WO2019072007A1 - 一种数据处理方法及装置 - Google Patents

一种数据处理方法及装置 Download PDF

Info

Publication number
WO2019072007A1
WO2019072007A1 PCT/CN2018/099542 CN2018099542W WO2019072007A1 WO 2019072007 A1 WO2019072007 A1 WO 2019072007A1 CN 2018099542 W CN2018099542 W CN 2018099542W WO 2019072007 A1 WO2019072007 A1 WO 2019072007A1
Authority
WO
WIPO (PCT)
Prior art keywords
search result
user
result item
search
item
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2018/099542
Other languages
English (en)
French (fr)
Inventor
王彦锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Publication of WO2019072007A1 publication Critical patent/WO2019072007A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24Querying
    • G06F16/245Query processing
    • G06F16/2457Query processing with adaptation to user needs

Definitions

  • the present application relates to the field of data processing technologies, and in particular, to a data processing method and apparatus.
  • the data query service can obtain relevant data from the platform according to the search term input by the user, and display the related data to the user.
  • the specific related data may include, for example, documents, product information, and the like.
  • the user data obtained by the above method can only obtain some data information that is literally related to the search term, but cannot satisfy the user's actual search appeal. Therefore, there is a need in the prior art for a data processing method that matches the actual search needs of the user.
  • the purpose of the embodiment of the present application is to provide a data processing method and device, which can make the search result item recommended to the user more in line with the user's needs and the user's identity information.
  • a data processing method comprising: acquiring identity information of a user, a search term, and a set of search result items matching the search term; acquiring historical access data of the search result item set by the user, and Historical access data of the search result item set by a plurality of other users matching the identity information of the user; according to historical access data of the user and historical access data of the plurality of other users At least one search result item recommended to the user is extracted from the set of search result items.
  • a data processing device comprising: a result item obtaining unit, configured to acquire identity information of a user, a search term, and a set of search result items matching the search term; a historical data obtaining unit, configured to acquire the Historical access data of the search result item set by the user, and historical access data of the search result item set by a plurality of other users matching the identity information of the user; a recommendation result item extracting unit, configured to: And extracting at least one search result item recommended to the user from the search result item set according to historical access data of the user and historical access data of the plurality of other users.
  • a data processing apparatus includes a processor and a memory for storing processor-executable instructions, the processor executing the instructions to: obtain user identity information, a search term, and a search that matches the search term a result item set; obtaining historical access data of the user for the search result item set, and historical access data of the plurality of other users matching the search result item set that matches the user's identity information; The historical access data of the user and the historical access data of the plurality of other users, and extracting at least one search result item recommended to the user from the set of search result items.
  • the data processing method and apparatus provided by the present application may be matched according to a search term of a user to a set of search result items, and then according to historical access data of the set of search result items by the user, and multiple matching with the identity information of the user.
  • the other user searches for the historical access data of the search result item set, and extracts at least one search result item recommended to the user from the search result item set.
  • the user and the historical data of the plurality of users that match the user identity information have an influence on the search result item recommended to the user, so that the search result item recommended to the user is more in line with the user's needs and the user's identity information.
  • the relevant data that users search within the organizational structure is more in line with the user's search needs and the identity of the user within the organizational structure.
  • FIG. 1 is a schematic flow chart of an embodiment of a data processing method provided by the present application.
  • FIG. 2 is a schematic flow chart of a method for extracting a search result item according to an embodiment of the present application
  • FIG. 3 is a flowchart of a method for acquiring an frequent item method according to an embodiment of the present application
  • FIG. 5 is a flowchart of a method of an embodiment of a method for recommending search result items provided by the present application
  • FIG. 6 is a schematic structural diagram of a module of an embodiment of a data processing apparatus provided by the present application.
  • FIG. 7 is a schematic block diagram showing another embodiment of a data processing apparatus provided by the present application.
  • FIG. 1 is a schematic flow chart of a method of an embodiment of a data processing method provided by the present application.
  • the present application provides method operational steps as illustrated in the following embodiments or figures, more or fewer operational steps may be included in the method based on routine or no inventive effort.
  • the order of execution of the steps is not limited to the execution order provided by the embodiment of the present application.
  • the method may be sequentially executed in parallel or in parallel (for example, a parallel processor or a multi-threaded environment) in the actual data processing.
  • the method may include:
  • S101 Acquire user identity information, a search term, and a set of search result items that match the search term.
  • the identity information of the user may include identity information of the user in an organizational structure.
  • the organizational structure may include, for example, an enterprise, a school, a library, etc., having an internal local area network in which the user can search for data provided by the organizational structure through search terms.
  • the identity information of the user may include the department, position (such as a manager, a developer, a clerk, etc.) of the user.
  • the user's identity information can include the college, position (professor, lecturer, counselor, doctoral, master, undergraduate, etc.) of the user.
  • the library it can have a similar identity information to the school, and will not go into details here.
  • the identity information may be obtained according to the identity information of the user.
  • the identity information that matches the identification information may be obtained according to the identification information of the user, such as a login account, for example, the identification information may be related to the identity information. Binding, after determining the identification information of the user, determining the identity information of the user according to the identification information.
  • the organization structure is not limited to having an internal local area network, and may also be operated in a non-local area network, which is not limited in this application.
  • a user may search for data within an organizational structure. Specifically, a user may input search terms such as a "convolutional neural network", a “medical VR”, and the like in a local area network of the organizational structure. After the search term is obtained, a set of search result items matching the search term may be acquired.
  • the search result item may include various data types, such as documents, pictures, videos, information, logs, and the like.
  • the search result item set may include one or more of the above data types.
  • the matching of the search result item with the search term may include that the search term may include the search term.
  • the searched document may include the search term, such as The search term appears in one or more places such as the title, abstract, body, and conclusion of the document.
  • the matching of the search result item with the search term may further include: the search result item includes a vocabulary having the same meaning or similar meaning as the search term. The application is not restricted here.
  • S103 Acquire historical access data of the search result item set by the user, and historical access data of the search result item set by a plurality of other users that match the identity information of the user.
  • historical access data of the search result item set by the user and historical access data of the search result item set by a plurality of other users matching the identity information of the user may also be acquired.
  • a user log of a user may be obtained from a database of the organization structure, and historical access data of the user to the set of search result items is obtained from the user log.
  • the other users that match the identity information of the user may include multiple other users that are the same as the identity information of the user, or are at the same level as the identity information of the user. Other users.
  • the historical access data may include a plurality of data such as an access frequency, an access time, and the like of the search result item by the user and the plurality of other users.
  • a total of 200 electronic documents related to the convolutional neural network are searched according to the search word "convolutional neural network" of the user A, and according to the user log of the background database, the user A can be obtained separately.
