WO2021196476A1 - 对象推荐方法、电子设备及存储介质 - Google Patents

对象推荐方法、电子设备及存储介质 Download PDF

Info

Publication number
WO2021196476A1
WO2021196476A1 PCT/CN2020/106016 CN2020106016W WO2021196476A1 WO 2021196476 A1 WO2021196476 A1 WO 2021196476A1 CN 2020106016 W CN2020106016 W CN 2020106016W WO 2021196476 A1 WO2021196476 A1 WO 2021196476A1
Authority
WO
WIPO (PCT)
Prior art keywords
preset
label
user
level node
initial object
Prior art date
Application number
PCT/CN2020/106016
Other languages
English (en)
French (fr)
Inventor
彭燕
Original Assignee
深圳壹账通智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳壹账通智能科技有限公司 filed Critical 深圳壹账通智能科技有限公司
Publication of WO2021196476A1 publication Critical patent/WO2021196476A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/03Credit; Loans; Processing thereof

Definitions

  • This application relates to the field of data processing technology, and in particular to an object recommendation method, electronic equipment, and computer-readable storage medium.
  • the requirements are publicly released on the platform after they are confirmed by the platform operators to be true and effective, and are visible to all technical service providers. This approach will cause the demand side to receive a large number of invalid bid proposals and increase rent-seeking costs.
  • technical service providers also need to screen and focus on many needs.
  • Requirements are distributed by platform operators. Manual distribution requires operators to have the ability to identify and set up corresponding operating procedures, which consumes operating manpower costs. To sum up, due to the mismatch of information between the two parties, it is impossible to quickly and accurately match the technology demander and the technology service provider, resulting in a serious reduction in the efficiency of docking.
  • This application provides an object recommendation method, which includes:
  • Receiving step receiving a technical requirement sent by the first user through the client, where the technical requirement includes the first label and requirement description information corresponding to the technical requirement;
  • Label determination step parse the requirement description information, determine the keywords corresponding to the requirement description information, and determine the second label and the third label corresponding to the technical requirement according to the keywords, and generate the technical requirement Collection of tags;
  • the first screening step obtain the knowledge map of the preset second user from the preset storage path, analyze the knowledge map of the preset second user to obtain the analysis result, and obtain the analysis result from the preset second user according to the analysis result Screening out a second user matching the first tag of the technical requirement as the initial object;
  • the second screening step Calculate the match between the initial object and the technical requirement based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement Degree, screen out the initial objects whose matching degree is greater than or equal to the preset threshold as the target object;
  • the recommended step feedback the target object to the client of the first user in the order of matching degree.
  • the present application also provides an electronic device that includes a memory and a processor.
  • the memory stores an object recommendation program that can run on the processor, and the object recommendation program is used by the processor.
  • the steps of the object recommendation method described below can be implemented:
  • Receiving step receiving a technical requirement sent by the first user through the client, where the technical requirement includes the first label and requirement description information corresponding to the technical requirement;
  • Label determination step parse the requirement description information, determine the keywords corresponding to the requirement description information, and determine the second label and the third label corresponding to the technical requirement according to the keywords, and generate the technical requirement Collection of tags;
  • the first screening step obtain the knowledge map of the preset second user from the preset storage path, analyze the knowledge map of the preset second user to obtain the analysis result, and obtain the analysis result from the preset second user according to the analysis result Screening out a second user matching the first tag of the technical requirement as the initial object;
  • the second screening step Calculate the match between the initial object and the technical requirement based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement Degree, screen out the initial objects whose matching degree is greater than or equal to the preset threshold as the target object;
  • the recommended step feedback the target object to the client of the first user in the order of matching degree.
  • the present application also provides a computer-readable storage medium, the computer-readable storage medium includes an object recommendation program, and when the object recommendation program is executed by a processor, the following object recommendation can be realized Method steps:
  • Receiving step receiving a technical requirement sent by the first user through the client, where the technical requirement includes the first label and requirement description information corresponding to the technical requirement;
  • Label determination step parse the requirement description information, determine the keywords corresponding to the requirement description information, and determine the second label and the third label corresponding to the technical requirement according to the keywords, and generate the technical requirement Collection of tags;
  • the first screening step obtain the knowledge map of the preset second user from the preset storage path, analyze the knowledge map of the preset second user to obtain the analysis result, and obtain the analysis result from the preset second user according to the analysis result Screening out a second user matching the first tag of the technical requirement as the initial object;
  • the second screening step Calculate the match between the initial object and the technical requirement based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement Degree, screen out the initial objects whose matching degree is greater than or equal to the preset threshold as the target object;
  • the recommended step feedback the target object to the client of the first user in the order of matching degree.
  • an object recommendation device which includes:
  • the receiving module is configured to receive the technical requirement sent by the first user through the client, and the technical requirement includes the first label and requirement description information corresponding to the technical requirement;
  • the label determination module is used to analyze the demand description information, determine the keywords corresponding to the demand description information, and determine the second label and the third label corresponding to the technical demand according to the keywords, and generate the Label collection of technical requirements;
  • the first screening module is configured to obtain a knowledge map of a preset second user from a preset storage path, analyze the knowledge map of the preset second user to obtain an analysis result, and obtain an analysis result from the preset second user based on the analysis result Second, among users, a second user matching the first tag of the technical requirement is selected as an initial object;
  • the second screening module is configured to calculate the initial object and the technical requirement based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement
  • the matching degree of screening out the initial object with the matching degree greater than or equal to the preset threshold as the target object;
  • the recommendation module is configured to feed back the target object to the client of the first user in the order of matching degree.
  • Fig. 1 is a flowchart of a preferred embodiment of a method for recommending an object of the application
  • Figure 2 is a schematic diagram of the knowledge map
  • FIG. 3 is a schematic diagram of a preferred embodiment of the electronic device of this application.
  • Fig. 4 is a schematic diagram of modules of the recommending device for the subject of the application.
  • This application provides an object recommendation method, an electronic device, and a computer-readable storage medium.
  • the electronic device can obtain requirement information corresponding to the technical requirement, and then obtain a second user (technical service to be matched)
  • the knowledge map is matched with the knowledge map according to the demand information, and the second user who matches is determined to be recommended to the first user, which can better meet the technical needs of the first user in the actual business, and avoid the first user to find the corresponding technology by himself
  • the cumbersome process of service improves the efficiency of the first user's selection of technical services, thereby improving the first user's experience.
  • FIG. 1 it is a flowchart of a preferred embodiment of a method for recommending an object of this application.
  • the method can be executed by a device, and the device can be implemented by software and/or hardware.
  • the object recommendation method includes steps S1-S5.
  • Step S1 receiving a technical requirement sent by a first user through a client, the technical requirement including a first label corresponding to the technical requirement and requirement description information.
  • the above-mentioned first label is used to determine the type of required service, the above-mentioned technical requirement may include one or more first labels, and the first label is selected from a preset first label set, for example, the preset first label set includes : The number one label for loans, payments, operations, and marketing.
  • the above-mentioned first user is the technology demander.
  • the above-mentioned client is a mobile terminal used by the first user, and a client APP is installed on the client.
  • the above-mentioned requirement description information is used to describe the issued technical requirements for the business system to analyze the technical requirements and determine the detailed technical tags of the technical requirements.
  • the technical demander selects the corresponding first label from the preset first label set according to its own needs, and enters detailed demand description information according to its own needs.
  • the technical demander sends a technical demand carrying the first label and demand description information to the business system through the client APP of the client.
  • Step S2 Analyze the requirement description information, determine the keywords corresponding to the requirement description information, and determine the second tag and the third tag corresponding to the technical requirement according to the keywords, and generate a description of the technical requirement Label collection.
  • the above-mentioned first label is a primary label
  • the second label is a secondary label
  • the third label is a tertiary label.
  • the above-mentioned relationship between the first label, the second label and the third label is: one or more second labels may be included under the first label, and one or more third labels may be included under the second label.
  • the above-mentioned second label is the specific technical requirement put forward by the technical demander.
  • the identified second label is the credit investigation and risk control engine
  • the above-mentioned third label is the specific technical requirement put forward by the technical demander.
  • the corresponding product, for example, the recognized third label is a score card, etc.
  • the foregoing analysis of the requirement description information to determine the keywords corresponding to the requirement description information includes:
  • the preset keyword set includes a large number of keywords in the industry.
  • the requirement description information in the technical requirement A is P.
  • the requirement description information P is segmented, stop words, modal particles, and deduplication are processed, a word set ⁇ A1, A2, A3, A4, A5 is obtained ⁇
  • the preset keyword set corresponding to the second tag is ⁇ A2, A3, B1, B2,..., BN ⁇ . It is determined that the words that can be matched with the keyword set include ⁇ A2, A3 ⁇ , then the words A2 and A3 are taken as
  • the keyword corresponding to the above-mentioned requirement description information P is the keyword corresponding to the technical service request A.
  • the label for analyzing and determining the requirement description information is not limited to the second label and the third label, and may also be the first label.
  • the above-mentioned preset keyword database needs to be updated in time, so as to improve the accuracy and completeness of the subsequent determination of the second tag and the third tag.
  • mapping data of preset words and tags from a preset storage path, and determine whether there are words matching the keywords in the mapping data;
  • the label corresponding to the word matched by the keyword is used as the label corresponding to the keyword, and the label corresponding to the keyword is used as the second label and the third label corresponding to the technical requirement.
  • the mapping data of the preset words and tags includes: the mapping data of the words and the first tag, the mapping data of the words and the second tag, and the mapping data of the words and the third tag.
  • the same tag may correspond to one or more keywords (words), and the same keyword (word) may also correspond to one or more tags.
  • words keywords
  • word keyword
  • mapping data of preset words and tags words ⁇ W1, W2, W3,... ⁇ correspond to the first tag W; words ⁇ Y1, Y2, Y3,... ⁇ correspond to the second tag Y; words ⁇ M1, M2 , M3,... ⁇ correspond to the third label M;....
  • the keywords extracted in the above steps include Y2 and M3
  • the mapping data of the words and tags is the mapping data corresponding to the second tag
  • the determined third label may not have a one-to-one correspondence with the second label, obtain the identified third label, and determine the second label corresponding to the identified third label, and determine The second label is added to the parsed second label. If the second label is confirmed and the third label corresponding to the second label is not parsed, the third label and the second label corresponding to the second label are determined.
