A personalized recommendation method for in-campus information resources for students

Abstract

The application discloses a student-oriented in-campus information resource personalized recommendation method and relates to the technical field of resource personalized recommendation. Personal information of a user is acquired, data collection is performed according to the personal information, and employment information composed of engaged information, step increment value, initial salary value and increment ratio is collected. The employment information is analyzed and sorted. According to the engaged information and non-engaged information, each corresponding step increment value, initial salary value and increment ratio are obtained, and popular engaged information, total engaged value, popular non-engaged information and total non-engaged value are obtained. According to the total engaged value and the total non-engaged value, in-campus recruitment resources of students are reasonably analyzed and recommended. The application is simple, effective and easy to use.

Description

Technical Field

[0001] This invention belongs to the field of personalized information recommendation technology, specifically a personalized recommendation method for on-campus information resources for students. Background Technology

[0002] Patent CN107862012A discloses an automatic information resource recommendation method for university students, comprising: acquiring user data and data to be recommended; calculating a neighbor set based on a user rating matrix using a user similarity model; calculating a neighbor set based on social network information using a similarity model based on social network information; calculating the score of the predicted item based on the two neighbor sets, and mixing the two results; selecting the top-N items with the highest scores to obtain the recommendation result. This invention integrates the user's social network into a traditional collaborative filtering algorithm, meeting the needs of university students for information sharing and communication. This method can provide more suitable recommendation results for vertical fields targeting university students.

[0003] However, for this system, and especially for college students, information resources for job hunting are difficult to obtain, and targeted recommendations are also difficult to make. Therefore, a solution is provided. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a personalized recommendation method for on-campus information resources for students.

[0005] To achieve the above objectives, according to an embodiment of the first aspect of the present invention, a personalized recommendation method for on-campus information resources for students is proposed, the method specifically including the following steps:

[0006] Step 1: First, obtain the user's personal information, which includes tag information, which corresponds to the user's professional direction;

[0007] Step Two: Next, obtain the tag information from the personal information, and collect data based on the tag information to collect employment information consisting of job information, job value-added, initial salary, and growth rate.

[0008] Step 3: Organize and analyze employment information. First, divide employment information into related information and unrelated information based on the employment information. Then, calculate the weight value of all related information based on the relationship between the corresponding value-added, initial salary, and growth rate of each related information. Based on the weight value, obtain the popular related information. Then, determine the value-added core value, initial salary core value, and growth rate core ratio based on the dispersion of each data point of value-added, initial salary, and growth rate of each related information, and thus calculate the core value of the total value of the related information.

[0009] Step 4: After obtaining the irrelevant information, process it in the same way as the related information. Mark the popular related information obtained here as popular irrelevant information, and mark the core-linked total value obtained here as the core-non-linked total value.

[0010] Step 5: Obtain the popular related information and total core value of related information, as well as the popular non-related information and total core value of non-related information;

[0011] Step Six: Recommend information based on the total value of nuclear joint and non-nuclear total values, specifically in the following manner:

[0012] When the total value of non-core information exceeds the total value of core information, a recommendation is generated. The recommendation message is: "The current non-relevant information is more relevant. Please check the popular non-relevant information."

[0013] Otherwise, the recommended information will be "The current related information is more relevant; please check the popular related information."

[0014] Furthermore, the specific method for data collection in step two is as follows:

[0015] Based on the tag information, we can obtain the employment information of previous graduates with the same tag information. The employment information includes job information, job value-added, initial salary, and growth rate.

[0016] "Information" refers to the corresponding industry, "level increase" refers to the number of promotions after entering the corresponding industry, "initial salary" refers to the initial comprehensive salary when entering the corresponding information industry, and "growth rate" is the value obtained by subtracting the initial salary from the student's current salary and then dividing by the number of years.

[0017] We obtain employment information consisting of information, job value-added, initial salary, and growth rate.

