WO2010034259A1 - 提供在线服务的方法和装置 - Google Patents

提供在线服务的方法和装置 Download PDF

Info

Publication number
WO2010034259A1
WO2010034259A1 PCT/CN2009/074253 CN2009074253W WO2010034259A1 WO 2010034259 A1 WO2010034259 A1 WO 2010034259A1 CN 2009074253 W CN2009074253 W CN 2009074253W WO 2010034259 A1 WO2010034259 A1 WO 2010034259A1
Authority
WO
WIPO (PCT)
Prior art keywords
personality
user
service
category
module
Prior art date
Application number
PCT/CN2009/074253
Other languages
English (en)
French (fr)
Inventor
丁在茂
陈元强
Original Assignee
腾讯科技(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 腾讯科技(深圳)有限公司 filed Critical 腾讯科技(深圳)有限公司
Publication of WO2010034259A1 publication Critical patent/WO2010034259A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/335Filtering based on additional data, e.g. user or group profiles
    • G06F16/337Profile generation, learning or modification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35Clustering; Classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35Clustering; Classification
    • G06F16/353Clustering; Classification into predefined classes
    • 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

Definitions

  • the present invention relates to the field of network communications, and in particular, to a method and apparatus for providing an online service. Background of the invention
  • the network-based online service is almost always based on the push distribution method of the existing classification, and the user can only passively accept the service published by the service content provider. content.
  • the relevance recommendation method is generally recommended based on topic relevance, content relevance, or a user's recommendation list.
  • Embodiments of the present invention provide a method for providing an online service, which can provide a network online service required by a user more accurately.
  • Embodiments of the present invention provide an apparatus for providing an online service, which can provide more accurate The online service required by the user.
  • a method of providing an online service comprising:
  • the user's personality category is searched, and the service corresponding to the personality category is provided to the user.
  • An apparatus for providing an online service comprising:
  • a personality classification module configured to perform personality classification on all collected users according to the collected user network behavior characteristic information and the established personality classification model
  • a service setting module connected to the personality classification module, configured to set a corresponding service for each personality category obtained by character classification of the personality classification module;
  • a search module which is respectively connected to the personality classification module and the service setting module, and is configured to: when a user applies for an online service, search for a personality category of the user in the result of the personality classification of the personality classification module, and Searching for a service corresponding to the personality category in a service set by the service setting module;
  • An online service module coupled to the lookup module, for providing a service found by the lookup module to the user.
  • Embodiment 1 is a flowchart of a method for providing an online service in Embodiment 1 of the present invention
  • Embodiment 2 is a flowchart of a method for providing an online service in Embodiment 2 of the present invention
  • FIG. 3 is a first structural diagram of an apparatus for providing an online service in Embodiment 3 of the present invention
  • FIG. 4 is a second structural diagram of an apparatus for providing an online service according to Embodiment 3 of the present invention
  • Figure 5 is a third structural diagram of an apparatus for providing an online service in Embodiment 3 of the present invention
  • Figure 6 is a fourth structural diagram of an apparatus for providing an online service in Embodiment 3 of the present invention. Mode for carrying out the invention
  • the invention mainly establishes a set of methods for classifying users' personality and providing online services according to personality classification according to the user's network behavior characteristics, and classifying users into different users by collecting and modeling user network behavior characteristic information. Personality categories, and provide different online services to meet their needs for each personality category, so as to accurately provide the online services required by users.
  • an embodiment of the present invention provides a method for providing an online service, including: Step 101: Perform personality classification on all collected users according to the collected user network behavior characteristic information and the established personality classification model;
  • the personality classification model may be a pre-established correspondence between the user network behavior characteristic information and the user personality according to the analysis and summary of the user network behavior characteristic information, according to the corresponding relationship, according to the collected user network. Behavioral characteristics information, which classifies user personality.
  • Step 102 Set the corresponding service for each personality category obtained by the personality classification.
  • Step 103 When a user applies for an online service, in the result of the personality classification, find the personality category of the user.
  • Step 104 Provide a service corresponding to the personality category to the user.
  • the method provided in this embodiment improves the service to the user by classifying the user and setting the corresponding service.
  • the user's personality category is searched and the corresponding service is provided to the user. Rate, overcoming the network in the prior art
  • the online service provides users with services that cannot accurately meet the defects of the user's needs.
  • an embodiment of the present invention further provides another method for providing an online service, including:
  • Step 201 Perform a calculation according to the pre-selected user network behavior characteristic information sample and the dynamic modeling engine to obtain a personality classification model.
  • the user network behavior characteristic information refers to attribute information related to the network behavior when the user performs the network behavior.
  • the user network behavior characteristic information may be various, including but not limited to: personal data filled by the user on the network, the type and quantity of the free service operated by the user, the type and quantity of the user browsing the service, the type, quantity and amount of the user trial service The delay time for the user to subscribe to the service, the time interval during which the user officially uses the service, the number of months the user continuously uses the service, the type and quantity of services that the user recommends to others, the type and amount of services recommended by others, and the use of the service by others. Evaluation, as well as the number of comments the user has made on the review, and so on.
  • the personal data filled in by the user on the network may include: personality, gender, age, blood type, hobbies, and regions, etc., which reflect basic information of the user.
  • the user network behavior characteristic information sample is the user and various information selected from the reference.
  • User network behavior characteristics information can be obtained through various channels, such as statistics on the network, non-network means to investigate, and so on.
  • the obtained user network behavior characteristic information samples may be stored in a database, such as establishing a modeling behavior database for storing the obtained user network behavior characteristic information samples.
  • the dynamic modeling engine in this embodiment can adopt various types, including but not limited to: mathematical modeling engines such as Bayesian, decision tree and support vector machine.
  • Step 202 Perform the operation on the collected user network behavior characteristic information by using the established personality classification model, obtain the collected personality category of each user and the probability that the user belongs to the personality category, and store the result of the operation.
  • the collected users are usually mass users, that is, the network behavior characteristic information of a large number of users is collected.