  • the tags are the same, for example, the third tag corresponding to the second tag "Wind Control Engine” is "Wind Control Engine”.
  • Step S3 Obtain the knowledge map of the preset second user from the preset storage path, analyze the knowledge map of the preset second user to obtain the analysis result, and filter from the preset second users according to the analysis result A second user who matches the first tag of the technical requirement is selected as an initial object.
  • the above-mentioned preset second user is all technical service providers in the business system.
  • the preset knowledge map of the second user is pre-calculated and stored in a preset storage path (for example, a database).
  • the above-mentioned knowledge map includes three levels: first-level node (large classification according to business scope, such as loan, operation, marketing), second-level node (subdivision of large business scope, for example: loan includes: credit investigation, advancement Software, risk control engine, collection system, etc.), and the third-level nodes are the products corresponding to the above-mentioned second-level nodes (for example, credit information includes: score cards, credit blacklists, etc.).
  • the above-mentioned preset knowledge map of the second user is determined through the following steps:
  • the preset type data determines the four-level node information
  • the knowledge map of the preset second user is generated based on the inclusion relationship of the first-level node information, the second-level node information, the third-level node information, and the fourth-level node information.
  • the foregoing preset time interval may be the previous month, the previous quarter, or the previous year, etc.
  • the foregoing first preset type data is the overall business scope of each preset second user
  • the foregoing second preset type data is each
  • the detailed business scope of the second user is preset
  • the third preset type data is product information within the business scope of each preset second user
  • the fourth preset type data is historical implementation case information corresponding to each product.
  • the implementation cases in the fourth preset type information include: the number of implementation cases and detailed information of each implementation case.
  • the detailed information of each implementation case includes the nature and scale of the implementation client's enterprise.
  • the foregoing determination of four-level node information based on the fourth preset type data includes: rating implementation cases, where the levels include: star implementation cases and common implementation cases, which are based on the nature of the implementing client's enterprise (for example, implementing client Comprehensive evaluation for central enterprises, state-owned enterprises, top 500, etc.) and scale (number of enterprises).
  • the technical service provider provides services related to the loan business.
  • the service includes incoming parts, risk control engine, collection, and credit reporting. Including two products: score card and credit blacklist, and there are 3 implementation cases for incoming items, 2 implementation cases for risk control engine, and no implementation case for collection. Then the knowledge map of the technical service provider is shown in Figure 2.
  • the above-mentioned analysis is performed on the knowledge map of the preset second user to obtain the analysis result, and the first tag with the technical requirement is selected from the preset second user according to the analysis result.
  • the matched second user is used as the initial object, including:
  • Analyze the knowledge map of the preset second user and respectively determine the first-level node information, the second-level node information, the third-level node information, and the fourth-level node information of the knowledge map;
  • the first-level node information of the knowledge map of the preset second user obtain the first label of the technical requirement from the label set of the technical requirement, and compare the first label with the first-level node The information is matched, and the second user who is successfully matched is selected as the initial object.
  • Step S4 Calculate the matching degree between the initial object and the technical requirement based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement.
  • the initial objects whose matching degree is greater than or equal to the preset threshold are screened out as target objects.
  • the foregoing calculation of the initial object and the technical requirement is performed based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement.
  • the matching degree includes:
  • the third-level node information of the knowledge map of the initial object is obtained from the analysis result, and the third-level node information of the initial object is obtained according to the third node information and the first Three tags determine the three-level related nodes of the initial object;
  • the fourth-level node information corresponding to the third-level related nodes of the initial object is extracted, and the matching degree between the initial object and the technical requirement is calculated according to the extracted fourth-level node information.
  • the number of secondary related nodes mentioned above is the number of nodes matching the second label of the first user in the secondary node information of the knowledge map of a certain initial object.
  • the above-mentioned three-level related nodes are nodes that match the third label of the first user in the three-level node information of the knowledge map of a certain initial object.
  • the foregoing preset condition is that the number of related nodes of the second-level node is greater than or equal to a preset percentage of the number of second-level demand nodes, for example, 30%.
  • the number of secondary demand nodes is the number of second tags of the first user.
  • the above-mentioned three-level related nodes are three-level nodes that match the third label of the first user.
  • the matching degree is match score (MS)
  • the knowledge map node matching degree is point score (PS)
  • the knowledge nodes at different levels are PS2, PS3, and PS4.
  • the calculation rules for matching degree are:
  • the PS4 calculation formula is:
  • PS3 represents the matching degree of the three-level correlation node of the initial object
  • PS4 represents the matching degree of a four-level node under a three-level correlation node of the initial object
  • represents the initial matching degree
  • n represents the four-level node.
  • the number of star implementation cases under the level node, ⁇ represents the matching degree of each star implementation case
  • m represents the number of common implementation cases under the four-level node
  • represents the matching degree of each common implementation case
  • step S5 the target object is fed back to the client of the first user according to the order of matching degree.
  • the selected technical service providers are sequentially recommended to the technical demanders according to the order of matching degree.
  • the feeding back the target object to the client of the first user in the order of matching degree includes:
  • a preset number of target objects with the highest matching degree ranking are selected and fed back to the client of the first user.
  • the feeding back the target object to the client of the first user in the order of matching degree further includes:
  • the target objects are matched with the white list, and the target objects with the highest matching degree and successful matching are selected first.
  • the technical service providers who have been blocked by the first user can be excluded before the recommendation, and the technical service providers that the first user is more satisfied with can be recommended first, thereby improving the user experience.
  • the object recommendation method further includes:
  • the knowledge map of the second user is updated and saved regularly.
  • an implementation case of the technical service provider is generated based on the requirement, and the historical business data and knowledge map of the technical service provider are updated and saved based on the implementation case.
  • the object recommendation method proposed in the above embodiment 1. Determine node information at all levels and generate a knowledge map according to the historical business information of each technical service provider (second user) in advance. During the object recommendation process, you can directly obtain the information of each technical service provider.
  • the knowledge map performs subsequent matching operations, which is equivalent to pre-characterizing the information data of each service provider, improving the efficiency of matching and recommendation; 2. After receiving the technical requirements carrying the requirement description information, analyze the requirement description information and Keyword matching, to obtain complete technical tags related to R&D requirements: first-level tags, second-level tags, and third-level tags for subsequent screening, calculation, and matching recommendations.
  • Labels at all levels Determine the number of demand nodes and the relevant nodes of the technical service providers at all levels, and calculate the matching degree of each technical service provider by analyzing the implementation case information of the second-level related nodes, the third-level related nodes and the fourth-level nodes of the technical service providers, and improve the calculation of the matching degree
  • This application does not use model recommendation. On the one hand, there is no need to obtain a large amount of data to train the model, which saves training time and improves matching efficiency. On the other hand, there is no problem of low accuracy of model training, which improves matching accuracy.
  • the dimensions considered in this application are more comprehensive, taking into account the quality and quantity of each implementation case, and improving the comprehensiveness and accuracy of the matching degree calculation, which is more in line with the demand side.
  • the application also provides an electronic device.
  • FIG. 3 is a schematic diagram of a preferred embodiment of the electronic device 1 of this application.
  • the electronic device 1 may be a terminal device with data processing functions such as a server, a smart phone, a tablet computer, a portable computer, a desktop computer, etc.
  • the server may be a rack server, a blade server, or a tower. Server or rack server.
  • the electronic device 1 includes a memory 11, a processor 12, and a network interface 13.
  • the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, and the like.
  • the memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as a hard disk of the electronic device 1.
  • the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in hard disk, a smart media card (SMC), and a secure digital (Secure Digital) equipped on the electronic device 1. ,SD) card, flash card (Flash Card), etc.
  • the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device.
  • the memory 11 can be used not only to store application software and various types of data installed in the electronic device 1, such as the object recommendation program 10, etc., but also to temporarily store data that has been output or will be output.
  • the processor 12 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip, for running program codes or processing stored in the memory 11 Data, such as object recommendation program 10, etc.
  • CPU central processing unit
  • controller microcontroller
  • microprocessor or other data processing chip, for running program codes or processing stored in the memory 11 Data, such as object recommendation program 10, etc.
  • the network interface 13 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface), and is usually used to establish a communication connection between the electronic device 1 and other electronic devices, for example, a client (not marked in the figure). ).
  • a standard wired interface and a wireless interface such as a WI-FI interface
  • FIG. 3 only shows the electronic device 1 with components 11-13. Those skilled in the art can understand that the structure shown in FIG. 3 does not constitute a limitation on the electronic device 1, and may include less or more Multiple components, or a combination of certain components, or different component arrangements.
  • the electronic device 1 may also include a user interface.
  • the user interface may include a display (Display) and an input unit such as a keyboard (Keyboard).
  • the optional user interface may also include a standard wired interface and a wireless interface.
  • the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an organic light-emitting diode (OLED) touch device, and the like.
  • the display may also be called a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display a visualized user interface.
  • the memory 11 as a computer storage medium stores the program code of the object recommendation program 10, and when the processor 12 executes the program code of the object recommendation program 10, the following steps are implemented:
  • Receiving step receiving a technical requirement sent by the first user through the client, where the technical requirement includes the first label and requirement description information corresponding to the technical requirement;
  • the above-mentioned first label is used to determine the type of required service, the above-mentioned technical requirement may include one or more first labels, and the first label is selected from a preset first label set, for example, the preset first label set includes : The number one label for loans, payments, operations, and marketing.
  • the above-mentioned first user is the technology demander.
  • the above-mentioned client is a mobile terminal used by the first user, and a client APP is installed on the client.
  • the above-mentioned requirement description information is used to describe the issued technical requirements for the business system to analyze the technical requirements and determine the detailed technical tags of the technical requirements.
  • the technical demander selects the corresponding first label from the preset first label set according to its own needs, and enters detailed demand description information according to its own needs.
  • the technical demander sends a technical demand carrying the first label and demand description information to the business system through the client APP of the client.
