A method and device for recommending digital solutions based on tags
By designing label models for users and digital solutions, calculating the weight value of the solution sub-labels, filtering and sorting, the problem of insufficient comprehensive recommendation results in the existing technology is solved, and the retrieval efficiency and accuracy of digital solutions are improved.
Patent Information
- Application Number
- CN202210157528.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-21
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2042-02-21
AI Technical Summary
The existing digital solution recommendation methods and systems lack comprehensive consideration of user behavior, preferences and solutions themselves, resulting in insufficient comprehensive recommendation results, affecting user retrieval efficiency and time cost.
By designing a tag model for users and digital solutions, combining user tags, solution tags and associated interactive tags, the program sub-label weight value is calculated, and recommended solutions that meet user expectations are selected, and sorted according to quality performance scores and relevance.
It realizes rapid and accurate screening of digital solutions that meet user expectations, improves search efficiency and the integrity of recommended solutions, and ensures the accuracy and relevance of the solution content.
Smart Images

Figure CN114610776B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing, and in particular to a tag-based digital solution recommendation method and device. Background Art
[0002] Currently, digital solutions exist in documents scattered in various storage systems, and cannot provide a complete solution for other users. Moreover, the relationship between digital solutions and users is a pull relationship. When users need to call or find a certain digital solution, they need to manually search the project results library when needed. For companies whose project results are not uniformly archived, it is even more difficult for digital practitioners to find the digital solution results they want.
[0003] Moreover, the existing personalized solution recommendation methods and system recommendation algorithms do not take into account comprehensive factors. They only consider the user's basic information and problem information, lack consideration of dynamic factors such as user behavior and preferences, and lack consideration of factors related to the solution itself and factors related to the interaction between users and solutions. As a result, the solutions ultimately recommended to users are not comprehensive enough, or the content of the solutions does not meet user expectations, which affects the user's search time and increases time costs. Summary of the invention
[0004] The embodiment of the present invention provides a tag-based digital solution recommendation method and device, which can quickly and accurately screen out recommended solutions that meet user expectations by modeling solutions and designing labels.
[0005] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application provides a tag-based digital solution recommendation method, the method comprising:
[0006] Attach a corresponding user label to each user according to the user label model, and attach a corresponding solution label to each digital solution according to the solution label model;
[0007] Combining the user tag and the solution tag, and attaching an associated basic tag and an associated interactive tag to each digital solution according to an associated tag model;
[0008] Obtain the target user's digital solution access history based on the associated basic tags and associated interactive tags corresponding to each digital solution;
[0009] According to the target user's digital solution access history, multiple solution sub-tag values corresponding to each accessed digital solution are counted, and a solution sub-tag weight value corresponding to each accessed digital solution is calculated according to a preset matching rule; the accessed digital solution is a digital solution accessed by the target user; the solution sub-tag weight value is calculated based on the occurrence frequency of all solution sub-tag values;
[0010] According to the weight value of the solution sub-label corresponding to each accessed digital solution, a preset number of digital solutions are selected as multiple recommended solutions.
[0011] In a possible implementation manner of the first aspect, a method for calculating the weight value of the scheme sub-label is as follows:
[0012] Count the values of multiple solution sub-tags corresponding to each accessed digital solution;
[0013] According to the multiple scheme sub-tag values, counting the number of occurrences of each scheme sub-tag value;
[0014] According to the order of the number of occurrences of each solution sub-label value from large to small, a preset number of solution sub-label values are selected as a label reference value set;
[0015] Calculate the frequency of occurrence of each solution sub-label value according to the number of occurrences of each solution sub-label value and the label reference value set;
[0016] The scheme subtag weight value is calculated according to the occurrence frequency of each scheme subtag value and the hit value of each scheme subtag value; the hit value of each scheme subtag value depends on the mode of various scheme subtag values in the tag reference value set.