  • There are a plurality of user network behavior characteristics information collected for example: user ID1, gender is male, age is 25 years old, 10 types of free service operations are operated, 100 types of browsing services are 100, and types of trial services are 50. And so on, this information is substituted into the established personality classification model, after the operation, it is concluded that the user's personality category is outward consumption, and the probability that the user belongs to the outward consumption user is 90%.
  • the user set corresponding to the user network behavior feature information collected in this step does not coincide with the user set corresponding to the user network behavior feature information sample selected in step 201.
  • the personality classification model is established in step 201, 100,000 users are selected.
  • network behavior characteristic information of 1 million users is collected, wherein the 100,000 users are not included in the one million users.
  • the collected user network behavior characteristic information may be separately stored in a database, such as a pre-established network behavior characteristic database.
  • the user has a variety of personality categories, including but not limited to: outward consumption, inward consumption, open consumption, conservative consumption, impulsive consumption, and cautious consumption.
  • the result of the personality classification can obtain a plurality of personality categories, and the probability of each personality category is different. For example, the probability that a user belongs to the outward consumption type is 80%, the probability of belonging to the open consumption type is 70%, the probability of belonging to the impulse consumption type is 50%, the probability of belonging to the inward consumption type is 30%, and so on.
  • the results of the personality classification may be stored in a database, such as establishing a personality classification database for storing the calculated personality categories of each user and corresponding probabilities.
  • Step 203 Set a corresponding service for each personality category obtained by the personality classification, and store the set service.
  • the corresponding service may be set for each personality category according to needs; the reasoning may also be performed according to the collected user network behavior characteristic information, and for any personality category, the user service information related information is analyzed, and the result is set according to the analysis result.
  • the service corresponding to the personality category For example, there are 10,000 users in the outward-sending personality category, of which 8,000 users like online game services and 6,000 users like chat services, and the services corresponding to the outgoing consumption type of personality categories include online games and chats.
  • the set service can contain a variety of content.
  • one personality category may correspond to one or more services, and different personality categories may also correspond to the same service, for example, the conservative consumption type and the inward consumption type correspond to the same service.
  • Step 204 When a user applies for an online service, in the result of the personality classification, find the personality category of the user.
  • the user who applies for the online service in this embodiment is one of the user sets corresponding to the feature information of the user network collected in step 202.
  • the personality category of the user is searched, and if the personality classification of the user is stored in the result of the personality classification, the plurality of personality categories may be queried.
  • Step 205 Provide a service corresponding to the personality category to the user.
  • This embodiment is described by taking a user's online service application as an example.
  • the process of providing each user with the service is the same as the process of providing a service to the user in this embodiment. The same, no longer repeat here.
  • whether the service is provided to one user or the service is provided to multiple users the foregoing is based on the foregoing personality classification and setting service, and the personality classification is based on the collected large number of users. Network behavior characteristics information obtained.
  • the plurality of personality categories of the user may be sorted according to probability, from high to low or low to high, and then selected in the service corresponding to the plurality of personality categories in the sorted order.
  • Some or all of the services are available to the user.
  • the user has three personality categories, namely: conservative consumption, cautious consumption, and inward consumption, the corresponding probabilities are 90%, 70%, and 80%, respectively, and the corresponding services are Al, A2, and A3. , sorted from high to low, and select some or all of Al, A3, and A2 to provide to the user.
  • the method may further include:
  • the feedback information includes but is not limited to: the user clicks on the view, the number of pages viewed by the user, whether the user evaluates the service, the user accepts the service, and the like.
  • the method may further include:
  • Step 206 Adjust the probability corresponding to the current personality category, or adjust the service corresponding to the current personality category, or adjust the current personality category correspondingly according to the user's feedback on the current personality category corresponding service, in the stored personality categories, probabilities, and services.
  • the probability and service, and the adjusted content is updated to the above stored character classifications and services.
  • the probability of adjusting the personality category may be specifically as follows: if the service corresponding to the current personality category is accepted by the user, the probability that the user belongs to the personality category is improved; if the service corresponding to the current personality category is rejected by the user, the user belongs to the personality category. Probability. Wherein, the magnitude of the probability adjustment can be set according to actual needs, and the increased amplitude and the reduced amplitude can be set to be the same or different.
  • the probability that the user belongs to the outward consumption type is 90%
  • the corresponding service is accepted by the user
  • the probability can be increased to 92%
  • the probability that the user belongs to the inward consumption type is 60%
  • the corresponding service is rejected by the user, Reduce the probability to 55 % and so on.
  • the services corresponding to the adjusted personality category can be as follows:
  • the online game service may be cancelled from the service corresponding to the outward consumption type; if the chat service is accepted by multiple inward consumption users, the corresponding inward consumption type may be adopted.
  • the service may be adjusted according to the result of multiple user feedbacks, and the number of the multiple users may take different values, such as 100 users or 500 users.
  • the adjustment service includes the cancellation of one or more services corresponding to the personality category, and one or more services corresponding to the new personality category.
  • information such as the number of successful application of the personality category, the number of failed applications, the number of corrections after the failure, the correction time interval, and the like may be recorded, and the recorded information may also be stored in the personality classification database.
  • the accuracy of the personality classification can be improved, so that the user can be provided with the corresponding service by using the more accurate personality classification result, and by adjusting the service, the user can be provided with the service that the user likes, thereby further better Provide services to users.
  • whether the user accepts or rejects the feedback information such as the service may be obtained through the service data display and processing interface, where the service data display and processing interface is the functional entity corresponding to the foregoing method in the embodiment and other functional entities in the actual system.
  • the interface of the user can obtain the application information, the feedback information, and the like of the user, and ensure the compatibility of the functional entity corresponding to the foregoing method in the embodiment with other functional entities.
  • the embodiment improves the manner of classifying the user's personality.
