CN118350904A - User clothing order behavior analysis method and system based on big data - Google Patents
User clothing order behavior analysis method and system based on big data Download PDFInfo
- Publication number
- CN118350904A CN118350904A CN202410750730.7A CN202410750730A CN118350904A CN 118350904 A CN118350904 A CN 118350904A CN 202410750730 A CN202410750730 A CN 202410750730A CN 118350904 A CN118350904 A CN 118350904A
- Authority
- CN
- China
- Prior art keywords
- user
- collaborative
- knowledge
- interest
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Recommending goods or services
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
- G06N5/025—Extracting rules from data
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0253—During e-commerce, i.e. online transactions
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/2866—Architectures; Arrangements
- H04L67/30—Profiles
- H04L67/306—User profiles
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Animal Behavior & Ethology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
技术领域Technical Field
本申请涉及大数据技术领域,具体而言,涉及一种基于大数据的用户服装订单行为分析方法及系统。The present application relates to the field of big data technology, and in particular to a method and system for analyzing user clothing order behavior based on big data.
背景技术Background technique
随着互联网技术的飞速发展,电子商务领域的数据量呈现爆炸式增长,特别是在服装电商领域,用户产生的订单行为数据尤为丰富。这些数据不仅记录了用户的购买行为,还包含了用户的浏览历史、评价内容等多维度信息,为分析用户兴趣和行为模式提供了宝贵的数据资源。With the rapid development of Internet technology, the amount of data in the field of e-commerce has exploded, especially in the field of clothing e-commerce, where the order behavior data generated by users is particularly rich. These data not only record the user's purchase behavior, but also include multi-dimensional information such as the user's browsing history and evaluation content, providing valuable data resources for analyzing user interests and behavior patterns.
然而,面对海量的用户服装订单行为数据,如何有效地提取有价值的信息,分析用户之间的协同兴趣状态和用户兴趣相关度,进而为用户提供更加个性化的服务,成为了一个亟待解决的问题。传统相关技术往往难以应对大规模、复杂的数据集,且难以捕捉用户之间的深层次联系。此外,然而在相关技术,往往只关注用户个体的购买行为,忽视了用户之间的协同兴趣和影响。However, faced with massive amounts of user clothing order behavior data, how to effectively extract valuable information, analyze the collaborative interest status and user interest relevance between users, and then provide users with more personalized services has become an urgent problem to be solved. Traditional related technologies often have difficulty coping with large-scale and complex data sets, and it is difficult to capture the deep connections between users. In addition, related technologies often only focus on individual user purchase behaviors, ignoring the collaborative interests and influences between users.
发明内容Summary of the invention
鉴于上述提及的问题,结合本申请的第一方面,本申请实施例提供一种基于大数据的用户服装订单行为分析方法,所述方法包括:In view of the above-mentioned problems, in combination with the first aspect of the present application, an embodiment of the present application provides a user clothing order behavior analysis method based on big data, the method comprising:
获取目标用户的用户服装订单行为大数据;Obtain big data on clothing order behavior of target users;
依据所述用户服装订单行为大数据确定所述目标用户之间的用户协同兴趣状态以及用户兴趣相关度,所述用户协同兴趣状态反映多个目标用户是否组成用户协同群组,所述用户兴趣相关度反映组成所述用户协同群组的多个所述目标用户之间的兴趣相关度;Determine the user collaborative interest state and user interest correlation between the target users based on the user clothing order behavior big data, the user collaborative interest state reflects whether multiple target users form a user collaborative group, and the user interest correlation reflects the interest correlation between the multiple target users forming the user collaborative group;
依据所述用户协同兴趣状态以及所述用户兴趣相关度生成目标用户协同知识图谱,所述目标用户协同知识图谱由知识成员以及所述知识成员之间的知识链路组成,其中,不同知识成员对应不同目标用户,且所述知识链路反映组成所述用户协同群组的多个所述目标用户之间的所述用户兴趣相关度;Generate a target user collaborative knowledge graph according to the user collaborative interest state and the user interest relevance, wherein the target user collaborative knowledge graph is composed of knowledge members and knowledge links between the knowledge members, wherein different knowledge members correspond to different target users, and the knowledge links reflect the user interest relevance between the multiple target users constituting the user collaborative group;
基于所述目标用户协同知识图谱,确定每个目标用户的协同兴趣路径,并基于所述每个目标用户的协同兴趣路径对所述每个目标用户进行信息推送。Based on the target user collaborative knowledge graph, a collaborative interest path of each target user is determined, and information is pushed to each target user based on the collaborative interest path of each target user.
在第一方面的一种可能的实施方式中,所述依据所述用户协同兴趣状态以及所述用户兴趣相关度生成目标用户协同知识图谱,包括:In a possible implementation of the first aspect, generating a target user collaborative knowledge graph according to the user collaborative interest state and the user interest relevance includes:
依据所述用户协同兴趣状态,确定各个所述目标用户的协同群组数量,所述协同群组数量表示由所述目标用户组成的所述用户协同群组的数量;Determining the number of collaborative groups of each of the target users according to the user collaborative interest state, wherein the number of collaborative groups represents the number of the user collaborative groups formed by the target users;
依据所述协同群组数量,确定所述目标用户对应知识成员的知识成员特征,所述知识成员特征包括知识成员活跃度和知识成员影响力中的至少一种,其中,所述知识成员活跃度与所述协同群组数量为正向关联关系;Determining, according to the number of collaborative groups, knowledge member characteristics of the knowledge member corresponding to the target user, wherein the knowledge member characteristics include at least one of knowledge member activity and knowledge member influence, wherein the knowledge member activity is positively correlated with the number of collaborative groups;
依据所述知识成员特征生成所述知识成员;generating the knowledge member according to the characteristics of the knowledge member;
依据所述目标用户之间的所述用户兴趣相关度,确定所述目标用户对应知识成员之间的偏离度,所述偏离度与所述用户兴趣相关度为负向关联关系;Determine the deviation between the knowledge members corresponding to the target users according to the user interest correlation between the target users, wherein the deviation is negatively correlated with the user interest correlation;
在依据所述偏离度完成知识成员位置优化时,在组成所述用户协同群组的所述目标用户对应的知识成员之间生成所述知识链路;When the knowledge member position optimization is completed according to the deviation, the knowledge link is generated between the knowledge members corresponding to the target user constituting the user collaboration group;
生成由所述知识成员和所述知识链路组成的所述目标用户协同知识图谱。Generate the target user collaborative knowledge graph consisting of the knowledge members and the knowledge links.
在第一方面的一种可能的实施方式中,所述依据所述用户协同兴趣状态以及所述用户兴趣相关度生成目标用户协同知识图谱之后,所述方法包括:In a possible implementation of the first aspect, after generating a target user collaborative knowledge graph according to the user collaborative interest state and the user interest relevance, the method includes:
基于对所述目标用户协同知识图谱中第一目标用户的强化表达指令,对所述目标用户协同知识图谱中的第一知识成员以及第二知识成员进行强化表达,所述第一知识成员为所述第一目标用户对应的知识成员,所述第二知识成员为第二目标用户对应的知识成员,所述第二目标用户与所述第一目标用户之间具有k级兴趣联系,k为正整数;Based on the enhanced expression instruction for the first target user in the target user collaborative knowledge graph, the first knowledge member and the second knowledge member in the target user collaborative knowledge graph are enhanced, wherein the first knowledge member is the knowledge member corresponding to the first target user, the second knowledge member is the knowledge member corresponding to the second target user, and there is a k-level interest connection between the second target user and the first target user, where k is a positive integer;
对所述第一知识成员与所述第二知识成员之间的知识链路,以及所述第二知识成员之间的知识链路进行强化表达。The knowledge link between the first knowledge member and the second knowledge member, and the knowledge link between the second knowledge members are enhanced and expressed.
在第一方面的一种可能的实施方式中,所述方法还包括:In a possible implementation of the first aspect, the method further includes:
基于对所述第一目标用户的选中指令,对所述第一知识成员以及所述第二知识成员以外的知识成员进行收缩优化;Based on the selection instruction of the first target user, shrinking and optimizing the knowledge members other than the first knowledge member and the second knowledge member;
其中,收缩条件下,所述第二知识成员中的目标知识成员与收缩知识成员连接,所述目标知识成员对应的所述第二目标用户与所述第一目标用户之间存在k级兴趣相关度,且所述收缩知识成员中表达与所述目标知识成员相连接且被收缩的知识成员的数量;Wherein, under the contraction condition, the target knowledge member in the second knowledge member is connected to the contraction knowledge member, there is a k-level interest correlation between the second target user corresponding to the target knowledge member and the first target user, and the contraction knowledge member expresses the number of knowledge members connected to the target knowledge member and contracted;
所述方法还包括:The method further comprises:
基于对所述收缩知识成员的选中指令,呈现与所述目标知识成员存在一级兴趣相关度的第三知识成员。Based on the selection instruction of the contracted knowledge member, a third knowledge member having a primary interest relevance to the target knowledge member is presented.
在第一方面的一种可能的实施方式中,所述方法还包括:In a possible implementation of the first aspect, the method further includes:
基于对所述目标用户协同知识图谱的过滤指令,获取过滤要求,所述过滤要求包括用户特征标签、门限用户兴趣相关度以及兴趣相关层级中的至少一种;Based on the filtering instruction of the target user collaborative knowledge graph, a filtering requirement is obtained, wherein the filtering requirement includes at least one of a user feature label, a threshold user interest relevance, and an interest relevance level;
对所述目标用户协同知识图谱中不满足所述过滤要求的知识成员进行削弱表达。The knowledge members in the target user collaborative knowledge graph that do not meet the filtering requirements are weakened in expression.
在第一方面的一种可能的实施方式中,所述方法还包括:In a possible implementation of the first aspect, the method further includes:
基于对所述目标用户协同知识图谱的冗余优化指令,依据所述冗余优化指令表征的冗余优化权重,确定所述目标用户协同知识图谱的优化参数量;Based on the redundant optimization instructions for the target user collaborative knowledge graph, and according to the redundant optimization weights represented by the redundant optimization instructions, determining the optimization parameter amount of the target user collaborative knowledge graph;
依据所述优化参数量优化所述知识成员和/或所述知识链路所生成的特征向量,其中,所述知识成员可展示的信息包括用户标志、用户名以及协同群组数目,所述知识链路可展示的信息包括用户兴趣相关度。The feature vector generated by optimizing the knowledge member and/or the knowledge link is/are optimized according to the optimization parameter amount, wherein the information that can be displayed by the knowledge member includes user logo, user name and number of collaborative groups, and the information that can be displayed by the knowledge link includes user interest relevance.
在第一方面的一种可能的实施方式中,所述依据所述用户服装订单行为大数据确定所述目标用户之间的用户协同兴趣状态以及用户兴趣相关度,包括:In a possible implementation of the first aspect, determining the user collaborative interest state and user interest relevance between the target users based on the user clothing order behavior big data includes:
确定参考用户集合,所述参考用户集合由多个所述目标用户组成;Determine a reference user set, where the reference user set consists of a plurality of the target users;
依据所述参考用户集合中各个所述目标用户的所述用户服装订单行为大数据,确定所述参考用户集合的协同特征数据;Determining collaborative feature data of the reference user set based on the user clothing order behavior big data of each target user in the reference user set;
在所述协同特征数据符合协同匹配要求时,确定所述参考用户集合中的多个所述目标用户组成所述用户协同群组;When the collaborative feature data meets the collaborative matching requirement, determining that a plurality of the target users in the reference user set form the user collaborative group;
依据所述协同特征数据确定所述用户协同群组的所述用户兴趣相关度;Determining the user interest relevance of the user collaborative group according to the collaborative feature data;
在第一方面的一种可能的实施方式中,所述协同特征数据包括协同购买率提升矢量、协同浏览率提升矢量以及用户偏好匹配度,所述协同购买率提升矢量用于表征所述参考用户集合对购买率的影响情况,所述协同浏览率提升矢量用于表征所述参考用户集合对商品浏览率的影响,所述用户偏好匹配度用于表征至少两个所述目标用户之间是否存在相似的服装偏好;In a possible implementation of the first aspect, the collaborative feature data includes a collaborative purchase rate improvement vector, a collaborative browsing rate improvement vector, and a user preference matching degree, wherein the collaborative purchase rate improvement vector is used to characterize the influence of the reference user set on the purchase rate, the collaborative browsing rate improvement vector is used to characterize the influence of the reference user set on the product browsing rate, and the user preference matching degree is used to characterize whether there are similar clothing preferences between at least two of the target users;
所述方法包括:The method comprises:
在所述协同购买率提升矢量或所述协同浏览率提升矢量为负提升矢量时,确定不符合所述协同匹配要求;或,When the collaborative purchase rate improvement vector or the collaborative browsing rate improvement vector is a negative improvement vector, determining that the collaborative matching requirement is not met; or,
在所述协同购买率提升矢量和所述协同浏览率提升矢量为正向提升矢量,且所述用户偏好匹配度表征存在相似偏好,且所述协同购买率提升矢量和所述协同浏览率提升矢量的显著提升矢量值大于门限值时,确定符合所述协同匹配要求;或,When the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates the existence of similar preferences, and the significant improvement vector values of the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are greater than a threshold value, it is determined that the collaborative matching requirement is met; or,
在所述协同购买率提升矢量和所述协同浏览率提升矢量为正向提升矢量,且所述用户偏好匹配度表征存在相似偏好,但所述协同购买率提升矢量和所述协同浏览率提升矢量的显著提升矢量值小于门限值时,确定不符合所述协同匹配要求;或,When the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates the existence of similar preferences, but the significant improvement vector values of the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are less than a threshold value, it is determined that the collaborative matching requirement is not met; or,
在所述协同购买率提升矢量和所述协同浏览率提升矢量为正向提升矢量,且所述用户偏好匹配度表征不存在相似偏好,但所述协同购买率提升矢量的显著提升矢量值大于门限值时,确定符合所述协同匹配要求;或,When the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates that there is no similar preference, but the significant improvement vector value of the collaborative purchase rate improvement vector is greater than a threshold value, it is determined that the collaborative matching requirement is met; or,
在所述协同购买率提升矢量和所述协同浏览率提升矢量为正向提升矢量,且所述用户偏好匹配度表征不存在相似偏好,且所述协同购买率提升矢量的显著提升矢量值小于门限值时,确定不符合所述协同匹配要求。When the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates that there is no similar preference, and the significant improvement vector value of the collaborative purchase rate improvement vector is less than a threshold value, it is determined that the collaborative matching requirement is not met.
