WO2020135001A1 - 一种小说推荐方法及设备 - Google Patents

一种小说推荐方法及设备 Download PDF

Info

Publication number
WO2020135001A1
WO2020135001A1 PCT/CN2019/123979 CN2019123979W WO2020135001A1 WO 2020135001 A1 WO2020135001 A1 WO 2020135001A1 CN 2019123979 W CN2019123979 W CN 2019123979W WO 2020135001 A1 WO2020135001 A1 WO 2020135001A1
Authority
WO
WIPO (PCT)
Prior art keywords
novel
recall
user
target
reading
Prior art date
Application number
PCT/CN2019/123979
Other languages
English (en)
French (fr)
Inventor
刘向前
康英永
Original Assignee
上海连尚网络科技有限公司
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 上海连尚网络科技有限公司 filed Critical 上海连尚网络科技有限公司
Publication of WO2020135001A1 publication Critical patent/WO2020135001A1/zh

Links

Images

Classifications

    • 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
    • G06F16/33—Querying
    • G06F16/335—Filtering based on additional data, e.g. user or group profiles

Definitions

  • This application relates to the field of computers, in particular to a novel recommendation method and device.
  • An object of the present application is to provide a novel recommendation method and device to improve the enthusiasm, experience, and user viscosity of users when reading novels.
  • a novel recommendation method is provided, wherein the method includes:
  • Sort the at least one second target novel and send the at least one second target novel to the user equipment according to the sorting result.
  • the method in response to the first target novel reading instruction, based on at least one recall method and its recall weight, before acquiring at least one second target novel, the method further includes:
  • At least one recall method is preset, and the recall weight of each of the recall methods is determined.
  • determining the recall weight of each of the recall methods includes:
  • obtaining at least one second target novel based on at least one recall method and its recall weight includes:
  • At least one second target novel is determined and obtained.
  • determining the novel to be recommended corresponding to each of the at least one recall method includes:
  • a recall is performed in the novel database to determine a novel to be recommended corresponding to each recall method in at least one recall method.
  • determining the novel to be recommended corresponding to each of the at least one recall method includes:
  • collaborative filtering is performed in the novel database to determine the novel to be recommended corresponding to each recall method in at least one recall method.
  • determining the novel to be recommended corresponding to each of the at least one recall method includes:
  • a recall is performed in the novel database based on the historical reading information of the user who reads the first target novel, and a novel to be recommended corresponding to each recall method in at least one recall method is determined.
  • sorting the at least one second target novel and sending the at least one second target novel to the user device according to the sorting result includes:
  • a computer readable medium on which computer readable instructions are stored, and when the computer readable instructions can be executed by a processor, the processor is implemented as described in the above novel method.
  • a device is also provided, wherein the device includes:
  • One or more processors are One or more processors;
  • Computer-readable medium for storing one or more computer-readable instructions
  • the one or more processors When the one or more computer-readable instructions are executed by the one or more processors, the one or more processors implement the novel recommendation method as described above.
  • the application obtains at least one second target novel according to at least one recall method and its recall weight; Sort the at least one second target novel, and send the at least one second target novel to the user equipment according to the sorting result, so that the user equipment presents the text information of the first target novel to the user
  • at least one second target novel can also be recommended to the user, so that the user can also view at least one second target novel when reading the first target novel, so as to increase the user's reading of the second target novel
  • Interests can not only improve the user's interest, enthusiasm, and experience in reading the first target novel, but also increase the user's viscosity for reading novels.
  • FIG. 1 shows a schematic flowchart of a novel recommendation method according to an aspect of this application
  • FIG. 2 shows a schematic diagram of a recommendation framework in a novel recommendation method according to an aspect of the present application.
  • the terminal, the device serving the network, and the trusted party all include one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • the memory may include non-permanent memory, random access memory (RAM) and/or non-volatile memory in a computer-readable medium, such as read only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
  • RAM random access memory
  • ROM read only memory
  • flash RAM flash memory
  • Computer readable media including permanent and non-permanent, removable and non-removable media, can store information by any method or technology.
  • the information may be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, read-only compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices.
  • computer-readable media does not include non-transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.
  • the flow of a novel recommendation method in one aspect of the present application illustrates the process of a second target novel associated with the first target novel.
  • the method includes: step S11, step S12, step S13, step S14, step S21, step S22, step S23 and step S24, specifically including the following steps:
  • the recall method includes but is not limited to: content-based recall method, based on The collaborative filtering recall method, the combined recall method based on content and collaborative filtering, and the recall method based on user history reading, wherein the collaborative filtering-based recall method can adopt a linear algorithm, the content-based and collaborative filtering-based
  • the combined recall method can use Context-Aware Network Embedding (CANE) algorithm and so on.
  • Step S12 Sort the at least one second target novel, and send the at least one second target novel to the user device according to the sorting result, so as to sort the acquired at least one second target novel, And send the second target novel to the user equipment according to the sorting result.
  • the above steps S11 and S12 not only realize the operation of triggering the acquisition of at least one second target novel when the user reads the first target novel, but also sorts the acquired at least one second target novel according to the sorting result
  • the at least one second target novel is sent to the user equipment corresponding to the user who reads the first target novel, so that the user equipment can present the text information of the first target novel to the user, and can also It is said that the user recommends at least one second target novel, so that the user can view at least one second target novel when reading the first target novel, so as to increase the user's interest in reading the second target novel, thereby not only improving the user
  • the interest, enthusiasm, and experience of reading the first target novel can also increase the user's viscosity to reading novels.
  • step S11 in response to the first target novel reading instruction, based on at least one recall method and its recall weight, before acquiring at least one second target novel, the method further includes:
  • At least one recall method is preset, and the recall weight of each of the recall methods is determined.
  • the recall weight is used to indicate that the novel to be recommended recalled based on the recall method can be determined as the proportion of the second target novel.