[0017] In a possible implementation manner of the first aspect, after selecting a preset number of digital solutions as the plurality of recommended solutions, the method further includes:
[0018] The recommended solutions are sorted in descending order according to the weight values of the solution sub-tags corresponding to each accessed digital solution.
[0019] In a possible implementation manner of the first aspect, the associated interaction tags include visit tags, like tags, follow tags, and comment tags.
[0020] In a possible implementation manner of the first aspect, after selecting a preset number of digital solutions as the plurality of recommended solutions, the method further includes:
[0021] Obtain access tags corresponding to multiple recommended solutions and count the number of user visits to each recommended solution within a unit time.
[0022] Calculate the correlation coefficients among the multiple recommendation solutions according to the number of user visits to each recommendation solution within the unit time;
[0023] The plurality of recommended solutions are sorted in descending order of the values of the correlation coefficients between the solutions.
[0024] In a possible implementation manner of the first aspect, after selecting a preset number of digital solutions as the plurality of recommended solutions, the method further includes:
[0025] Get the visit tags, like tags, follow tags and comment tags corresponding to each recommendation plan, and calculate the effective reading number, total reading time, average user reading time, number of replays, number of likes, number of comments and number of favorites for each recommendation plan;
[0026] Obtain the solution label of each recommended solution and evaluate the basic score of each recommended solution according to the preset standard;
[0027] Calculate the quality performance score of each solution based on the number of effective readings, total reading time, average user reading time, number of replays, number of likes, number of comments, number of favorites and basis of each recommended solution;
[0028] The multiple recommended solutions are sorted in descending order according to the quality performance score corresponding to each solution.
[0029] In a possible implementation manner of the first aspect, the preset standard includes a business analysis domain, a solution domain, and an effectiveness value domain;
[0030] The business analysis domain, the solution domain, and the effectiveness value domain are all defined according to the digital solution lifecycle for the electric power industry.
[0031] In a possible implementation of the first aspect, the business analysis domain includes a solution overview, business status, business processes, and business pain points;
[0032] The solution domain includes solution objectives and overall solutions;
[0033] The effectiveness value domain includes: system operation analysis, solution effectiveness and value analysis.
[0034] In a possible implementation of the first aspect, a key technology tag corresponding to each recommended solution is obtained, recommended solutions with the same key technology tag value are selected, and other recommended solutions are excluded; the solution tags include key technology tags.
[0035] A second aspect of an embodiment of the present application provides a tag-based digital solution recommendation device, including:
[0036] A user label module, used to label each user with a corresponding user label according to the user label model, and to label each digital solution with a corresponding solution label according to the solution label model;
[0037] An interactive label module, used to combine the user label and the solution label, and attach an associated basic label and an associated interactive label to each digital solution according to an associated label model;
[0038] An access history module, used to obtain the target user's digital solution access history according to the associated basic tags and associated interactive tags corresponding to each digital solution;
[0039] A weight calculation module is used to count multiple solution sub-tag values corresponding to each visited digital solution according to the digital solution access history of the target user, and calculate the solution sub-tag weight value corresponding to each visited digital solution according to a preset matching rule; the visited digital solution is a digital solution visited by the target user; the solution sub-tag weight value is calculated based on the occurrence frequency of all solution sub-tag values;
[0040] The solution selection module selects a preset number of digital solutions as multiple recommended solutions according to the weight values of the solution sub-labels corresponding to each accessed digital solution.
[0041] The embodiments of the present invention provide a label-based digital solution recommendation method and device, which establishes connections between digital solutions distributed in various locations and various users by attaching solution labels and associated interactive labels to each digital solution in the database, and evaluates the effectiveness of each digital solution based on the solution label and preset standards to ensure the solution integrity of the recommended solution.