在第一方面的一种可能的实施方式中,所述依据所述协同特征数据确定所述用户协同群组的所述用户兴趣相关度,包括:In a possible implementation of the first aspect, determining the user interest relevance of the user collaborative group according to the collaborative feature data includes:
依据所述协同购买率提升矢量和所述协同浏览率提升矢量,确定所述用户协同群组的所述用户兴趣相关度,所述用户兴趣相关度与所述协同购买率提升矢量为正向关联关系,所述用户兴趣相关度与所述协同浏览率提升矢量为正向关联关系。The user interest correlation of the user collaborative group is determined according to the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector, wherein the user interest correlation is positively correlated with the collaborative purchase rate improvement vector, and the user interest correlation is positively correlated with the collaborative browsing rate improvement vector.
在第一方面的一种可能的实施方式中,所述基于所述目标用户协同知识图谱,确定每个目标用户的协同兴趣路径,并基于所述每个目标用户的协同兴趣路径对所述每个目标用户进行信息推送的步骤,包括:In a possible implementation of the first aspect, the step of determining a collaborative interest path of each target user based on the target user collaborative knowledge graph, and pushing information to each target user based on the collaborative interest path of each target user includes:
对所述目标用户协同知识图谱中的每个知识成员进行活跃度分析,生成每个知识成员的活跃度,所述活跃度根据每个知识成员对应的目标用户在服装电商平台上的交互频率、订单数量、评价次数进行衡量;Performing activity analysis on each knowledge member in the target user collaborative knowledge graph to generate the activity of each knowledge member, wherein the activity is measured according to the interaction frequency, order quantity, and evaluation number of the target user corresponding to each knowledge member on the clothing e-commerce platform;
选择活跃度满足预设条件的知识成员作为协同兴趣路径的起始知识成员,从所述起始知识成员出发,沿着知识链路进行拓展,初步识别出与目标用户相关联的协同兴趣路径,所述协同兴趣路径由多个相互关联的知识成员和知识链路组成,反映了目标用户之间的兴趣传递和影响关系;Select knowledge members whose activity meets preset conditions as the starting knowledge members of the collaborative interest path, start from the starting knowledge members, expand along the knowledge links, and preliminarily identify the collaborative interest path associated with the target user. The collaborative interest path consists of multiple interrelated knowledge members and knowledge links, reflecting the interest transfer and influence relationship between the target users;
基于目标用户之间的兴趣相关度、购买行为的相似度和用户评价的一致性对初步识别的协同兴趣路径进行权重赋值,并根据赋值后的权重对所述协同兴趣路径进行优化和剪枝;Assigning weights to the initially identified collaborative interest paths based on the interest relevance, purchase behavior similarity, and user evaluation consistency among target users, and optimizing and pruning the collaborative interest paths according to the assigned weights;
在优化和剪枝后的协同兴趣路径中,识别出关键知识成员,并获取每个关键知识成员在协同兴趣路径中的位置特征,所述位置特征包括每个关键知识成员处于协同兴趣路径的具体位置以及每个关键知识成员与所述协同兴趣路径中其它知识成员的路径代价关系;In the optimized and pruned collaborative interest path, key knowledge members are identified, and the position characteristics of each key knowledge member in the collaborative interest path are obtained, wherein the position characteristics include the specific position of each key knowledge member in the collaborative interest path and the path cost relationship between each key knowledge member and other knowledge members in the collaborative interest path;
根据所述关键知识成员在所述协同兴趣路径中的用户行为和协同影响力,识别所述关键知识成员在协同兴趣路径中的角色特征,所述角色特征包括引领角色、传递角色和跟随角色中的一种;According to the user behavior and collaborative influence of the key knowledge member in the collaborative interest path, identifying the role characteristics of the key knowledge member in the collaborative interest path, wherein the role characteristics include one of a leading role, a transmitting role, and a following role;
根据所述每个关键知识成员在协同兴趣路径中的位置特征和角色特征,从所述每个关键知识成员与其它知识成员之间的协同行为数据中挖掘目标兴趣偏好数据,并根据所述目标兴趣偏好数据,生成与其它目标用户协同关联的个性化推送内容。According to the position characteristics and role characteristics of each key knowledge member in the collaborative interest path, target interest preference data is mined from the collaborative behavior data between each key knowledge member and other knowledge members, and personalized push content collaboratively associated with other target users is generated based on the target interest preference data.
再一方面,本申请实施例还提供一种基于大数据的用户服装订单行为分析系统,包括处理器、机器可读存储介质,所述机器可读存储介质和所述处理器连接,所述机器可读存储介质用于存储程序、指令或代码,所述处理器用于执行所述机器可读存储介质中的程序、指令或代码,以实现上述的方法。On the other hand, an embodiment of the present application also provides a user clothing order behavior analysis system based on big data, including a processor and a machine-readable storage medium, wherein the machine-readable storage medium is connected to the processor, the machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.
基于以上方面,本申请实施例通过获取目标用户的用户服装订单行为大数据,能够深入分析用户之间的协同兴趣状态和用户兴趣相关度,从而准确地识别出具有相似购买兴趣和行为的用户群体,即用户协同群组。进一步地,依据这些分析结果生成目标用户协同知识图谱,该目标用户协同知识图谱直观展示了用户之间的兴趣联系和协同关系,为深入了解用户行为提供了有力工具。基于目标用户协同知识图谱,能够精准地确定每个目标用户的协同兴趣路径,这意味着能够清晰地描绘出用户兴趣的传播和影响路径。通过利用这些协同兴趣路径,能够为用户提供高度个性化的信息推送服务,如向用户推荐与其兴趣高度匹配的服装款式或潮流趋势。这种基于协同兴趣路径的推送方式,不仅大大提高了信息推送的准确性和有效性,还显著提升了用户体验和满意度。由此,通过生成目标用户协同知识图谱和确定协同兴趣路径,为服装电商平台提供了有力的数据支持,帮助平台实现更精准的用户画像和个性化的服务策略,从而提升了平台的竞争力和用户黏性。Based on the above aspects, the embodiment of the present application can deeply analyze the collaborative interest status and user interest relevance between users by obtaining the user clothing order behavior big data of the target user, so as to accurately identify the user group with similar purchasing interests and behaviors, that is, the user collaborative group. Further, based on these analysis results, a target user collaborative knowledge graph is generated, which intuitively displays the interest connection and collaborative relationship between users, and provides a powerful tool for in-depth understanding of user behavior. Based on the target user collaborative knowledge graph, the collaborative interest path of each target user can be accurately determined, which means that the propagation and influence path of user interests can be clearly depicted. By utilizing these collaborative interest paths, highly personalized information push services can be provided to users, such as recommending clothing styles or fashion trends that are highly matched with their interests to users. This push method based on collaborative interest paths not only greatly improves the accuracy and effectiveness of information push, but also significantly improves user experience and satisfaction. Therefore, by generating a target user collaborative knowledge graph and determining a collaborative interest path, a strong data support is provided for the clothing e-commerce platform, which helps the platform to achieve more accurate user portraits and personalized service strategies, thereby improving the competitiveness and user stickiness of the platform.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
图1是本申请实施例提供的基于大数据的用户服装订单行为分析方法的执行流程示意图。FIG1 is a schematic diagram of the execution flow of a method for analyzing user clothing order behavior based on big data provided in an embodiment of the present application.
图2是本申请实施例提供的基于大数据的用户服装订单行为分析系统硬件架构示意图。FIG2 is a schematic diagram of the hardware architecture of a user clothing order behavior analysis system based on big data provided in an embodiment of the present application.
具体实施方式Detailed ways
下面结合说明书附图对本申请进行具体说明,图1是本申请一种实施例提供的基于大数据的用户服装订单行为分析方法的流程示意图,下面对该基于大数据的用户服装订单行为分析方法进行详细介绍。The present application is described in detail below in conjunction with the accompanying drawings in the specification. Figure 1 is a flow chart of a user clothing order behavior analysis method based on big data provided by an embodiment of the present application. The user clothing order behavior analysis method based on big data is introduced in detail below.
步骤S110,获取目标用户的用户服装订单行为大数据。Step S110, obtaining the clothing order behavior big data of the target user.
本实施例中,服务器首先连接到服装电商平台的数据库,从中提取目标用户在服装电商平台上的所有服装订单数据,这些服装订单数据包括但不限于用户浏览的服装品类、添加到购物车的商品、下单购买的商品、订单金额、购买时间、收货地址等等。此外,还包括用户对商品的评价、退货记录以及与客服的交互记录,这些服装订单数据构成了用户服装订单行为大数据的基础,为后续分析用户兴趣和协同行为提供了丰富的信息源。In this embodiment, the server first connects to the database of the clothing e-commerce platform and extracts all clothing order data of the target user on the clothing e-commerce platform. These clothing order data include but are not limited to the clothing categories browsed by the user, the goods added to the shopping cart, the goods ordered, the order amount, the purchase time, the delivery address, etc. In addition, it also includes the user's evaluation of the goods, the return record and the interaction record with the customer service. These clothing order data constitute the basis of the user's clothing order behavior big data, which provides a rich information source for the subsequent analysis of user interests and collaborative behaviors.
例如,服务器从电商平台的数据库中,获取了用户A、B、C等在过去一年内的所有服装订单记录,这些服装订单记录显示,用户A经常购买休闲风格的服装,用户B偏好商务正装,而用户C则对时尚潮流的服装更感兴趣。For example, the server obtains all clothing order records of users A, B, C, etc. in the past year from the database of the e-commerce platform. These clothing order records show that user A often buys casual-style clothing, user B prefers business formal wear, and user C is more interested in fashionable clothing.
步骤S120,依据所述用户服装订单行为大数据确定所述目标用户之间的用户协同兴趣状态以及用户兴趣相关度,所述用户协同兴趣状态反映多个目标用户是否组成用户协同群组,所述用户兴趣相关度反映组成所述用户协同群组的多个所述目标用户之间的兴趣相关度。Step S120, determining the user collaborative interest state and user interest correlation between the target users based on the user clothing order behavior big data, the user collaborative interest state reflects whether multiple target users form a user collaborative group, and the user interest correlation reflects the interest correlation between the multiple target users constituting the user collaborative group.
详细地,所述用户协同兴趣状态指的是通过分析用户的服装订单行为数据,判断多个目标用户之间是否存在相似的兴趣和行为模式,从而确定他们是否形成了一个用户协同群组。所述用户兴趣相关度则用于量化这个协同群组中用户之间的兴趣相似程度。例如,在分析了用户A、B、C的服装订单行为数据后,服务器发现用户A和用户B在某些特定品类(如户外运动装备)上有相似的购买记录和评价内容,因此判断他们之间存在协同兴趣,并将他们归入一个用户协同群组。同时,服务器还根据他们的购买频率、购买金额、评价一致性等因素计算出他们之间的用户兴趣相关度,比如可以是一个介于0到1之间的数值,数值越高表示兴趣相关度越高。In detail, the user collaborative interest state refers to determining whether there are similar interests and behavior patterns between multiple target users by analyzing the user's clothing order behavior data, thereby determining whether they have formed a user collaborative group. The user interest relevance is used to quantify the degree of interest similarity between users in this collaborative group. For example, after analyzing the clothing order behavior data of users A, B, and C, the server finds that users A and B have similar purchase records and evaluation content in certain specific categories (such as outdoor sports equipment), so it is determined that there is a collaborative interest between them and they are classified into a user collaborative group. At the same time, the server also calculates the user interest relevance between them based on factors such as their purchase frequency, purchase amount, and evaluation consistency. For example, it can be a value between 0 and 1. The higher the value, the higher the interest relevance.