  • a preset Or multiple recall methods which can be one or more of content-based recall methods, collaborative filtering-based recall methods, content-based and collaborative filtering-based recall methods, and user-based historical reading recall methods.
  • the relevance is the degree of relevance and fit between the second target novel and the first target novel;
  • multiple recall methods in order to make it clear that the specific number of novels to be recommended among the novels to be recommended recalled by each recall method can be determined as the second target novel, the recall weight of each of the recall methods needs to be determined in order to After the subsequent determination of the novel to be recommended corresponding to each recall mode is determined, the novel to be recommended corresponding to each recall mode can be selected based on the recall weight of each recall mode to be determined as the second target novel and the number.
  • the determination of the recall weight of each of the recall methods in step S11 includes:
  • the historical reading information of the user who reads the first target novel perform a weight evaluation on the at least one recall mode to obtain a recommendation weight corresponding to each of the recall modes.
  • the historical reading information of the user includes, but is not limited to, relevant information that can reflect the user's historical reading, such as the user's reading volume, reading type, reading time period, reading depth, and reading speed.
  • the step S11 will acquire the user who reads the first target novel in real time or periodically Based on the historical reading information of the user, and evaluate the weight of the at least one recall method according to the acquired historical reading information of the user to obtain the recommended weight corresponding to each of the recall methods, for example, if the preset recall method includes The content-based recall method, the collaborative filtering-based recall method and the user's historical reading-based recall method, when the user's historical reading information can be seen from the user's historical reading information, and the user is mainly biased towards reading by the user-based It can be seen from the novels recalled by the historical reading recall method and the text content information of the novels read by the user that the user is also inclined to read the novels recalled by the content-based recall method.
  • the content-based recall method and the collaborative filtering-based recall among the above three recall methods are obtained.
  • the recall weights of the method and the recall method based on the user’s historical reading are: 35%, 5%, and 60%, respectively, so that the second to be sent to the user device from the novel to be recommended recalled by the corresponding recall method according to the recall weight Target novel, to achieve the calculation and determination of the weight of each recall method.
  • the step S11 in response to the first target novel reading instruction, obtains at least one second target novel based on at least one recall method and its recall weight, including:
  • At least one second target novel is determined and obtained.
  • At least one preset recall method includes 4 recall methods, namely: content-based recall method, collaborative filtering-based recall method, content-based and collaborative filtering-based recall method, and user-based historical reading recall method
  • recall method based on content the recall method based on collaborative filtering
  • the combined recall method based on content and collaborative filtering the recall method based on user history reading
  • the recall method based on user history reading are respectively: 20%, 10%, 25% and 45 %
  • step S11 in response to the first target novel reading instruction, determining the novel to be recommended corresponding to each of the at least one recall method includes:
  • the user initial reading tag includes but is not limited to the initial reading tag set to the novel application when the user initially reads a novel, such as a war reading tag, a historical reading tag, and a martial arts reading tag, so that the novel application can subsequently send to the user Recommend a novel corresponding to the initial reading tag;
  • the novel tag is used to indicate the reading tag added by the user during the historical reading process or the tag of the novel type that the user is concerned about or the tag of the novel type that the user wants to read in the future, etc.
  • the user's initial reading tag set when the user initially logs in to the novel application the novel tag added by the user during the historical reading process, and the user's usual reading time
  • the novel label of the novel type concerned and the label of the novel type that the user wants to read in the future are recalled in the novel database to determine the novel to be recommended corresponding to each recall method in at least one recall method. Confirmation of the novel with recommendation letter corresponding to the recall method.
  • step S11 in response to the first target novel reading instruction, determining the novel to be recommended corresponding to each of the at least one recall method includes:
  • a friend user who has an association relationship with the user who reads the first target novel is determined; here, the friend user is used to indicate that there is an association relationship with the user who reads the first target novel
  • the association relationship may be either a friend relationship or a reading user recommended by the user's friend user to the user.
  • collaborative filtering is performed in the novel database to determine the novel to be recommended corresponding to each recall method in at least one recall method.
  • a friend user who has an associated relationship with the user is searched for according to the user currently reading the first target novel, and the friend user's Historical reading information; then, based on the historical reading information of the friend user, collaborative filtering is performed in the novel database to determine the novel to be recommended corresponding to each recall method in at least one recall method, thereby realizing the relationship between the user and the user Friends’ historical reading information to recall novels to be recommended, so as to subsequently determine the second target novel to be sent to the user device according to the novel to be recommended, so that the user can timely know the friends with which he exists while reading the first target novel Relationship friends may have read the second target novel, so as to improve the user's interest in reading, but also make the user understand the novels that their friends have read to stimulate the user's interest in reading the second target novel, and further improve the user's reading experience And user viscosity.
  • step S11 in response to the first target novel reading instruction, determining the novel to be recommended corresponding to each of the at least one recall method includes:
  • a recall is performed in the novel database based on the historical reading information of the user who reads the first target novel, and a novel to be recommended corresponding to each recall method in at least one recall method is determined.
  • the user's historical reading information can be clearly understood from the user's historical reading information, including but not limited to the user's history Reading volume of reading, length of reading novel, type of reading novel, frequency of reading novel and style of reading novel, etc.
