  3. 根据权利要求2所述的信息推荐方法,其中,所述预推荐用户的第二特征包括一个或多个因素;
    所述将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,包括:
    将该预推荐用户的第二特征中的一个或多个因素输入至每个决策树模型中;
    基于所述预推荐用户的第二特征与每个决策树模型中所述预推荐信息的第二特征,得到每个决策树模型输出的该预推荐用户与该决策树模型对应的预推荐信息之间的匹配概率。
  4. 根据权利要求1所述的信息推荐方法,其中,所述根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息,包括:
    将所述预推荐信息作为所述推荐信息。
  5. 根据权利要求4所述的信息推荐方法,其中,所述根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息,包括:
    所述预推荐用户的第二特征包括总资源W;
    将第M个决策树模型对应的预推荐信息预设权重记为θ M(M=1,2,3…M),第M个决策树模型输出的匹配概率记为P M
    从M个模型树对应的预推荐信息中选取N个所述预推荐信息,将选取的第N个所述预推荐信息对应的决策树模型输出的匹配概率记为P MN(N=1,2,3……);
    通过公式:W N=P MNθ MW×β/(P M1+P M2+…+P MN)得到第N个预推荐信息的建议分配资源W M,其中,β为预设系数。
  6. 根据权利要求5所述的信息推荐方法,其中,所述β的值为10%。
  7. 一种信息推荐装置,其中,包括:
    获取模块,用于获取用户资料库,获取所述用户资料库中每个用户的第一特征,根据所述用户资料库中每个用户的第一特征,通过神经网络模型确定所述用户资料库中的预推荐用户,用于所述预推荐用户为一个或多个;
    预推荐模块,对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个所述决策树模型对应一种预推荐信息;
    推荐模块,根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息。
  8. 根据权利要求7所述的信息推荐装置,其中,
    所述预推荐模块,还用于获取所述预推荐信息的第二特征,所述预推荐信息的第二特征包括一个或多个因素;根据所述预推荐信息的第二特征的一个或多个因素在所述预推荐信息中所占的权重,建立所述预推荐信息对应的决策树模型。
  9. 一种电子设备,其中,包括:
    一个或多个处理器;
    存储装置,用于存储一个或多个程序,当所述一个或多个程序被所述一个或多个处理器执行时,使得所述一个或多个处理器实现以下方法:
    获取用户资料库,获取所述用户资料库中每个用户的第一特征,根据所述用户资料库中每个用户的第一特征,通过神经网络模型确定所述用户资料库中的预推荐用户,所述预推荐用户为一个或多个;
    对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个所述决策树模型对应一种预推荐信息;
    根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息。
  10. 根据权利要求9所述的电子设备,其中,在将该预推荐用户的第二特征输入至一个或多个决策树模型中之前,所述一个或多个处理器还用于实现:
    获取所述预推荐信息的第二特征,所述预推荐信息的第二特征包括一个或多个因素;
    根据所述预推荐信息的第二特征的一个或多个因素在所述预推荐信息中所占的权重,建立所述预推荐信息对应的决策树模型。
  11. 根据权利要求10所述的电子设备,其中,所述预推荐用户的第二特征包括一个或多个因素;
    所述将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果时,具体实现:
    将该预推荐用户的第二特征中的一个或多个因素输入至每个决策树模型中;
    基于所述预推荐用户的第二特征与每个决策树模型中所述预推荐信息的第二特征,得到每个决策树模型输出的该预推荐用户与该决策树模型对应的预推荐信息之间的匹配概率。
  12. 根据权利要求9所述的电子设备,其中,所述根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息时,具体实现:
    将所述预推荐信息作为所述推荐信息。
  13. 根据权利要求12所述的电子设备,其中,所述根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息时,具体实现:
    所述预推荐用户的第二特征包括总资源W;
    将第M个决策树模型对应的预推荐信息预设权重记为θ M(M=1,2,3…M),第M个决策树模型输出的匹配概率记为P M
    从M个模型树对应的预推荐信息中选取N个所述预推荐信息,将选取的第N个所述 预推荐信息对应的决策树模型输出的匹配概率记为P MN(N=1,2,3……);
    通过公式:W N=P MNθ MW×β/(P M1+P M2+…+P MN)得到第N个预推荐信息的建议分配资源W M,其中,β为预设系数。
  14. 根据权利要求13所述的电子设备,其中,所述β的值为10%。
  15. 一种计算机程序介质,其上存储有计算机可读指令,当所述计算机可读指令被计算机的处理器执行时,使计算机执行以下方法:
    获取用户资料库,获取所述用户资料库中每个用户的第一特征,根据所述用户资料库中每个用户的第一特征,通过神经网络模型确定所述用户资料库中的预推荐用户,所述预推荐用户为一个或多个;
    对于每个预推荐用户,获取该预推荐用户的第二特征,将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果,其中,每个所述决策树模型对应一种预推荐信息;
    根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息。
  16. 根据权利要求15所述的计算机程序介质,其中,在将该预推荐用户的第二特征输入至一个或多个决策树模型中之前,当所述计算机可读指令被计算机的处理器执行时还使计算机执行:
    获取所述预推荐信息的第二特征,所述预推荐信息的第二特征包括一个或多个因素;
    根据所述预推荐信息的第二特征的一个或多个因素在所述预推荐信息中所占的权重,建立所述预推荐信息对应的决策树模型。
  17. 根据权利要求16所述的计算机程序介质,其中,所述预推荐用户的第二特征包括一个或多个因素;
    所述将该预推荐用户的第二特征输入至一个或多个决策树模型中,得到每个决策树模型输出的预推荐结果时,具体执行:
    将该预推荐用户的第二特征中的一个或多个因素输入至每个决策树模型中;
    基于所述预推荐用户的第二特征与每个决策树模型中所述预推荐信息的第二特征,得到每个决策树模型输出的该预推荐用户与该决策树模型对应的预推荐信息之间的匹配概率。
  18. 根据权利要求15所述的计算机程序介质,其中,所述根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息时,具体执行:
    将所述预推荐信息作为所述推荐信息。
  19. 根据权利要求18所述的计算机程序介质,其中,所述根据所述预推荐结果和所述预推荐用户的第二特征选取所述预推荐信息进行组合得到推荐信息时,具体执行:
    所述预推荐用户的第二特征包括总资源W;
    将第M个决策树模型对应的预推荐信息预设权重记为θ M(M=1,2,3…M),第M个决策树模型输出的匹配概率记为P M
    从M个模型树对应的预推荐信息中选取N个所述预推荐信息,将选取的第N个所述预推荐信息对应的决策树模型输出的匹配概率记为P MN(N=1,2,3……);
    通过公式:W N=P MNθ MW×β/(P M1+P M2+…+P MN)得到第N个预推荐信息的建议分配资 源W M,其中,β为预设系数。
  20. 根据权利要求19所述的计算机程序介质,其中,所述β的值为10%。
PCT/CN2020/134886 2020-03-26 2020-12-09 信息推荐方法、装置、电子设备及介质 WO2021189949A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010223996.8 2020-03-26
CN202010223996.8A CN111460286A (zh) 2020-03-26 2020-03-26 信息推荐方法、装置、电子设备及介质