  • the user's network behavior characteristic information sample is established by the dynamic modeling engine.
  • the personality classification model forms a personality classification model formation mechanism that can be easily adjusted and modified.
  • the embodiment further improves the classification method for classifying the user to a certain personality classification, and breaks the conventional probability of calculating the user belonging to a certain personality classification based on the personality classification model. Therefore, the user's personality classification is further refined, and the practical problems of user personality diversification are fully considered, and the ability to set different service strategies according to personality probabilities is realized, which makes the service provision more precise and more humanized.
  • the method provided in this embodiment improves the service to the user by classifying the user and setting the corresponding service.
  • the user's personality category is searched and the corresponding service is provided to the user.
  • the rate can provide the service required by the user more accurately, overcomes the defect that the online online service in the prior art is not accurate, and thus provides better service to the user.
  • Correcting existing probabilities and/or services based on the results of user feedback can update probabilities and adjust services to further improve the accuracy of personality classification, provide users with services that users like, and achieve better service to users.
  • an embodiment of the present invention further provides an apparatus for providing an online service, including:
  • the character classification module 301 is configured to perform personality classification on all collected users according to the collected user network behavior characteristic information and the established personality classification model;
  • the service setting module 302 is connected to the personality classification module 301, and is configured to set a corresponding service for each personality category obtained by the personality classification module 301, wherein different different personality categories generally have different services, and multiple personality categories may be used. Corresponding to the same service, a personality category can also correspond to multiple services;
  • the search module 303 is respectively connected to the personality classification module 301 and the service setting module 302, and is configured to search for the personality category of the user in the result of the character classification module 301 when the user applies for the online service, and in the service setting module. Find the service corresponding to the personality category in the service set by 302;
  • the online service module 304 is connected to the search module 303, and is configured to find the search module 303.
  • the service arrived is provided to the user.
  • the user network behavior characteristic information in this embodiment includes multiple types, which are the same as the user network behavior characteristic information described in Embodiment 2, and details are not described herein again.
  • the apparatus may further include:
  • the personality classification model building module 305 is configured to perform operations according to the pre-selected user network behavior feature information samples and the dynamic modeling engine to obtain the personality classification model.
  • Dynamic modeling engines can be used in a variety of ways, including but not limited to: mathematical modeling engines such as Bayesian, decision trees, and support vector machines.
  • the personality classification module 301 may specifically include:
  • the character classification unit 301a is configured to calculate the collected user network behavior characteristic information by using the established personality classification model, and obtain the collected personality category of each user and the probability that the user belongs to the personality category.
  • Each user's personality category may be one or more, each of which corresponds to a probability, and the probability of corresponding different personality categories is usually different.
  • the online service module 304 may specifically include:
  • the sorting unit 304a is configured to sort the plurality of personality categories of the user according to the probability of the found personality category after the search module 303 finds the plurality of personality categories of the user and the probability of each personality category. Including high to low or low to high;
  • the online service unit 304b is connected to the sorting unit 304a for selecting some or all of the services provided to the user in the service corresponding to the plurality of personality categories in the sorted order of the sorting unit 304a.
  • the above apparatus may further include a personality classification model establishing module 305.
  • the foregoing apparatus in this embodiment may further include:
  • the recording module 306 is connected to the online service module 304, and is configured to record feedback information of the user on the service after the online service module 304 provides the service to the user.
  • the feedback information includes but is not limited to: the user clicks on the view, the number of pages viewed by the user, whether the user evaluates the service, the user accepts the service, and the like.
  • the apparatus of Figure 4 includes both a personality classification model building module 305 and a recording module 306.
  • the embodiment of the present invention does not exclude that the apparatus includes only one of the personality classification model building module 305 and the recording module 306.
  • the foregoing apparatus may further include:
  • the storage module 307 is configured to store the personality category of each user obtained by the personality classification unit 301a and the probability that the user belongs to the personality category, and store the service set by the service setting module 302 for each personality category;
  • the modification module 308 is respectively connected to the storage module 307 and the online service module 304, and is configured to adjust the character category, probability and service stored in the storage module 307 according to the feedback of the user on the service provided by the online service module 304. At least one of a probability and a service corresponding to the user's personality category. That is, only the probability corresponding to the personality category can be adjusted, or only the service corresponding to the personality category can be adjusted, and the probability and service corresponding to the personality category can be adjusted at the same time.
  • the method may include: a probability correction unit 308a, configured to determine, according to the personality category obtained by the personality classification unit 301a, the feedback of the user providing service to the online service module 304. If the service corresponding to the personality category is accepted by the user, in the probability stored by the storage module 307, the probability that the user belongs to the personality category is increased; if the service corresponding to the personality category is rejected by the user, Among the probabilities stored by the storage module 307, the probability that the user belongs to the personality category is reduced.
  • the magnitude of the probability adjustment can be set according to actual needs, and The magnitude of the increase and the magnitude of the decrease can be set to be the same or different.
  • the above apparatus may further include at least one of a personality classification model establishing module 305 and a recording module 306.
  • the device provided in this embodiment classifies the user and sets the corresponding service.
  • the user searches for the personality category of the user, and provides the corresponding service to the user, which can improve the service success of the user.
  • the rate overcomes the shortcomings of the prior art network online service, so as to better provide services to users. Setting conditions and stopping the provision of services to users when conditions are met can avoid wasting network resources and improving efficiency.
  • the existing probability is corrected according to the result of the user feedback, and the probability can be updated to further improve the accuracy of the personality classification, so that the user can be provided with the service according to the updated probability, so as to achieve better service to the user.