  • the user's historical reading information is acquired; afterwards, in order to more accurately calculate and determine the novel to be recommended corresponding to each recall method, the step S11 Recall in the novel database according to the user's historical reading information, determine the novel to be recommended corresponding to each recall method in at least one recall method, and realize the determination of the to-be-recommended corresponding to each recall method through the user's historical reading information Novels, so as to subsequently determine that the second target novel sent to the user device according to the novel to be recommended is more in line with the user's reading needs, so as to meet the user's interest in reading the second target novel while learning the second interest of the user in time
  • the target novel saves the user's effort to find the novel that needs to be read after reading the first target novel, which not only improves the user's reading experience, but also strengthens the user's viscosity when reading the novel.
  • step S12 sorts the at least one second target novel, and sends the at least one second target novel to the user equipment according to the sorting result, including:
  • Sort the at least one second target novel based on a machine learning algorithm send the at least one second target novel to the user equipment in sequence according to the sorting result.
  • the machine learning algorithms include, but are not limited to: linear regression algorithm (Linear Regression, LR), XGBoost enhancement algorithm, LightGBM gradient lifting framework algorithm, etc.
  • the push effect is relatively high Poor users cannot receive many push novels at one time while reading.
  • the determined at least one second target novel is 8 second target novels, they are: Novel 1, Novel 2, Novel 3, Novel 4, Novel 5, Novel 6, Novel 7, and Novel 8, then sort the 8 novels according to the machine learning algorithm, and the sorting results are: Novel 6, Novel 2, Novel 4, Fiction 8, Fiction 1, Fiction 7, Fiction 3 and Fiction 5, then, according to the order of the eight novels: Fiction 6, Fiction 2, Fiction 4, Fiction 8, Fiction 1, Fiction 1, Fiction 7, Fiction 3 and Fiction 5 Send the 8 novels to the user equipment in sequence, for example, first send novel 6 to the user equipment, then send novel 2 to the user equipment, and finally send to the user equipment Sending novel 5 so that the user device recommends different second target novels to the user when reading different text content information of the novel during the user reading the novel, reaching the process of reading the novel by the user,
  • the purpose of pushing different second target novels to users in sequence further improves the user's interest, enthusiasm, and user viscosity in reading novels.
  • the preset recall methods include but are not limited to: content-based recall methods, collaborative filtering (relationship) recall methods, content-based and collaborative filtering (relationship)
  • the novel to be recommended corresponding to the content-based recall method are determined based on the user’s initial reading tag and the novel tag, and the corresponding novels to be recommended based on the collaborative filtering (relationship) recall method are determined using linear regression and other algorithms.
  • the novel to be recommended corresponding to the recall method based on content and collaborative filtering (relationship) recall are context-aware network embedding algorithm: CANE is used to determine the recall.
  • the user's historical reading information of the reading situation is supplemented and recalled to determine the novel to be recommended corresponding to the recall mode based on the user's historical reading; then, according to the recall weight of each recall mode, the novel to be recommended corresponding to the corresponding recall mode is extracted in proportion to determine Two target novels, and propose a middle proportion of the novel to be recommended corresponding to each recall method to determine the fusion as the second target novel to obtain at least one second target novel that needs to be finally sent to the user device; after that, a machine learning algorithm is adopted For example, linear regression algorithm (Linear Regression (LR), XGBoost enhancement algorithm and LightGBM gradient lifting framework algorithm, etc., sort at least one second target novel determined, and sort at least one second target novel according to the sorting result
  • LR Linear Regression
  • XGBoost enhancement algorithm and LightGB
  • a computer readable medium on which computer readable instructions are stored, and when the computer readable instructions can be executed by a processor, the processor is implemented as described in the above novel method.
  • a device is also provided, wherein the device includes:
  • One or more processors are One or more processors;
  • Computer-readable medium for storing one or more computer-readable instructions
  • the one or more processors When the one or more computer-readable instructions are executed by the one or more processors, the one or more processors implement the novel recommendation method as described above.
  • the application when the user triggers to read the first target novel, in response to the reading instruction for the first target novel, the application obtains at least one second target novel according to at least one recall method and its recall weight; At least one second target novel is sorted, and the at least one second target novel is sent to the user equipment according to the sorting result, so that the user equipment presents the text information of the first target novel to the user , It is also possible to recommend at least one second target novel to the user, so that the user can also view at least one second target novel when reading the first target novel, so as to increase the user's interest in reading the second target novel, Thus, not only can the user's interest, enthusiasm, and experience of reading the first target novel be improved, but also the user's viscosity to reading the novel can be increased.
  • the present application may be implemented in software and/or a combination of software and hardware, for example, it may be implemented using an application specific integrated circuit (ASIC), a general purpose computer, or any other similar hardware device.
  • ASIC application specific integrated circuit
  • the software program of the present application may be executed by a processor to implement the steps or functions described above.
  • the software programs of the present application can be stored in computer-readable recording media, such as RAM memory, magnetic or optical drives or floppy disks, and similar devices.
  • some steps or functions of the present application may be implemented by hardware, for example, as a circuit that cooperates with a processor to perform various steps or functions.
  • a part of the application can be applied as a computer program product, such as computer program instructions, and when it is executed by a computer, the method and/or technical solution according to the application can be invoked or provided through the operation of the computer.
  • the program instructions for invoking the method of the present application may be stored in a fixed or removable recording medium, and/or transmitted through a data stream in a broadcast or other signal-bearing medium, and/or stored in accordance with The program instructions run in the working memory of the computer device.
  • an embodiment according to the present application includes an apparatus including a memory for storing computer program instructions and a processor for executing program instructions, wherein, when the computer program instructions are executed by the processor, trigger
  • the device operates based on the aforementioned methods and/or technical solutions according to various embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • User Interface Of Digital Computer (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种小说推荐方法及设备，所述方法包括：在用户触发阅读第一目标小说时，响应于对第一目标小说的阅读指令，根据至少一种召回方式及其召回权重获取至少一部第二目标小说（S11）；对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备（S12）。所述方法可以使得所述用户设备将所述第一目标小说的文本信息呈现给用户的同时，也可以向所述用户推荐至少一部第二目标小说，使得用户在阅读第一目标小说时还可以查看到至少一部第二目标小说，以增加用户对该第二目标小说的阅读兴趣，从而不仅可以提高用户阅读第一目标小说时的趣味性、积极性及用户体验，还可以增加用户对阅读小说的用户粘度。