  • 200 access records of electronic documents such as the number of visits to various electronic documents.
  • the access records of the plurality of R&D personnel to the 200 electronic documents may be acquired.
  • S105 Extract at least one search result item recommended to the user from the set of search result items according to historical access data of the user and historical access data of the plurality of other users.
  • the historical access data may be obtained according to the user and the plurality of other users.
  • At least one search result item recommended to the user is extracted from the search result item set.
  • the extracting recommended information from the search result item set to the user according to the historical access data of the user and the historical access data of the plurality of other users can include:
  • S201 According to the historical access data of the user and the plurality of other users, respectively, the frequency of accessing the search result item by the user and the plurality of other users respectively.
  • S203 Extract a frequent item set corresponding to the identity information from the search result item set according to the access frequency.
  • S205 Extract, according to the historical access data of the user and the frequent item set, at least one search result item recommended to the user from the search result item set.
  • the frequency of access by the user and the plurality of other users to the search result items in the search result item set may be respectively counted according to historical access data of the user and the plurality of other users.
  • the access frequency of the user A to the document 3 is 3, that is, the user A has accessed the document 3 three times in total
  • the access frequency of the user B to the document 80 is 10, that is, the user B has accessed the document 10 times in total.
  • the frequent item set corresponding to the identity information may be extracted from the search result item according to the access frequency.
  • the frequent item set may include a set of multiple search result items that are accessed by the identity information and are frequently accessed.
  • the extracting the frequent item set corresponding to the identity information from the search result item set according to the access frequency may include:
  • S301 The total access frequency of the search result items in the search result item set is separately calculated according to the access frequency.
  • S303 Calculate a support value of the search result item according to a total access frequency of the search result item.
  • S305 Calculate a total support value of the search result item set according to the support value of the search result item.
  • S307 The set of search result items in the search result item set whose support value is not less than the total support value is used as the frequent item set corresponding to the identity information.
  • the total access frequency of each search result item may be counted according to the access frequency of the search result item by the user and the plurality of other users.
  • the total access frequency of the user A and other multiple developers to the document 3 can be statistically obtained, for example, according to statistics, the total access to the document 3 The frequency is 40 times.
  • the support value of each search result item may be calculated, and the support value may include the interest heat information of each search result item.
  • the identity information may further include multiple identity levels.
  • the identifier information is extracted from the search result item set according to the access frequency.
  • the corresponding frequent item set may include:
  • S401 Extract, according to the access frequency, a frequent item set corresponding to the multiple identity levels, respectively, from the search result item set.
  • the identity level may include vertical user level partitioning within the organizational structure. For example, for a typical enterprise, it can be divided into multiple departments, each with its own job level, such as decision-making, management, and common user layers. That is to say, for a user in the enterprise, the identity information of the user may include multiple identity level information.
  • the user A in the above example may include ⁇ R&D department, general R&D personnel ⁇ , that is, the identity information of User A includes at least two levels of identity.
  • the frequent item set corresponding to the multiple identity levels may be separately extracted from the search result item set according to the access frequency.
  • the intersection of the frequent items corresponding to the multiple identity levels may be used as the frequent item set corresponding to the identity information.
  • the frequent item set C t recommended to the search department user can be calculated.
  • the support degree of the total item D t in the click item can be greater than or equal to E(S t ).
  • the extracting the at least one search result item recommended to the user from the set of search result items may include:
  • S501 Extract at least one search participle from the search term
  • S503 Calculate, respectively, a proximity value of the plurality of search areas in the search result item by the at least one search participle;
  • S505 Calculate, according to the proximity value and the weight value of the multiple search areas, a proximity value of the search term corresponding to the search result item;
  • S507 Extracting a recommendation from the search result item set according to historical access data of the user, historical access data of the plurality of other users, and a proximity value of the search term corresponding to the search result item. At least one search result item of the user.
  • the search term may include one or more independent word segments.
  • at least one search segmentation word may be extracted from the search word.
  • the search may be extracted from the search word.
  • the proximity values of the plurality of search regions of the at least one search segment in the search result items may be separately calculated. Wherein, the greater the proximity value indicates that the greater the degree of association between the search segmentation and the search area, the more likely it is to meet the user's desired result.
  • the search area of the search result item may include various components of the search result item, for example, when the search result item is a document, the search area may include a title, a summary, a body, a conclusion, and the like of the document. Different components.
  • the proximity value corresponding to the search result item may be separately calculated.
  • the proximity values corresponding to the plurality of search particials are merged into the proximity values of the search terms corresponding to the search result items.
  • the historical access data of the user, the historical access data of the plurality of other users, and the search term may correspond to the search.
  • the historical access data according to the history of the user, the historical access data of the plurality of other users, and the proximity value of the search term corresponding to the search result item, Extracting at least one search result item recommended to the user from the set of search result items may include:
  • SS1 Calculate, according to historical access data of the user, historical access data of the plurality of other users, a personalized weight value corresponding to the search result item.
  • SS3 Calculate a score value of the search result item according to the proximity value of the search term corresponding to the search result item and the personalized weight value of the user corresponding to the search result item.
  • SS5 recommend, to the user, a search result item in which the score value in the search result item set is greater than a preset threshold.
  • the personalized weight value corresponding to each search result item may be separately calculated according to historical access data of the user and the plurality of other users. And calculating a score value of the search result item according to the proximity value of the search term corresponding to the search result item and the personalized weight value of the user corresponding to the search result item. Then, the search result item in the set of search result items whose score value is greater than a preset threshold is recommended to the user.
  • the search participle ⁇ t i ⁇ is obtained by Chinese word segmentation (where t i represents the i th search term in the search term q), and t i ⁇ q, the position of the search participle t i in a certain search area f is recorded as Pos(t i , f), the position of the search participle t i+1 is recorded as Pos(t i+1 , f), and the search participle t
  • the minimum distance between i and t i+1 is denoted as Dis min (t i , f), and the proximity value is denoted as prox(t i , t i+1 , f), then
  • the search result item d contains descriptions (such as title, content, path, etc.) of the plurality of search areas, and f ⁇ d, the weight of the search area is w(f), and the search term q corresponds to the search result item d
  • the proximity value is recorded as Prox(q,d), then,
  • the search term input by the user can be used, and the proximity value is used to influence the search result related to the user demand, and the search result can be accurately returned. Matching search results.
  • the complexity of the algorithm is not increased, and the faster search speed is guaranteed.
  • F(h j ) represents the user's click on the h j item
  • F'(h j ) and F'(v ij ) respectively represent the hits of the recommended h j , v ij items.
  • the ranking results of the search presentation are more in line with the characteristics of individualization.
  • the weight of the recommended item can be calculated offline, and the weights of the click item and the recommended item are regularly updated without affecting the retrieval speed.
  • the score sorting result fusion is to first calculate the score value of the search result item according to the proximity value of the query content and the keyword weight, and then combine the personalized weights of the result search result item to obtain the final sort result.