Description

提供在线服务的方法和装置
技术领域
本发明涉及网络通信领域, 特别涉及一种提供在线服务的方法和装 置。 发明背景
目前基于网络的在线服务, 其相关的服务内容尤其是带有用户付费 行为的服务内容, 其提供方式几乎都是基于已有分类的推送发布方式, 用户只能被动接受服务内容提供商发布的服务内容。
另外, 还有采用内容到内容的相关性推荐方式, 其中, 相关性推荐 方式一般是基于主题相关性、 内容相关性或用户的推荐列表等进行推 荐。
在现实生活中, 人的行为, 如购买的达成, 除了物品本身的因素外, 达成交易的过程中涉及到的环节, 如消费者的性格、 服务提供者的物品 导购技巧和肢体语言, 尤其是服务提供者对消费者性格的把握, 也是影 响是否能达成交易的关键。 用户提供个性化的服务, 使得所提供的在线服务往往无法准确地满足用 户的需求, 而无法达到预期的效果。 发明内容
本发明实施例提供一种提供在线服务的方法, 可以更加准确地提供 用户所需的网络在线服务。
本发明实施例提供一种提供在线服务的装置, 可以更加准确地提供 用户所需的网络在线服务。
本发明实施例的技术方案是这样实现的:
一种提供在线服务的方法, 所述方法包括:
根据收集的用户网络行为特征信息和建立的性格分类模型, 对所述 收集的所有用户进行性格分类;
为所述性格分类得到的每种性格类别设置对应的服务;
当有用户申请在线服务时, 在所述性格分类的结杲中, 查找所述用 户的性格类别 , 将所述性格类别对应的服务提供给所述用户。
一种提供在线服务的装置, 所述装置包括:
性格分类模块, 用于根据收集的用户网络行为特征信息和建立的性 格分类模型, 对所述收集的所有用户进行性格分类;
服务设置模块, 与所述性格分类模块相连, 用于为所述性格分类模 块进行性格分类得到的每种性格类别设置对应的服务;
查找模块, 与所述性格分类模块及服务设置模块分別相连, 用于当 有用户申请在线服务时, 在所述性格分类模块进行性格分类的结果中, 查找所述用户的性格类别, 并在所述服务设置模块设置的服务中查找所 述性格类别对应的服务;
在线服务模块, 与所述查找模块相连, 用于将所述查找模块找到的 服务提供给所述用户。 附图简要说明
图 1是本发明实施例 1中提供在线服务的方法流程图;
图 2是本发明实施例 2中提供在线服务的方法流程图;
图 3是本发明实施例 3中提供在线服务的装置第一结构图; 图 4是本发明实施例 3中提供在线服务的装置第二结构图; 图 5是本发明实施例 3中提供在线服务的装置第三结构图; 图 6是本发明实施例 3中提供在线服务的装置第四结构图。 实施本发明的方式
为使本发明的目的、技术方案和优点更加清楚明白, 以下举实施例, 并参照附图, 对本发明进一步详细说明。
本发明主要是根据用户的网络行为特征建立了一套对用户的性格 进行分类并根据性格分类来提供在线服务的方法, 通过对用户网络行为 特征信息的收集和建模, 将用户分为不同的性格类别, 并针对每种性格 类别分别提供满足其需求的不同的在线服务, 从而达到准确提供用户所 需在线服务的目的。
实施例 1
参见图 1, 本发明实施例提供了一种提供在线服务的方法, 包括: 步骤 101: 根据收集的用户网络行为特征信息和建立的性格分类模 型, 对收集的所有用户进行性格分类; 这里所建立的性格分类模型可以 是预先根据用户网络行为特征信息的分析和总结所预先建立的用户网 络行为特征信息与用户性格之间的对应关系, 才艮据这种对应关系, 就可 以根据收集的用户网络行为特征信息, 对用户性格进行分类。
步骤 102: 为性格分类得到的每种性格类别设置对应的服务。
步骤 103: 当有用户申请在线服务时, 在性格分类的结果中, 查找 所述用户的性格类别。
步骤 104: 将所述性格类别对应的服务提供给所述用户。
本实施例提供的方法通过对用户进行性格分类并设置对应的服务, 当有用户申请在线服务时, 查找该用户的性格类别, 并将对应的服务提 供给用户, 可以提高给用户提供服务的成功率, 克服了现有技术中网络 在线服务对于用户提供的服务无法准确满足用户需求的缺陷, 从而更好 地给用户提供服务。
实施例 2
参见图 2, 本发明实施例还提供了另一种提供在线服务的方法, 具 体包括:
步骤 201: 根据预先选取的用户网络行为特征信息样本和动态建模 引擎进行运算, 得到性格分类模型。
本实施例中, 用户网络行为特征信息是指用户在执行网络行为时, 与该网络行为相关的属性信息。 用户网络行为特征信息可以有多种, 包 括但不限于: 用户在网络上填写的个人资料, 用户操作免费业务的类别 及数量, 用户浏览业务的种类和数量, 用户试用业务的种类、 数量和金 额, 用户订购业务的延迟时间, 用户正式使用业务的时间间隔, 用户连 续使用业务的月份数, 用户推荐给别人使用的业务种类和数量, 被别人 推荐的业务种类和数量, 別人对该业务使用的评价, 以及用户对该评价 的留言次数等等。 其中, 用户在网络上填写的个人资料可以包括: 性格、 性别、 年龄、 血型、 爱好和地区等反应用户基本信息的数据。 而用户网 絡行为特征信息样本则是从中选出的作为参考的用户和各种信息。
用户网络行为特征信息可以通过多种渠道得到, 如在网络上进行统 计、 利用非网络手段进行调查等等。 得到的用户网络行为特征信息样本 可以存储在数据库中, 如建立一个建模行为数据库, 用于存储已获得的 用户网络行为特征信息样本。
本实施例中的动态建模引擎可以采用多种, 包括但不限于: 贝叶斯、 决策树和支持向量机等数学建模引擎。
将用户网络行为特征信息样本导入动态建模引擎, 即可得到性格分 类模型, 用于对之后收集到的用户网络行为特征信息进行计算并分类。 步骤 202: 利用建立的性格分类模型, 对收集的用户网络行为特征 信息进行运算, 得到收集的每个用户的性格类别和该用户属于该性格类 别的概率, 并存储该运算得到的结果。
本实施例中, 收集的用户通常为海量用户, 即收集大量用户的网络 行为特征信息。 其中, 收集的用户网絡行为特征信息有多项, 例如: 用 户 ID1、 性别为男性、 年龄为 25岁、 操作免费业务的类别为 10种、 浏 览业务的种类为 100种、 试用业务的种类为 50种等等, 将这些信息代 入已建立的性格分类模型中, 经过运算, 得出结论: 该用户的性格类别 为外向消费型, 且该用户属于外向消费型用户的概率为 90 % 。