在这个例子中,用户A和用户B的兴趣相关度可能较高,因为他们经常购买相似的户外运动装备,并且评价内容也相似。而用户C虽然可能也对时尚潮流服装感兴趣,但由于与用户A、B的购买记录和评价内容差异较大,因此他们之间的兴趣相关度可能较低。In this example, the interest correlation between user A and user B may be high because they often buy similar outdoor sports equipment and their evaluation content is similar. Although user C may also be interested in fashion clothing, the interest correlation between them may be low because of the large differences in their purchase records and evaluation content with users A and B.
也即,在本实施例中,服务器利用大数据分析技术,对用户服装订单行为大数据进行深入挖掘。通过比较不同用户的购买历史、浏览记录和评价内容,服务器能够识别出具有相似兴趣的用户群体,这些用户群体即为用户协同群组,其中的目标用户之间具有较高的兴趣相关度。That is, in this embodiment, the server uses big data analysis technology to conduct in-depth mining of user clothing order behavior big data. By comparing the purchase history, browsing history and evaluation content of different users, the server can identify user groups with similar interests. These user groups are user collaboration groups, in which the target users have a high degree of interest correlation.
例如,服务器分析识别到,尽管用户A和用户B在日常购买风格上有所不同,但他们在某些特定活动(如户外运动)时都会购买相同类型的服装。因此,服务器判断用户A和用户B在户外运动服装方面存在协同兴趣,并将他们归入一个用户协同群组。同时,根据他们的购买频率和金额,服务器计算出他们之间的用户兴趣相关度。For example, the server analysis recognizes that although user A and user B have different daily purchasing styles, they both buy the same type of clothing when engaging in certain activities (such as outdoor sports). Therefore, the server determines that user A and user B have a common interest in outdoor sports clothing and classifies them into a user collaboration group. At the same time, based on their purchase frequency and amount, the server calculates the user interest correlation between them.
步骤S130,依据所述用户协同兴趣状态以及所述用户兴趣相关度生成目标用户协同知识图谱,所述目标用户协同知识图谱由知识成员以及所述知识成员之间的知识链路组成,其中,不同知识成员对应不同目标用户,且所述知识链路反映组成所述用户协同群组的多个所述目标用户之间的所述用户兴趣相关度。Step S130, generating a target user collaborative knowledge graph based on the user collaborative interest status and the user interest relevance, wherein the target user collaborative knowledge graph is composed of knowledge members and knowledge links between the knowledge members, wherein different knowledge members correspond to different target users, and the knowledge links reflect the user interest relevance between the multiple target users constituting the user collaborative group.
本实施例中,基于上一步骤确定的用户协同兴趣状态和用户兴趣相关度,服务器开始构建目标用户协同知识图谱。在这个目标用户协同知识图谱中,每个知识成员代表一个目标用户,知识成员之间的知识链路则反映了用户之间的兴趣相关度。知识链路的粗细或颜色深浅可以用来表示兴趣相关度的强弱。In this embodiment, based on the user collaborative interest state and user interest relevance determined in the previous step, the server starts to build a target user collaborative knowledge graph. In this target user collaborative knowledge graph, each knowledge member represents a target user, and the knowledge links between knowledge members reflect the interest relevance between users. The thickness or color depth of the knowledge link can be used to indicate the strength of interest relevance.
例如,在目标用户协同知识图谱中,用户A、B和C被表示为三个知识成员。由于用户A和用户B在户外运动服装上有协同兴趣,他们之间形成了一条较粗的知识链路。而用户C与其他两位用户的兴趣相关度较低,因此与用户A和用户B之间的知识链路相对较细。For example, in the target user collaborative knowledge graph, users A, B, and C are represented as three knowledge members. Since users A and B have a collaborative interest in outdoor sportswear, a coarse knowledge link is formed between them. User C has a low interest relevance to the other two users, so the knowledge link between him and users A and B is relatively thin.
步骤S140,基于所述目标用户协同知识图谱,确定每个目标用户的协同兴趣路径,并基于所述每个目标用户的协同兴趣路径对所述每个目标用户进行信息推送。Step S140: determining a collaborative interest path of each target user based on the collaborative knowledge graph of the target users, and pushing information to each target user based on the collaborative interest path of each target user.
详细地,所述协同兴趣路径是基于目标用户协同知识图谱确定的,反映了用户之间兴趣传播和影响的具体路径,揭示了用户如何受到其他用户的影响,以及这种影响是如何在用户之间传递的。例如,假设用户A经常通过用户B了解到新的户外运动装备信息,并受到用户B的影响而购买相似的装备。那么,从用户B到用户A就形成了一条协同兴趣路径。服务器可以通过分析目标用户协同知识图谱中的知识成员和知识链路来确定这样的路径,并基于这些路径为每个用户定制个性化的信息推送策略。例如,当用户B浏览或购买新款户外运动装备时,服务器可以自动向用户A推送相关信息,以提高信息推送的准确性和用户满意度。In detail, the collaborative interest path is determined based on the collaborative knowledge graph of the target user, which reflects the specific path of interest propagation and influence between users, reveals how users are influenced by other users, and how this influence is transmitted between users. For example, suppose user A often learns about new outdoor sports equipment information through user B, and is influenced by user B to purchase similar equipment. Then, a collaborative interest path is formed from user B to user A. The server can determine such a path by analyzing the knowledge members and knowledge links in the collaborative knowledge graph of the target user, and customize personalized information push strategies for each user based on these paths. For example, when user B browses or purchases new outdoor sports equipment, the server can automatically push relevant information to user A to improve the accuracy of information push and user satisfaction.
也即,在本实施例中,服务器通过分析目标用户协同知识图谱中的知识成员和知识链路,确定每个目标用户的协同兴趣路径,这些协同兴趣路径揭示了用户之间兴趣的传播和影响关系。基于这些协同兴趣路径,服务器可以为每个用户定制个性化的信息推送策略。例如,服务器识别到,用户A经常通过用户B了解到新的户外运动服装款式。因此,当用户B浏览或购买新款户外运动服装时,服务器可以自动向用户A推送相关信息,这种基于协同兴趣路径的推送方式,大大提高了信息推送的准确性和用户满意度。That is, in this embodiment, the server determines the collaborative interest paths of each target user by analyzing the knowledge members and knowledge links in the collaborative knowledge graph of the target user. These collaborative interest paths reveal the propagation and influence relationship of interests between users. Based on these collaborative interest paths, the server can customize personalized information push strategies for each user. For example, the server recognizes that user A often learns about new outdoor sportswear styles through user B. Therefore, when user B browses or buys new outdoor sportswear, the server can automatically push relevant information to user A. This push method based on collaborative interest paths greatly improves the accuracy of information push and user satisfaction.
基于以上步骤,本申请实施例通过获取目标用户的用户服装订单行为大数据,能够深入分析用户之间的协同兴趣状态和用户兴趣相关度,从而准确地识别出具有相似购买兴趣和行为的用户群体,即用户协同群组。进一步地,依据这些分析结果生成目标用户协同知识图谱,该目标用户协同知识图谱直观展示了用户之间的兴趣联系和协同关系,为深入了解用户行为提供了有力工具。基于目标用户协同知识图谱,能够精准地确定每个目标用户的协同兴趣路径,这意味着能够清晰地描绘出用户兴趣的传播和影响路径。通过利用这些协同兴趣路径,能够为用户提供高度个性化的信息推送服务,如向用户推荐与其兴趣高度匹配的服装款式或潮流趋势。这种基于协同兴趣路径的推送方式,不仅大大提高了信息推送的准确性和有效性,还显著提升了用户体验和满意度。由此,通过生成目标用户协同知识图谱和确定协同兴趣路径,为服装电商平台提供了有力的数据支持,帮助平台实现更精准的用户画像和个性化的服务策略,从而提升了平台的竞争力和用户黏性。Based on the above steps, the embodiment of the present application can deeply analyze the collaborative interest status and user interest relevance between users by obtaining the user clothing order behavior big data of the target user, so as to accurately identify the user group with similar purchasing interests and behaviors, that is, the user collaborative group. Further, based on these analysis results, a target user collaborative knowledge graph is generated, which intuitively displays the interest connection and collaborative relationship between users, and provides a powerful tool for in-depth understanding of user behavior. Based on the target user collaborative knowledge graph, the collaborative interest path of each target user can be accurately determined, which means that the propagation and influence path of user interests can be clearly depicted. By utilizing these collaborative interest paths, highly personalized information push services can be provided to users, such as recommending clothing styles or fashion trends that are highly matched with their interests to users. This push method based on collaborative interest paths not only greatly improves the accuracy and effectiveness of information push, but also significantly improves user experience and satisfaction. Therefore, by generating a target user collaborative knowledge graph and determining a collaborative interest path, a strong data support is provided for the clothing e-commerce platform, which helps the platform to achieve more accurate user portraits and personalized service strategies, thereby improving the competitiveness and user stickiness of the platform.
在一种可能的实施方式中,步骤S130包括:In a possible implementation, step S130 includes:
步骤S131,依据所述用户协同兴趣状态,确定各个所述目标用户的协同群组数量,所述协同群组数量表示由所述目标用户组成的所述用户协同群组的数量。Step S131 : determining the number of collaborative groups of each target user according to the user collaborative interest state, where the number of collaborative groups represents the number of user collaborative groups formed by the target users.
步骤S132,依据所述协同群组数量,确定所述目标用户对应知识成员的知识成员特征,所述知识成员特征包括知识成员活跃度和知识成员影响力中的至少一种,其中,所述知识成员活跃度与所述协同群组数量为正向关联关系。Step S132, determining the knowledge member characteristics of the knowledge member corresponding to the target user according to the number of collaborative groups, wherein the knowledge member characteristics include at least one of knowledge member activity and knowledge member influence, wherein the knowledge member activity is positively correlated with the number of collaborative groups.
步骤S133,依据所述知识成员特征生成所述知识成员。Step S133: generating the knowledge member according to the knowledge member characteristics.
步骤S134,依据所述目标用户之间的所述用户兴趣相关度,确定所述目标用户对应知识成员之间的偏离度,所述偏离度与所述用户兴趣相关度为负向关联关系。Step S134, determining the deviation between the knowledge members corresponding to the target users according to the user interest correlation between the target users, wherein the deviation is negatively correlated with the user interest correlation.
步骤S135,在依据所述偏离度完成知识成员位置优化时,在组成所述用户协同群组的所述目标用户对应的知识成员之间生成所述知识链路。Step S135 , when the knowledge member position optimization is completed according to the deviation, the knowledge link is generated between the knowledge members corresponding to the target user constituting the user collaboration group.
步骤S136,生成由所述知识成员和所述知识链路组成的所述目标用户协同知识图谱。Step S136, generating the target user collaborative knowledge graph consisting of the knowledge members and the knowledge links.
本实施例中,服务器首先分析用户协同兴趣状态,这是基于之前步骤中从服装电商平台数据库中提取的用户服装订单行为大数据得出的,这些数据揭示了哪些用户之间存在相似的购物兴趣和行为。根据这些协同兴趣状态,服务器确定每个目标用户参与的协同群组数量。In this embodiment, the server first analyzes the user collaborative interest status, which is based on the user clothing order behavior big data extracted from the clothing e-commerce platform database in the previous step, which reveals which users have similar shopping interests and behaviors. Based on these collaborative interest states, the server determines the number of collaborative groups that each target user participates in.
例如,假设服务器识别出用户A参与了3个不同的协同群组,这些群组分别围绕“户外运动装备”、“商务正装”和“时尚潮流服饰”形成。用户B参与了2个协同群组:“户外运动装备”和“时尚配饰”。而用户C只参与了1个协同群组:“时尚潮流服饰”。For example, suppose the server identifies that user A is involved in three different collaborative groups, which are formed around "outdoor sports equipment", "business formal wear" and "fashion and trendy clothing". User B is involved in two collaborative groups: "outdoor sports equipment" and "fashion accessories". User C is involved in only one collaborative group: "fashion and trendy clothing".
接下来,服务器根据每个目标用户参与的协同群组数量,来确定对应知识成员的特征,这些特征主要包括知识成员活跃度和知识成员影响力。如果用户参与的协同群组多,说明其在不同兴趣领域都有涉猎,因此其知识成员活跃度会相对较高。Next, the server determines the characteristics of the corresponding knowledge members according to the number of collaborative groups each target user participates in. These characteristics mainly include knowledge member activity and knowledge member influence. If a user participates in many collaborative groups, it means that he or she has interests in different fields, so his or her knowledge member activity will be relatively high.