  • the specific method is as follows:
  • the result of the search result item is divided into two parts.
  • One is the weighted score Scorer(q,d) based on the Boolean and vector space models.
  • the calculation formula can be expressed as
  • f T (t) represents the word frequency of the query item t appearing in a certain search result item d
  • f d (t) represents the number of search result items (or the proportion of the total search result item) containing the query item t.
  • the score data ExtendScorer(q,d) is updated based on the weight of the result search result item.
  • the final result item set D is known.
  • the weight of the click item W(q, d) affects the score of the set D search result item.
  • the original item weight value is 1, and the fusion result is obtained by the formula.
  • the score is unchanged.
  • the score thus calculated can reflect the user's behavior characteristics, and also recommend the content of other users' attention to the user, which improves the user's experience and improves the user's personalized ranking result.
  • the data processing method provided by the present application may be matched to a search result item set according to a user's search term, and then according to historical access data of the user for the search result item set, and a plurality of other users matching the user identity information. And extracting, from the search result item set, at least one search result item recommended to the user for the historical access data of the search result item set.
  • the user and the historical data of the plurality of users that match the user identity information have an influence on the search result item recommended to the user, so that the search result item recommended to the user is more in line with the user's needs and the user's identity information.
  • the relevant data that users search within the organizational structure is more in line with the user's search needs and the identity of the user within the organizational structure.
  • FIG. 6 is a block diagram showing a structure of an embodiment of a data processing device provided by the present application. As shown in FIG. 6, the device 60 may include:
  • a result item obtaining unit 61 configured to acquire identity information of the user, a search term, and a set of search result items that match the search term;
  • the historical data obtaining unit 63 is configured to acquire historical access data of the search result item set by the user, and history of the search result item set by a plurality of other users that match the identity information of the user Access data;
  • a recommendation result item extracting unit 65 configured to extract, from the search result item set, at least one search result item recommended to the user according to historical access data of the user and historical access data of the plurality of other users .
  • the recommendation result item extracting unit 65 includes:
  • An access frequency statistics unit configured to separately count, according to the historical access data of the user and the plurality of other users, the access frequency of the user and the plurality of other users respectively to the search result item;
  • a frequent item set extracting unit configured to extract a frequent item set corresponding to the identity information from the search result item set according to the access frequency
  • a result item extracting unit configured to extract, from the search result item set, at least one search result item recommended to the user according to the historical access data of the user and the frequent item set.
  • the frequent item set extracting unit may include:
  • a total access frequency statistics unit configured to separately count, according to the access frequency, a total access frequency of the search result items in the search result item set;
  • a support degree calculation unit configured to calculate a support value of the search result item according to a total access frequency of the search result item
  • a total support calculation unit configured to calculate a total support value of the search result item set according to the support value of the search result item
  • a frequent item set extraction subunit configured to use, as the frequent item set corresponding to the identity information, a set of search result items whose support value is not less than the total support value in the search result item set.
  • the frequent item set extraction unit may include:
  • a plurality of frequent item set extracting units configured to respectively extract, according to the access frequency, the frequent item sets corresponding to the plurality of identity levels respectively;
  • a frequent item intersection calculation unit configured to use an intersection of the frequent item sets corresponding to the multiple identity levels as a frequent item set corresponding to the identity information.
  • the recommendation result item extracting unit may include:
  • a participle extraction unit configured to extract at least one search participle from the search term
  • a first proximity value calculation unit configured to separately calculate proximity values of the plurality of search areas of the at least one search part in the search result item
  • a second proximity value calculation unit configured to separately calculate, according to the proximity value and the weight value of the plurality of search regions, a proximity value corresponding to the search result item
  • a result item extracting subunit configured to obtain, according to historical access data of the user, historical access data of the plurality of other users, and a proximity value of the search term corresponding to the search result item, from the search result item At least one search result item recommended to the user is extracted from the collection.
  • the result item extraction subunit may include:
  • a weight value calculation unit configured to calculate a personalized weight value corresponding to the search result item by the user according to historical access data of the user and historical access data of the plurality of other users;
  • a score value calculation unit configured to calculate a score value of the search result item according to a proximity value corresponding to the search result item of the search term and a personalized weight value corresponding to the search result item by the user;
  • a result item recommendation unit configured to recommend, to the user, a search result item in which the score value in the search result item set is greater than a preset threshold.
  • FIG. 7 is a block diagram showing a structure of an embodiment of a data processing apparatus provided by the present application.
  • the apparatus 70 may include a processor and A memory storing instructions executable by the processor, the processor executing the instructions to:
  • PLD Programmable Logic Device
  • FPGA Field Programmable Gate Array
  • HDL Hardware Description Language
  • the controller can be implemented in any suitable manner, for example, the controller can take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (eg, software or firmware) executable by the (micro)processor.
  • computer readable program code eg, software or firmware
  • examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, The Microchip PIC18F26K20 and the Silicone Labs C8051F320, the memory controller can also be implemented as part of the memory's control logic.
  • the controller can be logically programmed by means of logic gates, switches, ASICs, programmable logic controllers, and embedding.
  • Such a controller can therefore be considered a hardware component, and the means for implementing various functions included therein can also be considered as a structure within the hardware component.
  • a device for implementing various functions can be considered as a software module that can be both a method of implementation and a structure within a hardware component.
  • the unit illustrated in the above embodiment may be implemented by a computer chip or an entity, or by a product having a certain function.
  • a typical implementation device is a computer.
  • the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or A combination of any of these devices.
  • embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.
  • a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • the memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory.
  • RAM random access memory
  • ROM read only memory
  • Memory is an example of a computer readable medium.
  • Computer readable media including both permanent and non-persistent, removable and non-removable media, can be stored by any method or technology.
  • the information can be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transportable media can be used to store information that can be accessed by a computing device.
  • computer readable media does not include temporary storage of computer readable media, such as modulated data signals and carrier waves.
  • embodiments of the present application can be provided as a method, system, or computer program product.
  • the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment in combination of software and hardware.
  • the application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • the application can be described in the general context of computer-executable instructions executed by a computer, such as a program module.
  • program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types.
  • the present application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected through a communication network.
  • program modules can be located in both local and remote computer storage media including storage devices.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种数据处理方法及装置。所述方法包括:获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合(S101);获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据(S103);根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项(S105)。利用本方案,可以使得推荐给用户的搜索结果项更加符合用户的需求以及用户的身份信息。