[0042] When screening recommended solutions, a preset number of digital solutions are selected as multiple recommended solutions according to the weight values of the solution sub-tags corresponding to each accessed digital solution. Since the weight values of the solution sub-tags can reflect the relevance of the digital solution to other solutions and the accuracy of the content, screening recommended solutions based on the weight values of the solution sub-tags can ensure that the solution content meets the user's expectations while ensuring that relevant solutions are not missed, because the weight values of the solution sub-tags are obtained by counting the frequency of occurrence of each solution sub-tag representing the solution content.
[0043] After the screening is completed, it can also be sorted according to user needs, using the solution's quality performance score and relevance as criteria, placing search results that meet user expectations at the front, further improving user search efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 It is a flowchart of a tag-based digital solution recommendation method provided by an embodiment of the present invention;
[0045] Figure 2 It is a schematic diagram of the composition of a user tag model provided by an embodiment of the present invention;
[0046] Figure 3 It is a schematic diagram of the composition of a solution label model provided by an embodiment of the present invention;
[0047] Figure 4 It is a schematic diagram of the composition of an association tag model provided by an embodiment of the present invention;
[0048] Figure 5 A schematic diagram of a sub-label weight value calculation process of a solution provided by an embodiment of the present invention;
[0049] Figure 6 It is a schematic diagram of the composition of a preset standard provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0051] See also Figure 1 An embodiment of the present invention provides a tag-based digital solution recommendation method, the method comprising:
[0052] S10. A corresponding user label is attached to each user according to the user label model, and a corresponding solution label is attached to each digital solution according to the solution label model.
[0053] S11. Combining the user tag and the solution tag, attaching an associated basic tag and an associated interactive tag to each digital solution according to an associated tag model.
[0054] S12. Obtain the target user's digital solution access history based on the associated basic tags and associated interactive tags corresponding to each digital solution.
[0055] S13. According to the digital solution access history of the target user, multiple solution sub-tag values corresponding to each visited digital solution are counted, and a solution sub-tag weight value corresponding to each visited digital solution is calculated according to a preset matching rule; the visited digital solution is a digital solution visited by the target user; the solution sub-tag weight value is calculated based on the occurrence frequency of all solution sub-tag values.
[0056] S14: Select a preset number of digital solutions as multiple recommended solutions according to the weight values of the solution sub-labels corresponding to each accessed digital solution.
[0057] Different from the existing digital solution recommendation scheme, this embodiment attaches a solution label representing the properties of the solution itself and an association label representing the properties of interaction with the user to the solution in S10-S12. By attaching the association label, solutions distributed in various places are connected with users, and the solution label itself can make it easy for users to see the basic content of the solution.
[0058] When labeling users, it is necessary to base it on the user label model, such as Figure 2 shown. Figure 2 This is a user label model adopted in this embodiment, which includes: basic labels, preference labels, behavior labels, and capability labels. Among them, user basic labels include user ID, name, gender, department, age, responsibilities, supervisor's work number, subordinate's work number; user capability labels include education, major, professional title, length of service in this major, previous major, honors, certificates and patents; user preference labels include channel preference, access path preference, business domain preference, capability domain preference and scenario preference; user behavior labels include last login time, number of logins in the past 30 days, common login time period, active status, interaction enthusiasm and terminal usage time.
[0059] When labeling solutions, we need to base it on the solution label model, such as Figure 3 shown. Figure 3It is a solution label model adopted in this embodiment, which includes basic labels, subordinate labels, interactive labels and key technical attribute labels. Among them, the basic labels of the solution include solution ID, solution name, business pain points, business person in charge, technical person in charge, solution conversion time, year, department, working group, related solutions, and system name; the subordinate labels of the solution include the scenario, the business domain, the department, and the construction unit; the evaluation labels of the solution include the number of likes, the number of comments, the number of collections, whether it is an excellent case, the degree of attention, the number of visits, the number of visits in the past month, the number of visits to this department, the number of visits to other departments, the number of attention to this business domain, and the number of attention to other business domains; the key technical labels of the solution include whether it involves artificial intelligence, whether it involves spatiotemporal maps, whether it involves RPA, whether it involves data analysis, whether it involves the Internet of Things, and whether it involves blockchain.