例如,由于用户A参与了3个协同群组,因此其知识成员活跃度被标记为“高”。用户B参与了2个群组,其活跃度被标记为“中”。而用户C只参与了1个群组,其活跃度被标记为“低”。For example, since user A participated in 3 collaborative groups, his knowledge member activity was marked as "high". User B participated in 2 groups, and his activity was marked as "medium". User C participated in only 1 group, and his activity was marked as "low".
基于上一步确定的知识成员特征,服务器开始生成知识成员。每个知识成员代表一个目标用户,并携带该用户的特征信息,如活跃度、影响力等。Based on the knowledge member characteristics determined in the previous step, the server starts to generate knowledge members. Each knowledge member represents a target user and carries the user's characteristic information, such as activity, influence, etc.
例如,服务器为用户A、B、C分别生成了三个知识成员,每个知识成员都包含了对应的活跃度标记和其他相关信息。For example, the server generates three knowledge members for users A, B, and C respectively, and each knowledge member contains a corresponding activity tag and other related information.
服务器继续分析目标用户之间的用户兴趣相关度,这是通过比较他们在服装订单行为上的相似性来得出的。基于这个相关度,服务器计算出知识成员之间的偏离度。偏离度是一个反映知识成员之间兴趣差异的指标,与用户兴趣相关度呈负向关联。The server continues to analyze the user interest relevance between target users, which is obtained by comparing the similarity of their clothing order behaviors. Based on this relevance, the server calculates the deviation between knowledge members. Deviation is an indicator that reflects the interest differences between knowledge members and is negatively correlated with user interest relevance.
例如,用户A和用户B在“户外运动装备”群组中有较高的兴趣相关度,因此他们对应的知识成员之间的偏离度较低。而用户A和用户C虽然在“时尚潮流服饰”群组中有交集,但他们的整体兴趣相关度不如A和B高,因此他们对应的知识成员之间的偏离度相对较高。For example, user A and user B have a high interest relevance in the "outdoor sports equipment" group, so the deviation between their corresponding knowledge members is low. Although user A and user C have an intersection in the "fashion trend clothing" group, their overall interest relevance is not as high as that of A and B, so the deviation between their corresponding knowledge members is relatively high.
在确定了偏离度后,服务器开始优化知识成员的位置,该优化过程考虑到了偏离度以及协同群组内的紧密程度。优化完成后,服务器在组成协同群组的目标用户对应的知识成员之间生成知识链路,这些链路直观地展示了用户之间的兴趣联系。After determining the deviation, the server starts to optimize the positions of knowledge members, which takes into account the deviation and the closeness within the collaborative group. After the optimization is completed, the server generates knowledge links between the knowledge members corresponding to the target users that make up the collaborative group. These links intuitively show the interest connection between users.
例如,在优化过程中,服务器可以调整知识成员的位置,使得具有更高兴趣相关度的知识成员在目标用户协同知识图谱中更靠近。例如,用户A和用户B对应的知识成员可能会被放置在相对较近的位置,并用较粗的知识链路连接,以反映他们之间的高兴趣相关度。For example, during the optimization process, the server can adjust the positions of knowledge members so that knowledge members with higher interest relevance are closer in the target user collaborative knowledge graph. For example, knowledge members corresponding to user A and user B may be placed relatively close and connected with coarser knowledge links to reflect the high interest relevance between them.
最后,服务器生成了由知识成员和知识链路组成的目标用户协同知识图谱,该图谱直观地展示了目标用户之间的协同关系和兴趣联系。Finally, the server generates a target user collaborative knowledge graph consisting of knowledge members and knowledge links, which intuitively displays the collaborative relationships and interest connections between target users.
例如,最终生成的目标用户协同知识图谱中,用户A、B、C分别由三个知识成员表示,他们之间通过知识链路相连,这些知识链路的粗细和颜色可能因用户之间的兴趣相关度不同而有所差异,从而提供了一个直观的视图来理解用户之间的协同兴趣关系。For example, in the target user collaborative knowledge graph that is finally generated, users A, B, and C are represented by three knowledge members respectively, and they are connected by knowledge links. The thickness and color of these knowledge links may vary depending on the interest relevance between users, thus providing an intuitive view to understand the collaborative interest relationship between users.
在一种可能的实施方式中,所述步骤S130之后,所述方法包括:In a possible implementation, after step S130, the method includes:
步骤A110,基于对所述目标用户协同知识图谱中第一目标用户的强化表达指令,对所述目标用户协同知识图谱中的第一知识成员以及第二知识成员进行强化表达,所述第一知识成员为所述第一目标用户对应的知识成员,所述第二知识成员为第二目标用户对应的知识成员,所述第二目标用户与所述第一目标用户之间具有k级兴趣联系,k为正整数。Step A110, based on the enhanced expression instruction for the first target user in the target user collaborative knowledge graph, the first knowledge member and the second knowledge member in the target user collaborative knowledge graph are enhanced. The first knowledge member is the knowledge member corresponding to the first target user, and the second knowledge member is the knowledge member corresponding to the second target user. There is a k-level interest relationship between the second target user and the first target user, where k is a positive integer.
步骤A120,对所述第一知识成员与所述第二知识成员之间的知识链路,以及所述第二知识成员之间的知识链路进行强化表达。Step A120: Strengthen the expression of the knowledge link between the first knowledge member and the second knowledge member, and the knowledge link between the second knowledge members.
本实施例中,服务器接收到一个强化表达指令,这个强化表达指令指定了要对目标用户协同知识图谱中的第一目标用户(假设为用户A)进行强化表达。根据这个强化表达指令,服务器首先识别出用户A对应的知识成员,即知识成员A。同时,服务器还会识别出与用户A具有k级兴趣联系的第二目标用户(假设为用户B)对应的知识成员,即知识成员B。In this embodiment, the server receives an enhanced expression instruction, which specifies that the first target user (assuming user A) in the target user collaborative knowledge graph should be enhanced. According to this enhanced expression instruction, the server first identifies the knowledge member corresponding to user A, that is, knowledge member A. At the same time, the server also identifies the knowledge member corresponding to the second target user (assuming user B) who has a k-level interest relationship with user A, that is, knowledge member B.
在这个场景中,k级兴趣联系可以理解为用户A和用户B之间通过k个中间用户或k条兴趣链路相连。服务器可以通过分析目标用户协同知识图谱中的知识链路来确定这种联系。In this scenario, the k-level interest connection can be understood as user A and user B being connected through k intermediate users or k interest links. The server can determine this connection by analyzing the knowledge links in the target user's collaborative knowledge graph.
一旦识别出需要强化表达的知识成员,服务器就会开始执行强化表达的操作,这可以包括增加知识成员的视觉突出度(如改变颜色、大小或形状),或者在知识成员旁边添加额外的标注或信息,以强调其重要性或特殊性。Once a knowledge member that needs to be enhanced is identified, the server begins to perform enhancement operations, which can include increasing the visual prominence of the knowledge member (such as changing the color, size, or shape), or adding additional annotations or information next to the knowledge member to emphasize its importance or uniqueness.
在完成对知识成员A和B的强化表达之后,服务器可以继续执行对这两个知识成员之间的知识链路的强化表达,这意味着服务器可以特别突出显示连接知识成员A和B的知识链路,以强调他们之间的联系。After completing the enhanced expression of knowledge members A and B, the server may continue to perform the enhanced expression of the knowledge link between the two knowledge members, which means that the server may particularly highlight the knowledge link connecting knowledge members A and B to emphasize the connection between them.
此外,如果知识成员B还与其他知识成员(假设为用户C对应的知识成员C)之间存在知识链路,服务器也会对这些链路进行强化表达,这是为了展示用户B在用户协同群组中的更广泛联系,以及这些联系如何与用户A相关联。In addition, if knowledge member B has knowledge links with other knowledge members (assuming it is knowledge member C corresponding to user C), the server will also enhance the expression of these links. This is to show the broader connections of user B in the user collaboration group and how these connections are associated with user A.
强化表达知识链路的方法可以包括使用更粗或更亮的线条来表示这些链路,或者在链路旁边添加额外的说明或标签,这样,当查看目标用户协同知识图谱时,用户可以更容易地注意到这些被强化的链路,从而更好地理解用户之间的协同兴趣和联系。Methods for strengthening the expression of knowledge links may include using thicker or brighter lines to represent these links, or adding additional descriptions or labels next to the links, so that when viewing the target user collaborative knowledge graph, users can more easily notice these strengthened links and better understand the collaborative interests and connections between users.
由此,通过强化表达特定的知识成员和知识链路,来突出显示目标用户协同知识图谱中的关键用户和联系,从而帮助用户更直观地理解用户之间的协同关系和兴趣联系。Therefore, by strengthening the expression of specific knowledge members and knowledge links, the key users and connections in the target user's collaborative knowledge graph are highlighted, helping users to more intuitively understand the collaborative relationships and interest connections between users.
在一种可能的实施方式中,所述方法还包括:基于对所述第一目标用户的选中指令,对所述第一知识成员以及所述第二知识成员以外的知识成员进行收缩优化。In a possible implementation, the method further includes: based on a selection instruction for the first target user, shrinking and optimizing the first knowledge member and knowledge members other than the second knowledge member.
其中,收缩条件下,所述第二知识成员中的目标知识成员与收缩知识成员连接,所述目标知识成员对应的所述第二目标用户与所述第一目标用户之间存在k级兴趣相关度,且所述收缩知识成员中表达与所述目标知识成员相连接且被收缩的知识成员的数量。Among them, under the contraction condition, the target knowledge member in the second knowledge member is connected to the contracted knowledge member, there is a k-level interest correlation between the second target user corresponding to the target knowledge member and the first target user, and the contracted knowledge member expresses the number of knowledge members connected to the target knowledge member and contracted.
所述方法还包括:The method further comprises:
基于对所述收缩知识成员的选中指令,呈现与所述目标知识成员存在一级兴趣相关度的第三知识成员。Based on the selection instruction of the contracted knowledge member, a third knowledge member having a primary interest relevance to the target knowledge member is presented.
本实施例中,服务器接收到了一个针对第一目标用户(例如,用户A)的选中指令,该选中指令可以是由用户通过界面操作发出的,表示用户想要更清晰地查看与用户A相关的协同关系。In this embodiment, the server receives a selection instruction for a first target user (eg, user A), and the selection instruction may be issued by the user through an interface operation, indicating that the user wants to view the collaborative relationship related to user A more clearly.
服务器在接收到这个选中指令后,会对目标用户协同知识图谱进行收缩优化。具体来说,服务器可以保持第一知识成员(代表用户A)和第二知识成员(与用户A具有直接兴趣联系的用户,比如用户B)不变,而对其他的知识成员进行收缩,这种收缩可能表现为将这些知识成员在图谱上的展示简化,或者将它们聚集到一个收缩节点中,以减少图谱的复杂性,并突出显示与用户A直接相关的用户。After receiving this selection instruction, the server will shrink and optimize the target user collaborative knowledge graph. Specifically, the server can keep the first knowledge member (representing user A) and the second knowledge member (users with direct interest connection with user A, such as user B) unchanged, and shrink other knowledge members. This shrinkage may be manifested as simplifying the display of these knowledge members on the graph, or aggregating them into a shrinking node to reduce the complexity of the graph and highlight users directly related to user A.
例如,如果原始图谱中包含了与用户A兴趣相关度较低的用户C、D、E等对应的知识成员,那么在收缩优化后,这些知识成员可能会被简化为一个收缩节点,以节省空间并突出核心关系。For example, if the original graph contains knowledge members corresponding to users C, D, E, etc. that have low relevance to the interests of user A, then after shrinkage optimization, these knowledge members may be simplified into a shrinkage node to save space and highlight the core relationship.
在收缩优化过程中,服务器可以特别注意第二知识成员(即与用户A有直接兴趣联系的用户对应的知识成员)与收缩知识成员之间的连接关系。如果某个第二知识成员(比如代表用户B的知识成员)与收缩节点中的某个知识成员(比如代表用户C的知识成员)原本存在连接关系,那么这种连接关系在收缩后仍然需要被保留并展示出来。During the shrinking optimization process, the server can pay special attention to the connection relationship between the second knowledge member (i.e., the knowledge member corresponding to the user who has a direct interest connection with user A) and the shrunken knowledge member. If a second knowledge member (such as the knowledge member representing user B) originally has a connection relationship with a knowledge member in the shrunken node (such as the knowledge member representing user C), then this connection relationship still needs to be retained and displayed after shrinking.