Description

一种数据处理方法及装置 技术领域
本申请涉及数据处理技术领域,特别涉及一种数据处理方法及装置。
背景技术
近几年,基于互联网平台提供的服务越来越多,其中,数据查询服务已经在各个行业中广泛应用。数据查询服务可以根据用户输入的搜索词从平台上获取相关数据,并将相关数据展示给用户,具体的相关数据例如可以包括文档、产品信息等数据。
在一些特定行业内,如互联网行业,随着企业不断的发展,企业内信息系统沉淀积累的数据信息较多。由于企业内部数据信息的保密性和不可公开性,诸多信息一般不允许对外部搜索引擎开放,仅允许在组织架构的局域网内访问。企业内部积累的数据信息对于企业用户来说,是具有价值的数据资料,对推动企业内部知识转移和学习具有重要意义。但是,目前很多行业内组织架构复杂,组织架构内用户在根据搜索词搜索企业内数据时,企业内组织架构往往仅根据搜索词匹配到与所述搜索词相似的内部数据,并将该数据展示给用户。
利用上述方式获取的用户数据往往只能获取一些字面上与搜索词相关的数据信息,但是却无法满足用户的实际搜索诉求。因此,现有技术中亟需一种与用户实际的搜索需求相匹配的数据处理方式。
发明内容
本申请实施例的目的在于提供一种数据处理方法及装置,可以使得推荐给用户的搜索结果项更加符合用户的需求以及用户的身份信息。
本申请实施例提供的一种数据处理方法及装置具体是这样实现的:
一种数据处理方法,所述方法包括:获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合;获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据;根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
一种数据处理装置,所述装置包括:结果项获取单元,用于获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合;历史数据获取单元,用于获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据;推荐结果项提取单元,用于根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
一种数据处理装置,包括处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时实现:获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合;获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据;根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
本申请提供的数据处理方法及装置,可以根据用户的搜索词匹配到搜索结果项集合,再根据用户对所述搜索结果项集合的历史访问数据,以及与所述用户身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。利用用户以及与用户身份信息相匹配的多个用户的历史数据,对推荐给用户的搜索结果项产生影响,使得推荐给用户的搜索结果项更加符合用户的需求以及用户的身份信息。尤其对于处于组织架构内的用户来说,用户在组织架构内搜索到的相关数据更加符合用户的搜索需求以及用户在组织架构内的身份。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请中记载的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是本申请提供的数据处理方法的一种实施例的流程示意图;
图2是本申请提供的提取搜索结果项方法的一种实施例的方法流程示意图;
图3是本申请提供的获取频繁项方法的一种实施例的方法流程图;
图4是本申请提供的获取频繁项方法的另一种实施例的方法流程图;
图5是本申请提供的推荐搜索结果项方法的一种实施例的方法流程图;
图6是本申请提供的数据处理装置的一种实施例的模块结构示意图;
图7是本申请提供的数据处理装置的另一种实施例的模块结构示意图。
具体实施方式
为了使本技术领域的人员更好地理解本申请中的技术方案,下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都应当属于本申请保护的范围。
下面结合附图对本申请所述的数据处理方法进行详细的说明。图1是本申请提供的数据处理方法的一种实施例的方法流程示意图。虽然本申请提供了如下述实施例或附图所示的方法操作步骤,但基于常规或者无需创造性的劳动在所述方法中可以包括更多或者更少的操作步骤。在逻辑性上不存在必要因果关系的步骤中,这些步骤的执行顺序不限于本申请实施例提供的执行顺序。所述方法在实际中的数据处理过程中,可以按照实施例或者附图所示的方法顺序执行或者并行执行(例如并行处理器或者多线程处理的环境)。
具体的本申请提供的一种数据处理方法的一种实施例如图1所示,所述方法可以包括:
S101:获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合。
本说明书中,所述用户的身份信息可以包括所述用户在组织架构中的身份信息。所述组织架构例如可以包括企业、学校、图书馆等具有内部局域网的机构,在所述组织结构中,用户可以通过搜索词搜索组织架构提供的数据。例如,对于企业来说,用户的身份信息可以包括该用户所在的部门、职位(如管理人员、研发人员、文员等等)。对于学校来说,用户的身份信息可以包括该用户所在的学院、职位(教授、讲师、辅导员、博士研究生、硕士研究生、本科生等等)。对于图书馆,可以具有与学校相似的身份信息划分,对此不再赘述。在获取所述用户的身份信息时,可以根据用户的标识信息获取所述身份信息。例如,当用户登录到组织架构的局域网时,可以根据所述用户的标识信息(如登录账号)获取到与所述标识信息相匹配的身份信息,例如,所述标识信息可以 与所述身份信息绑定,在确定用户的标识信息之后,可以根据所述标识信息确定用户的身份信息。当然,在其他实施例中,所述组织架构不限于具有内部局域网,也可以运行于非局域网中,本申请对此不作限制。
本说明书中,用户可以在组织架构内搜索数据,具体地,用户可以在所述组织架构的局域网内输入搜索词,如“卷积神经网络”、“医疗VR”等等搜索词。在获取到所述搜索词之后,可以获取与所述搜索词相匹配的搜索结果项集合。本说明书中,所述搜索结果项可以包括多种数据类型,如文档、图片、视频、资讯、日志等多种数据信息。所述搜索结果项集合中可以包括上述数据类型中的一种或者多种。其中,所述搜索结果项与所述搜索词相匹配可以包括所述搜索结果项中可以包含所述搜索词,例如,对于文档搜索来说,搜索到的文档中可以包含所述搜索词,如文档的标题、摘要、正文、结语等一处或者多处出现所述搜索词。当然,在说明书的其他实施例中,所述搜索结果项与所述搜索词相匹配还可以包括所述搜索结果项中包括与所述搜索词的同义或者相似含义的词汇,对此,本申请在此不做限制。
S103:获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据。
本说明书中,还可以获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据。本说明书中,可以从所述组织架构的数据库中获取用户的用户日志,并从所述用户日志中获取所述用户对所述搜索结果项集合的历史访问数据。本说明书中,所述与所述用户的所述身份信息相匹配的其他用户可以包括与所述用户的身份信息相同的多个其他用户,或者,与所述用户的身份信息位于同一等级的多个其他用户。例如,若用户A属于公司研发部门的研发人员,即与用户A相匹配的多个其他用户可以包括与所述用户A同样属于研发部门的多个研发人员。本说明书中,所述历史访问数据可以包括所述用户和所述多个其他用户对所述搜索结果项的访问频率、访问时间等多个数据。例如,在一个示例中,根据用户A的搜索词“卷积神经网络”搜索到共200篇与卷积神经网络相关的电子文档,根据后台数据库的用户日志,可以获取到用户A分别对所述200篇电子文档的访问记录,如对各篇电子文档的访问次数等数据。另外,在确定与用户A相匹配的多个其他用户,即与所述用户A同样属于研发部门的多个研发人员之后,可以获取所述多个研发人员的对上述200篇电子文档的访问记录。
S105:根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所 述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
本说明书中,在获取到所述用户以及与所述用户身份信息相匹配的多个其他用户的历史访问数据之后,可以根据所述用户以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。在一个实施例中,如图2所示,所述根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项可以包括:
S201:根据所述用户以及所述多个其他用户的历史访问数据,分别统计所述用户以及所述多个其他用户分别对所述搜索结果项的访问频率。
S203:根据所述访问频率从所述搜索结果项集合中提取所述身份信息所对应的频繁项集合。
S205:根据所述用户的历史访问数据以及所述频繁项集合,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
本说明书中,可以根据所述用户以及所述多个其他用户的历史访问数据,分别统计所述用户以及所述多个其他用户对所述搜索结果项集合中搜索结果项的访问频率。例如在上述举例中,用户A对文档3的访问频率为3,即用户A一共访问过3次文档3,而用户B对文档80的访问频率为10,即用户B一共访问过10次文档80。然后,可以根据所述访问频率从所述搜索结果项中提取所述身份信息所对应的频繁项集合。所述频繁项集合可以包括所述身份信息所对应的访问比较频繁的多个搜索结果项组成的集合。具体地,在一个实施例中,如图3所示,所述根据所述访问频率从所述搜索结果项集合中提取所述身份信息所对应的频繁项集合可以包括:
S301:根据所述访问频率分别统计所述搜索结果项集合中搜索结果项的总访问频率。
S303:根据所述搜索结果项的总访问频率计算所述搜索结果项的支持度值。
S305:根据所述搜索结果项的支持度值计算所述搜索结果项集合的总支持度值。
S307:将所述搜索结果项集合中支持度值不小于所述总支持度值的搜索结果项的集合作为所述身份信息对应的频繁项集合。
本说明书中,可以根据所述用户以及所述多个其他用户分别对所述搜索结果项的访问频率,统计各个搜索结果项的总访问频率。例如,对于上述示例中的文档3,根据各 个用户对文档3的访问频率,可以统计得到用户A以及其他多个研发人员对文档3的总访问频率,例如,据统计,对文档3的总访问频率为40次。然后,在获取到所述搜索结果项集合中各个搜索结果项的访问频率之后,可以计算各个搜索结果项的支持度值,所述支持度值可以包括各个搜索结果项的关注热度信息。
在本说明书的其他实施例中,所述身份信息还可以包括多个身份等级,此时,如图4所示,所述根据所述访问频率从所述搜索结果项集合中提取所述身份信息所对应的频繁项集合可以包括:
S401:根据所述访问频率分别从所述搜索结果项集合中分别提取所述多个身份等级对应的频繁项集合。
S403:将所述多个身份等级对应的频繁项集合的交集作为所述身份信息所对应的频繁项集合。
在本实施例中,所述身份等级可以包括组织架构内垂直的用户等级划分。例如,对于一般的企业来说,可以划分为多个部门,每个部门具有各自的职位等级,如决策层、管理层以及普通用户层。也就是说,对于企业内的一个用户来说,该用户的身份信息中可以包括多个身份等级信息。例如上述示例中用户A,其身份信息可以包括{研发部门,普通研发人员},即用户A的身份信息中至少包含两层身份等级。在获取到用户的身份等级之后,可以根据所述访问频率从所述搜索结果项集合中分别提取所述多个身份等级所对应的频繁项集合。在获取到所述多个身份等级所对应的频繁项集合之后,可以将所述多个身份等级所对应的频繁项结合的交集作为所述身份信息所对应的频繁项集合。
下面通过一个示例说明上述获取频繁项集合的过程,例如,根据搜索词搜索得到的搜索结果项集合可以表示为I={i 1,i 2,......,i m},i m表示为第m个搜索结果项。根据组织架构局域网中的用户日志等历史数据,统计得到组织架构中搜索部门的用户A实际访问的点击项集合为H i={h 1,h 2,......,h k},h k表示为第k个点击项,显然k≤m且
Figure PCTCN2018099542-appb-000001
对于该搜索部门内的所有u个用户,可以得到具有u个不同点击项集合的点击项总集合D t={H 1,H 2,......,H u}。统计所述计算点击项总集合D t中所有的点击项,得到点击项总集合D t中共包含p个点击项H t={h 1,h 2,......,h p},其中,h i表示H t一项,若设置所述p个点击项的支持度为S t={s 1,s 2,......,s p},其中,s i表示S t一项,则s i的计算公式可以为:
Figure PCTCN2018099542-appb-000002
其中,n≤m,0<s i<1且
Figure PCTCN2018099542-appb-000003
将[0,1]区间划分为n个等份,每个点击项的支持度S t分布在该区间各个子区间段上,那么求得部门总支持度值E(S t)为
Figure PCTCN2018099542-appb-000004
根据求出的总支持度E(S t)可以计算得到推荐给该搜索部门用户的频繁项集合C t,具体地,可以将点击项总集合D t中支持度大于等于E(S t)的点击项的集合作为所述频繁项集合C t。假设所述频繁项集合C t具有j项,且C t={c 1,c 2,......,c j},则C t中任意一个搜索结果项的支持度值大于等于E(S t)。
在本说明书的一个实施例中,如图5所示,所述从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项可以包括:
S501:从所述搜索词中提取至少一个搜索分词;
S503:分别计算所述至少一个搜索分词在所述搜索结果项中多个搜索区域的邻近度值;
S505:根据所述邻近度值以及所述多个搜索区域的权重值,分别计算所述搜索词对应于所述搜索结果项的邻近度值;
S507:根据所述用户的历史访问数据、所述多个其他用户的历史访问数据以及所述搜索词对应于所述搜索结果项的邻近度值,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
本实施例中,所述搜索词中可以包括一个或者多个独立的分词,此时,可以从所述搜索词中提取至少一个搜索分词,例如对于描述词“医疗VR”,可以从中提取出搜索分词“医疗”、“VR”。在确定所述搜索词中的至少一个搜索分词之后,可以分别计算所述至少一个搜索分词在所述搜索结果项中多个搜索区域的邻近度值。其中,所述邻近度值越大,表示所述搜索分词与所述搜索区域的关联程度越大,就越有可能符合用户的期望结果。所述搜索结果项的搜索区域可以包括所述搜索结果项的各个组成部分,例如,当所述搜索结果项为文档时,所述搜索区域可以包括文档的标题、摘要、正文、结语等多个不同的组成部分。
在获取到所述邻近度值以及所述多个搜索区域的权重值之后,可以分别计算所述搜索词对应于所述搜索结果项的邻近度值。即将多个搜索分词对应的邻近度值合并成所述搜索词对应于所述搜索结果项的邻近度值。在获取到所述搜索词对应于所述搜索结果项的邻近度值之后,可以根据所述用户的历史访问数据、所述多个其他用户的历史访问数据以及所述搜索词对应于所述搜索结果项的邻近度值,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
具体地,在本说明书的一个实施例中,所述根据所述用户的历史访问数据、所述多个其他用户的历史访问数据以及所述搜索词对应于所述搜索结果项的邻近度值,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项可以包括:
SS1:根据所述用户的历史访问数据、所述多个其他用户的历史访问数据,分别计算所述用户对应于所述搜索结果项的个性化权重值。
SS3:根据所述搜索词对应于所述搜索结果项的邻近度值以及所述用户对应于所述搜索结果项的个性化权重值,计算所述搜索结果项的评分值。
SS5:将所述搜索结果项集合中评分值大于预设阈值的搜索结果项推荐给所述用户。
本实施例中,可以根据所述用户以及所述多个其他用户的历史访问数据,分别计算所述用户对应于各个搜索结果项的个性化权重值。根据所述搜索词对应于所述搜索结果项的邻近度值以及所述用户对应于所述搜索结果项的个性化权重值,计算所述搜索结果项的评分值。然后,将所述搜索结果项集合中评分值大于预设阈值的搜索结果项推荐给所述用户。
下面通过一个示例具体说明上述实施例过程,对于用户提交的搜索词q,通过中文分词得到搜索分词{t i}(其中,t i表示搜索词q中的第i个搜索分词),且t i∈q,则在某一搜索区域f中搜索分词t i的位置记为Pos(t i,f),搜索分词t i+1的位置记为Pos(t i+1,f),搜索分词t i与t i+1的最小距离值记为Dis min(t i,f),其邻近度值记为prox(t i,t i+1,f),则
Dis min(t i,t i+1,f)=min|Pos(t i+1,f)-Pos(t i,f)|
Figure PCTCN2018099542-appb-000005
搜索结果项d中包含多个搜索区域的描述(如标题,内容,路径等),且f∈d,搜索区域的权重为w(f),则搜索词q对应于所述搜索结果项d的邻近度值记为Prox(q,d), 那么,
Figure PCTCN2018099542-appb-000006
通过搜索结果项d的邻近度值Prox(q,d)更新评分结果,可以从用户输入的搜索词入手,利用邻近度值对与用户需求相关的搜索结果产生影响,能够准确返回与搜索词相匹配的搜索结果。另外,从邻近度计算公式来看,算法的复杂度没有增加,保证较快的搜索速度。
在按照上述实施例方法,根据组织架构内用户点击项集合H i与其所在部门的权限组的频繁项集V rt之间不存在包属的关系,但H i∩V rt≠Φ,系统初始设置用户点击项与推荐项的权值比为常量θ=50%(0<θ<1),当θ=0时,忽略用户实际点击效果,搜索引擎展现给用户的是推荐的搜索结果;当θ=100%时,忽略系统推荐的搜索结果,仅体现用户点击效果。当0<θ<100%时,融合了用户点击和系统推荐的搜索结果。那么用户推荐点击项的权重为W(q,d)可以表示为:
Figure PCTCN2018099542-appb-000007
其中F(h j)表示用户对h j项的点击量,F'(h j),F'(v ij)分别表示推荐h j,v ij项的点击量。由于处于组织架构内不同部门、不同权限组的用户关注的内容不同,点击密度随时间、部门差别和权限组有较大变化,搜索呈现的排序结果更符合个性化的特点。而且推荐项的权重可以离线计算,定期更新点击项与推荐项的权重,不影响检索速度。
评分排序结果融合是先根据查询内容的邻近度值与关键词权重计算计算结果搜索结果项的评分值,然后融合结果搜索结果项的个性化权重得到最终排序结果。具体方法如下:
结果搜索结果项的评分分为两个部分,一是基于布尔和向量空间模型的加权评分 Scorer(q,d),计算公式可以表示为
Figure PCTCN2018099542-appb-000008
其中,f T(t)表示查询项t在某一搜索结果项d中出现的词频,f d(t)表示含有查询项t的搜索结果项数(或占总搜索结果项的比例)。
二是加权邻近度评分Prox(q,d)。因此,扩展布尔和向量空间模型后的评分公式ExtendScorer(q,d)为
ExtendScorer(q,d)=Scorer(q,d)·Prox(q,d)
根据结果搜索结果项的权重,更新评分数据ExtendScorer(q,d)。最终得到的结果项集合D已知,点击项的权重W(q,d)影响集合D搜索结果项的评分,原始项权重值为1,由公式得到融合后的评分结果为
Figure PCTCN2018099542-appb-000009
对于集合D不包含W(q,d)的搜索结果项,其评分不变。这样计算得到的评分,能够体现用户行为特征的同时,也向用户推荐了其它用户关注的内容,很好的改善了用户的体验感,从而实现了用户个性化排序结果的展现。