Description

一种小说推荐方法及设备 技术领域
本申请涉及计算机领域，尤其涉及一种小说推荐方法及设备。
背景技术
当前，各类的电子设备不断渗透到人们生活的方方面面。其中，由于携带的轻便性，越来越多的用户喜欢通过电子设备(比如手机、PAD等)阅读图书。但大多用户阅读小说具有随机性且缺乏耐心，这就使阅读小说变得走马观花，导致用户不仅阅读兴趣减退，还可能会流失一部分阅读用户。
发明内容
本申请的一个目的是提供一种小说推荐方法及设备，以提高用户阅读小说时的积极性、体验度和用户粘度。
根据本申请的一个方面，提供了一种小说推荐方法，其中，所述方法包括：
响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说；
对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备。
进一步地，上述小说推荐方法中，所述响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说之前，还包括：
预设至少一种召回方式，并确定每种所述召回方式的召回权重。
进一步地，上述小说推荐方法中，确定每种所述召回方式召回权重，包括：
根据阅读所述第一目标小说的用户的历史阅读信息，对所述至少一种召回方式进行权重评估，得到每种所述召回方式对应的推荐权重。
进一步地，上述小说推荐方法中，响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说，包括：
响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说；
基于所述至少一种召回方式中的每种所述召回方式对应的待推荐小说和所述召回权重，确定并获取至少一部第二目标小说。
进一步地，上述小说推荐方法中，响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
响应于第一目标小说阅读指令，基于用户初始阅读标签和小说标签，在小说数据库中进行召回，确定至少一种召回方式中的每种召回方式对应的待推荐小说。
进一步地，上述小说推荐方法中，响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
响应于第一目标小说阅读指令，确定与阅读所述第一目标小说的用户存在关联关系的好友用户；
基于所述好友用户的历史阅读信息，在小说数据库中进行协同过滤，确定至少一种召回方式中的每种召回方式对应的待推荐小说。
进一步地，上述小说推荐方法中，响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
响应于第一目标小说阅读指令，基于阅读所述第一目标小说的用户的历史阅读信息在小说数据库中进行召回，确定至少一种召回方式中的每种召回方式对应的待推荐小说。
进一步地，上述小说推荐方法中，对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备，包括：
基于机器学习算法对所述至少一部第二目标小说进行排序；
按照排序结果将所述至少一部第二目标小说按序发送给所述用户设备。
根据本申请的另一方面，还提供了一种计算机可读介质，其上存储有计算机可读指令，所述计算机可读指令可被处理器执行时，使所述处理器实现如上述小说推荐方法。
根据本申请的另一方面，还提供了一种设备，其中，该设备包括：
一个或多个处理器；
计算机可读介质，用于存储一个或多个计算机可读指令，
当所述一个或多个计算机可读指令被所述一个或多个处理器执行，使得所述一个或多个处理器实现如上述小说推荐方法。
与现有技术相比，本申请在用户触发阅读第一目标小说时，响应于对第一目标小说的阅读指令，根据至少一种召回方式及其召回权重获取至少一部第二目标小说；对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备，以使所述用户设备将所述第一目标小说的文本信息呈现给用户的同时，也可以向所述用户推荐至少一部第二目标小说，使得用户在阅读第一目标小说时还可以查看到至少一部第二目标小说，以增加用户对该第二目标小说的阅读兴趣，从而不仅可以提高用户阅读第一目标小说时的趣味性、积极性及于都体验，还可以增加用户对阅读小说的用户粘度。
附图说明
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述，本申请的其它特征、目的和优点将会变得更明显：
图1示出根据本申请一个方面的一种小说推荐方法的流程示意图；
图2示出根据本申请一个方面的一种小说推荐方法中推荐框架示意图。
附图中相同或相似的附图标记代表相同或相似的部件。
具体实施方式
下面结合附图对本申请作进一步详细描述。
在本申请一个典型的配置中，终端、服务网络的设备和可信方均包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器，随机存取存储器(RAM)和/或非易失性内存等形式，如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括，但不限于相变内存 (PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。按照本文中的界定，计算机可读介质不包括非暂存电脑可读媒体(transitory media)，如调制的数据信号和载波。
如图1所示，本申请一个方面的一种小说推荐方法的流程示意该第一目标小说相关联的第二目标小说的过程中。该方法包括：步骤S11、步骤S12、步骤S13、步骤S14、步骤S21、步骤S22、步骤S23及步骤S24，具体包括如下步骤：
在实际的应用场景中，用户在阅读小说时，为了便于用户在阅读该第一目标小说时也能够了解到该第一目标小说的关联信息，以提高用户阅读该第一目标小说的积极性，步骤S11，响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说；在此，所述召回方式包括但不限于：基于内容的召回方式、基于协同过滤的召回方式、基于内容与协同过滤的组合召回方式及基于用户历史阅读的召回方式，其中，所述基于协同过滤的召回方式可以采用线性(line)算法，所述基于内容与协同过滤的组合召回方式可以采用上下文感知网络嵌入(Context-Aware Network Embedding，CANE)算法等。
步骤S12，对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备，实现了对获取的至少一部第二目标小说的排序，并按照排序结果将第二目标小说发送给所述用户设备。
上述步骤S11和步骤S12，不仅实现了在用户阅读第一目标小说时，触发获取至少一部第二目标小说的操作，还将获取的至少一部第二目标小说进行排序后，按照排序结果将所述至少一部第二目标小说发送给阅读该第一目标小说的用户对应的用户设备，以使所述用户设备将所述第一目标小说的文本信息呈现给用户的同时，也可以向所述用户推荐至少一部第二目标小说，使得用户在阅读第一目标小说时还可以查看到至少一部第二目标小说，以增加用户对该第 二目标小说的阅读兴趣，从而不仅可以提高用户阅读第一目标小说时的趣味性、积极性及于都体验，还可以增加用户对阅读小说的用户粘度。