[0018] Furthermore, the specific method for organizing and analyzing in step three is as follows:

[0019] S1: First, based on the employment information, the employment information is reclassified into relevant information and irrelevant information;

[0020] S2: Obtain the updated related and irrelevant information;

[0021] S3: Analyze the related information to obtain the job information within the related information and its corresponding step value-added, initial salary value, and growth rate;

[0022] S4: Label the job information as Ci, i = 1, ..., n, indicating that there are n job information. Correspondingly label the step value increment, initial salary value, and growth rate as Zi, Ci, and Fi, i = 1, ..., n;

[0023] S5: Then, calculate the weighted value Qi using the formula:

[0024] Q i=0.33*Zi+0.32*Ci+0.35*Fi;

[0025] In the formula, 0.33, 0.32, and 0.35 are all preset weights;

[0026] S6: Then sort Ci according to Qi value from largest to smallest, and mark the top 30% as popular related information;

[0027] S7: After obtaining the order increment Zi, its mean is automatically calculated and marked as P. Then, the aggregation degree W is calculated using the formula:

[0028]

[0029] Next, the W value is compared with X1. When W≤X1, the mean P at this time is marked as the order-increasing kernel value; otherwise, data is deleted, and the final order-increasing kernel value is determined based on the data deletion.

[0030] S8: Then, following the same principle as step S7, process the initial salary value Ci and the growth rate ratio Fi, and mark the resulting values ​​as the initial salary value and the growth rate ratio.

[0031] S9: Then, calculate the total value of the associated information using the formula. The specific calculation formula is as follows:

[0032] Total core value = 0.33 * incremental core value + 0.32 * initial salary core value + 0.35 * growth rate core ratio.

[0033] Furthermore, in step S1, the specific method for distinguishing between related information and irrelevant information is as follows: when the job information matches the corresponding tag information, it is marked as related information; otherwise, it is marked as irrelevant information. Here, matching refers to the existence of corresponding tag information within the recruitment requirements of the job information.

[0034] Furthermore, the specific method for data deletion in step S7 is as follows:

[0035] Sort Zi values ​​in descending order of |Zi-P|. Then, select Zi values ​​sequentially. Delete each selected Zi value and recalculate the W value. If the W value is still greater than X1, select the next Zi value sequentially, delete it, and recalculate the W value until W ≤ X1. Obtain the number of Zi values ​​deleted at this point and divide it by n to get the deletion percentage. When the deletion percentage exceeds X2, X2 is a preset value and less than 1; a discrete signal is generated. Otherwise, the mean of the remaining Zi values ​​after deletion is automatically marked as the incremental kernel value.

[0036] When generating a discrete signal, the number of values ​​in Zi that are greater than the mean is obtained and marked as the upper digit. The number of values ​​in Zi that are less than the mean is marked as the lower digit. When the upper digit exceeds the lower digit, the median of the maximum value and the mean in Zi is marked as the order-increasing kernel value. Otherwise, the median of the minimum value and the mean in Zi is marked as the order-increasing kernel value, thus obtaining the order-increasing kernel value.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] This invention obtains users' personal information, then collects data based on that information, gathering employment information consisting of job information, job level, initial salary, and growth rate. The employment information is then organized and analyzed. Based on job information, it is first divided into related and irrelevant information. Then, based on the relationship between the corresponding job level, initial salary, and growth rate within each related and irrelevant information segment, popular related information, total related information value, popular irrelevant information, and total irrelevant information value are obtained.

[0039] Based on the total value of nuclear and non-nuclear recruitment, a rational analysis of students' on-campus recruitment resources is conducted and corresponding recommendations are made; this invention is simple, effective, and easy to use. Detailed Implementation

[0040] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] This application provides a personalized recommendation method for on-campus information resources for students, which specifically includes the following steps:

[0042] Step 1: First, obtain the user's personal information, which includes tag information, and the tag information corresponds to the user's professional direction;

[0043] Step Two: Next, obtain the tag information from the personal information, and collect data based on the tag information. The specific data collection method is as follows:

[0044] Based on the tag information, we can obtain the employment information of previous graduates with the same tag information. The employment information includes job information, job value-added, initial salary, and growth rate.

[0045] "Information" refers to the corresponding industry, "level increase" refers to the number of promotions after entering the corresponding industry, "initial salary" refers to the initial comprehensive salary when entering the corresponding information industry, and "growth rate" is the value obtained by subtracting the initial salary from the student's current salary and then dividing by the number of years.

[0046] We obtain employment information consisting of information, job value-added, initial salary, and growth rate.