本步骤中收集的用户网络行为特征信息对应的用户集合, 与步骤 201中选取的用户网络行为特征信息样本对应的用户集合不重合。例如, 步骤 201 中建立性格分类模型时选取 10万个用户, 本步骤中收集 100 万个用户的网络行为特征信息, 其中, 该 10 万个用户不包含在该 100 万个用户中。 其中, 收集的用户网络行为特征信息可以单独存储在一个 数据库中, 如预先建立的网絡行为特征数据库。
本实施例中, 用户的性格类别有多种, 包括但不限于: 外向消费型、 内向消费型、 开放消费型、保守消费型、 冲动消费型和谨慎消费型等等。
本实施例中, 对于一个用户来说, 性格分类的结果可以得到多种性 格类别, 且每种性格类别对应的概率不相同。 如某个用户属于外向消费 型的概率为 80 % , 属于开放消费型的概率为 70 % , 属于冲动消费型的 概率为 50 % , 属于内向消费型的概率为 30 %等等。
本实施例中, 性格分类的结果可以存储在数据库中, 如建立一个性 格分类数据库, 用于存储计算出的每个用户的性格类别及相应的概率。
步骤 203: 为性格分类得到的每种性格类别设置对应的服务, 并存 储设置的服务。 具体地, 可以根据需要为每种性格类别设置对应的服务; 还可以根 据收集的用户网络行为特征信息进行推理, 对于任一种性格类别, 分析 其中用户的服务行为相关信息, 根据分析的结果设置该性格类别对应的 服务。 例如, 外向消费型的性格类别中有 1万个用户, 其中有 8000个 用户喜欢网絡游戏服务、 6000个用户喜欢聊天服务, 则设置外向消费型 的性格类别对应的服务有网络游戏和聊天。
其中, 设置的服务可以包含多种内容。 例如, 显示产品名称、 筒要 功能、 金额、 客户评价、 外形图片、 销量排行榜等等。 本实施例中一种 性格类别可以对应一个或多个服务, 不同的性格类别也可以对应同一个 服务, 如保守消费型和内向消费型对应同一个服务。
步骤 204: 当有一个用户申请在线服务时, 在性格分类的结果中, 查找所述用户的性格类别。
本实施例中申请在线服务的用户, 为步骤 202中收集的用户网络行 为特征信息对应的用户集合中的一个用户。 可以根据该用户的 ID, 在性 格分类的结果(如性格分类数据库) 中, 查找该用户的性格类别, 如果 性格分类的结果中存储了该用户的多个性格类别, 则可以查询到该多个 性格类别, 以及其中每个性格类别对应的概率和服务等等。
步骤 205: 将所述性格类别对应的服务提供给所述用户。
本实施例是以一个用户提出在线服务申请为例进行说明的, 当有多 个用户提出在线服务申请时, 给其中每个用户提供服务的过程都与本实 施例中给一个用户提供服务的过程相同, 此处不再赘述。 在本发明实施 例中, 无论是给一个用户提供服务, 还是给多个用户提供服务, 都是在 上述性格分类和设置服务的基础上进行的, 而该性格分类则是根据收集 的大量用户的网络行为特征信息得到的。
当查询到该用户的多个性格类别, 以及每种性格类别对应的概率和 服务后, 具体地, 可以按照概率对该用户的多个性格类别进行排序, 从 高到低或者从低到高, 然后按照排序后的顺序, 在所述多个性格类别对 应的服务中, 选取部分或全部服务提供给该用户。 例如, 查询到该用户 有三种性格类别, 分别为: 保守消费型、 谨慎消费型和内向消费型, 对 应的概率分别为 90 %、 70 %和 80 %, 对应的服务分别为 Al、 A2和 A3, 按照从高到低的顺序排序后, 从 Al、 A3和 A2中选取部分或全部提供 给该用户。
进一步地, 对用户提供服务之后, 还可以包括:
记录用户对当前服务的反馈信息。 该反馈信息包括但不限于: 用户 点击查看、 用户浏览的页数、 用户是否对该服务进行评价、 用户接受该 服务等等。
为了提高给用户提供服务的成功率, 进一步地, 步骤 205中给用户 提供服务之后, 还可以包括:
步骤 206: 根据用户对当前性格类别对应服务的反馈, 在上述存储 的性格类别、 概率和服务中, 调整当前性格类别对应的概率, 或者调整 当前性格类别对应的服务, 或者同时调整当前性格类别对应的概率和服 务, 并将调整后的内容更新到上述存储的性格分类和服务中。
调整性格类别对应的概率可以具体如下: 如果当前性格类别对应的 服务被用户接受, 则提高用户属于该性格类别的概率; 如果当前性格类 别对应的服务被用户拒绝, 则降低用户属于该性格类别的概率。 其中, 概率调整的幅度可以根据实际需要进行设置, 并且提高的幅度和降低的 幅度可以设置得相同或不相同。
例如,用户属于外向消费型的概率为 90 % ,对应的服务被用户接受, 则可以将该概率提高为 92 % ; 用户属于内向消费型的概率为 60 % , 对 应的服务被用户拒绝, 则可以将该概率降低为 55 %等等。 调整性格类别对应的服务可以具体如下:
例如, 网络游戏服务被多个外向消费型的用户拒绝, 则可以将外向 消费型对应的服务中取消网络游戏服务; 聊天服务被多个内向消费型的 用户接受,则可以在内向消费型对应的服务中增加聊天服务等等。其中, 可以根据多个用户反馈的结果对服务进行调整, 该多个用户的个数可以 取不同的值, 如 100个用户或 500个用户等等。 调整服务包括将性格类 別对应的一个或多个服务取消 , 以及新增性格类别对应的一个或多个服 务。
进一步地, 在给用户提供服务后, 还可以记录性格类别成功应用次 数、 失败应用次数、 失败后修正次数、 修正时间间隔等信息, 而且还可 以将所记录的信息存储到上述性格分类数据库中。
通过对概率进行修正, 可以提高性格分类的准确性, 从而可以利用 更精确的性格分类结果给用户提供相应的服务, 通过对服务进行调整, 可以给用户提供用户喜欢的服务, 从而进而更好地给用户提供服务。
本实施例中, 用户是否接受或拒绝服务等反馈信息, 可以通过业务 数据展示与处理接口得到, 该业务数据展示与处理接口为本实施例中上 述方法对应的功能实体与实际系统中其它功能实体的接口, 通过该接口 可以获取到用户的申请信息、 反馈信息等等, 并且保证了本实施例中上 述方法对应的功能实体与其它功能实体的兼容性。
本实施例相比实施例 1 , 改进了对用户性格进行分类的方式, 不是 采用预先建立的固定的性格分类模型对用户性格进行分类, 而是根据用 户网络行为特征信息样本通过动态建模引擎建立性格分类模型, 形成可 方便调整和修改的性格分类模型形成机制。
