Search

CN-115757972-B - Student-oriented in-school information resource personalized recommendation method

CN115757972BCN 115757972 BCN115757972 BCN 115757972BCN-115757972-B

Abstract

The invention discloses a student-oriented in-school information resource personalized recommendation method, which relates to the technical field of resource personalized recommendation, and comprises the steps of acquiring personal information of a user, collecting employment information consisting of engaged information, step increment value, primary salary value and amplification ratio according to the personal information, carrying out sorting analysis on the employment information, dividing the employment information into associated information and non-related information according to the engaged information, and obtaining trending associated information, core linkage total value, trending non-related information and core non-total value according to the data relationship between each corresponding step increment value, primary salary value and amplification ratio in the associated information and the non-related information.

Inventors

  • ZHANG WU
  • WANG YANJUN

Assignees

  • 浙江正元智慧科技股份有限公司

Dates

Publication Date
20260508
Application Date
20221201

Claims (5)

  1. 1. The student-oriented in-school information resource personalized recommendation method is characterized by comprising the following steps of: firstly, personal information of a user is acquired, wherein the personal information comprises tag information; Step two, data collection is carried out according to the label information, employment information of students with the same label information in the past graduation is obtained according to the label information, the employment information comprises the employment information, the step increment value, the primary salary value and the increment ratio, wherein the employment information is corresponding business, the step increment value is the number of job promotion after the corresponding business, the primary salary value is corresponding to the initial comprehensive salary when the corresponding business is just entered, the increment ratio is the value obtained by dividing the current salary of the corresponding students by the age after subtracting the primary salary value; Sorting and analyzing employment information, namely firstly dividing the employment information into associated information and non-relevant information according to the engaged information, then calculating to obtain weight values of all the associated information according to the data relation among the step increment value, the primary firework value and the amplification ratio corresponding to each associated information, obtaining hot associated information according to the weight values, and determining a step increment core value, a primary firework value and the amplification core ratio according to the dispersion condition of each data of the associated information to obtain a core linkage total value; Step four, according to the principle of the step three, obtaining hot non-closing information and a nuclear non-total value for the non-closing information; step five, recommending information according to the total value of the core linkage and the non-total value of the core; The specific mode of the arrangement analysis is as follows: S1, firstly, dividing the employment information into associated information and non-relevant information according to the engaged information in the employment information, wherein the manner of dividing the associated information and the non-relevant information is that when the engaged information is matched with the corresponding label information, the engaged information is marked as the associated information, otherwise, the engaged information is marked as the non-relevant information, and the matching refers to the fact that the corresponding label information exists in the recruitment requirement of the corresponding engaged information; s2, obtaining updated associated information and non-related information; S3, analyzing the associated information to obtain the engaged information in the associated information and the corresponding step increment value, the primary firewood value and the amplification ratio; s4, marking the engaged information as Ci, i=1, & gt, n, representing that n engaged information exists, and correspondingly marking the step increment value, the preliminary firewood value and the amplification ratio as Zi, ci and Fi in sequence; s5, calculating a weight value Qi by using a formula, wherein the specific calculation formula is as follows: Qi=0.33*Zi+0.32*Ci+0.35*Fi; Wherein, 0.33, 0.32 and 0.35 are all preset weights; s6, sequencing Ci according to the sequence from the high value to the low value of Qi, and marking thirty percent of the top ranking as hot associated information; S7, obtaining an order increment value Zi, automatically solving the average value of the order increment value Zi, marking the average value as P, and calculating the polymerization degree W by using a formula, wherein the specific calculation formula is as follows: ; Then comparing the W value with X1, and when W is less than or equal to X1, marking the average value P at the moment as a step-by-step kernel value, otherwise, deleting data, and determining a final step-by-step kernel value according to the data deletion; s8, processing the primary firewood value Ci and the amplification ratio Fi according to the same principle of the step S7, and correspondingly marking the obtained value as a primary firewood core value and an amplification core ratio; s9, calculating a core link total value of the associated information by using a formula, wherein the specific calculation formula is as follows: total core = 0.33 order core increment + 0.32: the primary core value +0.35 amplified core ratio.
  2. 2. The personalized recommendation method for the in-school information resources for students according to claim 1, wherein the specific mode of deleting the data in the step S7 is as follows: Sequencing Zi according to the sequence of the values of I Zi-P I from large to small, sequentially selecting Zi values according to the sequence, deleting each Zi value, recalculating a W value after deleting, sequentially selecting the next Zi value if the W value is more than X1, recalculating the W value after deleting until W is less than or equal to X1, obtaining the number of the deleted Zi values at the moment, dividing the number by n to obtain a deletion duty ratio, generating a discrete signal when the deletion duty ratio exceeds X2, and automatically marking the average value of the deleted residual Zi as a step-increase core value, wherein X2 is a preset value and less than 1.
  3. 3. The method for personalized recommendation of in-school information resources for students according to claim 2, wherein when discrete signals are generated, the number of values larger than the average value in the Zi is obtained and marked as an upper number, the number of values smaller than the average value in the Zi is marked as a lower number, when the upper number exceeds the lower number, the median of the maximum value and the average value in the Zi is marked as a step-by-step value, otherwise, the median of the minimum value and the average value in the Zi is marked as a step-by-step value, and the step-by-step value is obtained.
  4. 4. The personalized recommendation method for the in-school information resources for students according to claim 1, wherein the information recommendation method based on the total value of the core and the non-total value of the core is as follows: When the core non-total value exceeds the core linkage total value, generating recommendation information, wherein the content of the recommendation information is that the current non-related information is provided with a reference value, and please check the hot non-related information; Otherwise, the recommended information content is "the current associated information has a reference value, please check the hot associated information".
  5. 5. The personalized recommendation method for student-oriented in-school information resources according to claim 4, wherein the recommended information content is displayed at a user side.