  • Label determination step parse the requirement description information, determine the keywords corresponding to the requirement description information, and determine the second label and the third label corresponding to the technical requirement according to the keywords, and generate the technical requirement Collection of tags;
  • the above-mentioned first label is a primary label
  • the second label is a secondary label
  • the third label is a tertiary label.
  • the above-mentioned relationship between the first label, the second label and the third label is: one or more second labels may be included under the first label, and one or more third labels may be included under the second label.
  • the above-mentioned second label is the specific technical requirement put forward by the technical demander.
  • the identified second label is the credit investigation and risk control engine
  • the above-mentioned third label is the specific technical requirement put forward by the technical demander.
  • the corresponding product, for example, the recognized third label is a score card, etc.
  • the foregoing analysis of the requirement description information to determine the keywords corresponding to the requirement description information includes:
  • the preset keyword set includes a large number of keywords in the industry.
  • the requirement description information in the technical requirement A is P.
  • the requirement description information P is segmented, stop words, modal particles, and deduplication are processed, a word set ⁇ A1, A2, A3, A4, A5 is obtained ⁇
  • the preset keyword set corresponding to the second tag is ⁇ A2, A3, B1, B2,..., BN ⁇ . It is determined that the words that can be matched with the keyword set include ⁇ A2, A3 ⁇ , then the words A2 and A3 are taken as
  • the keyword corresponding to the above-mentioned requirement description information P is the keyword corresponding to the technical service request A.
  • the label for analyzing and determining the requirement description information is not limited to the second label and the third label, and may also be the first label.
  • the above-mentioned preset keyword database needs to be updated in time, so as to improve the accuracy and completeness of the subsequent determination of the second tag and the third tag.
  • mapping data of preset words and tags from a preset storage path, and determine whether there are words matching the keywords in the mapping data;
  • the label corresponding to the word matched by the keyword is used as the label corresponding to the keyword, and the label corresponding to the keyword is used as the second label and the third label corresponding to the technical requirement.
  • the mapping data of the preset words and tags includes: the mapping data of the words and the first tag, the mapping data of the words and the second tag, and the mapping data of the words and the third tag.
  • the same tag may correspond to one or more keywords (words), and the same keyword (word) may also correspond to one or more tags.
  • words keywords
  • word keyword
  • mapping data of preset words and tags words ⁇ W1, W2, W3,... ⁇ correspond to the first tag W; words ⁇ Y1, Y2, Y3,... ⁇ correspond to the second tag Y; words ⁇ M1, M2 , M3,... ⁇ correspond to the third label M;....
  • the keywords extracted in the above steps include Y2 and M3
  • the mapping data of the words and tags is the mapping data corresponding to the second tag
  • the determined third label may not have a one-to-one correspondence with the second label, obtain the identified third label, and determine the second label corresponding to the identified third label, and determine The second label is added to the parsed second label. If the second label is confirmed and the third label corresponding to the second label is not parsed, the third label and the second label corresponding to the second label are determined.
  • the tags are the same, for example, the third tag corresponding to the second tag "Wind Control Engine” is "Wind Control Engine”.
  • the first screening step obtain the knowledge map of the preset second user from the preset storage path, analyze the knowledge map of the preset second user to obtain the analysis result, and obtain the analysis result from the preset second user according to the analysis result Screening out a second user matching the first tag of the technical requirement as the initial object;
  • the above-mentioned preset second user is all technical service providers in the business system.
  • the preset knowledge map of the second user is pre-calculated and stored in a preset storage path (for example, a database).
  • the above-mentioned knowledge map includes three levels: first-level node (large classification according to business scope, such as loan, operation, marketing), second-level node (subdivision of large business scope, for example: loan includes: credit investigation, advancement Software, risk control engine, collection system, etc.), and the third-level nodes are the products corresponding to the above-mentioned second-level nodes (for example, credit information includes: score cards, credit blacklists, etc.).
  • the above-mentioned preset knowledge map of the second user is determined through the following steps:
  • the preset type data determines the four-level node information
  • the knowledge map of the preset second user is generated based on the inclusion relationship of the first-level node information, the second-level node information, the third-level node information, and the fourth-level node information.
  • the foregoing preset time interval may be the previous month, the previous quarter, or the previous year, etc.
  • the foregoing first preset type data is the overall business scope of each preset second user
  • the foregoing second preset type data is each
  • the detailed business scope of the second user is preset
  • the third preset type data is product information within the business scope of each preset second user
  • the fourth preset type data is historical implementation case information corresponding to each product.
  • the implementation cases in the fourth preset type information include: the number of implementation cases and detailed information of each implementation case.
  • the detailed information of each implementation case includes the nature and scale of the implementation client's enterprise.
  • the foregoing determination of four-level node information based on the fourth preset type data includes: rating implementation cases, where the levels include: star implementation cases and common implementation cases, which are based on the nature of the implementing client's enterprise (for example, implementing client Comprehensive evaluation for central enterprises, state-owned enterprises, top 500, etc.) and scale (number of enterprises).
  • the technical service provider provides services related to the loan business.
  • the service includes incoming parts, risk control engine, collection, and credit reporting. Including two products: score card and credit blacklist, and there are 3 implementation cases for incoming items, 2 implementation cases for risk control engine, and no implementation case for collection. Then the knowledge map of the technical service provider is shown in Figure 2.
  • the above-mentioned analysis is performed on the knowledge map of the preset second user to obtain the analysis result, and the first tag with the technical requirement is selected from the preset second user according to the analysis result.
  • the matched second user is used as the initial object, including:
  • the first-level node information of the knowledge map of the preset second user obtain the first label of the technical requirement from the label set of the technical requirement, and compare the first label with the first-level node The information is matched, and the second user who is successfully matched is selected as the initial object.
  • the second screening step Calculate the match between the initial object and the technical requirement based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement Degree, screen out the initial objects whose matching degree is greater than or equal to the preset threshold as the target object;
  • the foregoing calculation of the initial object and the technical requirement is performed based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement.
  • the matching degree includes:
  • the third-level node information of the knowledge map of the initial object is obtained from the analysis result, and the third-level node information of the initial object is obtained according to the third node information and the first Three tags determine the three-level related nodes of the initial object;
  • the fourth-level node information corresponding to the third-level related nodes of the initial object is extracted, and the matching degree between the initial object and the technical requirement is calculated according to the extracted fourth-level node information.
  • the number of secondary related nodes mentioned above is the number of nodes matching the second label of the first user in the secondary node information of the knowledge map of a certain initial object.
  • the above-mentioned three-level related nodes are nodes that match the third label of the first user in the three-level node information of the knowledge map of a certain initial object.
  • the foregoing preset condition is that the number of related nodes of the second-level node is greater than or equal to a preset percentage of the number of second-level demand nodes, for example, 30%.
  • the number of secondary demand nodes is the number of second tags of the first user.
  • the above-mentioned three-level related nodes are three-level nodes that match the third label of the first user.
  • the matching degree is match score (MS)
  • the knowledge map node matching degree is point score (PS)
  • the knowledge nodes at different levels are PS2, PS3, and PS4.
  • the calculation rules for matching degree are:
  • the PS4 calculation formula is:
  • PS3 represents the matching degree of the three-level correlation node of the initial object
  • PS4 represents the matching degree of a four-level node under a three-level correlation node of the initial object
  • represents the initial matching degree
  • n represents the four-level node.
  • the number of star implementation cases under the level node, ⁇ represents the matching degree of each star implementation case
  • m represents the number of common implementation cases under the four-level node
  • represents the matching degree of each common implementation case
  • the recommended step feedback the target object to the client of the first user in the order of matching degree.
  • the selected technical service providers are sequentially recommended to the technical demanders according to the order of matching degree.
  • the feeding back the target object to the client of the first user in the order of matching degree includes:
  • a preset number of target objects with the highest matching degree ranking are selected and fed back to the client of the first user.
  • the feeding back the target object to the client of the first user in the order of matching degree further includes:
  • the target objects are matched with the white list, and the target objects with the highest matching degree and successful matching are selected first.
  • the technical service providers who have been blocked by the first user can be excluded before the recommendation, and the technical service providers that the first user is more satisfied with can be recommended first, thereby improving the user experience.
  • the knowledge map of the second user is updated and saved regularly.
  • an implementation case of the technical service provider is generated based on the requirement, and the historical business data and knowledge map of the technical service provider are updated and saved based on the implementation case.
  • the object recommendation program 10 may also be divided into one or more modules, and the one or more modules are stored in the memory 11 and run by one or more processors (this embodiment It is executed by the processor 12) to complete this application.
  • the module referred to in this application refers to a series of computer program instruction segments that can complete specific functions.
  • the object recommendation device 100 can be divided into modules 110-150, and the functions or operation steps implemented by the modules 110-150 are the same as those described above. Similar, not detailed here, exemplarily, for example, where:
  • the receiving module 110 is configured to receive a technical requirement sent by a first user through the client, and the technical requirement includes the first label and requirement description information corresponding to the technical requirement;
  • the label determination module 120 is configured to analyze the demand description information, determine the keywords corresponding to the demand description information, and determine the second label and the third label corresponding to the technical demand according to the keywords, and generate all A collection of labels describing technical requirements;
  • the first screening module 130 is configured to obtain a knowledge map of a preset second user from a preset storage path, analyze the knowledge map of the preset second user to obtain an analysis result, and obtain an analysis result from the preset storage path according to the analysis result. From the second users, a second user matching the first tag of the technical requirement is selected as the initial object;
  • the second screening module 140 is configured to calculate the initial object and the technology based on the preset matching degree calculation rule, the analysis result of the knowledge map of the initial object, and the second label and the third label corresponding to the technical requirement.
  • the recommendation module 150 is configured to feed back the target object to the client of the first user according to the order of matching degree.
  • an embodiment of the present application also proposes a computer-readable storage medium, the computer-readable storage medium includes an object recommendation program 10, and when the object recommendation program 10 is executed by a processor, the object recommendation method described above is implemented. step.
  • the computer-readable storage medium may be non-volatile or volatile.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • General Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Technology Law (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种对象推荐方法、电子设备及计算机存储介质,该方法包括:接收第一用户通过客户端发出的技术需求后,解析需求描述信息确定对应的关键词,并根据关键词确定对应的第二标签及第三标签;获取预设第二用户的知识地图并进行解析,根据解析结果对预设第二用户进行初筛确定初始对象;计算初始对象与所述技术需求的匹配度,根据匹配度对初始对象进行筛选确定目标对象;按照匹配度高低顺序将目标对象反馈至第一用户。利用该方法,可对技术需求与技术服务进行精准匹配,从而实现对象的快速精准推荐。