Publications (1)

Publication Number Publication Date
WO2021189949A1 true WO2021189949A1 (zh) 2021-09-30

Family

ID=71683279

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/134886 WO2021189949A1 (zh) 2020-03-26 2020-12-09 信息推荐方法、装置、电子设备及介质

Country Status (2)

Country Link
CN (1) CN111460286A (zh)
WO (1) WO2021189949A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115794898A (zh) * 2023-01-05 2023-03-14 中信建投证券股份有限公司 一种金融资讯推荐方法、装置、电子设备及存储介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111460286A (zh) * 2020-03-26 2020-07-28 平安科技(深圳)有限公司 信息推荐方法、装置、电子设备及介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140114694A1 (en) * 2011-02-28 2014-04-24 Hartford Fire Insurance Company Systems and methods for determining insurance decisions based on social network data
US20150206220A1 (en) * 2014-01-20 2015-07-23 Ecole Polytechnique Federale De Lausanne (Epfl) Recommendation Strategy Portfolios
CN108711110A (zh) * 2018-08-14 2018-10-26 中国平安人寿保险股份有限公司 保险产品推荐方法、装置、计算机设备和存储介质
CN109242631A (zh) * 2018-09-17 2019-01-18 平安科技(深圳)有限公司 产品智能推荐方法、服务器及存储介质
CN109389490A (zh) * 2018-09-26 2019-02-26 深圳壹账通智能科技有限公司 贷款产品匹配方法、装置、计算机设备和存储介质
CN110910257A (zh) * 2019-11-15 2020-03-24 泰康保险集团股份有限公司 信息预测方法、装置、电子设备和计算机可读介质
CN111460286A (zh) * 2020-03-26 2020-07-28 平安科技(深圳)有限公司 信息推荐方法、装置、电子设备及介质