本申请提供的数据处理方法,可以根据用户的搜索词匹配到搜索结果项集合,再根据用户对所述搜索结果项集合的历史访问数据,以及与所述用户身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。利用用户以及与用户身份信息相匹配的多个用户的历史数据,对推荐给用户的搜索结果项产生影响,使得推荐给用户的搜索结果项更加符合用户的需求以及用户的身份信息。尤其对于处于组织架构内的用户来说,用户在组织架构内搜索到的相关数据更加符合用户的搜索需求以及用户在组织架构内的身份。
本申请另一方面还提供一种时间序列中异常点检测装置,图6是本申请提供的数据处理装置的一种实施例的模块结构示意图,如图6所示,所述装置60可以包括:
结果项获取单元61,用于获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合;
历史数据获取单元63,用于获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据;
推荐结果项提取单元65,用于根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
可选的,在本申请的一个实施例中,所述推荐结果项提取单元65包括:
访问频率统计单元,用于根据所述用户以及所述多个其他用户的历史访问数据,分别统计所述用户以及所述多个其他用户分别对所述搜索结果项的访问频率;
频繁项集合提取单元,用于根据所述访问频率从所述搜索结果项集合中提取所述身份信息所对应的频繁项集合;
结果项提取单元,用于根据所述用户的历史访问数据以及所述频繁项集合,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
可选的,在本申请的一个实施例中,所述频繁项集合提取单元可以包括:
总访问频率统计单元,用于根据所述访问频率分别统计所述搜索结果项集合中搜索结果项的总访问频率;
支持度计算单元,用于根据所述搜索结果项的总访问频率计算所述搜索结果项的支持度值;
总支持度计算单元,用于根据所述搜索结果项的支持度值计算所述搜索结果项集合的总支持度值;
频繁项集合提取子单元,用于将所述搜索结果项集合中支持度值不小于所述总支持度值的搜索结果项的集合作为所述身份信息对应的频繁项集合。
可选的,在本申请的一个实施例中,当所述身份信息包括多个身份等级时,所述频繁项集合提取单元可以包括:
多个频繁项集合提取单元,用于根据所述访问频率分别从所述搜索结果项集合中分别提取所述多个身份等级对应的频繁项集合;
频繁项交集计算单元,用于将所述多个身份等级对应的频繁项集合的交集作为所述身份信息所对应的频繁项集合。
可选的,在本申请的一个实施例中,所述推荐结果项提取单元可以包括:
分词提取单元,用于从所述搜索词中提取至少一个搜索分词;
第一邻近度值计算单元,用于分别计算所述至少一个搜索分词在所述搜索结果项中多个搜索区域的邻近度值;
第二邻近度值计算单元,用于根据所述邻近度值以及所述多个搜索区域的权重值,分别计算所述搜索词对应于所述搜索结果项的邻近度值;
结果项提取子单元,用于根据所述用户的历史访问数据、所述多个其他用户的历史访问数据以及所述搜索词对应于所述搜索结果项的邻近度值,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
可选的,在本申请的一个实施例中,所述结果项提取子单元可以包括:
权重值计算单元,用于根据所述用户的历史访问数据、所述多个其他用户的历史访问数据,分别计算所述用户对应于所述搜索结果项的个性化权重值;
评分值计算单元,用于根据所述搜索词对应于所述搜索结果项的邻近度值以及所述用户对应于所述搜索结果项的个性化权重值,计算所述搜索结果项的评分值;
结果项推荐单元,用于将所述搜索结果项集合中评分值大于预设阈值的搜索结果项推荐给所述用户。
本申请另一方面还提供一种数据处理装置,图7是本申请提供的数据处理装置的一种实施例的模块结构示意图,如图7所示,所述装置70可以包括处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时可以实现:
获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合;
获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据;
根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
在20世纪90年代,对于一个技术的改进可以很明显地区分是硬件上的改进(例如,对二极管、晶体管、开关等电路结构的改进)还是软件上的改进(对于方法流程的改进)。然而,随着技术的发展,当今的很多方法流程的改进已经可以视为硬件电路结构的直接改进。设计人员几乎都通过将改进的方法流程编程到硬件电路中来得到相应的硬件电路 结构。因此,不能说一个方法流程的改进就不能用硬件实体模块来实现。例如,可编程逻辑器件(Programmable Logic Device,PLD)(例如现场可编程门阵列(Field Programmable Gate Array,FPGA))就是这样一种集成电路,其逻辑功能由用户对器件编程来确定。由设计人员自行编程来把一个数字系统“集成”在一片PLD上,而不需要请芯片制造厂商来设计和制作专用的集成电路芯片。而且,如今,取代手工地制作集成电路芯片,这种编程也多半改用“逻辑编译器(logic compiler)”软件来实现,它与程序开发撰写时所用的软件编译器相类似,而要编译之前的原始代码也得用特定的编程语言来撰写,此称之为硬件描述语言(Hardware Description Language,HDL),而HDL也并非仅有一种,而是有许多种,如ABEL(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby Hardware Description Language)等,目前最普遍使用的是VHDL(Very-High-Speed Integrated Circuit Hardware Description Language)与Verilog。本领域技术人员也应该清楚,只需要将方法流程用上述几种硬件描述语言稍作逻辑编程并编程到集成电路中,就可以很容易得到实现该逻辑方法流程的硬件电路。
控制器可以按任何适当的方式实现,例如,控制器可以采取例如微处理器或处理器以及存储可由该(微)处理器执行的计算机可读程序代码(例如软件或固件)的计算机可读介质、逻辑门、开关、专用集成电路(Application Specific Integrated Circuit,ASIC)、可编程逻辑控制器和嵌入微控制器的形式,控制器的例子包括但不限于以下微控制器:ARC 625D、Atmel AT91SAM、Microchip PIC18F26K20以及Silicone Labs C8051F320,存储器控制器还可以被实现为存储器的控制逻辑的一部分。本领域技术人员也知道,除了以纯计算机可读程序代码方式实现控制器以外,完全可以通过将方法步骤进行逻辑编程来使得控制器以逻辑门、开关、专用集成电路、可编程逻辑控制器和嵌入微控制器等的形式来实现相同功能。因此这种控制器可以被认为是一种硬件部件,而对其内包括的用于实现各种功能的装置也可以视为硬件部件内的结构。或者甚至,可以将用于实现各种功能的装置视为既可以是实现方法的软件模块又可以是硬件部件内的结构。
上述实施例阐明的单元,具体可以由计算机芯片或实体实现,或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的,计算机例如可以为个人计算机、膝上型计算机、蜂窝电话、相机电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这些设备中的任何设备 的组合。
为了描述的方便,描述以上客户端或服务器时以功能分为各种单元分别描述。当然,在实施本申请时可以把各单元的功能在同一个或多个软件和/或硬件中实现。
本领域内的技术人员应明白,本发明的实施例可提供为方法、系统、或计算机程序产品。因此,本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中,计算设备包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器,随机存取存储器(RAM)和/或非易失性内存等形式,如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或 技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
还需要说明的是,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、商品或者设备中还存在另外的相同要素。
本领域技术人员应明白,本申请的实施例可提供为方法、系统或计算机程序产品。因此,本申请可采用完全硬件实施例、完全软件实施例或结合软件和硬件方面的实施例的形式。而且,本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请可以在由计算机执行的计算机可执行指令的一般上下文中描述,例如程序模块。一般地,程序模块包括执行特定任务或实现特定抽象数据类型的例程、程序、对象、组件、数据结构等等。也可以在分布式计算环境中实践本申请,在这些分布式计算环境中,由通过通信网络而被连接的远程处理设备来执行任务。在分布式计算环境中,程序模块可以位于包括存储设备在内的本地和远程计算机存储介质中。
本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于系统实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
以上所述仅为本申请的实施例而已,并不用于限制本申请。对于本领域技术人员来说,本申请可以有各种更改和变化。凡在本申请的精神和原理之内所作的任何修改、等同替换、改进等,均应包含在本申请的权利要求范围之内。