[0060] When attaching association tags (association basic tags and association interaction tags) to the scheme, it is necessary to base it on the scheme tag model, such as Figure 4 shown. Figure 4 This is a solution-related tag model adopted in this embodiment, including basic tags, access tags, like tags, attention tags and evaluation tags. Among them, the basic tags include the visitor ID, the visited solution ID, the access time, whether to pay attention to the solution, the number of visits in the past week, the access path, and the access channel; the conversion tag includes the converter ID, the solution ID, and the conversion time; the like tag includes the liker ID, the solution ID, the like time, and whether to cancel; the attention tag includes the attention time, whether to cancel, and the number of personal attention entries; the comment tag includes the evaluator ID, the solution ID, the comment time, the comment content, the hit keyword, and the number of comments.
[0061] It should be noted that the above user tag model, solution tag model and association tag model are not unique, and the attributes and types of each sub-tag in their models can be adjusted according to business scenarios. The one given here is just a more practical general model.
[0062] After labeling users and digital solutions, when selecting recommended solutions for users, we can use this as a basis and combine the label weight value to quickly filter out digital solutions related to the user's search target. The calculation of solution label weight value is essentially based on the similarity calculation between solutions. The following is an example of a weight calculation method.
[0063] Exemplarily, the weight value of the scheme sub-label is calculated as follows:
[0064] Count the values of multiple solution sub-tags corresponding to each accessed digital solution.
[0065] According to the multiple scheme sub-tag values, the number of occurrences of each scheme sub-tag value is counted.
[0066] According to the order of the number of occurrences of each solution sub-label value from large to small, a preset number of solution sub-label values are selected as a label reference value set.
[0067] The occurrence frequency of each solution sub-label value is calculated according to the number of occurrences of each solution sub-label value and the label reference value set.
[0068] A scheme subtag weight value is calculated according to the occurrence frequency of each scheme subtag value and the hit value of each scheme subtag value; the hit value of each scheme subtag value depends on whether the scheme subtag value exists in the mode of various scheme subtag values in the tag reference value set.
[0069] See also Figure 5 . In this example, after analyzing the historical access records of the target users, the sub-tag values contained in the solution labels of each visited digital solution are listed first. In the figure, "label value 1-1" means that the first seed label value is 1, "label value 1-2" means that the first seed label value is 2, "label value 2-2" means that the second seed label value is 2, and so on: "label value XY", X represents the sub-tag type, and Y represents the sub-tag value. It should be noted that the historical access records used here are obtained by statistics based on the associated basic tags and associated interactive tags corresponding to each digital solution.
[0070] Assuming that the user has browsed n cases in the past, and each solution has m sub-label attributes, an n*m label matrix is constructed. Then the number of times the label attribute value of each column of the matrix appears is calculated, and the record is stored. In this embodiment, the number of occurrences of each solution sub-label value is counted, and the solution sub-label values of a preset number of types (the preset number here is 5, which means selecting 5 sub-labels, 5 types) are selected as the label reference value set. Calculate the proportion of the top five label values in the label reference value set. The specific calculation method is: the frequency of occurrence of this label value / the sum of the frequencies of occurrence of the top five label values. In this embodiment, the number of elements in the label reference value set is the sum of the frequencies of occurrence of the top five label values.
[0071] Calculate the weights of the top 5 label values of all solutions to be recommended. The calculation formula is ∑ label weight * whether the label value is hit (hit is 1, otherwise it is 0). Because the top 5 label values in each solution are different, take the first label value as an example. The label value of solution 1 is 1. The mode of the first label value in the label reference value set is 1, which is the same as the mode, so the hit value is 1. The label value of solution 2 is 2, and the mode of the first label value in the label reference value set is 1, so the hit value is 0.