另外, 本实施例还改进了将用户固定归于某一性格分类的分类方 法, 打破常规地基于性格分类模型计算用户属于某一性格分类的概率, 从而进一步将用户的性格分类细致化, 充分考虑到用户性格多样化的现 实问题, 实现了根据性格概率设置不同服务策略的能力, 使得服务提供 更加精准, 更加人性化。
本实施例提供的方法通过对用户进行性格分类并设置对应的服务, 当有用户申请在线服务时, 查找该用户的性格类别, 并将对应的服务提 供给用户, 可以提高给用户提供服务的成功率, 能够更加准确地提供用 户所需的服务, 克服了现有技术中网络在线服务准确性不高的缺陷, 从 而更好地给用户提供服务。 设定条件并在满足条件时停止对用户提供服 务, 可以避免浪费网絡资源, 提高效率。 根据用户反馈的结果修正已有 的概率和 /或服务, 可以更新概率及调整服务, 进一步提高性格分类的准 确性,给用户提供用户喜欢的服务,达到更好地给用户提供服务的目的。
实施例 3
参见图 3 , 本发明实施例还提供了一种提供在线服务的装置, 具体 包括:
性格分类模块 301, 用于根据收集的用户网络行为特征信息和建立 的性格分类模型, 对收集的所有用户进行性格分类;
服务设置模块 302, 与性格分类模块 301相连, 用于为性格分类模 块 301进行性格分类得到的每种性格类别设置对应的服务, 其中, 通常 不同的性格类别设置的服务不同, 多个性格类别可以对应同一个服务, 一个性格类别也可以对应多个服务;
查找模块 303 ,与性格分类模块 301及服务设置模块 302分别相连, 用于当有用户申请在线服务时, 在性格分类模块 301进行性格分类的结 果中, 查找用户的性格类别, 并在服务设置模块 302设置的服务中查找 该性格类別对应的服务;
在线服务模块 304, 与查找模块 303相连, 用于将查找模块 303找 到的服务提供给所述用户。
本实施例中的用户网络行为特征信息包括多种, 具体与实施例 2中 描述的用户网络行为特征信息相同, 此处不再赘述。
进一步地, 参见图 4, 提供了另一结构的提供在线服务的装置, 该 装置相比实施例 3中的提供在线服务的装置还可以包括:
性格分类模型建立模块 305 , 用于根据预先选取的用户网络行为特 征信息样本和动态建模引擎进行运算, 得到所述性格分类模型。
其中, 预先选取的用户网络行为特征信息样本对应的用户集合, 与 上述收集的用户网络行为特征信息对应的用户集合不重合。 动态建模引 擎可以采用多种, 包括但不限于: 贝叶斯、 决策树和支持向量机等数学 建模引擎。
本实施例中, 性格分类模块 301可以具体包括:
性格分类单元 301a, 用于利用建立的性格分类模型, 对收集的用户 网络行为特征信息进行运算, 得到收集的每个用户的性格类別和该用户 属于该性格类别的概率。
其中, 每个用户的性格类别可以为一种或多种, 每一种都对应一个 概率, 不同的性格类别对应的概率通常不相同。
进一步地, 参见图 5 , 在线服务模块 304可以具体包括:
排序单元 304a,用于当查找模块 303查找到所述用户的多个性格类 别, 以及每种性格类别的概率后, 根据找到的性格类别的概率, 对所述 用户的多个性格类别进行排序, 包括从高到低或从低到高等方式;
在线服务单元 304b,与排序单元 304a相连,用于按照排序单元 304a 排序后的顺序, 在所述多个性格类别对应的服务中, 选取部分或全部服 务提供给所述用户。 此时, 进一步地, 上述装置还可以包括性格分类模 型建立模块 305。 进一步地, 参见图 4, 本实施例中上述装置还可以包括:
记录模块 306,与在线服务模块 304相连,用于当在线服务模块 304 给所述用户提供服务后, 记录所述用户对该服务的反馈信息。 该反馈信 息包括但不限于: 用户点击查看、 用户浏览的页数、 用户是否对该服务 进行评价、 用户接受该服务等等。
图 4中装置同时包含性格分类模型建立模块 305和记录模块 306, 本发明实施例不排除该装置只包含性格分类模型建立模块 305和记录模 块 306中的一个。
本实施例中, 当性格分类模块 301包括性格分类单元 301a时,参见 图 6, 上述装置还可以包括:
存储模块 307 ,用于存储性格分类单元 301a得到的每个用户的性格 类别和该用户属于该性格类别的概率, 并存储服务设置模块 302为每种 性格类别设置的服务;
修正模块 308 , 与存储模块 307及在线服务模块 304分別相连, 用 于根据所述用户对在线服务模块 304提供的服务的反馈,在存储模块 307 存储的性格类别、 概率和服务中, 调整所述用户的性格类别对应的概率 和服务中的至少一种。 即可以只调整性格类别对应的概率, 或者只调整 性格类别对应的服务, 还可以同时调整性格类别对应的概率和服务。
进一步地, 当修正模块 308只对概率进行调整时, 可以具体包括: 概率修正单元 308a, 用于根据性格分类单元 301a得到的性格类别, 对所述用户对在线服务模块 304提供服务的反馈进行判断, 如果该性格 类别对应的服务被所述用户接受, 则在存储模块 307存储的概率中, 提 高所述用户属于该性格类别的概率; 如果该性格类别对应的服务被所述 用户拒绝, 则在存储模块 307存储的概率中, 降低所述用户属于该性格 类别的概率。 其中, 概率调整的幅度可以根据实际需要进行设置, 并且 提高的幅度和降低的幅度可以设置得相同或不相同。 此时, 进一步地, 上述装置还可以包括性格分类模型建立模块 305和记录模块 306中的至 少一个。
本实施例提供的装置通过对用户进行性格分类并设置对应的服务, 当有用户申请在线服务时, 查找该用户的性格类别, 并将对应的服务提 供给用户, 可以提高给用户提供服务的成功率, 克服了现有技术中网络 在线服务比较局限的缺陷, 从而更好地给用户提供服务。 设定条件并在 满足条件时停止对用户提供服务, 可以避免浪費网络资源, 提高效率。 根据用户反馈的结果修正已有的概率, 可以更新概率, 进一步提高性格 分类的准确性, 从而可以根据更新后的概率给用户提供服务, 达到更好 地给用户提供服务的目的。
以上所述仅为本发明的较佳实施例, 并不用以限制本发明, 凡在本 发明的精神和原则之内, 所作的任何修改、 等同替换、 改进等, 均应包 含在本发明的保护范围之内。