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008225629A (ja) * 2007-03-09 2008-09-25 Ricoh Co Ltd 商品推薦システム、商品推薦方法、及び商品推薦方法を記録した記録媒体
CN104717124B (zh) * 2013-12-13 2019-07-16 腾讯科技(深圳)有限公司 一种好友推荐方法、装置及服务器
CA2998249A1 (en) * 2017-03-17 2018-09-17 Edatanetworks Inc. Artificial intelligence engine incenting merchant transaction with consumer affinity
JP2019061525A (ja) * 2017-09-27 2019-04-18 トッパン・フォームズ株式会社 コンテンツレコメンドシステム、コンテンツレコメンド方法、及びプログラム
CN108076154B (zh) * 2017-12-21 2019-12-31 Oppo广东移动通信有限公司 应用信息推荐方法、装置及存储介质和服务器
CN108154420A (zh) * 2017-12-26 2018-06-12 泰康保险集团股份有限公司 产品推荐方法及装置、存储介质、电子设备
CN108711107A (zh) * 2018-05-25 2018-10-26 上海钱智金融信息服务有限公司 智能融资服务推荐方法及其系统
CN110189209A (zh) * 2019-06-05 2019-08-30 浙江米奥兰特商务会展股份有限公司 一种展商与采购商双向推荐方法、装置、设备及存储介质
CN110502691A (zh) * 2019-07-05 2019-11-26 平安科技(深圳)有限公司 基于客户分类的产品推送方法、装置及可读存储介质
CN110659318B (zh) * 2019-08-15 2024-05-03 中国平安财产保险股份有限公司 基于大数据的策略推送方法、系统及计算机设备
CN111125512B (zh) * 2019-11-18 2021-04-27 口碑(上海)信息技术有限公司 业务推荐处理方法、装置及系统
CN111177569B (zh) * 2020-01-07 2021-06-11 腾讯科技(深圳)有限公司 基于人工智能的推荐处理方法、装置及设备

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140114694A1 (en) * 2011-02-28 2014-04-24 Hartford Fire Insurance Company Systems and methods for determining insurance decisions based on social network data
US20150206220A1 (en) * 2014-01-20 2015-07-23 Ecole Polytechnique Federale De Lausanne (Epfl) Recommendation Strategy Portfolios
CN108711110A (zh) * 2018-08-14 2018-10-26 中国平安人寿保险股份有限公司 保险产品推荐方法、装置、计算机设备和存储介质
CN109242631A (zh) * 2018-09-17 2019-01-18 平安科技(深圳)有限公司 产品智能推荐方法、服务器及存储介质
CN109389490A (zh) * 2018-09-26 2019-02-26 深圳壹账通智能科技有限公司 贷款产品匹配方法、装置、计算机设备和存储介质
CN110910257A (zh) * 2019-11-15 2020-03-24 泰康保险集团股份有限公司 信息预测方法、装置、电子设备和计算机可读介质
CN111460286A (zh) * 2020-03-26 2020-07-28 平安科技(深圳)有限公司 信息推荐方法、装置、电子设备及介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115794898A (zh) * 2023-01-05 2023-03-14 中信建投证券股份有限公司 一种金融资讯推荐方法、装置、电子设备及存储介质
CN115794898B (zh) * 2023-01-05 2023-04-25 中信建投证券股份有限公司 一种金融资讯推荐方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
CN111460286A (zh) 2020-07-28

Similar Documents

Publication Publication Date Title
US10395330B2 (en) Evaluating vendor communications for accuracy and quality
US10217048B2 (en) Predictive assignments that relate to genetic information and leverage machine learning models
US20170255790A1 (en) Systems and methods for processing requests for genetic data based on client permission data
CN110909222B (zh) 基于聚类的用户画像建立方法、装置、介质及电子设备
WO2021068513A1 (zh) 异常对象识别方法、装置、介质及电子设备
US20200111550A1 (en) Continuous Health Care Plan Coordination Between Patient and Patient Care Team
WO2020168851A1 (zh) 行为识别
WO2021189949A1 (zh) 信息推荐方法、装置、电子设备及介质
US10642231B1 (en) Switch terminal system with an activity assistant
US20200151627A1 (en) Adherence monitoring through machine learning and computing model application
US20180181722A1 (en) Eliciting Habit Formation Via Coordination Between Patient and Patient Care Team
US11514334B2 (en) Maintaining a knowledge database based on user interactions with a user interface
US11061934B1 (en) Method and system for characterizing time series
US20200272432A1 (en) Modality transformations
US20180181711A1 (en) Continuous Health Care Plan Coordination and Habit Eliciting Patient Communications
CN112348658A (zh) 资源分配方法、装置及电子设备
CN111210109A (zh) 基于关联用户预测用户风险的方法、装置和电子设备
US20220391453A1 (en) Machine learning aided automatic taxonomy for web data
US10019295B2 (en) Method and system for managing and linking software applications
CN113094602B (zh) 酒店推荐方法、系统、设备及介质
CN112542244A (zh) 辅助信息的生成方法、相关装置及计算机程序产品
Gorfine et al. A fully nonparametric estimator of the marginal survival function based on case–control clustered age-at-onset data
CN114792568B (zh) 住院风险预测方法及装置、存储介质及电子设备
CN113010784B (zh) 用于生成预测信息的方法、装置、电子设备和介质
CN109218411A (zh) 数据处理方法及装置、计算机可读存储介质、电子设备

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: 20927191

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: 20927191

Country of ref document: EP

Kind code of ref document: A1