[0072] Exemplarily, after selecting a preset number of digital solutions as the plurality of recommended solutions, the method further includes:
[0073] The recommended solutions are sorted in descending order according to the weight values of the solution sub-tags corresponding to each accessed digital solution.
[0074] Exemplarily, the associated interactive tags include visit tags, like tags, follow tags and comment tags.
[0075] Exemplarily, after selecting a preset number of digital solutions as the plurality of recommended solutions, the method further includes:
[0076] Obtain access tags corresponding to multiple recommended solutions and count the number of user visits to each recommended solution within a unit time.
[0077] Calculate the correlation coefficients among the multiple recommendation solutions according to the number of user visits to each recommendation solution within the unit time;
[0078] The plurality of recommended solutions are sorted in descending order of the values of the correlation coefficients between the solutions.
[0079] Assuming that there are N(x) users browsing the current solution x and N(y) users browsing the solution y, then the correlation coefficient between solutions x and y is Finally, solutions are recommended to users based on the correlation coefficient from high to low.
[0080] Exemplarily, after selecting a preset number of digital solutions as the plurality of recommended solutions, the method further includes:
[0081] Get the visit tags, like tags, follow tags and comment tags corresponding to each recommendation plan, and calculate the effective reading number, total reading time, average user reading time, number of replays, number of likes, number of comments and number of favorites for each recommendation plan;
[0082] Obtain the solution label of each recommended solution and evaluate the basic score of each recommended solution according to the preset standard;
[0083] Calculate the quality performance score of each solution based on the number of effective readings, total reading time, average user reading time, number of replays, number of likes, number of comments, number of favorites and basis of each recommended solution;
[0084] The multiple recommended solutions are sorted in descending order according to the quality performance score corresponding to each solution.
[0085] Take a digital solution with a basic score of 60 as an example. The corresponding effective reading number B, total reading time (minutes) C, average user reading time (minutes) D, replay number E, like number F, comment number G, and favorite number H are added up to get the solution quality performance score. The calculation formula is: Q = A + 5B + C + D + 3E + 2F + G + 2H. It should be noted that the examples of A, B, C, D, E, and G in the above formula can be set according to actual needs.
[0086] See also Figure 6 , exemplarily, the preset standards include a business analysis domain, a solution domain, and an effectiveness value domain.
[0087] The business analysis domain, the solution domain, and the effectiveness value domain are all defined according to the digital solution lifecycle for the electric power industry.
[0088] Exemplarily, the business analysis domain includes solution overview, business status, business process and business pain points;
[0089] The solution domain includes solution objectives and overall solutions;
[0090] The effectiveness value domain includes: system operation analysis, solution effectiveness and value analysis.
[0091] Exemplarily, the key technology tag corresponding to each recommended solution is obtained, the recommended solutions with the same key technology tag value are selected, and other recommended solutions are excluded; the solution tags include the key technology tags.
[0092] Using key technology tags for secondary screening can further narrow the scope of recommendation results and reduce users' reading and retrieval time.
[0093] The embodiments of the present invention provide a label-based digital solution recommendation method and device, which establishes connections between digital solutions distributed in various locations and various users by attaching solution labels and associated interactive labels to each digital solution in the database, and evaluates the effectiveness of each digital solution based on the solution label and preset standards to ensure the solution integrity of the recommended solution.
[0094] When screening recommended solutions, a preset number of digital solutions are selected as multiple recommended solutions according to the weight values of the solution sub-tags corresponding to each accessed digital solution. Since the weight values of the solution sub-tags can reflect the relevance of the digital solution to other solutions and the accuracy of the content, screening recommended solutions based on the weight values of the solution sub-tags can ensure that the solution content meets the user's expectations while ensuring that relevant solutions are not missed, because the weight values of the solution sub-tags are obtained by counting the frequency of occurrence of each solution sub-tag representing the solution content.
[0095] After the screening is completed, it can also be sorted according to user needs, using the solution's quality performance score and relevance as criteria, placing search results that meet user expectations at the front, further improving user search efficiency.