Description

对象推荐方法、电子设备及存储介质
本申请要求于2020年04月01日提交中国专利局、申请号为202010249649.2、发明名称为“对象推荐方法、电子装置及存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及数据处理技术领域,尤其涉及一种对象推荐方法、电子设备及计算机可读存储介质。
背景技术
随着互联网技术的发展,技术研发需求越来越多,一部分技术型公司可以自己满足技术研发需求,而另一部分非技术性公司则需通过技术服务商提供技术支持以满足技术研发需求。同时,部分公司为了减少技术研发支出,也会选择向技术服务商寻求技术支持。
发明人意识到,目前在通用的研发需求综合类开放平台上,需求分发有2种方式:1.需求经过平台运营人员确认真实有效后公开发布在平台上,所有技术服务商均可见。这种方式会导致需求方收到大量无效的应标方案增加寻租成本,同时技术服务商也需要在众多需求中进行筛选关注。2.需求由平台运营人员进行分发,人工分发需要运营人员具备甄别能力,且需要设置相应的运营流程,耗费运营人力成本。综上,由于双方信息不对等,无法快速准确地匹配技术需求方和技术服务商,使得对接效率严重降低。
目前的服务商与需求的匹配大多利用模型进行预测匹配,然而模型训练需要大量的数据,且耗时长、准确率不高,因此,亟需提供一种能精准匹配技术需求与技术服务以快速向技术需求方推荐技术服务商的方法。
发明内容
本申请提供一种对象推荐方法,该方法包括:
接收步骤:接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
标签确定步骤:对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
第一筛选步骤:从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
第二筛选步骤:基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
推荐步骤:按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
此外,本申请还提供一种电子设备,该电子设备包括:存储器、处理器,所述存储器上存储有可在所述处理器上运行的对象推荐程序,所述对象推荐程序被所述处理器执行时,可实现如下所述对象推荐方法的步骤:
接收步骤:接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
标签确定步骤:对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
第一筛选步骤:从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
第二筛选步骤:基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
推荐步骤:按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
此外,为实现上述目的,本申请还提供一种计算机可读存储介质,所述计算机可读存储介质中包括对象推荐程序,所述对象推荐程序被处理器执行时,可实现如下所述对象推荐方法的步骤:
接收步骤:接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
标签确定步骤:对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
第一筛选步骤:从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
第二筛选步骤:基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
推荐步骤:按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
此外,为实现上述目的,本申请还提供一种对象推荐装置,所述装置包括:
接收模块,用于接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
标签确定模块,用于对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
第一筛选模块,用于从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
第二筛选模块,用于基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
推荐模块,用于按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
附图说明
图1为本申请对象推荐方法较佳实施例的流程图;
图2为知识地图的示意图;
图3为本申请电子设备较佳实施例的示意图;
图4为本申请对象推荐装置的模块示意图。
本申请目的的实现、功能特点及优点将结合实施例,参照附图做进一步说明。
具体实施方式
以下结合附图对本公开的具体实施方式进行详细说明。应当理解的是,此处所描述的具体实施例仅仅用于说明和解释本申请,并不用于限制本申请。
本申请提供一种对象推荐方法、电子设备及计算机可读存储介质,在第一用户输入技术需求时,电子设备可以获取技术需求对应的需求信息,然后获取第二用户(待匹配的技术服务)的知识地图,根据需求信息与知识地图进行匹配,确定匹配的第二用户推荐给第一用户,更能满足第一用户的实际业务中的技术需求,也避免了第一用户自己查找对应的技术服务的繁琐过程,从而提高了第一用户选择技术服务的效率,进而提高第一用户的体验。
下面结合附图对本申请的具体实施方式进行详细说明。
参照图1所示,为本申请对象推荐方法较佳实施例的流程图。该方法可以由一个装置执行,该装置可以由软件和/或硬件实现。
在本实施例中,所述对象推荐方法包括步骤S1-S5。
步骤S1,接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息。
上述第一标签用于确定需求业务类型,上述技术需求中可以包括一个或多个第一标签,所述第一标签是从预设第一标签集中选择的,例如,预设第一标签集中包括:贷款、支付、运营、营销等第一标签。
上述第一用户为技术需求方。上述客户端为第一用户使用的移动终端,客户端上安装有客户端APP。
上述需求描述信息用于对发出的技术需求进行描述,以供业务系统对技术需求进行分析,确定技术需求的细化技术标签。
技术需求方根据自身需求从预设第一标签集中选择对应的第一标签,并根据自身需求输入详细的需求描述信息。技术需求方通过客户端的客户端APP向业务系统发出携带第一标签及需求描述信息的技术需求。
步骤S2,对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合。
其中,上述第一标签为一级标签,第二标签为二级标签,第三标签为三级标签。上述第一标签、第二标签及第三标签之间的关系为:第一标签下可包括一个或多个第二标签,第二标签下可包括一个或多个第三标签。
上述第二标签为技术需求方提出的具体技术需求,以第一标签为贷款为例,识别出的第二标签为征信及风控引擎,上述第三标签为技术需求方提出的具体技术需求对应的产品,例如,识别出的第三标签为评分卡等。
在本实施例中,上述对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,包括:
对所述需求描述信息进行分词,得到所述需求描述信息对应的词语集合;及
将所述词语集合中各词语与预设关键词集合进行匹配,将与所述预设关键词集合中的关键词匹配的词语作为所述需求描述信息对应的关键词。
其中,上述预设关键词集合包括行业内大量的关键词。例如,技术需求A中的需求描述信息为P,在对需求描述信息P进行分词、去停用词、语气词、去重等处理后,得到一个词语集合{A1、A2、A3、A4、A5},第二标签对应的预设关键词集合为{A2、A3、B1、B2、…、BN},确定可与关键词集合匹配的词语包括{A2、A3},则将词语A2、A3作为上述需求描述信息P对应的关键词,即该技术服务请求A对应的关键词。
需要说明的是,对所述需求描述信息进行解析确定的标签并不仅限于第二标签、第三标签,也可能是第一标签。
为了提高关键词提取的准确性及完整性,需及时更新上述预设关键词库,进而提高后续确定第二标签及第三标签的准确性及完整性。
上述根据所述关键词确定所述技术需求对应的第二标签及第三标签,包括:
从预设存储路径获取预设词语与标签的映射数据,判断所述映射数据中是否存在与所述关键词匹配的词语;及
若存在,则将所述关键词匹配的词语对应的标签作为所述关键词对应的标签,并将所述关键词对应的标签作为所述技术需求对应的第二标签及第三标签。
上述预设词语与标签的映射数据中包括:词语与第一标签的映射数据、词语与第二标签的映射数据、词语与第三标签的映射数据。
可以理解的是,在实际应用过程中,由于业务、技术的交叉,同一个标签可以对应一个或多个关键词(词语),同一个关键词(词语)也可能对应一个或多个标签。例如,上述预设词语与标签的映射数据中:词语{W1、W2、W3、…}对应第一标签W;词语{Y1、Y2、Y3、…}对应第二标签Y;词语{M1、M2、M3、…}对应第三标签M;…。
若上述步骤提取出的关键词包括Y2、M3,词语与标签的映射数据为第二标签对应的映射数据,则可确定上述技术需求A对应的第二标签包括Y,对应的第三标签包括M。
需要说明的是,并不是所有的第二标签下都包括第三标签。在确认标签的过程中,确定的第三标签与第二标签可能不是一一对应的关系,获取识别出的第三标签,并确定与识别出的第三标签对应的第二标签,并将确定出的第二标签添加至解析出的第二标签中,若确认的是第二标签且未解析出该第二标签对应的第三标签,则确定该第二标签对应的第三标签与第二标签相同,例如,第二标签“风控引擎”对应的第三标签为“风控引擎”。
步骤S3,从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象。
上述预设第二用户为业务系统中所有的技术服务提供商。
在本实施例中,预设第二用户的知识地图为预先计算并保存在预设存储路径(例如,数据库)中的。上述知识地图包括三个层级:一级节点(按照业务范围进行的大分类,例如:贷款、运营、营销)、二级节点(对大业务范围进行细分,例如:贷款包括:征信、进件、风控引擎、催收系统等)、三级节点为上述二级节点对应的产品(例如,征信包括:评分卡、信贷黑名单等)。
在本实施例中,上述预设第二用户的知识地图通过以下步骤确定:
从预设数据库中获取所述预设第二用户预设时间间隔内的历史业务数据,根据所述历史业务数据确定所述预设第二用户的第一预设类型数据、第二预设类型数据、第三预设类型数据及第四预设类型数据;
根据所述第一预设类型数据确定一级节点信息,根据所述第二预设类型数据确定二级节点信息,根据所述第三预设类型数据确定三级节点信息,根据所述第四预设类型数据确定四级节点信息;及
基于所述一级节点信息、二级节点信息、三级节点信息及四级节点信息的包含关系生成所述预设第二用户的知识地图。
例如,上述预设时间间隔可以是上个月、上一季度或者上一年度等;上述第一预设类型数据为各预设第二用户的整体业务范围,上述第二预设类型数据为各预设第二用户的细化业务范围,上述第三预设类型数据为各预设第二用户在业务范围内的产品信息,上述第四预设类型数据为各产品对应的历史实施案例信息。
上述第四预设类型信息中的实施案例包括:实施案例数量及各实施案例的详细信息。例如,各实施案例的详细信息包括实施客户的企业性质、规模等。上述根据所述第四预设类型数据确定四级节点信息,包括:对实施案例进行等级评定,其中,等级包括:明星实施案例及普通实施案例,其根据实施客户的企业性质(例如,实施客户为央企、国企、500强等)和规模(企业人数)综合评定。
例如,通过分析某技术服务商在预设时间间隔内的历史业务数据可知,该技术服务商提供贷款业务相关的服务,该服务包括进件、风控引擎、催收、征信,其中,征信包括:评分卡、信贷黑名单这两个产品,且进件有3个实施案例,风控引擎有2个实施案例,催收暂无实施案例。那么该技术服务商的知识地图如图2所示。
在本实施例中,上述对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象,包括:
对所述预设第二用户的知识地图进行解析,分别确定知识地图的一级节点信息、二级节点信息、三级节点信息及四级节点信息;及