Description

Student-oriented in-school information resource personalized recommendation method Technical Field The invention belongs to the technical field of information personalized recommendation, and particularly relates to a student-oriented school information resource personalized recommendation method. Background The patent with the publication number of CN107862012A discloses an automatic information resource recommendation method for college student groups, which comprises the steps of obtaining user data and data to be recommended, calculating a neighbor set based on a user evaluation matrix by adopting a user similarity model, calculating a neighbor set based on social network information by adopting a similarity model based on social network information, calculating scores of predicted items according to the two neighbor sets, mixing the two results, and screening TOP-N items with the highest scores to obtain a recommendation result. The social network of the user is fused into the traditional collaborative filtering algorithm, so that the needs of college student information sharing and communication are met. The method can provide more proper recommendation results for the vertical field facing college students. But for the system, the information resources of employment and finding work are difficult to obtain and targeted recommendation is difficult to carry out for college students, and a solution is provided based on the information resources. Disclosure of Invention The invention aims to at least solve one of the technical problems in the prior art, and therefore, the invention provides a student-oriented in-school information resource personalized recommendation method. To achieve the above object, according to an embodiment of the first aspect of the present invention, a student-oriented method for personalized recommendation of in-school information resources is provided, which specifically includes the following steps: Firstly, personal information of a user is obtained, wherein the personal information comprises tag information, and the tag information corresponds to the professional direction of the user; acquiring tag information in the personal information, and collecting employment information consisting of engaged information, step increment value, primary salary value and amplification ratio according to the tag information; Step three, the employment information is arranged and analyzed, the employment information is firstly divided into related information and non-related information according to the engaged information, then the weight of all the related information is calculated according to the relation of data among each corresponding step increment value, primary salary value and amplification ratio in the related information, hot related information is obtained according to the weight, and then the step increment value, primary salary value and amplification ratio are determined according to the dispersion condition of each data of the related information, so that the total nuclear association value is calculated; Step four, obtaining non-related information, processing the non-related information in the same mode of processing the related information, marking the hot related information obtained here as hot non-related information, and marking the core-link total value obtained here as core non-total value; step five, obtaining hot associated information and core associated total values of the associated information, and hot non-closed information and core non-total values of the non-closed information; Step six, recommending information according to the total value and the non-total value of the core, wherein the specific mode is as follows: When the core non-total value exceeds the core linkage total value, generating recommendation information, wherein the content of the recommendation information is that the current non-related information is provided with a reference value, and please check the hot non-related information; Otherwise, the recommended information content is "the current associated information has a reference value, please check the hot associated information". Further, the specific data collection mode in the second step is as follows: Acquiring employment information of students with the same label information, which have graduated by the students, according to the label information, wherein the employment information comprises engaged information, step increment value, primary firewood value and increment ratio; the engaged information is the corresponding engaged industry, the step increment is the number of post-job promotion after the corresponding engaged industry, the primary salary value is the initial comprehensive salary when the corresponding engaged information is just entered into the corresponding industry, and the increment ratio is the value obtained by dividing the current salary of the corresponding student by the age after subtracting the pri