[0096] A second aspect of an embodiment of the present application provides a tag-based digital solution recommendation device, including: a user tag module, an interactive tag module, an access history module, a weight calculation module and a solution selection module.
[0097] The user label module is used to attach a corresponding user label to each user according to the user label model, and to attach a corresponding solution label to each digital solution according to the solution label model.
[0098] The interactive label module is used to combine the user label and the solution label and attach an associated basic label and an associated interactive label to each digital solution according to an associated label model.
[0099] The access history module is used to obtain the target user's digital solution access history based on the associated basic tags and associated interactive tags corresponding to each digital solution.
[0100] A weight calculation module is used to count multiple solution sub-tag values corresponding to each visited digital solution based on the digital solution access history of the target user, and calculate the solution sub-tag weight value corresponding to each visited digital solution according to a preset matching rule; the visited digital solution is a digital solution visited by the target user; the solution sub-tag weight value is calculated based on the occurrence frequency of all solution sub-tag values.
[0101] The solution selection module selects a preset number of digital solutions as multiple recommended solutions according to the weight values of the solution sub-labels corresponding to each accessed digital solution.
[0102] An embodiment of the present invention provides a label-based digital solution recommendation device, which establishes connections between digital solutions distributed in various locations and various users by attaching solution labels and associated interactive labels to each digital solution in the database, and evaluates the effectiveness of each digital solution based on the solution label and preset standards to ensure the solution integrity of the recommended solution.
[0103] When screening recommended solutions, a preset number of digital solutions are selected as multiple recommended solutions according to the weight values of the solution sub-tags corresponding to each accessed digital solution. Since the weight values of the solution sub-tags can reflect the relevance of the digital solution to other solutions and the accuracy of the content, screening recommended solutions based on the weight values of the solution sub-tags can ensure that the solution content meets the user's expectations while ensuring that relevant solutions are not missed, because the weight values of the solution sub-tags are obtained by counting the frequency of occurrence of each solution sub-tag representing the solution content.
[0104] After the screening is completed, it can also be sorted according to user needs, using the solution's quality performance score and relevance as criteria, placing search results that meet user expectations at the front, further improving user search efficiency.
[0105] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A tag-based digital solution recommendation method, characterized in that: Attach a corresponding user label to each user according to the user label model, and attach a corresponding solution label to each digital solution according to the solution label model; The user tag model includes: basic tags, preference tags, behavior tags and capability tags; the solution tag model includes basic tags, subordinate tags, interactive tags and key technical attribute tags; the association tag model includes basic tags, visit tags, like tags, attention tags and evaluation tags; Combining the user tag and the solution tag, and attaching an associated basic tag and an associated interactive tag to each digital solution according to an associated tag model; Obtain the target user's digital solution access history based on the associated basic tags and associated interactive tags corresponding to each digital solution; According to the target user's digital solution access history, multiple solution sub-tag values corresponding to each accessed digital solution are counted, and a solution sub-tag weight value corresponding to each accessed digital solution is calculated according to a preset matching rule; the accessed digital solution is a digital solution accessed by the target user; the solution sub-tag weight value is calculated based on the occurrence frequency of all solution sub-tag values; According to the weight value of the solution sub-label corresponding to each accessed digital solution, a preset number of digital solutions are selected as multiple recommended solutions.
2. The tag-based digital solution recommendation method according to claim 1, characterized in that: The calculation method of the weight value of the scheme sub-label is as follows: Count the values of multiple solution sub-tags corresponding to each accessed digital solution; According to the multiple scheme sub-tag values, counting the number of occurrences of each scheme sub-tag value; According to the order of the number of occurrences of each solution sub-label value from large to small, multiple solution sub-label values of a preset number of types are selected as a label reference value set; Calculate the frequency of occurrence of each solution sub-label value according to the number of occurrences of each solution sub-label value and the label reference value set; Calculate a solution sub-label weight value according to the occurrence frequency of each solution sub-label value and the hit value of each solution sub-label value; The hit value of each solution sub-label value depends on the mode of various solution sub-label values in the label reference value set.