获取所述预设第二用户的知识地图的一级节点信息,从所述技术需求的标签集合中获取所述技术需求的所述第一标签,对所述第一标签与所述一级节点信息进行匹配,筛选出匹配成功的所述第二用户作为初始对象。
为了节省后续处理的时间,减少计算量,对于业务范围不包含本技术需求的技术服务商直接剔除,避免计算能力浪费。
步骤S4,基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象。
在本实施例中,上述基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,包括:
从所述解析结果中获取所述初始对象的知识地图的二级节点信息,根据所述二级节点信息及所述第二标签确定所述初始对象的二级相关节点数量,判断所述初始对象的二级相关节点数量是否满足预设条件;
当所述初始对象的二级相关节点数量不满足预设条件时,判断所述初始对象与所述技术需求的匹配度为0;
当所述初始对象的二级相关节点数量满足预设条件时,则从所述解析结果中获取所述初始对象的知识地图的三级节点信息,并根据所述第三节点信息及所述第三标签确定所述初始对象的三级相关节点;及
提取出与所述初始对象的三级相关节点对应的四级节点信息,根据提取出的所述四级节点信息计算所述初始对象与所述技术需求的匹配度。
上述二级相关节点数量为某一个初始对象的知识地图的二级节点信息中与第一用户的第二标签匹配的节点数量。
上述三级相关节点为某一个初始对象的知识地图的三级节点信息中与第一用户的第三标签匹配的节点。
例如,上述预设条件为二级节点相关节点数量大于或等于二级需求节点数量的预设百分比,例如,30%。其中,二级需求节点数量为第一用户的第二标签的数量。上述三级相关节点为与第一用户的第三标签匹配的三级节点。
其中,匹配度为match score(MS),知识地图节点匹配度为point score(PS),不同层级的知识节点为PS2、PS3、PS4。匹配度的计算规则为:
若PS2相关节点数量<需求节点数量*30%,MS=0;
若PS2相关节点数量>=需求节点数量*30%,则MS的计算公式为:
MS=sum(PS3)=sum(sum(PS4))
在本实施例中,PS4的计算公式为:
PS4=α+n*β+m*γ
其中,PS3表示所述初始对象的三级相关节点的匹配度,PS4表示所述初始对象的一个三级相关节点下的一个四级节点的匹配度,α表示起始匹配度,n表示该四级节点下的明星实施案例的数量,β表示每个明星实施案例的匹配度,m表示该四级节点下普通实施案例的数量,γ表示每个普通实施案例的匹配度,n、m为大于或等于0的整数。例如,α=0.5,β=1,γ=0.2。若该末梢节点无实施案例,则PS4=0.5;若该末梢节点每增加一个明星实施案例PS4=0.5+1,每增加一个普通实施案例PS4=0.5+0.2。
需要说明的是,每个节点添加实施案例的数量有限(例如,最多3个),优先取明星实施案例。
步骤S5,按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
为了便于第一用户甄别,将筛选出的技术服务商按照匹配度的高低顺序依次推荐给技术需求方。
在其他实施例中,所述按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端,包括:
当所述目标对象的数量超过预设阈值时,筛选出预设数量的匹配度排序靠前的目标对象反馈至所述第一用户的客户端。
例如,若匹配度满足要求的目标对象超过10个时,仅取匹配度排序靠前的10个(或者8个)目标对象反馈给第一用户。
在其他实施例中,所述按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端,还包括:
获取第一用户对应的黑名单及白名单,将所述目标对象与所述黑名单进行匹配,从所述目标对象中删除匹配成功的目标对象;或者
当存在至少两个匹配度相同的目标对象时,将所述至少两个匹配度相同的目标对象与所述白名单进行匹配,将匹配成功的目标对象排在匹配失败的目标对象前;或者
当所述目标对象的数量超过预设阈值时,将所述目标对象与所述白名单进行匹配,优先筛选出匹配度靠前且匹配成功的目标对象。
通过比对黑名单与白名单,可以在推荐前排除掉第一用户拉黑过的技术服务商,可以优先推荐第一用户较满意的技术服务商,进而提高用户的使用体验。
在其他实施例中,所述对象推荐方法还包括:
定期对所述第二用户的知识地图进行更新保存。
具体地,当一个技术服务商完成一次需求后,基于该次需求生成该技术服务商的一个实施案例,并基于该实施案例对该技术服务商的历史业务数据及知识地图进行更新并保存。
上述实施例提出的对象推荐方法,1.预先根据各技术服务商(第二用户)的历史业务信息确定各级节点信息且生成知识地图,在对象推荐过程中,可直接获取各技术服务商的知识地图进行后续匹配操作,相当于预先对各服务商的信息数据进行特征处理,提高了匹配及推荐的效率;2.在接收到携带需求描述信息的技术需求后,对需求描述信息进行解析及关键词匹配,得到与研发需求相关的完整的技术标签:一级标签、二级标签及三级标签,以用于后续进行筛选、计算及匹配推荐,基于该技术特征,能自动、全面获取用户的需求,从而避免后续推荐的技术服务商不满足用户技术需求的情况,提高匹配的精准度,提升用户使用体验;3.在确定各技术服务商对应的各级节点信息后,根据一级节点信息与技术需求的一级标签对技术服务商进行初步筛选,过滤掉不满足条件的服务商,可以减少后续的 计算量,节省计算资源,提高匹配效率;4.根据技术需求对应的各级标签确定需求节点数量及技术服务商的各级相关节点,通过分析技术服务商的二级相关节点、三级相关节点及四级节点的实施案例信息计算各技术服务商的匹配度,提高匹配度计算的客观性准确性。综上,本申请未采用模型推荐,一方面不需要获取大量数据对模型进行训练,节省训练时长,提高匹配效率,另一方面也不存在模型训练的准确性不高的问题,提高了匹配准确度,进而提升用户体验;相较于现有技术,本申请考虑的维度更全面,考虑到了各实施案例的质量、数量等,提高了匹配度计算的全面性、准确性,更能符合需求方的技术研发需求,提升技术需求方的满意度。
本申请还提供一种电子设备。参照图3所示,为本申请电子设备1较佳实施例的示意图。
在本实施例中,电子设备1可以是服务器、智能手机、平板电脑、便携计算机、桌上型计算机等具有数据处理功能的终端设备,所述服务器可以是机架式服务器、刀片式服务器、塔式服务器或机柜式服务器。
该电子设备1包括存储器11、处理器12,及网络接口13。
其中,存储器11至少包括一种类型的可读存储介质,所述可读存储介质包括闪存、硬盘、多媒体卡、卡型存储器(例如,SD或DX存储器等)、磁性存储器、磁盘、光盘等。存储器11在一些实施例中可以是所述电子设备1的内部存储单元,例如该电子设备1的硬盘。存储器11在另一些实施例中也可以是所述电子设备1的外部存储设备,例如该电子设备1上配备的插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)等。进一步地,存储器11还可以既包括该电子设备1的内部存储单元也包括外部存储设备。
存储器11不仅可以用于存储安装于该电子设备1的应用软件及各类数据,例如对象推荐程序10等,还可以用于暂时地存储已经输出或者将要输出的数据。
处理器12在一些实施例中可以是一中央处理器(Central Processing Unit,CPU)、控制器、微控制器、微处理器或其他数据处理芯片,用于运行存储器11中存储的程序代码或处理数据,例如对象推荐程序10等。
网络接口13可选的可以包括标准的有线接口、无线接口(如WI-FI接口),通常用于在该电子设备1与其他电子设备之间建立通信连接,例如,客户端(图中未标识)。
图3仅示出了具有组件11-13的电子设备1,本领域技术人员可以理解的是,图3示出的结构并不构成对电子设备1的限定,可以包括比图示更少或者更多的部件,或者组合某些部件,或者不同的部件布置。
可选地,该电子设备1还可以包括用户接口,用户接口可以包括显示器(Display)、输入单元比如键盘(Keyboard),可选的用户接口还可以包括标准的有线接口、无线接口。
可选地,在一些实施例中,显示器可以是LED显示器、液晶显示器、触控式液晶显示器以及有机发光二极管(Organic Light-Emitting Diode,OLED)触摸器等。其中,显示器也可以称为显示屏或显示单元,用于显示在电子设备1中处理的信息以及用于显示可视化的用户界面。
在图3所示的电子设备1实施例中,作为一种计算机存储介质的存储器11中存储对象推荐程序10的程序代码,处理器12执行对象推荐程序10的程序代码时,实现如下步骤:
接收步骤:接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
上述第一标签用于确定需求业务类型,上述技术需求中可以包括一个或多个第一标签,所述第一标签是从预设第一标签集中选择的,例如,预设第一标签集中包括:贷款、支付、 运营、营销等第一标签。
上述第一用户为技术需求方。上述客户端为第一用户使用的移动终端,客户端上安装有客户端APP。
上述需求描述信息用于对发出的技术需求进行描述,以供业务系统对技术需求进行分析,确定技术需求的细化技术标签。
技术需求方根据自身需求从预设第一标签集中选择对应的第一标签,并根据自身需求输入详细的需求描述信息。技术需求方通过客户端的客户端APP向业务系统发出携带第一标签及需求描述信息的技术需求。
标签确定步骤:对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
其中,上述第一标签为一级标签,第二标签为二级标签,第三标签为三级标签。上述第一标签、第二标签及第三标签之间的关系为:第一标签下可包括一个或多个第二标签,第二标签下可包括一个或多个第三标签。
上述第二标签为技术需求方提出的具体技术需求,以第一标签为贷款为例,识别出的第二标签为征信及风控引擎,上述第三标签为技术需求方提出的具体技术需求对应的产品,例如,识别出的第三标签为评分卡等。
在本实施例中,上述对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,包括:
对所述需求描述信息进行分词,得到所述需求描述信息对应的词语集合;及
将所述词语集合中各词语与预设关键词集合进行匹配,将与所述预设关键词集合中的关键词匹配的词语作为所述需求描述信息对应的关键词。
其中,上述预设关键词集合包括行业内大量的关键词。例如,技术需求A中的需求描述信息为P,在对需求描述信息P进行分词、去停用词、语气词、去重等处理后,得到一个词语集合{A1、A2、A3、A4、A5},第二标签对应的预设关键词集合为{A2、A3、B1、B2、…、BN},确定可与关键词集合匹配的词语包括{A2、A3},则将词语A2、A3作为上述需求描述信息P对应的关键词,即该技术服务请求A对应的关键词。
需要说明的是,对所述需求描述信息进行解析确定的标签并不仅限于第二标签、第三标签,也可能是第一标签。
为了提高关键词提取的准确性及完整性,需及时更新上述预设关键词库,进而提高后续确定第二标签及第三标签的准确性及完整性。
上述根据所述关键词确定所述技术需求对应的第二标签及第三标签,包括:
从预设存储路径获取预设词语与标签的映射数据,判断所述映射数据中是否存在与所述关键词匹配的词语;及
若存在,则将所述关键词匹配的词语对应的标签作为所述关键词对应的标签,并将所述关键词对应的标签作为所述技术需求对应的第二标签及第三标签。
上述预设词语与标签的映射数据中包括:词语与第一标签的映射数据、词语与第二标签的映射数据、词语与第三标签的映射数据。
可以理解的是,在实际应用过程中,由于业务、技术的交叉,同一个标签可以对应一个或多个关键词(词语),同一个关键词(词语)也可能对应一个或多个标签。例如,上述预设词语与标签的映射数据中:词语{W1、W2、W3、…}对应第一标签W;词语{Y1、Y2、Y3、…}对应第二标签Y;词语{M1、M2、M3、…}对应第三标签M;…。
若上述步骤提取出的关键词包括Y2、M3,词语与标签的映射数据为第二标签对应的映射数据,则可确定上述技术需求A对应的第二标签包括Y,对应的第三标签包括M。
需要说明的是,并不是所有的第二标签下都包括第三标签。在确认标签的过程中,确 定的第三标签与第二标签可能不是一一对应的关系,获取识别出的第三标签,并确定与识别出的第三标签对应的第二标签,并将确定出的第二标签添加至解析出的第二标签中,若确认的是第二标签且未解析出该第二标签对应的第三标签,则确定该第二标签对应的第三标签与第二标签相同,例如,第二标签“风控引擎”对应的第三标签为“风控引擎”。