服务器可以通过数据分析来确定哪些连接关系是重要的,并在收缩图谱中明确展示这些关系,这样,用户就可以清晰地看到与用户A有直接兴趣联系的用户B,以及用户B与收缩节点中其他用户的关系。The server can determine which connection relationships are important through data analysis and clearly display these relationships in the contraction graph, so that users can clearly see user B who has a direct interest connection with user A, as well as the relationship between user B and other users in the contraction node.
当用户想要进一步了解收缩节点中的某个用户(比如用户C)与用户A的直接兴趣联系用户(比如用户B)之间的关系时,可以通过界面操作选中收缩节点中代表用户C的知识成员。服务器在接收到这个选中指令后,会展开收缩节点,并呈现与用户C存在一级兴趣相关度的其他知识成员(这里称之为第三知识成员)。When the user wants to learn more about the relationship between a user in the collapsed node (such as user C) and the user of direct interest to user A (such as user B), the user can select the knowledge member representing user C in the collapsed node through the interface operation. After receiving the selection instruction, the server will expand the collapsed node and present other knowledge members that have a first-level interest relevance to user C (here referred to as the third knowledge member).
这些第三知识成员可以是与用户C有直接协同兴趣关系的其他用户对应的知识成员。服务器可以通过数据分析来确定这些关系,并在图谱中清晰地展示出来,这样,用户就可以直观地看到用户C与用户B之间的关系,以及他们之间的兴趣相关度。These third knowledge members may be knowledge members corresponding to other users who have direct collaborative interest relationships with user C. The server can determine these relationships through data analysis and clearly display them in the graph, so that users can intuitively see the relationship between user C and user B, as well as the interest correlation between them.
由此,通过收缩和展开知识图谱中的特定部分,帮助用户更清晰地理解和探索目标用户之间的协同兴趣和联系。Therefore, by shrinking and expanding specific parts of the knowledge graph, it helps users to more clearly understand and explore the collaborative interests and connections between target users.
在一种可能的实施方式中,所述方法还包括:In a possible implementation, the method further includes:
步骤B110,基于对所述目标用户协同知识图谱的过滤指令,获取过滤要求,所述过滤要求包括用户特征标签、门限用户兴趣相关度以及兴趣相关层级中的至少一种。Step B110, based on the filtering instruction of the target user collaborative knowledge graph, obtain the filtering requirements, the filtering requirements including at least one of user feature labels, threshold user interest relevance and interest relevance level.
步骤B120,对所述目标用户协同知识图谱中不满足所述过滤要求的知识成员进行削弱表达。Step B120, weakening the expression of the knowledge members in the target user collaborative knowledge graph that do not meet the filtering requirements.
本实施例中,服务器接收一个针对目标用户协同知识图谱的过滤指令,该过滤指令可能来自于用户界面的操作,用户希望通过过滤来更精确地查看图谱中的特定信息。服务器在接收到过滤指令后,会解析并获取具体的过滤要求,这些过滤要求可以包括用户特征标签(如年龄、性别、地理位置等)、门限用户兴趣相关度(即用户之间兴趣相关度的最低标准)以及兴趣相关层级(表示用户之间兴趣相关性的深度或广度)中的至少一种。In this embodiment, the server receives a filtering instruction for the target user collaborative knowledge graph, which may come from the operation of the user interface, and the user hopes to view specific information in the graph more accurately through filtering. After receiving the filtering instruction, the server will parse and obtain specific filtering requirements, which may include at least one of user feature tags (such as age, gender, geographic location, etc.), threshold user interest relevance (i.e., the minimum standard for interest relevance between users), and interest relevance level (indicating the depth or breadth of interest relevance between users).
例如,用户可能想要查看年龄在25-35岁之间、与用户A的兴趣相关度高于0.8的用户群体。在这种情况下,服务器可以获取到这些具体的过滤要求。For example, the user may want to view a user group between the ages of 25 and 35 whose relevance to the interests of user A is higher than 0.8. In this case, the server can obtain these specific filtering requirements.
在获取了过滤要求后,服务器可以对目标用户协同知识图谱进行遍历,识别出那些不满足过滤要求的知识成员,这些知识成员可能代表的用户与用户A的兴趣相关度低于设定的门限值,或者它们的用户特征标签与指定的标签不匹配,或者它们不在用户设定的兴趣相关层级内。After obtaining the filtering requirements, the server can traverse the target user's collaborative knowledge graph and identify those knowledge members that do not meet the filtering requirements. These knowledge members may represent users whose interest relevance to user A is lower than the set threshold, or their user feature labels do not match the specified labels, or they are not within the interest relevance level set by the user.
对于这些不满足过滤要求的知识成员,服务器可以进行削弱表达。削弱表达的方式可以是降低这些知识成员在图谱中的视觉突出度(如减小节点大小、淡化颜色等),或者将它们从图谱中暂时隐藏起来,这样做的目的是使用户能够更清晰地看到满足过滤要求的知识成员和它们之间的关系,从而更容易地获取和分析所需的信息。For those knowledge members that do not meet the filtering requirements, the server can weaken their expression. The weakening expression can be done by reducing the visual prominence of these knowledge members in the graph (such as reducing the node size, fading the color, etc.), or temporarily hiding them from the graph. The purpose of this is to enable users to more clearly see the knowledge members that meet the filtering requirements and the relationships between them, so as to more easily obtain and analyze the required information.
例如,如果服务器发现用户B虽然与用户A有连接,但用户B的年龄超出了用户设定的范围,那么服务器就会削弱用户B在图谱中的表达,以便用户能够更专注于符合其过滤条件的其他用户。For example, if the server finds that user B is connected to user A, but user B's age is outside the range set by the user, the server will weaken user B's expression in the graph so that the user can focus more on other users who meet his filtering criteria.
在一种可能的实施方式中,所述方法还包括:In a possible implementation, the method further includes:
步骤C110,基于对所述目标用户协同知识图谱的冗余优化指令,依据所述冗余优化指令表征的冗余优化权重,确定所述目标用户协同知识图谱的优化参数量。Step C110, based on the redundant optimization instructions for the target user collaborative knowledge graph and according to the redundant optimization weights represented by the redundant optimization instructions, determine the optimization parameter amount of the target user collaborative knowledge graph.
步骤C120,依据所述优化参数量优化所述知识成员和/或所述知识链路所生成的特征向量,其中,所述知识成员可展示的信息包括用户标志、用户名以及协同群组数目,所述知识链路可展示的信息包括用户兴趣相关度。Step C120, optimizing the feature vector generated by the knowledge member and/or the knowledge link according to the optimization parameter, wherein the information that can be displayed by the knowledge member includes user logo, user name and number of collaborative groups, and the information that can be displayed by the knowledge link includes user interest relevance.
本实施例中,服务器可以接收到一个针对目标用户协同知识图谱的冗余优化指令,该冗余优化指令的目的是为了减少图谱中的冗余信息,提升图谱的效率和准确性。In this embodiment, the server may receive a redundant optimization instruction for the target user collaborative knowledge graph, the purpose of which is to reduce redundant information in the graph and improve the efficiency and accuracy of the graph.
冗余优化指令中会包含一个冗余优化权重,这个冗余优化权重指示了服务器在优化过程中应该多大程度上减少冗余。服务器可以根据这个权重来确定具体的优化参数量,这些优化参数量可以包括需要删除或合并的冗余知识成员的数量、需要简化的知识链路的数量等。The redundancy optimization instruction will include a redundancy optimization weight, which indicates to what extent the server should reduce redundancy during the optimization process. The server can determine the specific optimization parameter amount based on this weight, which may include the number of redundant knowledge members to be deleted or merged, the number of knowledge links to be simplified, etc.
例如,如果冗余优化权重设置得较高,服务器可以确定需要删除更多的冗余知识成员和简化更多的知识链路,以达到更高的优化效果。For example, if the redundancy optimization weight is set higher, the server may determine that more redundant knowledge members need to be deleted and more knowledge links need to be simplified to achieve a higher optimization effect.
在确定了优化参数量之后,服务器可以开始对知识成员和/或知识链路进行优化,该优化的过程可以包括删除冗余的知识成员、合并相似的知识成员、简化复杂的知识链路等。After determining the optimization parameter amount, the server may start to optimize the knowledge members and/or knowledge links. The optimization process may include deleting redundant knowledge members, merging similar knowledge members, simplifying complex knowledge links, etc.
在优化的过程中,服务器可以重新计算和优化知识成员和知识链路所生成的特征向量。特征向量是表示知识成员或知识链路特性的一组数值,它包含了关于用户标志、用户名、协同群组数目(对于知识成员)以及用户兴趣相关度(对于知识链路)等信息。During the optimization process, the server can recalculate and optimize the feature vectors generated by knowledge members and knowledge links. A feature vector is a set of values that represent the characteristics of a knowledge member or knowledge link. It contains information about user identifiers, user names, the number of collaborative groups (for knowledge members), and user interest relevance (for knowledge links).
例如,对于知识成员,服务器可以根据优化参数量,删除或合并一些冗余的用户,并重新计算剩余用户的特征向量,以确保它们能够更准确地反映用户的特性。对于知识链路,服务器可以简化一些复杂的链路,并重新计算链路的特征向量,以更准确地表示用户之间的兴趣相关度。For example, for knowledge members, the server can delete or merge some redundant users according to the optimization parameter amount, and recalculate the feature vectors of the remaining users to ensure that they can more accurately reflect the characteristics of the users. For knowledge links, the server can simplify some complex links and recalculate the feature vectors of the links to more accurately represent the interest relevance between users.
优化完成后,服务器可以更新目标用户协同知识图谱,使其更加简洁、高效,并能够更好地支持用户的分析和决策。After the optimization is completed, the server can update the target user collaborative knowledge graph to make it more concise and efficient, and better support user analysis and decision-making.
在一种可能的实施方式中,步骤S120可以包括:In a possible implementation, step S120 may include:
步骤S121,确定参考用户集合,所述参考用户集合由多个所述目标用户组成。Step S121: determining a reference user set, where the reference user set consists of a plurality of the target users.
步骤S122,依据所述参考用户集合中各个所述目标用户的所述用户服装订单行为大数据,确定所述参考用户集合的协同特征数据。Step S122, determining the collaborative feature data of the reference user set based on the user clothing order behavior big data of each target user in the reference user set.
步骤S123,在所述协同特征数据符合协同匹配要求时,确定所述参考用户集合中的多个所述目标用户组成所述用户协同群组。Step S123: when the collaborative feature data meets the collaborative matching requirement, determine that a plurality of the target users in the reference user set form the user collaborative group.
步骤S124,依据所述协同特征数据确定所述用户协同群组的所述用户兴趣相关度。Step S124: determining the user interest relevance of the user collaborative group according to the collaborative feature data.
本实施例中,服务器首先会从海量的用户数据中筛选出一部分用户作为目标用户,这些目标用户通常是活跃用户,即有一定数量的服装订单行为数据的用户。接下来,服务器可以根据特定的条件或算法,比如用户的购买频率、购买品类等,从这些目标用户中进一步挑选出一组用户,形成一个参考用户集合。In this embodiment, the server first selects a portion of users from the massive user data as target users. These target users are usually active users, that is, users with a certain amount of clothing order behavior data. Next, the server can further select a group of users from these target users based on specific conditions or algorithms, such as the user's purchase frequency, purchase category, etc., to form a reference user set.
例如,服务器可以选择在过去一个月内至少购买过三次服装的用户,组成一个包含100名用户的参考用户集合。For example, the server may select users who have purchased clothing at least three times in the past month to form a reference user set of 100 users.
服务器接下来会分析这个参考用户集合中每个目标用户的服装订单行为大数据,这些数据可以包括用户的购买时间、购买品类、购买价格、购买频率等。通过对这些数据的深入分析和挖掘,服务器能够提取出这些用户的共同特征,即协同特征数据。The server will then analyze the clothing order behavior big data of each target user in the reference user set, which may include the user's purchase time, purchase category, purchase price, purchase frequency, etc. Through in-depth analysis and mining of these data, the server can extract the common features of these users, namely collaborative feature data.
比如,服务器发现这100名用户中有80名用户在最近一个月内都购买了冬季保暖外套,且大部分选择在晚上下单,这就是一个显著的协同特征。For example, the server found that 80 of the 100 users had purchased warm winter coats in the past month, and most of them chose to place orders at night, which is a significant collaborative feature.
服务器可以预设一些协同匹配要求,比如协同特征数据需要达到一定的显著性水平,或者协同用户数量需要超过某个阈值等。当提取出的协同特征数据满足这些要求时,服务器可以判断这些用户之间存在协同兴趣状态。The server may preset some collaborative matching requirements, such as the collaborative feature data needs to reach a certain significance level, or the number of collaborative users needs to exceed a certain threshold, etc. When the extracted collaborative feature data meets these requirements, the server may determine that there is a collaborative interest state between these users.