3. The tag-based digital solution recommendation method according to claim 1, characterized in that: After selecting a preset number of digital solutions as multiple recommended solutions, the method further includes: The recommended solutions are sorted in descending order according to the weight values of the solution sub-labels corresponding to each of the accessed digital solutions.
4. The tag-based digital solution recommendation method according to claim 1, characterized in that: The associated interactive tags include visit tags, like tags, follow tags and comment tags.
5. The tag-based digital solution recommendation method according to claim 4, characterized in that: After selecting a preset number of digital solutions as multiple recommended solutions, the method further includes: Obtain access tags corresponding to multiple recommended solutions and count the number of user visits to each recommended solution within a unit time. Calculate the correlation coefficients among the multiple recommendation solutions according to the number of user visits to each recommendation solution within the unit time; The plurality of recommended solutions are sorted in descending order of the values of the correlation coefficients between the solutions.
6. The tag-based digital solution recommendation method according to claim 4, characterized in that: After selecting a preset number of digital solutions as multiple recommended solutions, the method further includes: Get the visit tags, like tags, follow tags and comment tags corresponding to each recommendation plan, and calculate the effective reading number, total reading time, average user reading time, number of replays, number of likes, number of comments and number of favorites for each recommendation plan; Obtain the solution label of each recommended solution and evaluate the basic score of each recommended solution according to the preset standard; Calculate the quality performance score of each solution based on the number of effective readings, total reading time, average user reading time, number of replays, number of likes, number of comments, number of favorites and basis of each recommended solution; The multiple recommended solutions are sorted in descending order according to the quality performance score corresponding to each solution.
7. The tag-based digital solution recommendation method according to claim 6, characterized in that: The preset standards include business analysis domain, solution domain and effectiveness value domain; The business analysis domain, the solution domain, and the effectiveness value domain are all defined according to the digital solution lifecycle for the electric power industry.
8. The tag-based digital solution recommendation method according to claim 7, characterized in that: The business analysis domain includes solution overview, business status, business processes and business pain points; The solution domain includes solution objectives and overall solutions; The effectiveness value domain includes: system operation analysis, solution effectiveness and value analysis.
9. The tag-based digital solution recommendation method according to claim 1, characterized in that: After selecting a preset number of digital solutions as multiple recommended solutions, the method further includes: The key technology tag corresponding to each recommended solution is obtained, the recommended solutions with the same key technology tag value are selected, and other recommended solutions are excluded; the solution tags include the key technology tags.
10. A tag-based digital solution recommendation device, characterized in that: include: A user label module, used to label each user with a corresponding user label according to the user label model, and to label each digital solution with a corresponding solution label according to the solution label model; The user tag model includes: basic tags, preference tags, behavior tags and capability tags; the solution tag model includes basic tags, subordinate tags, interactive tags and key technical attribute tags; the association tag model includes basic tags, visit tags, like tags, attention tags and evaluation tags; An interactive label module, used to combine the user label and the solution label, and attach an associated basic label and an associated interactive label to each digital solution according to an associated label model; An access history module, used to obtain the target user's digital solution access history according to the associated basic tags and associated interactive tags corresponding to each digital solution; A weight calculation module is used to count multiple solution sub-tag values corresponding to each visited digital solution according to the digital solution access history of the target user, and calculate the solution sub-tag weight value corresponding to each visited digital solution according to a preset matching rule; the visited digital solution is a digital solution visited by the target user; the solution sub-tag weight value is calculated based on the occurrence frequency of all solution sub-tag values; The solution selection module selects a preset number of digital solutions as multiple recommended solutions according to the weight values of the solution sub-labels corresponding to each accessed digital solution.