第一筛选步骤:从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
上述预设第二用户为业务系统中所有的技术服务提供商。
在本实施例中,预设第二用户的知识地图为预先计算并保存在预设存储路径(例如,数据库)中的。上述知识地图包括三个层级:一级节点(按照业务范围进行的大分类,例如:贷款、运营、营销)、二级节点(对大业务范围进行细分,例如:贷款包括:征信、进件、风控引擎、催收系统等)、三级节点为上述二级节点对应的产品(例如,征信包括:评分卡、信贷黑名单等)。
在本实施例中,上述预设第二用户的知识地图通过以下步骤确定:
从预设数据库中获取所述预设第二用户预设时间间隔内的历史业务数据,根据所述历史业务数据确定所述预设第二用户的第一预设类型数据、第二预设类型数据、第三预设类型数据及第四预设类型数据;
根据所述第一预设类型数据确定一级节点信息,根据所述第二预设类型数据确定二级节点信息,根据所述第三预设类型数据确定三级节点信息,根据所述第四预设类型数据确定四级节点信息;及
基于所述一级节点信息、二级节点信息、三级节点信息及四级节点信息的包含关系生成所述预设第二用户的知识地图。
例如,上述预设时间间隔可以是上个月、上一季度或者上一年度等;上述第一预设类型数据为各预设第二用户的整体业务范围,上述第二预设类型数据为各预设第二用户的细化业务范围,上述第三预设类型数据为各预设第二用户在业务范围内的产品信息,上述第四预设类型数据为各产品对应的历史实施案例信息。
上述第四预设类型信息中的实施案例包括:实施案例数量及各实施案例的详细信息。例如,各实施案例的详细信息包括实施客户的企业性质、规模等。上述根据所述第四预设类型数据确定四级节点信息,包括:对实施案例进行等级评定,其中,等级包括:明星实施案例及普通实施案例,其根据实施客户的企业性质(例如,实施客户为央企、国企、500强等)和规模(企业人数)综合评定。
例如,通过分析某技术服务商在预设时间间隔内的历史业务数据可知,该技术服务商提供贷款业务相关的服务,该服务包括进件、风控引擎、催收、征信,其中,征信包括:评分卡、信贷黑名单这两个产品,且进件有3个实施案例,风控引擎有2个实施案例,催收暂无实施案例。那么该技术服务商的知识地图如图2所示。
在本实施例中,上述对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象,包括:
对所述预设第二用户的知识地图进行解析,分别确定知识地图的一级节点信息、二级节点信息、三级节点信息及四级节点信息;
获取所述预设第二用户的知识地图的一级节点信息,从所述技术需求的标签集合中获取所述技术需求的所述第一标签,对所述第一标签与所述一级节点信息进行匹配,筛选出匹配成功的所述第二用户作为初始对象。
为了节省后续处理的时间,减少计算量,对于业务范围不包含本技术需求的技术服务商直接剔除,避免计算能力浪费。
第二筛选步骤:基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;
在本实施例中,上述基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,包括:
从所述解析结果中获取所述初始对象的知识地图的二级节点信息,根据所述二级节点信息及所述第二标签确定所述初始对象的二级相关节点数量,判断所述初始对象的二级相关节点数量是否满足预设条件;
当所述初始对象的二级相关节点数量不满足预设条件时,判断所述初始对象与所述技术需求的匹配度为0;
当所述初始对象的二级相关节点数量满足预设条件时,则从所述解析结果中获取所述初始对象的知识地图的三级节点信息,并根据所述第三节点信息及所述第三标签确定所述初始对象的三级相关节点;及
提取出与所述初始对象的三级相关节点对应的四级节点信息,根据提取出的所述四级节点信息计算所述初始对象与所述技术需求的匹配度。
上述二级相关节点数量为某一个初始对象的知识地图的二级节点信息中与第一用户的第二标签匹配的节点数量。
上述三级相关节点为某一个初始对象的知识地图的三级节点信息中与第一用户的第三标签匹配的节点。
例如,上述预设条件为二级节点相关节点数量大于或等于二级需求节点数量的预设百分比,例如,30%。其中,二级需求节点数量为第一用户的第二标签的数量。上述三级相关节点为与第一用户的第三标签匹配的三级节点。
其中,匹配度为match score(MS),知识地图节点匹配度为point score(PS),不同层级的知识节点为PS2、PS3、PS4。匹配度的计算规则为:
若PS2相关节点数量<需求节点数量*30%,MS=0;
若PS2相关节点数量>=需求节点数量*30%,则MS的计算公式为:
MS=sum(PS3)=sum(sum(PS4))
在本实施例中,PS4的计算公式为:
PS4=α+n*β+m*γ
其中,PS3表示所述初始对象的三级相关节点的匹配度,PS4表示所述初始对象的一个三级相关节点下的一个四级节点的匹配度,α表示起始匹配度,n表示该四级节点下的明星实施案例的数量,β表示每个明星实施案例的匹配度,m表示该四级节点下普通实施案例的数量,γ表示每个普通实施案例的匹配度,n、m为大于或等于0的整数。例如,α=0.5,β=1,γ=0.2。若该末梢节点无实施案例,则PS4=0.5;若该末梢节点每增加一个明星实施案例PS4=0.5+1,每增加一个普通实施案例PS4=0.5+0.2。
需要说明的是,每个节点添加实施案例的数量有限(例如,最多3个),优先取明星实施案例。
推荐步骤:按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
为了便于第一用户甄别,将筛选出的技术服务商按照匹配度的高低顺序依次推荐给技术需求方。
在其他实施例中,所述按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端,包括:
当所述目标对象的数量超过预设阈值时,筛选出预设数量的匹配度排序靠前的目标对象反馈至所述第一用户的客户端。
例如,若匹配度满足要求的目标对象超过10个时,仅取匹配度排序靠前的10个(或者8个)目标对象反馈给第一用户。
在其他实施例中,所述按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端,还包括:
获取第一用户对应的黑名单及白名单,将所述目标对象与所述黑名单进行匹配,从所述目标对象中删除匹配成功的目标对象;或者
当存在至少两个匹配度相同的目标对象时,将所述至少两个匹配度相同的目标对象与所述白名单进行匹配,将匹配成功的目标对象排在匹配失败的目标对象前;或者
当所述目标对象的数量超过预设阈值时,将所述目标对象与所述白名单进行匹配,优先筛选出匹配度靠前且匹配成功的目标对象。
通过比对黑名单与白名单,可以在推荐前排除掉第一用户拉黑过的技术服务商,可以优先推荐第一用户较满意的技术服务商,进而提高用户的使用体验。
在其他实施例中,处理器12执行对象推荐程序10的程序代码时,还实现如下步骤:
定期对所述第二用户的知识地图进行更新保存。
具体地,当一个技术服务商完成一次需求后,基于该次需求生成该技术服务商的一个实施案例,并基于该实施案例对该技术服务商的历史业务数据及知识地图进行更新并保存。
可选地,在其他的实施例中,对象推荐程序10还可以被分割为一个或者多个模块,一个或者多个模块被存储于存储器11中,并由一个或多个处理器(本实施例为处理器12)所执行,以完成本申请,本申请所称的模块是指能够完成特定功能的一系列计算机程序指令段。
参照图4所示,为本申请对象推荐装置的模块示意图,该实施例中,对象推荐装置100可以被分割为模块110-150,且模块110-150所实现的功能或操作步骤均与上文类似,此处不再详述,示例性地,例如其中:
接收模块110,用于接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
标签确定模块120,用于对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
第一筛选模块130,用于从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
第二筛选模块140,用于基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
推荐模块150,用于按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
此外,本申请实施例还提出一种计算机可读存储介质,所述计算机可读存储介质中包括对象推荐程序10,所述对象推荐程序10被处理器执行时实现如上所述的对象推荐方法的步骤。
所述计算机可读存储介质可以是非易失性,也可以是易失性。
本申请之计算机可读存储介质的具体实施方式与上述对象推荐方法的具体实施方式大致相同,在此不再赘述。
上述本申请实施例序号仅仅为了描述,不代表实施例的优劣。
需要说明的是,在本文中,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、装置、物品或者方法不仅包括那些要素,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、装置、物品或者方法所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、装置、物品或者方法中还存在另外的相同要素。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品存储在如上所述的一个存储介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台终端设备(可以是手机,计算机,服务器,或者网络设备等)执行本申请各个实施例所述的方法。
以上仅为本申请的优选实施例,并非因此限制本申请的专利范围,凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换,或直接或间接运用在其它相关的技术领域,均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种对象推荐方法,应用于电子设备,其中,所述方法包括:
    接收步骤:接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
    标签确定步骤:对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
    第一筛选步骤:从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
    第二筛选步骤:基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
    推荐步骤:按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
  2. 根据权利要求1所述的对象推荐方法,其中,所述对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,包括:
    对所述需求描述信息进行分词,得到所述需求描述信息对应的词语集合;及
    将所述词语集合中各词语与预设关键词集合进行匹配,将与所述预设关键词集合中的关键词匹配的词语作为所述需求描述信息对应的关键词。
  3. 根据权利要求1所述的对象推荐方法,其中,所述根据所述关键词确定所述技术需求对应的第二标签及第三标签,包括:
    从预设存储路径获取预设词语与标签的映射数据,判断所述映射数据中是否存在与所述关键词匹配的词语;及
    若存在,则将所述关键词匹配的词语对应的标签作为所述关键词对应的标签,并将所述关键词对应的标签作为所述技术需求对应的第二标签及第三标签。
  4. 根据权利要求1所述的对象推荐方法,其中,所述预设第二用户的知识地图通过以下步骤确定:
    从预设数据库中获取所述预设第二用户预设时间间隔内的历史业务数据,根据所述历史业务数据确定所述预设第二用户的第一预设类型数据、第二预设类型数据、第三预设类型数据及第四预设类型数据;
    根据所述第一预设类型数据确定一级节点信息,根据所述第二预设类型数据确定二级节点信息,根据所述第三预设类型数据确定三级节点信息,根据所述第四预设类型数据确定四级节点信息;及