以之前的例子来说,如果服务器设定的协同匹配要求是至少有70%的用户具有相同的购买行为,那么80名用户购买冬季保暖外套这一协同特征就符合了协同匹配要求。Taking the previous example, if the collaborative matching requirement set by the server is that at least 70% of users have the same purchasing behavior, then the collaborative feature of 80 users buying warm winter jackets meets the collaborative matching requirement.
一旦协同特征数据符合协同匹配要求,服务器就会将这些具有协同兴趣状态的目标用户组成一个用户协同群组。同时,服务器还会根据协同特征数据的显著性和一致性等因素,计算出这个用户协同群组的用户兴趣相关度。Once the collaborative feature data meets the collaborative matching requirements, the server will group these target users with collaborative interest status into a user collaborative group. At the same time, the server will also calculate the user interest relevance of this user collaborative group based on factors such as the significance and consistency of the collaborative feature data.
在上述例子中,那80名具有相同购买行为的用户就会被组成一个用户协同群组,而他们的用户兴趣相关度可能会根据具体的购买品类、购买时间等因素来计算,比如可以是一个介于0到1之间的数值,数值越高表示用户之间的兴趣相关度越大。In the above example, the 80 users with the same purchasing behavior will be formed into a user collaboration group, and their user interest relevance may be calculated based on factors such as the specific purchase category and purchase time. For example, it can be a value between 0 and 1. The higher the value, the greater the interest relevance between users.
这样,服务器就能够依据用户服装订单行为大数据,准确地确定出目标用户之间的协同兴趣状态以及用户兴趣相关度,为后续的个性化推荐、市场营销等提供有力的数据支持。In this way, the server can accurately determine the collaborative interest status and user interest relevance between target users based on the big data of user clothing order behavior, providing strong data support for subsequent personalized recommendations, marketing, etc.
在一种可能的实施方式中,所述协同特征数据包括协同购买率提升矢量、协同浏览率提升矢量以及用户偏好匹配度,所述协同购买率提升矢量用于表征所述参考用户集合对购买率的影响情况,所述协同浏览率提升矢量用于表征所述参考用户集合对商品浏览率的影响,所述用户偏好匹配度用于表征至少两个所述目标用户之间是否存在相似的服装偏好。In a possible implementation, the collaborative feature data includes a collaborative purchase rate improvement vector, a collaborative browsing rate improvement vector, and a user preference matching degree. The collaborative purchase rate improvement vector is used to characterize the influence of the reference user set on the purchase rate, the collaborative browsing rate improvement vector is used to characterize the influence of the reference user set on the product browsing rate, and the user preference matching degree is used to characterize whether there are similar clothing preferences between at least two of the target users.
所述方法包括:The method comprises:
在所述协同购买率提升矢量或所述协同浏览率提升矢量为负提升矢量时,确定不符合所述协同匹配要求。或,When the collaborative purchase rate improvement vector or the collaborative browsing rate improvement vector is a negative improvement vector, it is determined that the collaborative matching requirement is not met. Or,
在所述协同购买率提升矢量和所述协同浏览率提升矢量为正向提升矢量,且所述用户偏好匹配度表征存在相似偏好,且所述协同购买率提升矢量和所述协同浏览率提升矢量的显著提升矢量值大于门限值时,确定符合所述协同匹配要求。或,When the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates the existence of similar preferences, and the significant improvement vector values of the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are greater than a threshold value, it is determined that the collaborative matching requirement is met. Or,
在所述协同购买率提升矢量和所述协同浏览率提升矢量为正向提升矢量,且所述用户偏好匹配度表征存在相似偏好,但所述协同购买率提升矢量和所述协同浏览率提升矢量的显著提升矢量值小于门限值时,确定不符合所述协同匹配要求。或,When the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates the existence of similar preferences, but the significant improvement vector values of the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are less than the threshold value, it is determined that the collaborative matching requirement is not met. Or,
在所述协同购买率提升矢量和所述协同浏览率提升矢量为正向提升矢量,且所述用户偏好匹配度表征不存在相似偏好,但所述协同购买率提升矢量的显著提升矢量值大于门限值时,确定符合所述协同匹配要求。或,When the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates that there is no similar preference, but the significant improvement vector value of the collaborative purchase rate improvement vector is greater than the threshold value, it is determined that the collaborative matching requirement is met. Or,
在所述协同购买率提升矢量和所述协同浏览率提升矢量为正向提升矢量,且所述用户偏好匹配度表征不存在相似偏好,且所述协同购买率提升矢量的显著提升矢量值小于门限值时,确定不符合所述协同匹配要求。When the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates that there is no similar preference, and the significant improvement vector value of the collaborative purchase rate improvement vector is less than a threshold value, it is determined that the collaborative matching requirement is not met.
本实施例中,所述方法涉及对协同特征数据的详细分析,以确定是否符合协同匹配要求,这些协同特征数据包括协同购买率提升矢量、协同浏览率提升矢量以及用户偏好匹配度。以下是对各种情况的详细场景举例说明:In this embodiment, the method involves a detailed analysis of the collaborative feature data to determine whether the collaborative matching requirements are met. The collaborative feature data includes a collaborative purchase rate improvement vector, a collaborative browsing rate improvement vector, and a user preference matching degree. The following are detailed scenario examples for various situations:
情况一:协同购买率提升矢量或协同浏览率提升矢量为负提升矢量Case 1: The co-purchase rate lift vector or the co-browsing rate lift vector is a negative lift vector
服务器分析了一个由100名用户组成的参考用户集合。在观察期内,这个用户集合对某款新上市的连衣裙的协同购买率提升矢量为-5%,协同浏览率提升矢量为-3%,这意味着,与其他用户群体相比,这个用户集合在购买和浏览这款连衣裙的频率上都有所下降。The server analyzes a reference user set of 100 users. During the observation period, this user set had a -5% co-purchase rate lift vector and a -3% co-browsing rate lift vector for a newly launched dress, which means that this user set has decreased the frequency of purchasing and browsing this dress compared to other user groups.
由于协同购买率提升矢量和协同浏览率提升矢量均为负提升矢量,服务器确定这个用户集合不符合协同匹配要求,这可以是因为这款连衣裙并不符合这个用户集合的普遍审美或需求。Since both the co-purchase rate lift vector and the co-browsing rate lift vector are negative lift vectors, the server determines that this user set does not meet the co-matching requirements, which may be because the dress does not meet the general aesthetics or needs of this user set.
情况二:协同购买率提升矢量和协同浏览率提升矢量为正向提升矢量,且用户偏好匹配度表征存在相似偏好,且显著提升矢量值大于门限值Case 2: The collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, and the user preference matching degree indicates similar preferences, and the significant improvement vector value is greater than the threshold value.
服务器观察了另一个由200名热爱户外运动的用户组成的参考用户集合。在春季新款户外夹克上市时,这个用户集合的协同购买率提升矢量为15%,协同浏览率提升矢量为12%,且用户偏好匹配度显示这些用户普遍对户外运动装备有高度兴趣。The server observes another reference user set consisting of 200 users who love outdoor sports. When the new outdoor jackets were launched in spring, the co-purchase rate lift vector of this user set was 15%, the co-browsing rate lift vector was 12%, and the user preference matching showed that these users were generally highly interested in outdoor sports equipment.
由于协同购买率提升矢量和协同浏览率提升矢量均为正向提升矢量,且用户偏好匹配度表征存在相似偏好,同时这些显著提升矢量值超过了服务器设定的门限值(例如10%),因此服务器确定这个用户集合符合协同匹配要求,这表明新款户外夹克非常符合这个用户集合的兴趣和需求。Since the co-purchase rate lift vector and the co-browsing rate lift vector are both positive lift vectors, and the user preference matching indicates similar preferences, and these significant lift vector values exceed the threshold value set by the server (e.g., 10%), the server determines that this user set meets the co-matching requirements, indicating that the new outdoor jacket is very consistent with the interests and needs of this user set.
情况三:协同购买率提升矢量和协同浏览率提升矢量为正向提升矢量,但显著提升矢量值小于门限值Case 3: The collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, but the significant improvement vector value is less than the threshold value
对于一个由150名年轻女性用户组成的参考用户集合,服务器注意到某款新发布的时尚手提包的协同购买率提升矢量为5%,协同浏览率提升矢量为4%。尽管这些都是正向提升矢量,且用户偏好匹配度显示这些用户对时尚配饰有一定兴趣,但这些显著提升矢量值并未达到服务器的门限值(例如8%)。For a reference user set of 150 young female users, the server noticed that the co-purchase rate lift vector for a newly released fashion handbag was 5%, and the co-browsing rate lift vector was 4%. Although these are positive lift vectors and the user preference match shows that these users have some interest in fashion accessories, these significant lift vector values do not meet the server's threshold (for example, 8%).
尽管用户集合对这款手提包表现出了一定的兴趣增加,但由于显著提升矢量值小于门限值,服务器确定这个用户集合不符合协同匹配要求,这可以是因为这款手提包的设计或价格并未完全抓住这个用户集合的普遍喜好。Although the user set has shown some increased interest in this handbag, the server determines that this user set does not meet the co-matching requirements because the significant improvement vector value is less than the threshold value. This may be because the design or price of this handbag does not fully capture the general preferences of this user set.
情况四和情况五:用户偏好匹配度表征不存在相似偏好时的判断Case 4 and Case 5: User preference matching represents the judgment when there is no similar preference
这两种情况主要考虑的是,即使协同购买率提升矢量和协同浏览率提升矢量为正向提升矢量,但如果用户之间不存在相似的服装偏好,那么协同匹配要求的判断将主要依赖于协同购买率提升矢量的显著提升矢量值是否大于门限值。The main consideration for these two cases is that even if the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector are positive improvement vectors, if there are no similar clothing preferences between users, then the judgment of the collaborative matching requirements will mainly depend on whether the significant improvement vector value of the collaborative purchase rate improvement vector is greater than the threshold value.
场景举例(情况四):Scenario example (Case 4):
一个由不同年龄段和职业背景的用户组成的参考用户集合在购买某款多功能手表时表现出正向的协同购买率提升矢量(10%)和协同浏览率提升矢量(7%),但用户偏好匹配度显示他们之间并没有明显的相似服装或配饰偏好。然而,由于协同购买率提升矢量的显著提升矢量值超过了门限值(例如8%),服务器仍然确定这个用户集合符合协同匹配要求。A reference user set consisting of users of different age groups and professional backgrounds shows positive co-purchase rate lift vector (10%) and co-browsing rate lift vector (7%) when purchasing a certain multi-function watch, but the user preference matching shows that they do not have obvious similar clothing or accessories preferences. However, since the significant lift vector value of the co-purchase rate lift vector exceeds the threshold value (e.g., 8%), the server still determines that this user set meets the co-matching requirements.
场景举例(情况五):Scenario example (Case 5):
另一个多样化的用户集合在某款新款运动鞋上市时表现出正向但较小的协同购买率提升矢量(3%)和协同浏览率提升矢量(2%),且用户之间没有明显的相似偏好。由于协同购买率提升矢量的显著提升矢量值未达到门限值,服务器确定这个用户集合不符合协同匹配要求。Another diverse user set shows a positive but small co-purchase rate lift vector (3%) and co-browsing rate lift vector (2%) when a new pair of sneakers is launched, and there is no obvious similarity between users. Since the significant lift vector value of the co-purchase rate lift vector does not reach the threshold, the server determines that this user set does not meet the co-matching requirements.
在一种可能的实施方式中,步骤S124包括:依据所述协同购买率提升矢量和所述协同浏览率提升矢量,确定所述用户协同群组的所述用户兴趣相关度,所述用户兴趣相关度与所述协同购买率提升矢量为正向关联关系,所述用户兴趣相关度与所述协同浏览率提升矢量为正向关联关系。In a possible implementation, step S124 includes: determining the user interest correlation of the user collaborative group according to the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector, wherein the user interest correlation is positively correlated with the collaborative purchase rate improvement vector, and the user interest correlation is positively correlated with the collaborative browsing rate improvement vector.
本实施例中,服务器首先会收集并整理协同购买率提升矢量和协同浏览率提升矢量的数据,这些数据是通过分析用户服装订单行为大数据得出的,反映了用户协同群组在购买和浏览服装方面的整体趋势。In this embodiment, the server first collects and organizes the data of the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector. These data are obtained by analyzing the big data of user clothing order behavior and reflect the overall trend of the user collaborative group in purchasing and browsing clothing.
例如,假设服务器已经通过前面的步骤确定了一个由100名用户组成的用户协同群组,该群组中的用户在过去的一个月内,对某类户外运动服装的购买率和浏览率都有显著提升。服务器收集到了这些用户在这段时间内的购买和浏览数据,并据此计算出了协同购买率提升矢量和协同浏览率提升矢量。For example, suppose the server has determined a user collaborative group consisting of 100 users through the previous steps. The users in this group have significantly increased their purchase rate and browsing rate for a certain type of outdoor sportswear in the past month. The server collects the purchase and browsing data of these users during this period, and calculates the collaborative purchase rate increase vector and the collaborative browsing rate increase vector based on this data.