Claims

权利要求书
1. 一种提供在线服务的方法, 其特征在于, 所述方法包括: 根据收集的用户网络行为特征信息和建立的性格分类模型, 对所述 收集的所有用户进行性格分类;
为所述性格分类得到的每种性格类别设置对应的服务;
当有用户申请在线服务时, 在所述性格分类的结果中, 查找所述用 户的性格类别, 将所述性格类别对应的服务提供给所述用户。
2. 根据权利要求 1所述的提供在线服务的方法, 其特征在于, 所述 性格分类模型根据预先选取的用户网络行为特征信息样本和动态建模 引擎进行运算得到。
3. 根据权利要求 1所述的提供在线服务的方法, 其特征在于, 根据 收集的用户网络行为特征信息和建立的性格分类模型, 对所述收集的所 有用户进行性格分类, 具体包括:
利用建立的性格分类模型, 对收集的用户网络行为特征信息进行运 算, 得到所述收集的每个用户的性格类别和该用户属于该性格类别的概 率。
4. 根据权利要求 3所述的提供在线服务的方法, 其特征在于, 查找 所述用户的性格类别, 将所述性格类别对应的服务提供给所述用户, 具 体包括:
查找到所述用户的多个性格类别, 以及每种性格类别的概率; 根据所述性格类别的概率, 对所述用户的多个性格类别进行排序; 按照所述排序后的顺序, 在所述多个性格类别对应的服务中, 选取 部分或全部服务提供给所述用户。
5. 根据权利要求 1所述的提供在线服务的方法, 其特征在于, 将所 述性格类别对应的服务提供给所述用户之后, 还包括: 记录所述用户对所述服务的反馈信息。
6. 根据权利要求 3所述的提供在线服务的方法, 其特征在于, 所述 方法还包括: 存储所述每个用户的性格类别和该用户属于该性格类别的 概率, 并存储所述每种性格类别对应的服务;
将所述性格类别对应的服务提供给所述用户之后, 还包括: 根据所述用户对所述服务的反馈, 在所述存储的性格类別、 概率和 服务中, 调整所述用户的性格类别对应的概率和服务中的至少一种。
7. 根据权利要求 6所述的提供在线服务的方法, 其特征在于, 调整 所述用户的性格类别对应的概率, 具体包括:
如果所述性格类别对应的服务被所述用户接受, 则提高所述用户属 于所述性格类别的概率;
如果所述性格类别对应的服务被所述用户拒绝, 则降低所述用户属 于所述性格类别的概率。
8. 一种提供在线服务的装置, 其特征在于, 所述装置包括: 性格分类模块, 用于根据收集的用户网络行为特征信息和建立的性 格分类模型, 对所述收集的所有用户进行性格分类;
服务设置模块, 与所述性格分类模块相连, 用于为所述性格分类模 块进行性格分类得到的每种性格类别设置对应的服务;
查找模块, 与所述性格分类模块及服务设置模块分别相连, 用于当 有用户申请在线服务时, 在所述性格分类模块进行性格分类的结果中, 查找所述用户的性格类别, 并在所述服务设置模块设置的服务中查找所 述性格类别对应的服务;
在线服务模块, 与所述查找模块相连, 用于将所述查找模块找到的 服务提供给所述用户。
9. 根据权利要求 8所述的提供在线服务的装置, 其特征在于, 所述 装置还包括:
性格分类模型建立模块, 用于根据预先选取的用户网络行为特征信 息样本和动态建模引擎进行运算, 得到所述性格分类模型。
10. 根据权利要求 8所述的提供在线服务的装置, 其特征在于, 所 述性格分类模块具体包括:
性格分类单元, 用于利用建立的性格分类模型, 对收集的用户网络 行为特征信息进行运算, 得到所述收集的每个用户的性格类别和该用户 属于该性格类别的概率。
11. 根据权利要求 10所述的提供在线服务的装置, 其特征在于, 所 述在线服务模块具体包括:
排序单元, 用于当所述查找模块查找到所述用户的多个性格类别, 以及每种性格类别的概率后, 根据所述性格类别的概率, 对所述用户的 多个性格类别进行排序;
在线服务单元, 与所述排序单元相连, 用于按照所述排序单元排序 后的顺序, 在所述多个性格类别对应的服务中, 选取部分或全部服务提 供给所述用户。
12. 根据权利要求 8所述的提供在线服务的装置, 其特征在于, 所 述装置还包括:
记录模块, 用于当所述在线服务模块给所述用户提供服务后, 记录 所述用户对所述服务的反馈信息。
13. 根据权利要求 10所述的提供在线服务的装置, 其特征在于, 所 述装置还包括:
存储模块, 用于存储所述性格分类单元得到的所述每个用户的性格 类别和该用户属于该性格类别的概率, 并存储所述服务设置模块为所述 每种性格类别设置的服务;
修正模块, 与所述存储模块及在线服务模块分别相连, 用于根据所 述用户对所述在线服务模块提供的服务的反馈, 在所述存储模块存储的 性格类别、 概率和服务中, 调整所述用户的性格类别对应的概率和服务 中的至少一种。
14. 根据权利要求 13所述的提供在线服务的装置, 其特征在于, 所 述修正模块具体包括:
概率修正单元, 用于根据所述性格分类单元得到的性格类别, 对所 述用户对所述在线服务模块提供服务的反馈进行判断, 如果所述性格类 别对应的服务被所述用户接受, 则在所述存储模块存储的概率中, 提高 所述用户属于所述性格类别的概率; 如果所述性格类别对应的服务被所 述用户拒绝, 则在所述存储模块存储的概率中, 降低所述用户属于所述 性格类别的概率。
PCT/CN2009/074253 2008-09-28 2009-09-27 提供在线服务的方法和装置 WO2010034259A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200810223246.X 2008-09-28
CN200810223246XA CN101359995B (zh) 2008-09-28 2008-09-28 提供在线服务的方法和装置