本实施例中，所述步骤S11响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说之前，还包括：
预设至少一种召回方式，并确定每种所述召回方式的召回权重，在此，该召回权重用于指示基于所述召回方式召回的待推荐小说能够确定为第二目标小说的占比。
例如，为了便于用户在阅读小说时能够向用户推荐对应的第二目标小说，同时为了便于推荐的第二目标小说契合用户平时所阅读的小说，在所述步骤S11之前，会事先预设一种或多种召回方式，该召回方式可以是基于内容的召回方式、基于协同过滤的召回方式、基于内容与协同过滤的组合召回方式及基于用户历史阅读的召回方式中的一种或多种，以提高后续基于该一种或多种召回方式确定的第二目标小说的关联度，该关联度是该第二目标小说与第一目标小说之间的关联性和契合度；在预设了一种或多种召回方式后，为了便于清楚每种召回方式所召回的待推荐小说中的具体多少的待推荐小说可以确定为第二目标小说，还需确定每种所述召回方式的召回权重，以便在后续确定了每种召回方式对应所召回的待推荐小说后，可以基于每种召回方式的召回权重筛选出每种召回方式对应的待推荐小说可以确定为第二目标小说及数量。
接着本申请的上述实施例，所述步骤S11中的确定每种所述召回方式召回权重，包括：
根据阅读所述第一目标小说的用户的历史阅读信息，对所述至少一种召回方式进行权重评估，得到每种所述召回方式对应的推荐权重。在此，所述用户的历史阅读信息包括但不限于：用户阅读量、阅读类型、阅读时间段、阅读深度及阅读速度等能够反映用户历史阅读的相关信息。
例如，在随着用户阅读小说的阅读量的增加，用户需要阅读的小说的类型和阅读偏好也会随着改变，则所述步骤S11会实时或周期性地获取阅读该第一目标小说的用户的历史阅读信息，并根据获取的所述用户的历史阅读信息，对所述至少一种召回方式进行权重评估，得到每种所述召回方式对应的推荐权重，比如，若预设的召回方式包括基于内容的召回方式、基于协同过滤的召回方式 和基于用户历史阅读的召回方式，当从所述用户的历史阅读信息中可以看出用户的阅读量很大且用户主要偏向于阅读由该基于用户历史阅读的召回方式所召回的小说及从用户的阅读的小说文本内容信息等中可以看出用户也偏向于阅读基于内容的召回方式所召回的小说，则在根据所述用户的历史阅读信息对所述用户阅读所述至少一种召回方式中的每种所述召回方式所召回的小说的阅读偏好进行权重评估后，得到上述三种召回方式中的基于内容的召回方式、基于协同过滤的召回方式和基于用户历史阅读的召回方式的召回权重分别为：35％、5％及60％，使得后续按照该召回权重从对应的召回方式所召回的待推荐小说中确定出向用户设备发送的第二目标小说，实现对每种召回方式的权重的计算和确定。
本实施例中，所述步骤S11响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说，包括：
响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说；
基于所述至少一种召回方式中的每种所述召回方式对应的待推荐小说和所述召回权重，确定并获取至少一部第二目标小说。
例如，若预设的至少一种召回方式包括4种召回方式，分别为：基于内容的召回方式、基于协同过滤的召回方式、基于内容与协同过滤的组合召回方式及基于用户历史阅读的召回方式，同时，若基于内容的召回方式、基于协同过滤的召回方式、基于内容与协同过滤的组合召回方式及基于用户历史阅读的召回方式的召回权重分别为：20％、10％、25％及45％，则在用户阅读该第一目标小说时，响应于第一目标小说阅读指令，若基于内容的召回方式所召回的待推荐小说的数量为10部，基于协同过滤的召回方式所召回的待推荐小说的数量为20部，基于内容与协同过滤的组合召回方式所召回的待推荐小说的数量为8部，及基于用户历史阅读的召回方式所召回的待推荐小说的数量为40部；接着，分别根据每种召回方式对应的待推荐小说可以作为第二目标小说的数量，其中，基于内容的召回方式对应的10部小说中的2部说被确定为第二目标小说，基于协同过滤的召回方式对应的20部小说中的2部说被确定为第二目标小说，基于内容与协同过滤的召回方式对应的8部小说中的2部说被确定为第 二目标小说，及基于用户历史阅读的召回方式对应的40部小说中的18部说被确定为第二目标小说，进而确定基于上述4中召回方式确定的第二目标小说有2+2+2+18＝24部，并获取该24部第二目标小说，实现了根据至少一种召回方式及其召回权重，来确定并获取至少一部第二目标小说。
接着本申请的上述实施例，所述步骤S11中的响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
响应于第一目标小说阅读指令，基于用户初始阅读标签和小说标签，在小说数据库中进行召回，确定至少一种召回方式中的每种召回方式对应的待推荐小说。在此，所述用户初始阅读标签包括但不限于用户初始阅读小说时，向小说应用程序设置的初始阅读标签，比如战争阅读标签、历史阅读标签及武侠阅读标签，以便小说应用程序后续能够向用户推荐与该初始阅读标签对应的小说；所述小说标签用于指示所述用户在历史阅读过程中添加的阅读标签或者该用户关注的小说类型的标签或用户未来想要阅读的小说类型的标签等。例如，在用户在阅读小说时，响应于第一目标小说阅读指令，可以根据用户初始登录说小说应用程序时设置的用户初始阅读标签和用户在历史阅读过程中添加的小说标签、用户平时阅读时所关注的小说类型的小说标签及用户未来想要阅读的小说类型的标签，在小说数据库中进行召回，确定至少一种召回方式中的每种召回方式对应的待推荐小说，实现了对每种召回方式对应的带推荐信小说的确定。
接着本申请的上述实施例，所述步骤S11中的响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
响应于第一目标小说阅读指令，确定与阅读所述第一目标小说的用户存在关联关系的好友用户；在此，所述好友用户用于指示与阅读该第一目标小说的用户存在关联关系的用户，该关联关系既可以是好友关系也可以是该用户的好友用户推荐给所述用户的阅读用户。
基于所述好友用户的历史阅读信息，在小说数据库中进行协同过滤，确定至少一种召回方式中的每种召回方式对应的待推荐小说。
例如，在用户在阅读小说时，响应于第一目标小说阅读指令，根据当前阅读第一目标小说的用户查找与该用户存在关联关系的好友用户，在查找到该好 友用户后获取该好友用户的历史阅读信息；接着，根据所述好友用户的历史阅读信息，在小说数据库中进行协同过滤，确定至少一种召回方式中的每种召回方式对应的待推荐小说，实现了根据与用户具有好友关系的好友用户的历史阅读信息来召回待推荐小说，以便后续根据所述待推荐小说确定向用户设备发送的第二目标小说，使得用户在阅读第一目标小说的同时，能够及时了解到与其存在好友关系的好友可能阅读过的第二目标小说，从而提高用户阅读趣味性的同时，还使用户了解其好友用户所阅读过的小说以激发用户阅读第二目标小说的兴趣，进一步提高用户阅读体验度和用户粘度。
接着本申请的上述实施例，所述步骤S11中的响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
响应于第一目标小说阅读指令，基于阅读所述第一目标小说的用户的历史阅读信息在小说数据库中进行召回，确定至少一种召回方式中的每种召回方式对应的待推荐小说。