接下来,服务器可以根据协同购买率提升矢量和协同浏览率提升矢量来计算用户协同群组的用户兴趣相关度。由于用户兴趣相关度与协同购买率提升矢量和协同浏览率提升矢量均为正向关联关系,因此,当这两个矢量增加时,用户兴趣相关度也会随之增加。Next, the server can calculate the user interest relevance of the user collaborative group according to the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector. Since the user interest relevance is positively correlated with the collaborative purchase rate improvement vector and the collaborative browsing rate improvement vector, when these two vectors increase, the user interest relevance will also increase accordingly.
例如,服务器使用特定的算法(如加权求和、机器学习模型等)来结合协同购买率提升矢量和协同浏览率提升矢量,以得出用户兴趣相关度。例如,如果协同购买率提升矢量为10%,协同浏览率提升矢量为8%,服务器可以给这两个矢量分配不同的权重(如购买行为更重要,因此权重更高),然后计算出一个综合的用户兴趣相关度。For example, the server uses a specific algorithm (such as weighted sum, machine learning model, etc.) to combine the co-purchase rate lift vector and the co-browsing rate lift vector to obtain the user interest relevance. For example, if the co-purchase rate lift vector is 10% and the co-browsing rate lift vector is 8%, the server can assign different weights to the two vectors (such as purchase behavior is more important, so the weight is higher), and then calculate a comprehensive user interest relevance.
最后,服务器可以确定用户兴趣相关度的具体数值,这个数值将用于后续的推荐、广告投放等个性化服务。例如,经过计算,服务器得出该用户协同群组的用户兴趣相关度为0.85(假设用户兴趣相关度的取值范围为0到1,其中1表示兴趣完全相关),该数值表明,这个用户协同群组对户外运动服装有较高的兴趣,是后续个性化服务的重要参考。Finally, the server can determine the specific value of the user interest relevance, which will be used for subsequent personalized services such as recommendation and advertising. For example, after calculation, the server finds that the user interest relevance of the user collaborative group is 0.85 (assuming that the value range of user interest relevance is 0 to 1, where 1 indicates that the interests are completely relevant). This value indicates that this user collaborative group has a high interest in outdoor sportswear, which is an important reference for subsequent personalized services.
由此,通过收集协同特征数据、计算用户兴趣相关度并确定其具体数值,能够更准确地理解用户协同群组的兴趣偏好,从而提供更精准的个性化服务。Therefore, by collecting collaborative feature data, calculating user interest relevance and determining its specific value, we can more accurately understand the interest preferences of user collaborative groups, thereby providing more precise personalized services.
在一个详细的说明中,所述用户协同兴趣状态是通过分析用户服装订单行为大数据来确定的,主要依据的是协同特征数据。这些协同特征数据包括协同购买率提升矢量、协同浏览率提升矢量以及用户偏好匹配度。In a detailed description, the user collaborative interest state is determined by analyzing the user clothing order behavior big data, mainly based on collaborative feature data. These collaborative feature data include collaborative purchase rate improvement vector, collaborative browsing rate improvement vector and user preference matching degree.
协同购买率提升矢量:表示用户协同群组对购买率的影响。计算公式可以是:协同购买率提升矢量 = (协同群组内用户的平均购买率 - 总用户平均购买率) / 总用户平均购买率 * 100%。Collaborative purchase rate lift vector: represents the impact of user collaborative groups on purchase rate. The calculation formula can be: Collaborative purchase rate lift vector = (average purchase rate of users in the collaborative group - average purchase rate of all users) / average purchase rate of all users * 100%.
协同浏览率提升矢量:表示用户协同群组对商品浏览率的影响。计算公式可以是:协同浏览率提升矢量 = (协同群组内用户的平均浏览率 - 总用户平均浏览率) / 总用户平均浏览率 * 100%。Co-browsing rate lift vector: represents the impact of user collaborative groups on product browsing rate. The calculation formula can be: Co-browsing rate lift vector = (average browsing rate of users in the collaborative group - average browsing rate of all users) / average browsing rate of all users * 100%.
用户偏好匹配度:可以通过比较用户购买和浏览的服装品类、款式、颜色等来计算。具体算法可以是余弦相似度、皮尔逊相关系数等,用于量化用户之间的兴趣相似度。User preference matching: It can be calculated by comparing the clothing categories, styles, colors, etc. purchased and browsed by users. The specific algorithm can be cosine similarity, Pearson correlation coefficient, etc., which is used to quantify the similarity of interests between users.
综合以上三个协同特征数据,可以基于上述例子的一系列规则来判断用户协同群组是否符合协同匹配要求,从而确定用户协同兴趣状态。例如,当符合协同匹配要求时,用户协同兴趣状态对应的状态值为1,当不符合协同匹配要求时,用户协同兴趣状态对应的状态值为0。Based on the above three collaborative feature data, it is possible to determine whether the user collaborative group meets the collaborative matching requirements based on a series of rules in the above examples, thereby determining the user collaborative interest status. For example, when the collaborative matching requirements are met, the state value corresponding to the user collaborative interest status is 1, and when the collaborative matching requirements are not met, the state value corresponding to the user collaborative interest status is 0.
所述用户兴趣相关度是用于量化用户协同群组内用户对特定服装品类或款式的兴趣程度,主要依据协同购买率提升矢量和协同浏览率提升矢量来计算。The user interest correlation is used to quantify the interest of users in a user collaboration group in a specific clothing category or style, and is mainly calculated based on the collaborative purchase rate increase vector and the collaborative browsing rate increase vector.
一种可能的计算公式是:One possible calculation formula is:
用户兴趣相关度 = w1 * 协同购买率提升矢量 + w2 * 协同浏览率提升矢量User interest relevance = w1 * collaborative purchase rate improvement vector + w2 * collaborative browsing rate improvement vector
其中,w1 和 w2 是权重系数,可以根据实际情况进行调整。例如,如果认为购买行为比浏览行为更能反映用户兴趣,可以设置 w1 > w2。Among them, w1 and w2 are weight coefficients, which can be adjusted according to actual conditions. For example, if it is believed that purchasing behavior reflects user interests better than browsing behavior, w1 > w2 can be set.
这个公式将协同购买率提升矢量和协同浏览率提升矢量加权求和,得出一个综合的用户兴趣相关度指标。这个用户兴趣相关度的指标数值越高,表示用户协同群组对特定服装品类或款式的兴趣越浓厚。This formula weights and sums the co-purchase rate lift vector and the co-browsing rate lift vector to produce a comprehensive user interest relevance index. The higher the value of this user interest relevance index, the stronger the interest of the user collaborative group in a specific clothing category or style.
在一种可能的实施方式中,步骤S140包括:In a possible implementation, step S140 includes:
步骤S141,对所述目标用户协同知识图谱中的每个知识成员进行活跃度分析,生成每个知识成员的活跃度,所述活跃度根据每个知识成员对应的目标用户在服装电商平台上的交互频率、订单数量、评价次数进行衡量。Step S141, performing activity analysis on each knowledge member in the target user collaborative knowledge graph to generate the activity of each knowledge member, wherein the activity is measured based on the interaction frequency, order quantity, and evaluation times of the target user corresponding to each knowledge member on the clothing e-commerce platform.
本实施例中,服务器首先对目标用户协同知识图谱中的每个知识成员进行活跃度分析。活跃度是根据每个知识成员对应的目标用户在服装电商平台上的交互频率、订单数量、评价次数等指标进行衡量的。In this embodiment, the server first performs activity analysis on each knowledge member in the target user collaborative knowledge graph. The activity is measured based on indicators such as the interaction frequency, order quantity, and evaluation times of the target user corresponding to each knowledge member on the clothing e-commerce platform.
例如,假设知识成员A在过去一个月内交互频率高、订单数量多且评价次数频繁,那么服务器可以判断知识成员A的活跃度较高。相反,如果知识成员B在这些指标上表现平平,则其活跃度相对较低。For example, if knowledge member A has a high interaction frequency, a large number of orders, and frequent reviews in the past month, the server can determine that knowledge member A is more active. On the contrary, if knowledge member B performs mediocrely on these indicators, its activity is relatively low.
步骤S142,选择活跃度满足预设条件的知识成员作为协同兴趣路径的起始知识成员,从所述起始知识成员出发,沿着知识链路进行拓展,初步识别出与目标用户相关联的协同兴趣路径,所述协同兴趣路径由多个相互关联的知识成员和知识链路组成,反映了目标用户之间的兴趣传递和影响关系。Step S142, select knowledge members whose activity meets preset conditions as the starting knowledge members of the collaborative interest path, start from the starting knowledge members, expand along the knowledge links, and preliminarily identify the collaborative interest path associated with the target user. The collaborative interest path is composed of multiple interrelated knowledge members and knowledge links, reflecting the interest transfer and influence relationship between the target users.
本实施例中,服务器选择活跃度满足预设条件的知识成员作为协同兴趣路径的起始知识成员。从起始知识成员出发,沿着知识链路进行拓展,初步识别出与目标用户相关联的协同兴趣路径。In this embodiment, the server selects knowledge members whose activity meets preset conditions as the starting knowledge members of the collaborative interest path. Starting from the starting knowledge members, the knowledge links are expanded to preliminarily identify the collaborative interest path associated with the target user.
例如,以活跃度高的知识成员A为起始点,服务器发现A经常与知识成员C、D、E产生交互,形成了一条协同兴趣路径A-C-D-E,这条协同兴趣路径反映了目标用户之间的兴趣传递和影响关系。For example, taking the highly active knowledge member A as the starting point, the server finds that A often interacts with knowledge members C, D, and E, forming a collaborative interest path A-C-D-E. This collaborative interest path reflects the interest transfer and influence relationship between target users.
步骤S143,基于目标用户之间的兴趣相关度、购买行为的相似度和用户评价的一致性对初步识别的协同兴趣路径进行权重赋值,并根据赋值后的权重对所述协同兴趣路径进行优化和剪枝。Step S143, weights are assigned to the initially identified collaborative interest paths based on the interest relevance between target users, the similarity of purchase behaviors, and the consistency of user evaluations, and the collaborative interest paths are optimized and pruned according to the assigned weights.
本实施例中,服务器基于目标用户之间的兴趣相关度、购买行为的相似度和用户评价的一致性,对初步识别的协同兴趣路径进行权重赋值。根据赋值后的权重,对协同兴趣路径进行优化和剪枝。In this embodiment, the server assigns weights to the initially identified collaborative interest paths based on the interest relevance between target users, the similarity of purchase behaviors, and the consistency of user evaluations, and optimizes and prunes the collaborative interest paths based on the assigned weights.
例如,在路径A-C-D-E中,如果C与A的兴趣相关度很高,购买行为相似,且用户评价一致,那么C将获得较高的权重。反之,如果E在这些方面的表现较差,其权重就会相对较低。根据权重,服务器可以对路径进行剪枝,比如去除权重较低的E,优化后的路径变为A-C-D。For example, in the path A-C-D-E, if C and A have a high interest relevance, similar purchase behavior, and consistent user evaluation, then C will get a higher weight. Conversely, if E performs poorly in these aspects, its weight will be relatively low. Based on the weight, the server can prune the path, such as removing E with a lower weight, and the optimized path becomes A-C-D.
步骤S144,在优化和剪枝后的协同兴趣路径中,识别出关键知识成员,并获取每个关键知识成员在协同兴趣路径中的位置特征,所述位置特征包括每个关键知识成员处于协同兴趣路径的具体位置以及每个关键知识成员与所述协同兴趣路径中其它知识成员的路径代价关系。Step S144, in the optimized and pruned collaborative interest path, key knowledge members are identified, and the position characteristics of each key knowledge member in the collaborative interest path are obtained, wherein the position characteristics include the specific position of each key knowledge member in the collaborative interest path and the path cost relationship between each key knowledge member and other knowledge members in the collaborative interest path.
本实施例中,在优化和剪枝后的协同兴趣路径中,服务器识别出关键知识成员,并获取每个关键知识成员在协同兴趣路径中的位置特征。In this embodiment, in the collaborative interest path after optimization and pruning, the server identifies key knowledge members and obtains the position feature of each key knowledge member in the collaborative interest path.
例如,在路径A-C-D中,C可以是一个关键知识成员,因为它连接了A和D,起到了桥梁的作用。C的位置特征包括它处于路径的中间位置,以及与A和D的路径代价关系(如距离、交互次数等)。For example, in the path A-C-D, C can be a key knowledge member because it connects A and D and acts as a bridge. The positional characteristics of C include its being in the middle of the path and the path cost relationship with A and D (such as distance, number of interactions, etc.).