Publications (1)

Publication Number Publication Date
WO2010034259A1 true WO2010034259A1 (zh) 2010-04-01

Family

ID=40332344

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/074253 WO2010034259A1 (zh) 2008-09-28 2009-09-27 提供在线服务的方法和装置

Country Status (2)

Country Link
CN (1) CN101359995B (zh)
WO (1) WO2010034259A1 (zh)

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10489795B2 (en) * 2007-04-23 2019-11-26 The Nielsen Company (Us), Llc Determining relative effectiveness of media content items
CN101359995B (zh) * 2008-09-28 2011-05-04 腾讯科技(深圳)有限公司 提供在线服务的方法和装置
WO2011106925A1 (en) * 2010-03-01 2011-09-09 Nokia Corporation Method and apparatus for estimating user characteristics based on user interaction data
CN102456199A (zh) * 2010-10-18 2012-05-16 北京学之途网络科技有限公司 互联网用户样本集的扩充、属性参数获取方法及装置
CN102456201B (zh) * 2010-10-26 2016-09-14 阿里巴巴集团控股有限公司 一种客户操作数据的分析方法和装置
CN103905489B (zh) * 2012-12-27 2015-04-29 腾讯科技(深圳)有限公司 网络信息服务处理方法和系统
CN103440864A (zh) * 2013-07-31 2013-12-11 湖南大学 基于语音的人格特征预测方法
CN103927675B (zh) * 2014-04-18 2017-07-11 北京京东尚科信息技术有限公司 判断用户年龄段的方法及装置
CN104216965B (zh) * 2014-08-21 2018-11-20 北京金山安全软件有限公司 信息推荐方法和装置
CN104461016B (zh) * 2014-12-23 2018-02-13 杭州云造科技有限公司 产品的机器性格表现方法及装置
CN105957414A (zh) * 2016-06-27 2016-09-21 安徽科成信息科技有限公司 一种班班通的教育系统
CN106515939A (zh) * 2016-08-23 2017-03-22 杭州云造科技有限公司 自行车智能人机交互系统及其交互方法
CN106878392B (zh) * 2017-01-11 2021-09-07 浙江工商大学 一种基于学生成果的在线服务方法和装置
CN107592346B (zh) * 2017-08-31 2020-09-01 江西博瑞彤芸科技有限公司 基于用户行为分析的用户分类方法
CN108021864A (zh) * 2017-11-02 2018-05-11 平安科技(深圳)有限公司 人物性格分析方法、装置及存储介质
CN108776846A (zh) * 2018-05-15 2018-11-09 中国平安人寿保险股份有限公司 推荐方法、装置、计算机设备及存储介质
CN109598576A (zh) * 2018-10-25 2019-04-09 阿里巴巴集团控股有限公司 服务推荐方法、装置及设备
CN114117054B (zh) * 2022-01-24 2022-05-17 华南师范大学 基于个性词的学生期末评价方法、系统、装置和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1932871A (zh) * 2005-09-13 2007-03-21 北京搜畅网络技术有限公司 用于创建用户个性化页面的方法
CN1967533A (zh) * 2006-07-17 2007-05-23 北京航空航天大学 采用元推荐引擎的门户个性化推荐服务方法和系统
CN101266620A (zh) * 2008-04-07 2008-09-17 北京大学 向用户提供目标信息的方法及设备
CN101359995A (zh) * 2008-09-28 2009-02-04 腾讯科技(深圳)有限公司 提供在线服务的方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1932871A (zh) * 2005-09-13 2007-03-21 北京搜畅网络技术有限公司 用于创建用户个性化页面的方法
CN1967533A (zh) * 2006-07-17 2007-05-23 北京航空航天大学 采用元推荐引擎的门户个性化推荐服务方法和系统
CN101266620A (zh) * 2008-04-07 2008-09-17 北京大学 向用户提供目标信息的方法及设备
CN101359995A (zh) * 2008-09-28 2009-02-04 腾讯科技(深圳)有限公司 提供在线服务的方法和装置

Also Published As

Publication number Publication date
CN101359995A (zh) 2009-02-04
CN101359995B (zh) 2011-05-04

Similar Documents

Publication Publication Date Title
WO2010034259A1 (zh) 提供在线服务的方法和装置
Wu et al. A two-stage social trust network partition model for large-scale group decision-making problems
US9317592B1 (en) Content-based classification
US8700551B2 (en) Systems and methods for identifying provider noncustomers as likely acquisition targets
US7752159B2 (en) System and method for classifying text
US11763145B2 (en) Article recommendation method and apparatus, computer device, and storage medium
CN110059182A (zh) 面向客服的话术推荐方法和装置
US20070038646A1 (en) Ranking blog content
US20090327224A1 (en) Automatic Classification of Search Engine Quality
US20130166374A1 (en) Managing reputations
US20090138565A1 (en) Method and System for Facilitating Content Analysis and Insertion
CN107580704A (zh) 上下文人物推荐
EP2666101A1 (en) Method and system of selecting responders
CN108198017B (zh) 个性化客服处理方法及装置
CN113742492B (zh) 保险方案生成方法、装置、电子设备及存储介质
CN108027814A (zh) 停用词识别方法与装置
WO2020135642A1 (zh) 一种基于生成对抗网络的模型训练方法及设备
CN107832421A (zh) 一种自适应的组合数据挖掘模型系统及方法
WO2021179481A1 (zh) 数据内容个性化推送冷启动方法、装置、设备和存储介质
US11748426B2 (en) Personalized comment recommendation method based on link prediction model of graph bidirectional aggregation network
CN111695084A (zh) 模型生成方法、信用评分生成方法、装置、设备及存储介质
CN113111250A (zh) 业务推荐方法、装置、相关设备及存储介质
CN102959539A (zh) 一种业务交叉时的项目推荐方法及系统
CN109636682A (zh) 一种教学资源自动收集系统
CN112328868A (zh) 一种基于信息数据的信用评估与授信申请系统及方法

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: 09815628

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 8344/CHENP/2010

Country of ref document: IN

NENP Non-entry into the national phase

Ref country code: DE

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 30/08/2011)

122 Ep: pct application non-entry in european phase

Ref document number: 09815628

Country of ref document: EP

Kind code of ref document: A1