例如，随着阅读所述第一目标小说的用户的阅读量的不断提升，可以从用户的历史阅读信息中清楚地了解到用户在历史阅读过程中的信息，该信息包括但不限于该用户历史阅读的阅读量、阅读小说篇幅、阅读小说类型、阅读小说频率及阅读小说风格等。在用户在阅读第一目标小说时，响应于第一目标小说阅读指令，获取该用户的历史阅读信息；之后，为了更精确地计算并确定每种召回方式对应的待推荐小说，所述步骤S11根据该用户的历史阅读信息在小说数据库中进行召回，确定至少一种召回方式中的每种召回方式对应的待推荐小说，实现了通过用户的历史阅读信息来确定每种召回方式对应的待推荐小说，以便后续根据所述待推荐小说确定向用户设备发送的第二目标小说更符合用户的阅读需求，以满足用户在阅读第一目标小说的同时，能够及时了解到该用户所感兴趣的第二目标小说，从而节省用户在阅读完第一目标小说后查找需要阅读的小说的精力，不仅提高了用户的阅读体验度，还加强了用户阅读小说时的用户粘度。
接着本申请的上述实施例，所述步骤S12对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备，包括：
基于机器学习算法对所述至少一部第二目标小说进行排序；按照排序结果 将所述至少一部第二目标小说按序发送给所述用户设备。在此，所述机器学习算法包括但不限于：线性回归算法(Linear Regression，LR)、XGBoost增强算法及LightGBM梯度提升框架算法等。
例如，在用户阅读第一目标小说时，确定了需要向该用户推送的至少一部第二目标小说后，为了避免向用户一次性推送确定的至少一部第二目标小说而导致的推送效果较差和用户无法在阅读时一次性接收很多推送小说，本申请在确定了至少一部第二目标小说后，若该确定的至少一部第二目标小说为8部第二目标小说，分别为：小说1、小说2、小说3、小说4、小说5、小说6、小说7及小说8，则根据机器学习算法对折8部小说进行排序，得到排序结果为：小说6、小说2、小说4、小说8、小说1、小说7、小说3及小说5，之后，按照这8部小说的排序结果为：小说6、小说2、小说4、小说8、小说1、小说7、小说3及小说5将所述8部小说按序发送给所述用户设备，比如，先向所述用户设备发送小说6，再向所述用户设备发送小说2，......，最后向所述用户设备发送小说5，以便用户设备在用户阅读小说的过程中，随着阅读该小说的不同的文本内容信息时，向所述用户推荐不同的第二目标小说，达到了在用户阅读小说的过程中，按序向用户推送不同的第二目标小说的目的，进一步提高了用户阅读小说过程中的趣味性、积极性和用户粘度。
在本申请一实际应用场景中如图2所示，预先设置的召回方式包括但不限于：基于内容召回的召回方式、基于协同过滤(关系)召回的召回方式、基于内容和协同过滤(关系)召回的召回方式及基于用户真实阅读情况的用户历史阅读进行补充召回的召回方式中的一项或多项，在预置用于进行小说召回的至少一种召回方式后，对每种召回方式进行权重评估，得到每种召回方式的召回权重。其中，基于内容的召回方式对应的待推荐小说根据用户初始阅读标签和小说标签进行召回确定的，基于协同过滤(关系)召回的召回方式对应的待推荐小说是采用线性回归等算法进行召回确定的，基于内容和协同过滤(关系)召回的召回方式对应的待推荐小说是采用上下文感知网络嵌入算法：CANE进行召回确定的，为了保证召回的待推荐小说的全面性和准确定，还根据用户真实阅读情况的用户历史阅读信息进行补充召回来确定基于用户历史阅读的召回方式对应的待推荐小说；接着，根据每种召回方式的召回权重进行等比例提 取对应召回方式对应的待推荐小说确定为第二目标小说，并将每种召回方式对应的待推荐小说中等比例提出来确定为第二目标小说进行融合，得到最后需要向用户设备发送的至少一部第二目标小说；之后，采用机器学习算法，比如，线性回归算法(Linear Regression，LR)、XGBoost增强算法及LightGBM梯度提升框架算法等对确定出的至少一部第二目标小说进行排序，并按照排序结果将至少一部第二目标小说按序发送给用户设别，以使用户设备在随着阅读该第一目标小说的不同的文本内容信息时，向所述用户推荐不同的第二目标小说，达到了在用户阅读小说的过程中，按序向用户推送不同的第二目标小说的目的，进一步提高了用户阅读小说过程中的趣味性、积极性和用户粘度。
根据本申请的另一方面，还提供了一种计算机可读介质，其上存储有计算机可读指令，所述计算机可读指令可被处理器执行时，使所述处理器实现如上述小说推荐方法。
根据本申请的另一方面，还提供了一种设备，其中，该设备包括：
一个或多个处理器；
计算机可读介质，用于存储一个或多个计算机可读指令，
当所述一个或多个计算机可读指令被所述一个或多个处理器执行，使得所述一个或多个处理器实现如上述小说推荐方法。
在此，所述用于小说推荐的设备中的各实施例的详细内容，具体可参见上述的小说推荐方法实施例的对应部分，在此，不再赘述。
综上所述，本申请在用户触发阅读第一目标小说时，响应于对第一目标小说的阅读指令，根据至少一种召回方式及其召回权重获取至少一部第二目标小说；对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备，以使所述用户设备将所述第一目标小说的文本信息呈现给用户的同时，也可以向所述用户推荐至少一部第二目标小说，使得用户在阅读第一目标小说时还可以查看到至少一部第二目标小说，以增加用户对该第二目标小说的阅读兴趣，从而不仅可以提高用户阅读第一目标小说时的趣味性、积极性及于都体验，还可以增加用户对阅读小说的用户粘度。
需要注意的是，本申请可在软件和/或软件与硬件的组合体中被实施，例如，可采用专用集成电路(ASIC)、通用目的计算机或任何其他类似硬件设备来实 现。在一个实施例中，本申请的软件程序可以通过处理器执行以实现上文所述步骤或功能。同样地，本申请的软件程序(包括相关的数据结构)可以被存储到计算机可读记录介质中，例如，RAM存储器，磁或光驱动器或软磁盘及类似设备。另外，本申请的一些步骤或功能可采用硬件来实现，例如，作为与处理器配合从而执行各个步骤或功能的电路。
另外，本申请的一部分可被应用为计算机程序产品，例如计算机程序指令，当其被计算机执行时，通过该计算机的操作，可以调用或提供根据本申请的方法和/或技术方案。而调用本申请的方法的程序指令，可能被存储在固定的或可移动的记录介质中，和/或通过广播或其他信号承载媒体中的数据流而被传输，和/或被存储在根据所述程序指令运行的计算机设备的工作存储器中。在此，根据本申请的一个实施例包括一个装置，该装置包括用于存储计算机程序指令的存储器和用于执行程序指令的处理器，其中，当该计算机程序指令被该处理器执行时，触发该装置运行基于前述根据本申请的多个实施例的方法和/或技术方案。
对于本领域技术人员而言，显然本申请不限于上述示范性实施例的细节，而且在不背离本申请的精神或基本特征的情况下，能够以其他的具体形式实现本申请。因此，无论从哪一点来看，均应将实施例看作是示范性的，而且是非限制性的，本申请的范围由所附权利要求而不是上述说明限定，因此旨在将落在权利要求的等同要件的含义和范围内的所有变化涵括在本申请内。不应将权利要求中的任何附图标记视为限制所涉及的权利要求。此外，显然“包括”一词不排除其他单元或步骤，单数不排除复数。装置权利要求中陈述的多个单元或装置也可以由一个单元或装置通过软件或者硬件来实现。第一，第二等词语用来表示名称，而并不表示任何特定的顺序。