[0047] Step 3: Organize and analyze the employment information. The specific methods for organizing and analyzing the information are as follows:

[0048] S1: First, based on the job information, the job information is reclassified into related information and irrelevant information. Specifically, when the job information matches the corresponding tag information, it is marked as related information; otherwise, it is marked as irrelevant information. Here, matching means that the corresponding job information has corresponding tag information in the recruitment requirements.

[0049] S2: Obtain the updated related and irrelevant information;

[0050] S3: Analyze the related information to obtain the job information within the related information and its corresponding step value-added, initial salary value, and growth rate;

[0051] S4: Label the job information as Ci, i = 1, ..., n, indicating that there are n job information. Correspondingly label the step value increment, initial salary value, and growth rate as Zi, Ci, and Fi, i = 1, ..., n;

[0052] S5: Then, calculate the weighted value Qi using the formula:

[0053] Q i=0.33*Zi+0.32*Ci+0.35*Fi;

[0054] In the formula, 0.33, 0.32, and 0.35 are all preset weights;

[0055] S6: Then sort Ci according to Qi value from largest to smallest, and mark the top 30% as popular related information;

[0056] S7: After obtaining the order increment Zi, its mean is automatically calculated and marked as P. Then, the aggregation degree W is calculated using the formula:

[0057]

[0058] Next, compare the W value with X1. If W ≤ X1, mark the mean P at this time as the order-increasing kernel value; otherwise, delete the data. The specific method for data deletion is as follows:

[0059] Sort Zi values ​​in descending order of |Zi-P|. Then, select Zi values ​​sequentially. Each selected Zi value is deleted, and the W value is recalculated. If the W value is still greater than X1, select the next Zi value in sequence, delete it, and recalculate the W value until W ≤ X1. Obtain the number of Zi values ​​deleted at this point, divide it by n to get the deletion percentage. When the deletion percentage exceeds X2, X2 is a preset value less than 1, usually 0.2. A discrete signal is generated; otherwise, the mean of the remaining Zi values ​​after deletion is automatically marked as the incremental kernel value.

[0060] When generating a discrete signal, the number of values ​​in Zi that are greater than the mean is obtained and marked as the upper digit, and the number of values ​​in Zi that are less than the mean is marked as the lower digit. When the upper digit exceeds the lower digit, the median of the maximum value and the mean in Zi is marked as the order-increasing kernel value; otherwise, the median of the minimum value and the mean in Zi is marked as the order-increasing kernel value, thus obtaining the order-increasing kernel value.

[0061] S8: Then, following the same principle as step S7, process the initial salary value Ci and the growth rate ratio Fi, and mark the resulting values ​​as the initial salary value and the growth rate ratio.

[0062] S9: Then, calculate the total value of the associated information using the formula. The specific calculation formula is as follows:

[0063] Total core value = 0.33 * incremental core value + 0.32 * initial salary core value + 0.35 * growth rate core ratio;

[0064] Step 4: After obtaining the irrelevant information, process the irrelevant information according to the same principle as S3-S9 in Step 3. Mark the popular related information obtained here as popular irrelevant information, and mark the core-linked total value obtained here as the core-non-linked total value.

[0065] Step 5: Obtain the popular related information and total core value of related information, as well as the popular non-related information and total core value of non-related information;

[0066] Step Six: Recommend information based on the total value of nuclear joint and non-nuclear total values, specifically in the following manner:

[0067] When the total value of non-core information exceeds the total value of core information, a recommendation is generated. The recommendation message is: "The current non-relevant information is more relevant. Please check the popular non-relevant information."

[0068] Otherwise, the recommended information will be "The current related information is more relevant; please check the popular related information."

[0069] Step 7: Complete the information recommendation, and the recommended information content is displayed on the user's device.