    基于所述一级节点信息、二级节点信息、三级节点信息及四级节点信息的包含关系生成所述预设第二用户的知识地图。
  5. 根据权利要求1至4中任意一项所述的对象推荐方法,其中,所述基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,包括:
    从所述解析结果中获取所述初始对象的知识地图的二级节点信息,根据所述二级节点信息及所述第二标签确定所述初始对象的二级相关节点数量,判断所述初始对象的二级相关节点数量是否满足预设条件;
    当所述初始对象的二级相关节点数量不满足预设条件时,判断所述初始对象与所述技术需求的匹配度为0;
    当所述初始对象的二级相关节点数量满足预设条件时,则从所述解析结果中获取所述 初始对象的知识地图的三级节点信息,并根据所述第三节点信息及所述第三标签确定所述初始对象的三级相关节点;及
    提取出与所述初始对象的三级相关节点对应的四级节点信息,根据提取出的所述四级节点信息计算所述初始对象与所述技术需求的匹配度,匹配度的计算公式为:
    MS=sum(PS3)=sum(sum(PS4))
    PS4=α+n*β+m*γ
    其中,PS3表示所述初始对象的三级相关节点的匹配度,PS4表示所述初始对象的一个三级相关节点下的一个四级节点的匹配度,α表示起始匹配度,n表示该四级节点下的明星实施案例的数量,β表示每个明星实施案例的匹配度,m表示该四级节点下普通实施案例的数量,γ表示每个普通实施案例的匹配度,n、m为大于或等于0的整数。
  6. 根据权利要求1所述的对象推荐方法,其中,所述对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象,包括:
    对所述预设第二用户的知识地图进行解析,分别确定知识地图的一级节点信息、二级节点信息、三级节点信息及四级节点信息;及
    获取所述预设第二用户的知识地图的一级节点信息,从所述技术需求的标签集合中获取所述技术需求的所述第一标签,对所述第一标签与所述一级节点信息进行匹配,筛选出匹配成功的所述第二用户作为初始对象。
  7. 根据权利要求1所述的对象推荐方法,其中,所述按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端,包括:
    当所述目标对象的数量超过预设阈值时,筛选出预设数量的匹配度排序靠前的目标对象反馈至所述第一用户的客户端。
  8. 根据权利要求4所述的对象推荐方法,其中,所述对象推荐方法还包括:
    定期对所述第二用户的知识地图进行更新保存。
  9. 一种电子设备,其中,该电子设备包括:存储器、处理器,所述存储器上存储有可在所述处理器上运行的对象推荐程序,所述对象推荐程序被所述处理器执行时,可实现如下所述的对象推荐方法的步骤:
    接收步骤:接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
    标签确定步骤:对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
    第一筛选步骤:从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
    第二筛选步骤:基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
    推荐步骤:按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
  10. 根据权利要求9所述的电子设备,其中,所述对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,包括:
    对所述需求描述信息进行分词,得到所述需求描述信息对应的词语集合;及
    将所述词语集合中各词语与预设关键词集合进行匹配,将与所述预设关键词集合中的关键词匹配的词语作为所述需求描述信息对应的关键词。
  11. 根据权利要求9所述的电子设备,其中,所述根据所述关键词确定所述技术需求 对应的第二标签及第三标签,包括:
    从预设存储路径获取预设词语与标签的映射数据,判断所述映射数据中是否存在与所述关键词匹配的词语;及
    若存在,则将所述关键词匹配的词语对应的标签作为所述关键词对应的标签,并将所述关键词对应的标签作为所述技术需求对应的第二标签及第三标签。
  12. 根据权利要求9所述的电子设备,其中,所述预设第二用户的知识地图通过以下步骤确定:
    从预设数据库中获取所述预设第二用户预设时间间隔内的历史业务数据,根据所述历史业务数据确定所述预设第二用户的第一预设类型数据、第二预设类型数据、第三预设类型数据及第四预设类型数据;
    根据所述第一预设类型数据确定一级节点信息,根据所述第二预设类型数据确定二级节点信息,根据所述第三预设类型数据确定三级节点信息,根据所述第四预设类型数据确定四级节点信息;及
    基于所述一级节点信息、二级节点信息、三级节点信息及四级节点信息的包含关系生成所述预设第二用户的知识地图。
  13. 根据权利要求9至12中任意一项所述的电子设备,其中,所述基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,包括:
    从所述解析结果中获取所述初始对象的知识地图的二级节点信息,根据所述二级节点信息及所述第二标签确定所述初始对象的二级相关节点数量,判断所述初始对象的二级相关节点数量是否满足预设条件;
    当所述初始对象的二级相关节点数量不满足预设条件时,判断所述初始对象与所述技术需求的匹配度为0;
    当所述初始对象的二级相关节点数量满足预设条件时,则从所述解析结果中获取所述初始对象的知识地图的三级节点信息,并根据所述第三节点信息及所述第三标签确定所述初始对象的三级相关节点;及
    提取出与所述初始对象的三级相关节点对应的四级节点信息,根据提取出的所述四级节点信息计算所述初始对象与所述技术需求的匹配度,匹配度的计算公式为:
    MS=sum(PS3)=sum(sum(PS4))
    PS4=α+n*β+m*γ
    其中,PS3表示所述初始对象的三级相关节点的匹配度,PS4表示所述初始对象的一个三级相关节点下的一个四级节点的匹配度,α表示起始匹配度,n表示该四级节点下的明星实施案例的数量,β表示每个明星实施案例的匹配度,m表示该四级节点下普通实施案例的数量,γ表示每个普通实施案例的匹配度,n、m为大于或等于0的整数。
  14. 根据权利要求9所述的电子设备,其中,所述对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象,包括:
    对所述预设第二用户的知识地图进行解析,分别确定知识地图的一级节点信息、二级节点信息、三级节点信息及四级节点信息;及
    获取所述预设第二用户的知识地图的一级节点信息,从所述技术需求的标签集合中获取所述技术需求的所述第一标签,对所述第一标签与所述一级节点信息进行匹配,筛选出匹配成功的所述第二用户作为初始对象。
  15. 一种计算机可读存储介质,其中,所述计算机可读存储介质中包括对象推荐程序,所述对象推荐程序被处理器执行时,可实现如下所述的对象推荐方法的步骤:
    接收步骤:接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需 求对应的第一标签及需求描述信息;
    标签确定步骤:对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
    第一筛选步骤:从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
    第二筛选步骤:基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
    推荐步骤:按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
  16. 根据权利要求15所述的计算机可读存储介质,其中,所述对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,包括:
    对所述需求描述信息进行分词,得到所述需求描述信息对应的词语集合;及
    将所述词语集合中各词语与预设关键词集合进行匹配,将与所述预设关键词集合中的关键词匹配的词语作为所述需求描述信息对应的关键词。
  17. 根据权利要求15所述的计算机可读存储介质,其中,所述根据所述关键词确定所述技术需求对应的第二标签及第三标签,包括:
    从预设存储路径获取预设词语与标签的映射数据,判断所述映射数据中是否存在与所述关键词匹配的词语;及
    若存在,则将所述关键词匹配的词语对应的标签作为所述关键词对应的标签,并将所述关键词对应的标签作为所述技术需求对应的第二标签及第三标签。
  18. 根据权利要求15至17中任意一项所述的计算机可读存储介质,其中,所述基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,包括:
    从所述解析结果中获取所述初始对象的知识地图的二级节点信息,根据所述二级节点信息及所述第二标签确定所述初始对象的二级相关节点数量,判断所述初始对象的二级相关节点数量是否满足预设条件;
    当所述初始对象的二级相关节点数量不满足预设条件时,判断所述初始对象与所述技术需求的匹配度为0;
    当所述初始对象的二级相关节点数量满足预设条件时,则从所述解析结果中获取所述初始对象的知识地图的三级节点信息,并根据所述第三节点信息及所述第三标签确定所述初始对象的三级相关节点;及
    提取出与所述初始对象的三级相关节点对应的四级节点信息,根据提取出的所述四级节点信息计算所述初始对象与所述技术需求的匹配度,匹配度的计算公式为:
    MS=sum(PS3)=sum(sum(PS4))
    PS4=α+n*β+m*γ
    其中,PS3表示所述初始对象的三级相关节点的匹配度,PS4表示所述初始对象的一个三级相关节点下的一个四级节点的匹配度,α表示起始匹配度,n表示该四级节点下的明星实施案例的数量,β表示每个明星实施案例的匹配度,m表示该四级节点下普通实施案例的数量,γ表示每个普通实施案例的匹配度,n、m为大于或等于0的整数。
  19. 根据权利要求15所述的计算机可读存储介质,其中,所述对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象,包括:
    对所述预设第二用户的知识地图进行解析,分别确定知识地图的一级节点信息、二级 节点信息、三级节点信息及四级节点信息;及
    获取所述预设第二用户的知识地图的一级节点信息,从所述技术需求的标签集合中获取所述技术需求的所述第一标签,对所述第一标签与所述一级节点信息进行匹配,筛选出匹配成功的所述第二用户作为初始对象.
  20. 一种对象推荐装置,其中,所述装置包括:
    接收模块,用于接收第一用户通过客户端发出的技术需求,所述技术需求中包括该技术需求对应的第一标签及需求描述信息;
    标签确定模块,用于对所述需求描述信息进行解析,确定所述需求描述信息对应的关键词,并根据所述关键词确定所述技术需求对应的第二标签及第三标签,生成所述技术需求的标签集合;
    第一筛选模块,用于从预设存储路径中获取预设第二用户的知识地图,对所述预设第二用户的知识地图进行解析得到解析结果,并根据所述解析结果从预设第二用户中筛选出与所述技术需求的所述第一标签匹配的第二用户作为初始对象;
    第二筛选模块,用于基于预设匹配度计算规则、所述初始对象的知识地图的解析结果及所述技术需求对应的第二标签及第三标签,计算所述初始对象与所述技术需求的匹配度,筛选出匹配度大于或等于预设阈值的初始对象,作为目标对象;及
    推荐模块,用于按照匹配度高低顺序将所述目标对象反馈至所述第一用户的客户端。
PCT/CN2020/106016 2020-04-01 2020-07-30 对象推荐方法、电子设备及存储介质 WO2021196476A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010249649.2 2020-04-01
CN202010249649.2A CN111552870A (zh) 2020-04-01 2020-04-01 对象推荐方法、电子装置及存储介质