Claims (10)

  1. 一种小说推荐方法，其中，所述方法包括：
    响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说；
    对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备。
  2. 根据权利要求1所述的方法，其中，所述响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说之前，还包括：
    预设至少一种召回方式，并确定每种所述召回方式的召回权重。
  3. 根据权利要求2所述的方法，其中，确定每种所述召回方式召回权重，包括：
    根据阅读所述第一目标小说的用户的历史阅读信息，对所述至少一种召回方式进行权重评估，得到每种所述召回方式对应的推荐权重。
  4. 根据权利要求3所述的方法，其中，响应于第一目标小说阅读指令，基于至少一种召回方式及其召回权重，获取至少一部第二目标小说，包括：
    响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说；
    基于所述至少一种召回方式中的每种所述召回方式对应的待推荐小说和所述召回权重，确定并获取至少一部第二目标小说。
  5. 根据权利要求4所述的方法，其中，响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
    响应于第一目标小说阅读指令，基于用户初始阅读标签和小说标签，在小说数据库中进行召回，确定至少一种召回方式中的每种召回方式对应的待推荐小说。
  6. 根据权利要求5所述的方法，其中，响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
    响应于第一目标小说阅读指令，确定与阅读所述第一目标小说的用户存在关联关系的好友用户；
    基于所述好友用户的历史阅读信息，在小说数据库中进行协同过滤，确定至少 一种召回方式中的每种召回方式对应的待推荐小说。
  7. 根据权利要求5或6所述的方法，其中，响应于第一目标小说阅读指令，确定至少一种召回方式中的每种召回方式对应的待推荐小说，包括：
    响应于第一目标小说阅读指令，基于阅读所述第一目标小说的用户的历史阅读信息在小说数据库中进行召回，确定至少一种召回方式中的每种召回方式对应的待推荐小说。
  8. 根据权利要求7所述的方法，其中，对所述至少一部第二目标小说进行排序，并按照排序结果将所述至少一部第二目标小说发送给用户设备，包括：
    基于机器学习算法对所述至少一部第二目标小说进行排序；
    按照排序结果将所述至少一部第二目标小说按序发送给所述用户设备。
  9. 一种计算机可读介质，其上存储有计算机可读指令，所述计算机可读指令可被处理器执行时，使所述处理器实现如权利要求1至8中任一项所述的方法。
  10. 一种设备，其中，该设备包括：
    一个或多个处理器；
    计算机可读介质，用于存储一个或多个计算机可读指令，
    当所述一个或多个计算机可读指令被所述一个或多个处理器执行，使得所述一个或多个处理器实现如权利要求1至8中任一项所述的方法。
PCT/CN2019/123979 2018-12-27 2019-12-09 一种小说推荐方法及设备 WO2020135001A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811614711.2 2018-12-27
CN201811614711.2A CN109739972A (zh) 2018-12-27 2018-12-27 一种小说推荐方法及设备

