Post recommendation method and system based on archive document analysis

Abstract

The invention discloses a post recommendation method and system based on archive document analysis. The method comprises the following steps: obtaining archive documents; carrying out identification and information extraction on the archive document, and summarizing to obtain preliminary information; performing format processing on the preliminary information to obtain archive information; and matching the file information subjected to format processing based on the post information of the recommended post to obtain a matching result. According to the method, the archive information after format processing is obtained by normalizing the archive document, the information of the archive document is greatly simplified, then the recommendation result is obtained through matching processing, the matching depth and direction can be adjusted through the content after format processing, and the accuracy of post recommendation is guaranteed.

Description

Technical Field

[0001] The present application relates to a job recommendation method and system based on archival document analysis. Background Art

[0002] When analyzing the capabilities of public institutions or civil servants, positions are usually matched based on the analysis of personnel files. At present, the analysis of personnel files and the recommendation of positions are generally linear recommendations. That is, the original work scope is used as an important reference, and then in-depth research is conducted based on the work scope to finally obtain a suitable position. The main problem with this method is that it cannot objectively reflect the capabilities of personnel, and thus cannot obtain suitable recommended positions. Summary of the invention

[0003] In order to solve the above problems, the present application discloses a job recommendation method based on archival document analysis, comprising the following steps:

[0004] Obtain the archival documents of the persons to be recommended;

[0005] Identify and extract information from archival documents and summarize them to obtain preliminary information;

[0006] Format the preliminary information to obtain archival information;

[0007] The formatted archive information is matched with the job information of the recommended job to obtain the matching result. This application obtains the formatted archive information by regularizing the archive documents, greatly simplifies the information of the archive documents, and then obtains the recommendation result through matching processing. The matching depth and direction can be adjusted through the formatted content to ensure the accuracy of the job recommendation.

[0008] Preferably, the formatting process refers to setting all required information contents and then retrieving the archive information from the preliminary information.

[0009] Preferably, the profile information includes identity information, education information, and work experience.

[0010] Preferably, the identity information includes comparable information and non-comparable information. The non-comparable information includes information used to determine the identity of a person but unrelated to the person's experience, including name, ID number, and identity information; the comparable information includes gender, age, and ethnicity;

[0011] The educational information includes graduation year, graduation school, graduation school location, school category, school ranking information in China, graduation major, major ranking information in China, study subjects, academic performance, academic qualifications obtained, school-level awards received during school, provincial-level awards received during school, and national-level awards received during school;

[0012] The work experience shall include the positions held, duration of service, location of service, work accomplished, awards received and specific reasons for the awards.

[0013] Preferably, the matching is performed as follows:

[0014] Obtain sample file information and on-the-job performance scores for all current personnel;

[0015] Compare the information of the person to be recommended with the sample file information for similarity;

[0016] Determine whether to recommend based on similarity.

[0017] Preferably, the scoring is performed according to the following steps:

[0018] Obtain sample file information and on-the-job performance scores of all existing personnel. The performance scores are divided into A, B, C, D, and E according to whether they meet the job requirements, where A represents the top 3% of personnel in terms of performance competence, B represents the top 10% of personnel in terms of performance competence, C represents the top 30% of personnel in terms of performance competence, D represents the top 50% of personnel in terms of performance competence, and E represents the rest of the personnel.

[0019] Preferably, the similarity comparison is performed according to the following steps:

[0020] For the job information of the recommended job, a subset of sample file information of existing job personnel who are the same or similar to the recommended job is obtained;

[0021] Compare the information of the person to be recommended with the person assigned score A in the sample file information subset to obtain the final similarity, and calculate the percentage a of the similarity that exceeds the similarity threshold;

[0022] Compare the information of the person to be recommended with the person assigned with score B in the sample file information subset to obtain the final similarity, and calculate the percentage b of the similarity that exceeds the similarity threshold;

[0023] Compare the information of the person to be recommended with the person with a score of C in the sample file information subset to obtain the final similarity, and calculate the percentage c of the similarity that exceeds the similarity threshold;

[0024] Compare the information of the person to be recommended with the person with a score of D in the sample file information subset to obtain the final similarity, and calculate the percentage d of the similarity that exceeds the similarity threshold;

[0025] The information of the person to be recommended is compared with the person with a score of E in the sample file information subset to obtain the final similarity, and the percentage e of the similarity above the similarity threshold is calculated.

[0026] Preferably, the recommendation is determined as follows:

[0027] If (a+b) / 2>(c+d+e) / 3, the recommended person is considered as a basic recommendation;

[0028] On the basis of satisfying (a+b) / 2>(c+d+e) / 3, a>b and a>2c and a>2d and a>2e, the recommended person is considered highly recommended. This application is based on a huge amount of existing sample file information to compare and judge whether the new person is suitable for the position, which has good reference value and can avoid problems caused by subjectivity. In addition, through this scoring comparison method, the recommendation results obtained are relatively accurate and have strong reference value.

[0029] Preferably, the similarity comparison is calculated using the Jaccard similarity coefficient algorithm; the similarity threshold is λ, 0.8≤λ≤0.9;

[0030] The similarity threshold λ is determined as follows:

[0031] Set λ to 0.8. If the calculated values ​​of a, b, c, d, e are all greater than 0.4, then increase λ by 0.01 each time, and calculate until at least two of a, b, c, d, e are less than 0.4. Then take λ at this time as the similarity threshold. If λ is set to 0.8, and at least one of a, b, c, d, e is not greater than 0.4, then set λ to 0.8. This application is based on the principle of relative comparison. Therefore, a similarity threshold with more reference value is obtained through trial calculation, so that when making comparisons, the control is not too strict. Even if there is inconsistent information, the change of the similarity threshold can achieve the purpose of minimizing the impact on the recommendation results.