Claims (13)

  1. 一种数据处理方法,所述方法包括:
    获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合;
    获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据;
    根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
  2. 根据权利要求1所述的方法,其特征在于,所述根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项包括:
    根据所述用户以及所述多个其他用户的历史访问数据,分别统计所述用户以及所述多个其他用户分别对所述搜索结果项的访问频率;
    根据所述访问频率从所述搜索结果项集合中提取所述身份信息所对应的频繁项集合;
    根据所述用户的历史访问数据以及所述频繁项集合,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
  3. 根据权利要求2所述的方法,其特征在于,所述根据所述访问频率从所述搜索结果项集合中提取所述身份信息所对应的频繁项集合包括:
    根据所述访问频率分别统计所述搜索结果项集合中搜索结果项的总访问频率;
    根据所述搜索结果项的总访问频率计算所述搜索结果项的支持度值;
    根据所述搜索结果项的支持度值计算所述搜索结果项集合的总支持度值;
    将所述搜索结果项集合中支持度值不小于所述总支持度值的搜索结果项的集合作为所述身份信息对应的频繁项集合。
  4. 根据权利要求2所述的方法,其特征在于,当所述身份信息包括多个身份等级时,所述根据所述访问频率从所述搜索结果项集合中提取所述身份信息所对应的频繁项集合包括:
    根据所述访问频率分别从所述搜索结果项集合中分别提取所述多个身份等级对应的频繁项集合;
    将所述多个身份等级对应的频繁项集合的交集作为所述身份信息所对应的频繁项集合。
  5. 根据权利要求1所述的方法,其特征在于,所述从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项包括:
    从所述搜索词中提取至少一个搜索分词;
    分别计算所述至少一个搜索分词在所述搜索结果项中多个搜索区域的邻近度值;
    根据所述邻近度值以及所述多个搜索区域的权重值,分别计算所述搜索词对应于所述搜索结果项的邻近度值;
    根据所述用户的历史访问数据、所述多个其他用户的历史访问数据以及所述搜索词对应于所述搜索结果项的邻近度值,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
  6. 根据权利要求5所述的方法,其特征在于,所述根据所述用户的历史访问数据、所述多个其他用户的历史访问数据以及所述搜索词对应于所述搜索结果项的邻近度值,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项包括:
    根据所述用户的历史访问数据、所述多个其他用户的历史访问数据,分别计算所述用户对应于所述搜索结果项的个性化权重值;
    根据所述搜索词对应于所述搜索结果项的邻近度值以及所述用户对应于所述搜索结果项的个性化权重值,计算所述搜索结果项的评分值;
    将所述搜索结果项集合中评分值大于预设阈值的搜索结果项推荐给所述用户。
  7. 一种数据处理装置,所述装置包括:
    结果项获取单元,用于获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合;
    历史数据获取单元,用于获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据;
    推荐结果项提取单元,用于根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
  8. 根据权利要求7所述的装置,其特征在于,所述推荐结果项提取单元包括:
    访问频率统计单元,用于根据所述用户以及所述多个其他用户的历史访问数据,分别统计所述用户以及所述多个其他用户分别对所述搜索结果项的访问频率;
    频繁项集合提取单元,用于根据所述访问频率从所述搜索结果项集合中提取所述身份信息所对应的频繁项集合;
    结果项提取单元,用于根据所述用户的历史访问数据以及所述频繁项集合,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
  9. 根据权利要求8所述的装置,其特征在于,所述频繁项集合提取单元包括:
    总访问频率统计单元,用于根据所述访问频率分别统计所述搜索结果项集合中搜索结果项的总访问频率;
    支持度计算单元,用于根据所述搜索结果项的总访问频率计算所述搜索结果项的支持度值;
    总支持度计算单元,用于根据所述搜索结果项的支持度值计算所述搜索结果项集合的总支持度值;
    频繁项集合提取子单元,用于将所述搜索结果项集合中支持度值不小于所述总支持度值的搜索结果项的集合作为所述身份信息对应的频繁项集合。
  10. 根据权利要求8所述的装置,其特征在于,当所述身份信息包括多个身份等级时,所述频繁项集合提取单元包括:
    多个频繁项集合提取单元,用于根据所述访问频率分别从所述搜索结果项集合中分别提取所述多个身份等级对应的频繁项集合;
    频繁项交集计算单元,用于将所述多个身份等级对应的频繁项集合的交集作为所述身份信息所对应的频繁项集合。
  11. 根据权利要求7所述的装置,其特征在于,所述推荐结果项提取单元包括:
    分词提取单元,用于从所述搜索词中提取至少一个搜索分词;
    第一邻近度值计算单元,用于分别计算所述至少一个搜索分词在所述搜索结果项中多个搜索区域的邻近度值;
    第二邻近度值计算单元,用于根据所述邻近度值以及所述多个搜索区域的权重值, 分别计算所述搜索词对应于所述搜索结果项的邻近度值;
    结果项提取子单元,用于根据所述用户的历史访问数据、所述多个其他用户的历史访问数据以及所述搜索词对应于所述搜索结果项的邻近度值,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
  12. 根据权利要求11所述的装置,其特征在于,所述结果项提取子单元包括:
    权重值计算单元,用于根据所述用户的历史访问数据、所述多个其他用户的历史访问数据,分别计算所述用户对应于所述搜索结果项的个性化权重值;
    评分值计算单元,用于根据所述搜索词对应于所述搜索结果项的邻近度值以及所述用户对应于所述搜索结果项的个性化权重值,计算所述搜索结果项的评分值;
    结果项推荐单元,用于将所述搜索结果项集合中评分值大于预设阈值的搜索结果项推荐给所述用户。
  13. 一种数据处理装置,包括处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时实现:
    获取用户的身份信息、搜索词以及与所述搜索词相匹配的搜索结果项集合;
    获取所述用户对所述搜索结果项集合的历史访问数据,以及与所述用户的所述身份信息相匹配的多个其他用户对所述搜索结果项集合的历史访问数据;
    根据所述用户的历史访问数据以及所述多个其他用户的历史访问数据,从所述搜索结果项集合中提取出推荐给所述用户的至少一个搜索结果项。
PCT/CN2018/099542 2017-10-12 2018-08-09 一种数据处理方法及装置 Ceased WO2019072007A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710946879.2A CN107908653A (zh) 2017-10-12 2017-10-12 一种数据处理方法及装置
CN201710946879.2 2017-10-12