Publications (1)

Publication Number Publication Date
WO2020135001A1 true WO2020135001A1 (zh) 2020-07-02

Family

ID=66361472

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/123979 WO2020135001A1 (zh) 2018-12-27 2019-12-09 一种小说推荐方法及设备

Country Status (2)

Country Link
CN (1) CN109739972A (zh)
WO (1) WO2020135001A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109739972A (zh) * 2018-12-27 2019-05-10 上海连尚网络科技有限公司 一种小说推荐方法及设备
CN110502299B (zh) * 2019-08-12 2021-05-14 南京大众书网图书文化有限公司 一种用于提供小说信息的方法与设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106611050A (zh) * 2016-12-22 2017-05-03 掌阅科技股份有限公司 电子书推荐方法、装置和服务器
CN106682012A (zh) * 2015-11-06 2017-05-17 阿里巴巴集团控股有限公司 商品对象信息搜索方法及装置
EP3287913A1 (en) * 2016-08-18 2018-02-28 Spotify AB Systems, methods, and computer-readable products for recommending music tracks
CN109739972A (zh) * 2018-12-27 2019-05-10 上海连尚网络科技有限公司 一种小说推荐方法及设备

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8244564B2 (en) * 2009-03-31 2012-08-14 Richrelevance, Inc. Multi-strategy generation of product recommendations
CN105117474B (zh) * 2012-06-27 2019-03-05 北京奇虎科技有限公司 在网页的阅读模式中进行推荐信息加载的方法和装置
US20160188734A1 (en) * 2014-12-30 2016-06-30 Socialtopias, Llc Method and apparatus for programmatically synthesizing multiple sources of data for providing a recommendation
CN106327227A (zh) * 2015-06-19 2017-01-11 北京航天在线网络科技有限公司 一种信息推荐系统及信息推荐方法
JP6457358B2 (ja) * 2015-09-04 2019-01-23 株式会社東芝 アイテム推薦システム、アイテム推薦方法およびプログラム
CN106897284B (zh) * 2015-12-17 2020-06-16 腾讯科技（深圳）有限公司 电子书籍的推荐方法及装置
CN105488662B (zh) * 2016-01-07 2021-09-03 北京华品博睿网络技术有限公司 一种基于双向推荐的在线招聘系统
CN106997347A (zh) * 2016-01-22 2017-08-01 华为技术有限公司 信息推荐方法及服务器
CN107656938B (zh) * 2016-07-26 2022-01-11 北京搜狗科技发展有限公司 一种推荐方法和装置、一种用于推荐的装置
CN108509534B (zh) * 2018-03-15 2022-03-25 华南理工大学 基于深度学习的个性化音乐推荐系统及其实现方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106682012A (zh) * 2015-11-06 2017-05-17 阿里巴巴集团控股有限公司 商品对象信息搜索方法及装置
EP3287913A1 (en) * 2016-08-18 2018-02-28 Spotify AB Systems, methods, and computer-readable products for recommending music tracks
CN106611050A (zh) * 2016-12-22 2017-05-03 掌阅科技股份有限公司 电子书推荐方法、装置和服务器
CN109739972A (zh) * 2018-12-27 2019-05-10 上海连尚网络科技有限公司 一种小说推荐方法及设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LIN, ZHAOSHENG: "Design and Implementation of Internet Novels Recommendation System Base on Hybrid Recommendation", MASTER THESIS, no. 7, 15 April 2017 (2017-04-15), pages 1 - 75, XP009521702, ISSN: 1674-0246, DOI: 20200219121525X *

Also Published As

Publication number Publication date
CN109739972A (zh) 2019-05-10

Similar Documents

Publication Publication Date Title
TWI694401B (zh) 融合用戶關係資料的搜尋方法和裝置
CN107679211B (zh) 用于推送信息的方法和装置
US8290927B2 (en) Method and apparatus for rating user generated content in search results
US20160259818A1 (en) Searching for Software Applications Based on Application Attributes
JP5984917B2 (ja) 提案される語を提供するための方法および装置
US20110320715A1 (en) Identifying trending content items using content item histograms
US8880513B2 (en) Presentation of items based on a theme
EP2407897A1 (en) Device for determining internet activity
US11356498B2 (en) Method and a device for sharing a hosted application
WO2019169978A1 (zh) 资源推荐方法及装置
US20170214646A1 (en) Systems and methods for providing social media location information
US20150269615A1 (en) Predicting Visibility of Content Items
JP2013544000A (ja) 候補製品のリンクリストを判定する方法および装置
JP6932360B2 (ja) オブジェクト検索方法、装置およびサーバ
CN110457610B (zh) 信息推荐方法、装置、终端、服务器及存储介质
US20230146563A1 (en) Automated image processing and insight presentation
WO2020135001A1 (zh) 一种小说推荐方法及设备
CN104090899A (zh) 一种反馈展示内容信息的方法和装置
US20150154287A1 (en) Method for providing recommend information for mobile terminal browser and system using the same
US8825698B1 (en) Showing prominent users for information retrieval requests
US20150286658A1 (en) Image Processing
US20240037134A1 (en) Method and apparatus for searching for clipping template
US10817519B2 (en) Automatic conversion stage discovery
CN110827101A (zh) 一种店铺推荐的方法和装置
WO2021051744A1 (zh) 菜品推荐方法、推荐内容上传方法、装置及电子设备

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: 19905620

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: 19905620

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 14/10/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 19905620

Country of ref document: EP

Kind code of ref document: A1