Publications (1)

Publication Number Publication Date
WO2021196476A1 true WO2021196476A1 (zh) 2021-10-07

Family

ID=72007374

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/106016 WO2021196476A1 (zh) 2020-04-01 2020-07-30 对象推荐方法、电子设备及存储介质

Country Status (2)

Country Link
CN (1) CN111552870A (zh)
WO (1) WO2021196476A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113744066A (zh) * 2021-09-01 2021-12-03 杭州隆埠科技有限公司 信息推送方法及装置
CN114925289A (zh) * 2022-05-23 2022-08-19 中国平安财产保险股份有限公司 员工推荐方法、装置、设备及可读存储介质
CN117495142A (zh) * 2023-11-18 2024-02-02 北京连华永兴科技发展有限公司 一种企业水处理方案推荐方法及系统
CN117519483A (zh) * 2023-11-22 2024-02-06 艾迪普科技股份有限公司 基于数字虚拟的媒体动态交互方法、系统和介质
CN117708412A (zh) * 2023-10-31 2024-03-15 灵犀科技有限公司 一种企业动态推送方法、装置、电子设备和可读介质

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112100524B (zh) * 2020-09-17 2024-10-18 北京百度网讯科技有限公司 一种信息推荐的方法、装置、设备以及存储介质
CN112288406B (zh) * 2020-11-04 2022-03-01 深圳崎点数据有限公司 一种基于科技信息众创平台的服务对象检索方法
CN112598471A (zh) * 2020-12-25 2021-04-02 北京知因智慧科技有限公司 一种产品的推荐方法、推荐装置及电子设备
CN112948705A (zh) * 2020-12-30 2021-06-11 广州博士信息技术研究院有限公司 一种基于政策大数据的智能匹配方法、装置及介质
CN112784155A (zh) * 2020-12-31 2021-05-11 北京橙色云科技有限公司 匹配服务方的方法、装置以及存储介质
CN112686559B (zh) * 2021-01-06 2023-05-19 郑州铁路职业技术学院 基于大数据和人工智能的成果转化在线供需匹配方法
CN114358814B (zh) * 2021-11-29 2024-10-18 国网北京市电力公司 数据处理方法、装置、电子设备及计算机可读存储介质
CN113934901B (zh) * 2021-12-17 2022-09-23 北京因数健康科技有限公司 基于行为认知的信息匹配方法、装置、设备、系统及介质
CN116304277B (zh) * 2023-03-01 2023-12-15 张素愿 一种基于ai的智能匹配方法、系统和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106997341A (zh) * 2017-03-22 2017-08-01 山东大学 一种创新方案匹配方法、装置、服务器及系统
CN107861971A (zh) * 2017-09-15 2018-03-30 广州唯品会研究院有限公司 一种商品搜索方法和装置
CN108846679A (zh) * 2018-05-22 2018-11-20 北京易知创新数据科技有限公司 一种应用于b2b企业服务的客户推荐方法
CN110443666A (zh) * 2018-05-04 2019-11-12 阿里巴巴集团控股有限公司 信息处理方法、装置、终端及服务器

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105989549A (zh) * 2015-02-06 2016-10-05 白忠泽 一种基于移动互联网的生活服务功能的方法和系统
CN106372087B (zh) * 2015-07-23 2019-12-13 北京大学 一种面向信息检索的信息地图生成方法及其动态更新方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106997341A (zh) * 2017-03-22 2017-08-01 山东大学 一种创新方案匹配方法、装置、服务器及系统
CN107861971A (zh) * 2017-09-15 2018-03-30 广州唯品会研究院有限公司 一种商品搜索方法和装置
CN110443666A (zh) * 2018-05-04 2019-11-12 阿里巴巴集团控股有限公司 信息处理方法、装置、终端及服务器
CN108846679A (zh) * 2018-05-22 2018-11-20 北京易知创新数据科技有限公司 一种应用于b2b企业服务的客户推荐方法

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113744066A (zh) * 2021-09-01 2021-12-03 杭州隆埠科技有限公司 信息推送方法及装置
CN113744066B (zh) * 2021-09-01 2024-04-02 杭州隆埠科技有限公司 信息推送方法及装置
CN114925289A (zh) * 2022-05-23 2022-08-19 中国平安财产保险股份有限公司 员工推荐方法、装置、设备及可读存储介质
CN114925289B (zh) * 2022-05-23 2024-04-30 中国平安财产保险股份有限公司 员工推荐方法、装置、设备及可读存储介质
CN117708412A (zh) * 2023-10-31 2024-03-15 灵犀科技有限公司 一种企业动态推送方法、装置、电子设备和可读介质
CN117495142A (zh) * 2023-11-18 2024-02-02 北京连华永兴科技发展有限公司 一种企业水处理方案推荐方法及系统
CN117519483A (zh) * 2023-11-22 2024-02-06 艾迪普科技股份有限公司 基于数字虚拟的媒体动态交互方法、系统和介质

Also Published As

Publication number Publication date
CN111552870A (zh) 2020-08-18

Similar Documents

Publication Publication Date Title
WO2021196476A1 (zh) 对象推荐方法、电子设备及存储介质
US11281860B2 (en) Method, apparatus and device for recognizing text type
US10348550B2 (en) Method and system for processing network media information
US8949243B1 (en) Systems and methods for determining a rating for an item from user reviews
CN111210335B (zh) 用户风险识别方法、装置及电子设备
CN109245996B (zh) 邮件推送方法、装置、计算机设备和存储介质
US9384278B2 (en) Methods and systems for assessing excessive accessory listings in search results
WO2008144444A1 (en) Ranking online advertisements using product and seller reputation
US10002187B2 (en) Method and system for performing topic creation for social data
KR20150016973A (ko) 탐색 결과들을 생성하는 방법
US20190220909A1 (en) Collaborative Filtering to Generate Recommendations
WO2019179030A1 (zh) 产品购买预测方法、服务器及存储介质
CN113297287B (zh) 用户策略自动部署方法、装置及电子设备
WO2018171295A1 (zh) 一种给文章标注标签的方法、装置、终端及计算机可读存储介质
CN111966886A (zh) 对象推荐方法、对象推荐装置、电子设备及存储介质
US20190050229A1 (en) Similarity scores of rules in information technology workflows
WO2020164204A1 (zh) 文本模板识别方法、装置及计算机可读存储介质
CN111310032A (zh) 资源推荐方法、装置、计算机设备及可读存储介质
US20230259991A1 (en) Machine learning text interpretation model to determine customer scenarios
KR102299525B1 (ko) 제품 평가 마이닝 방법 및 이를 수행하는 장치
CN114036921A (zh) 一种政策信息匹配方法和装置
CN111199454A (zh) 实时用户转化评估方法、装置及电子设备
US20150235281A1 (en) Categorizing data based on cross-category relevance
CN116048463A (zh) 基于标签管理的需求项内容智能推荐方法及装置
US20110208738A1 (en) Method for Determining an Enhanced Value to Keywords Having Sparse Data

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20928288

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

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

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 19/01/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20928288

Country of ref document: EP

Kind code of ref document: A1