Publications (1)

Publication Number Publication Date
WO2019072007A1 true WO2019072007A1 (zh) 2019-04-18

Family

ID=61840465

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/099542 Ceased WO2019072007A1 (zh) 2017-10-12 2018-08-09 一种数据处理方法及装置

Country Status (3)

Country Link
CN (1) CN107908653A (zh)
TW (1) TWI710917B (zh)
WO (1) WO2019072007A1 (zh)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107908653A (zh) * 2017-10-12 2018-04-13 阿里巴巴集团控股有限公司 一种数据处理方法及装置
CN109255037B (zh) * 2018-08-31 2022-03-08 北京字节跳动网络技术有限公司 用于输出信息的方法和装置
CN109614507B (zh) * 2018-11-22 2020-08-04 浙江大学 一种基于频繁项挖掘的遥感图像推荐装置
CN109919193B (zh) * 2019-01-31 2023-05-12 中国科学院上海光学精密机械研究所 一种大数据的智能分级方法、系统及终端
CN110287881A (zh) * 2019-06-26 2019-09-27 上海交通大学 图书识别系统、图书识别方法、电子装置及储存介质
CN110717038B (zh) * 2019-09-17 2022-10-04 腾讯科技(深圳)有限公司 对象分类方法及装置
CN111026574B (zh) * 2019-11-25 2024-03-12 中盈优创资讯科技有限公司 诊断Elasticsearch集群问题的方法及装置
CN115002691B (zh) * 2021-03-01 2023-08-15 中国移动通信集团四川有限公司 消息发送方法、装置、设备和计算机可读存储介质
CN120179176B (zh) * 2025-03-17 2025-08-26 昱格电子科技有限公司 基于云计算的数据全闪存储优化方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030004838A1 (en) * 2001-06-29 2003-01-02 International Business Machines Corporation Information search system, information search method, call center system, server and information search apparatus
US20040249790A1 (en) * 2003-03-31 2004-12-09 Toshiba Tec Kabushiki Kaisha Search device, search system, and search method
CN101770474A (zh) * 2008-12-26 2010-07-07 新奥特硅谷视频技术有限责任公司 一种基于历史检索记录检索的方法和装置
CN102930022A (zh) * 2012-10-31 2013-02-13 中国运载火箭技术研究院 面向用户的信息搜索引擎系统及方法
CN103425650A (zh) * 2012-05-15 2013-12-04 腾讯科技(深圳)有限公司 推荐搜索方法和系统
CN107908653A (zh) * 2017-10-12 2018-04-13 阿里巴巴集团控股有限公司 一种数据处理方法及装置

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110029403A1 (en) * 2009-07-29 2011-02-03 Ebay Inc. System and method for targeted merchandising to returning users
CN102567326B (zh) * 2010-12-14 2014-09-03 中国移动通信集团湖南有限公司 一种信息搜索、及信息搜索排序装置和方法
CN103942279B (zh) * 2014-04-01 2018-07-10 百度(中国)有限公司 搜索结果的展现方法和装置
CN104239458A (zh) * 2014-09-02 2014-12-24 百度在线网络技术(北京)有限公司 搜索结果的展现方法和装置
CN104598511A (zh) * 2014-10-23 2015-05-06 腾讯科技(深圳)有限公司 搜索结果推荐方法、装置及系统
CN106156103B (zh) * 2015-04-02 2019-11-26 广州爱九游信息技术有限公司 一种搜索处理方法及装置
CN105912549A (zh) * 2015-12-15 2016-08-31 乐视网信息技术(北京)股份有限公司 一种内容推荐方法及其装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030004838A1 (en) * 2001-06-29 2003-01-02 International Business Machines Corporation Information search system, information search method, call center system, server and information search apparatus
US20040249790A1 (en) * 2003-03-31 2004-12-09 Toshiba Tec Kabushiki Kaisha Search device, search system, and search method
CN101770474A (zh) * 2008-12-26 2010-07-07 新奥特硅谷视频技术有限责任公司 一种基于历史检索记录检索的方法和装置
CN103425650A (zh) * 2012-05-15 2013-12-04 腾讯科技(深圳)有限公司 推荐搜索方法和系统
CN102930022A (zh) * 2012-10-31 2013-02-13 中国运载火箭技术研究院 面向用户的信息搜索引擎系统及方法
CN107908653A (zh) * 2017-10-12 2018-04-13 阿里巴巴集团控股有限公司 一种数据处理方法及装置

Also Published As

Publication number Publication date
CN107908653A (zh) 2018-04-13
TWI710917B (zh) 2020-11-21
TW201923629A (zh) 2019-06-16

Similar Documents

Publication Publication Date Title
WO2019072007A1 (zh) 一种数据处理方法及装置
US12141177B2 (en) Data loss prevention system for cloud security based on document discourse analysis
US10726208B2 (en) Consumer insights analysis using word embeddings
US11544550B2 (en) Analyzing spatially-sparse data based on submanifold sparse convolutional neural networks
CN106599278B (zh) 应用搜索意图的识别方法及装置
US11182806B1 (en) Consumer insights analysis by identifying a similarity in public sentiments for a pair of entities
JP6177871B2 (ja) 製品情報の公開
US10558759B1 (en) Consumer insights analysis using word embeddings
US10515374B2 (en) Keyword generation method and apparatus
US20190188285A1 (en) Image Search with Embedding-based Models on Online Social Networks
US10049148B1 (en) Enhanced text clustering based on topic clusters
US20190197119A1 (en) Language-agnostic understanding
WO2019192261A1 (zh) 一种支付方式推荐方法、装置及设备
US10509863B1 (en) Consumer insights analysis using word embeddings
US10956469B2 (en) System and method for metadata correlation using natural language processing
CN107958014B (zh) 搜索引擎
CN112334889A (zh) 用于用户与助理系统交互的个性化手势识别
US11734322B2 (en) Enhanced intent matching using keyword-based word mover's distance
CN111149100A (zh) 基于命名实体的解析和识别确定跨文档的修辞相互关系
US11361030B2 (en) Positive/negative facet identification in similar documents to search context
US10803248B1 (en) Consumer insights analysis using word embeddings
US20200065422A1 (en) Document Entity Linking on Online Social Networks
WO2020135247A1 (zh) 法律文书解析方法及装置
US20240256787A1 (en) Techniques for generating multi-modal discourse trees
CN112132238A (zh) 一种识别隐私数据的方法、装置、设备和可读介质

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18866010

Country of ref document: EP

Kind code of ref document: A1