[0070] The data in the above formula are all calculated by removing the dimensions and taking the numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0071] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A personalized recommendation method for on-campus information resources for students, characterized in that, The method specifically includes the following steps: Step 1: First, obtain the user's personal information, which includes tag information; Step Two: Data Collection Based on Tag Information. This involves obtaining employment information from previous graduates with the same tag information. Employment information includes industry information, career advancement, initial salary, and salary increase rate. Industry information refers to the corresponding industry; career advancement refers to the number of promotions within that industry; initial salary is the student's initial total salary upon entering the industry; and salary increase rate is the student's current salary minus their initial salary, divided by the number of years of service. The collected employment information comprises industry information, career advancement, initial salary, and salary increase rate. Step 3: Organize and analyze employment information. First, divide employment information into related information and unrelated information based on the employment information. Then, calculate the weight value of all related information based on the relationship between the data of the grade value-added, initial salary value, and growth rate of each related information. Based on the weight value, obtain the popular related information. Then, determine the grade value-added core value, initial salary core value, and growth rate core value based on the dispersion of each data of grade value-added, initial salary value, and growth rate core value to obtain the core value of the total related information. Step 4: Process irrelevant information according to the principle of Step 3 to obtain popular irrelevant information and the total value of core irrelevant information; Step 5: Recommend information based on the total value of core-linked data and the total value of non-core data; The specific methods for organizing and analyzing are as follows: S1: First, based on the job information, the job information is reclassified into related information and irrelevant information. The way to classify related information and irrelevant information is as follows: when the job information matches the corresponding tag information, it is marked as related information; otherwise, it is marked as irrelevant information. Here, matching means that the corresponding job information has corresponding tag information in the recruitment requirements. S2: Obtain the updated related and irrelevant information; S3: Analyze the related information to obtain the job information within the related information and its corresponding step value-added, initial salary value, and growth rate; S4: Label the employment information as Ci, i=1, ..., n, indicating that there are n employment information, and label the step value increase, initial salary value, and growth rate as Zi, Ci and Fi respectively; S5: Then, calculate the weighted value Qi using the formula: Qi = 0.33 * Zi + 0.32 * Ci + 0.35 * Fi; In the formula, 0.33, 0.32, and 0.35 are all preset weights; S6: Then sort Ci according to Qi value from largest to smallest, and mark the top 30% as popular related information; S7: After obtaining the order increment Zi, its mean is automatically calculated and labeled as P. Then, the aggregation degree W is calculated using the formula: ； Next, the W value is compared with X1. When W≤X1, the mean P at this time is marked as the order-increasing kernel value; otherwise, data is deleted, and the final order-increasing kernel value is determined based on the data deletion. S8: Then, following the same principle as step S7, process the initial salary value Ci and the growth rate ratio Fi, and mark the resulting values ​​as the initial salary value and the growth rate ratio. S9: Then, calculate the total value of the associated information using the formula. The specific calculation formula is as follows: Total core value = 0.33 * incremental core value + 0.32 * initial salary core value + 0.35 * growth rate core ratio.

2. The personalized recommendation method for on-campus information resources for students according to claim 1, characterized in that, The specific method for data deletion in step S7 is as follows: Sort Zi values ​​in descending order of |Zi-P|. Then select Zi values ​​sequentially. Delete each selected Zi value and recalculate the W value. If the W value is still greater than X1, select the next Zi value sequentially, delete it, and recalculate the W value until W ≤ X1. Obtain the number of deleted Zi values ​​and divide it by n to get the deletion percentage. When the deletion percentage exceeds X2, generate a discrete signal; otherwise, automatically mark the mean of the remaining deleted Zi values ​​as the incremental kernel value. X2 is a preset value less than 1.

3. The personalized recommendation method for on-campus information resources for students according to claim 2, characterized in that, When generating a discrete signal, the number of values ​​in Zi that are greater than the mean is obtained and marked as the upper digit. The number of values ​​in Zi that are less than the mean is marked as the lower digit. When the upper digit exceeds the lower digit, the median of the maximum value and the mean in Zi is marked as the order-increasing kernel value. Otherwise, the median of the minimum value and the mean in Zi is marked as the order-increasing kernel value, thus obtaining the order-increasing kernel value.

4. The personalized recommendation method for on-campus information resources for students according to claim 1, characterized in that, The method for recommending information based on the total value of nuclear joint and non-nuclear total values ​​is as follows: When the total value of non-core information exceeds the total value of core information, a recommendation is generated. The recommendation message is: "The current non-relevant information is more relevant. Please check the popular non-relevant information." Otherwise, the recommended information will be "The current related information is more valuable for reference. Please check the popular related information." 5. The personalized recommendation method for on-campus information resources for students according to claim 4, characterized in that, The recommended information is displayed on the user's device.

Images