步骤S145,根据所述关键知识成员在所述协同兴趣路径中的用户行为和协同影响力,识别所述关键知识成员在协同兴趣路径中的角色特征,所述角色特征包括引领角色、传递角色和跟随角色中的一种。Step S145, identifying the role characteristics of the key knowledge member in the collaborative interest path according to the user behavior and collaborative influence of the key knowledge member in the collaborative interest path, wherein the role characteristics include one of a leading role, a transmitting role and a following role.
本实施例中, 服务器根据关键知识成员在协同兴趣路径中的用户行为和协同影响力,识别其在协同兴趣路径中的角色特征。In this embodiment, the server identifies the role characteristics of the key knowledge members in the collaborative interest path according to the user behaviors and collaborative influences of the key knowledge members in the collaborative interest path.
例如,如果C经常引领新的购物趋势,并影响其他用户的购买决策,那么C可能扮演引领角色。如果C主要是传递信息,将A的影响扩散到D,那么它可能扮演传递角色。如果C主要是跟随A的购物趋势,那么它可能扮演跟随角色。For example, if C often leads new shopping trends and influences other users’ purchasing decisions, then C may play a leading role. If C mainly transmits information and spreads the influence of A to D, then it may play a transmitting role. If C mainly follows A’s shopping trends, then it may play a following role.
步骤S146,根据所述每个关键知识成员在协同兴趣路径中的位置特征和角色特征,从所述每个关键知识成员与其它知识成员之间的协同行为数据中挖掘目标兴趣偏好数据,并根据所述目标兴趣偏好数据,生成与其它目标用户协同关联的个性化推送内容。Step S146, based on the position characteristics and role characteristics of each key knowledge member in the collaborative interest path, target interest preference data is mined from the collaborative behavior data between each key knowledge member and other knowledge members, and personalized push content collaboratively associated with other target users is generated based on the target interest preference data.
最后,服务器根据每个关键知识成员在协同兴趣路径中的位置特征和角色特征,从其与其它知识成员之间的协同行为数据中挖掘目标兴趣偏好数据。根据这些数据,生成与其它目标用户协同关联的个性化推送内容。Finally, the server mines the target interest preference data from the collaborative behavior data between each key knowledge member and other knowledge members according to the position characteristics and role characteristics of each key knowledge member in the collaborative interest path. Based on these data, the server generates personalized push content that is collaboratively associated with other target users.
例如,如果C扮演引领角色,并且其位置特征表明它对D有重要影响,那么服务器可以向D推送与C相似的服装款式或潮流趋势,这样的推送内容既考虑了D的个人兴趣,又利用了C的引领作用,提高了推送的准确性和有效性。For example, if C plays a leading role and its location characteristics indicate that it has a significant influence on D, the server can push clothing styles or fashion trends similar to C to D. Such push content takes into account D's personal interests and utilizes C's leading role, thereby improving the accuracy and effectiveness of the push.
图2示出了本申请实施例提供的用于实现上述的基于大数据的用户服装订单行为分析方法的基于大数据的用户服装订单行为分析系统100的硬件结构意图,如图2所示,基于大数据的用户服装订单行为分析系统100可包括处理器110、机器可读存储介质120、总线130以及通信单元140。Figure 2 shows the hardware structure intention of the big data-based user clothing order behavior analysis system 100 provided in an embodiment of the present application for implementing the above-mentioned big data-based user clothing order behavior analysis method. As shown in Figure 2, the big data-based user clothing order behavior analysis system 100 may include a processor 110, a machine-readable storage medium 120, a bus 130 and a communication unit 140.
一种可能的设计中,基于大数据的用户服装订单行为分析系统100可以是单个服务器,也可以是服务器组。所述服务器组可以是集中式的,也可以是分布式的(例如,基于大数据的用户服装订单行为分析系统100可以是分布式的系统)。在一些实施例中,基于大数据的用户服装订单行为分析系统100可以是本地的,也可以是远程的。例如,基于大数据的用户服装订单行为分析系统100可以经由网络访问存储于机器可读存储介质120中的信息和/或数据。又例如,基于大数据的用户服装订单行为分析系统100可以直接连接到机器可读存储介质120以访问存储的信息和/或数据。在一些实施例中,基于大数据的用户服装订单行为分析系统100可以在基于大数据的用户服装订单行为分析系统上实施。仅作为示例,该基于大数据的用户服装订单行为分析系统可以包括私有云、公共云、混合云、社区云、分布云、内部云、多层云等或其任意组合。In a possible design, the user clothing order behavior analysis system 100 based on big data can be a single server or a server group. The server group can be centralized or distributed (for example, the user clothing order behavior analysis system 100 based on big data can be a distributed system). In some embodiments, the user clothing order behavior analysis system 100 based on big data can be local or remote. For example, the user clothing order behavior analysis system 100 based on big data can access information and/or data stored in a machine-readable storage medium 120 via a network. For another example, the user clothing order behavior analysis system 100 based on big data can be directly connected to the machine-readable storage medium 120 to access the stored information and/or data. In some embodiments, the user clothing order behavior analysis system 100 based on big data can be implemented on a user clothing order behavior analysis system based on big data. As an example only, the user clothing order behavior analysis system based on big data can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc. or any combination thereof.
机器可读存储介质120可以存储数据和/或指令。在一些实施例中,机器可读存储介质120可以存储从外部终端获取的数据。在一些实施例中,机器可读存储介质120可以存储基于大数据的用户服装订单行为分析系统100用来执行或使用来完成本申请中描述的示例性方法的数据及/或指令。The machine-readable storage medium 120 may store data and/or instructions. In some embodiments, the machine-readable storage medium 120 may store data acquired from an external terminal. In some embodiments, the machine-readable storage medium 120 may store data and/or instructions that the user clothing order behavior analysis system 100 based on big data uses to execute or use to complete the exemplary method described in this application.
在具体实现过程中,一个或多个处理器110执行机器可读存储介质120存储的计算机可执行指令,使得处理器110可以执行如上方法实施例的基于大数据的用户服装订单行为分析方法,处理器110、机器可读存储介质120以及通信单元140通过总线130连接,处理器110可以用于控制通信单元140的收发动作。In the specific implementation process, one or more processors 110 execute computer executable instructions stored in the machine-readable storage medium 120, so that the processor 110 can execute the user clothing order behavior analysis method based on big data in the above method embodiment. The processor 110, the machine-readable storage medium 120 and the communication unit 140 are connected through the bus 130, and the processor 110 can be used to control the sending and receiving actions of the communication unit 140.
处理器110的具体实现过程可参见上述基于大数据的用户服装订单行为分析系统100执行的各个方法实施例,其实现原理和技术效果类似,本实施例此处不再赘述。The specific implementation process of the processor 110 can refer to the various method embodiments executed by the above-mentioned big data-based user clothing order behavior analysis system 100. The implementation principles and technical effects are similar, and this embodiment will not be repeated here.
此外,本申请实施例还提供一种可读存储介质,所述可读存储介质中预设有计算机可执行指令,当处理器执行所述计算机可执行指令时,实现如上基于大数据的用户服装订单行为分析方法。In addition, an embodiment of the present application also provides a readable storage medium, in which computer executable instructions are preset. When a processor executes the computer executable instructions, the above-mentioned user clothing order behavior analysis method based on big data is implemented.
应当注意的是,为了简化本申请披露的表述,从而帮助对一个或以上发明实施例的理解,前文对本申请实施例的描述中,有时会将多种特征归并至一个实施例、附图或对其的描述中。It should be noted that in order to simplify the description disclosed in the present application and thereby facilitate the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of the present application, various features are sometimes grouped into one embodiment, drawing or description thereof.
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410750730.7A CN118350904B (en) | 2024-06-12 | 2024-06-12 | User clothing order behavior analysis method and system based on big data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410750730.7A CN118350904B (en) | 2024-06-12 | 2024-06-12 | User clothing order behavior analysis method and system based on big data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN118350904A true CN118350904A (en) | 2024-07-16 |
CN118350904B CN118350904B (en) | 2024-09-17 |
Family
ID=91814216
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410750730.7A Active CN118350904B (en) | 2024-06-12 | 2024-06-12 | User clothing order behavior analysis method and system based on big data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN118350904B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN119106099A (en) * | 2024-08-23 | 2024-12-10 | 正大康地农牧集团有限公司 | Aquaculture feedback big data mining method and system based on artificial intelligence |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030105682A1 (en) * | 1998-09-18 | 2003-06-05 | Dicker Russell A. | User interface and methods for recommending items to users |
CN107633430A (en) * | 2017-09-20 | 2018-01-26 | 哈尔滨工业大学 | A kind of Method of Commodity Recommendation based on community of colony |
CN114511353A (en) * | 2022-01-28 | 2022-05-17 | 中国工商银行股份有限公司 | Data analysis method and apparatus |
CN116402569A (en) * | 2023-03-09 | 2023-07-07 | 浙江创邻科技有限公司 | Commodity recommendation method, device and system based on knowledge graph and storage medium |
CN118071400A (en) * | 2024-01-26 | 2024-05-24 | 世纪恒通科技股份有限公司 | Application method and system based on graph computing technology in information consumption field |
-
2024
- 2024-06-12 CN CN202410750730.7A patent/CN118350904B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030105682A1 (en) * | 1998-09-18 | 2003-06-05 | Dicker Russell A. | User interface and methods for recommending items to users |
CN107633430A (en) * | 2017-09-20 | 2018-01-26 | 哈尔滨工业大学 | A kind of Method of Commodity Recommendation based on community of colony |
CN114511353A (en) * | 2022-01-28 | 2022-05-17 | 中国工商银行股份有限公司 | Data analysis method and apparatus |
CN116402569A (en) * | 2023-03-09 | 2023-07-07 | 浙江创邻科技有限公司 | Commodity recommendation method, device and system based on knowledge graph and storage medium |
CN118071400A (en) * | 2024-01-26 | 2024-05-24 | 世纪恒通科技股份有限公司 | Application method and system based on graph computing technology in information consumption field |
Non-Patent Citations (1)
Title |
---|
程芳;: "农业电子商务中基于用户兴趣变化的协同过滤推荐技术研究", 农业网络信息, no. 05, 26 May 2016 (2016-05-26) * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN119106099A (en) * | 2024-08-23 | 2024-12-10 | 正大康地农牧集团有限公司 | Aquaculture feedback big data mining method and system based on artificial intelligence |
CN119106099B (en) * | 2024-08-23 | 2025-08-01 | 正大康地农牧集团有限公司 | Artificial intelligence-based cultivation feedback big data mining method and system |
Also Published As
Publication number | Publication date |
---|---|
CN118350904B (en) | 2024-09-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11928428B2 (en) | Emoji understanding in online experiences | |
CN102708130B (en) | Calculate the easily extensible engine that fine point of user is mated for offer | |
CN103729359B (en) | A kind of method and system recommending search word | |
US10956928B2 (en) | Cognitive fashion product advertisement system and method | |
US11386301B2 (en) | Cluster and image-based feedback system | |
CN116762089A (en) | Product recommendation system and method based on multi-modal fashion knowledge graph | |
CN113032668A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
CN118350904B (en) | User clothing order behavior analysis method and system based on big data | |
KR20220001616A (en) | Method, Apparatus and System for Constructing Bigdata Based on Generating United Identifier of Customer | |
CN118822700B (en) | A commodity search method based on big data and electronic shopping mall | |
CN115272722A (en) | Object detection method, commodity detection method and similarity prediction model training method | |
KR20210131198A (en) | Method, apparatus and computer program for advertising recommended product | |
WO2023236742A9 (en) | Search result display method and apparatus, computer device, and storage medium | |
CN117522504A (en) | Commodity recommendation method and equipment | |
CN116523598A (en) | Medical instrument recommendation method, system and storage medium | |
CN119719315A (en) | Intelligent customer service session processing method, device and program product | |
Yeruva et al. | Apparel recommendation system using content-based filtering | |
CN118864008B (en) | A personalized advertising push method based on product data and related device | |
Zelenetska et al. | SEO-Optimization of Product Content on a Marketplace Platform | |
CN118379118A (en) | Personalized recommendation method and system for jackets based on big data | |
CN117370669A (en) | Search hotword determining method and device and electronic equipment | |
KR102576123B1 (en) | Product marketing linkage system and method | |
CN110020136A (en) | Object recommendation method and relevant device | |
CN116308687B (en) | Commodity information recommendation method based on knowledge graph, electronic equipment and storage medium | |
US20250238819A1 (en) | Systems and methods for automated content creation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP03 | Change of name, title or address | ||
CP03 | Change of name, title or address |
Address after: No. 8333, Yifangfu South Road, Yunmenshan Street, Qingzhou City, Weifang City, Shandong Province, China 262500 Patentee after: Tambor Group Co.,Ltd. Country or region after: China Address before: 262500 No. 3588, Zhaode South Road, Qingzhou City, Weifang City, Shandong Province Patentee before: QINGZHOU CITY TANBOER CLOTHING Corp.,Ltd. Country or region before: China |