[0032] On the other hand, a system for a job recommendation method based on archival document analysis is also disclosed, comprising the following modules:

[0033] The information acquisition module is used to obtain the archival documents of the person to be recommended, identify and extract information from the archival documents, and summarize the preliminary information;

[0034] A format processing module is used to process the preliminary information in a format to obtain archival information;

[0035] The matching module is used to match the formatted archive information with the job information of the recommended job to obtain a matching result.

[0036] This application can bring the following beneficial effects:

[0037] 1. This application obtains formatted archival information by regularizing archival documents, greatly simplifies the information of archival documents, and then obtains recommendation results through matching processing. The matching depth and direction can be adjusted through the formatted content to ensure the accuracy of job recommendations.

[0038] 2. This application is based on a huge amount of existing sample file information to compare and judge whether new employees are suitable for the position. It has good reference value and can avoid problems caused by subjectivity. Moreover, through this scoring comparison method, the recommendation results obtained are more accurate and have strong reference value.

[0039] 3. This application is based on the principle of relative comparison. Therefore, a similarity threshold with greater reference value is obtained through trial calculation, so that when making comparisons, the control will not be too strict. Even if there is inconsistent information, the change of the similarity threshold can achieve the purpose of minimizing the impact on the recommendation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0041] Figure 1 It is a schematic diagram of the first embodiment.

[0042] Figure 2 Schematic diagram of the second embodiment. DETAILED DESCRIPTION

[0043] In order to clearly illustrate the technical features of the present solution, the present application is described in detail below through specific implementation methods and in conjunction with the accompanying drawings.

[0044] In the first embodiment, if Figure 1 As shown, a job recommendation method based on archival document analysis includes the following steps:

[0045] S101: Obtain the archive documents of the person to be recommended, identify and extract information from the archive documents, and summarize to obtain preliminary information;

[0046] S102 performs format processing on the preliminary information to obtain archival information;

[0047] The formatting process refers to setting all the required information content and then retrieving the archive information from the preliminary information.

[0048] The file information includes identity information, education information, and work experience.

[0049] The identity information includes comparable information and non-comparable information. The non-comparable information includes information used to determine the identity of a person but not related to the person's experience, including name, ID number, and identity information; the comparable information includes gender, age, and ethnicity;

[0050] The educational information includes graduation year, graduation school, graduation school location, school category, school ranking information in China, graduation major, major ranking information in China, study subjects, academic performance, academic qualifications obtained, school-level awards received during school, provincial-level awards received during school, and national-level awards received during school;

[0051] The work experience shall include the positions held, duration of service, location of service, work accomplished, awards received and specific reasons for the awards.

[0052] S103 matches the formatted archive information with the job information of the recommended job to obtain a matching result.

[0053] The matching in S1031 is performed as follows:

[0054] Obtain sample file information and on-the-job performance scores for all current personnel;

[0055] The scoring is performed according to the following steps:

[0056] Obtain sample file information and on-the-job performance scores of all existing personnel. The performance scores are divided into A, B, C, D, and E according to whether they meet the job requirements, where A represents the top 3% of personnel in terms of performance competence, B represents the top 10% of personnel in terms of performance competence, C represents the top 30% of personnel in terms of performance competence, D represents the top 50% of personnel in terms of performance competence, and E represents the rest of the personnel.

[0057] S1032 compares the information of the person to be recommended with the sample file information for similarity;

[0058] The similarity comparison is performed according to the following steps:

[0059] For the job information of the recommended job, a subset of sample file information of existing job personnel who are the same or similar to the recommended job is obtained;

[0060] Compare the information of the person to be recommended with the person assigned score A in the sample file information subset to obtain the final similarity, and calculate the percentage a of the similarity that exceeds the similarity threshold;

[0061] Compare the information of the person to be recommended with the person assigned with score B in the sample file information subset to obtain the final similarity, and calculate the percentage b of the similarity that exceeds the similarity threshold;

[0062] Compare the information of the person to be recommended with the person with a score of C in the sample file information subset to obtain the final similarity, and calculate the percentage c of the similarity that exceeds the similarity threshold;

[0063] Compare the information of the person to be recommended with the person with a score of D in the sample file information subset to obtain the final similarity, and calculate the percentage d of the similarity that exceeds the similarity threshold;

[0064] The information of the person to be recommended is compared with the person with a score of E in the sample file information subset to obtain the final similarity, and the percentage e of the similarity above the similarity threshold is calculated.

[0065] The similarity comparison is calculated using the Jaccard similarity coefficient algorithm; the similarity threshold is λ, 0.8≤λ≤0.9;

[0066] The similarity threshold λ is determined as follows:

[0067] Set λ to 0.8. If the calculated values ​​of a, b, c, d, and e are all greater than 0.4, then increase λ by 0.01 each time until at least two of a, b, c, d, and e are less than 0.4. Take the λ at this time as the similarity threshold. If λ is set to 0.8 and at least one of a, b, c, d, and e is not greater than 0.4, then set λ to 0.8.

[0068] S1033 determines whether to recommend based on similarity;

[0069] The decision on whether to recommend is made as follows:

[0070] If (a+b) / 2>(c+d+e) / 3, the recommended person is considered as a basic recommendation;

[0071] On the basis that (a+b) / 2>(c+d+e) / 3, a>b and a>2c and a>2d and a>2e, the recommended person is deemed to be highly recommended.

[0072] In the second embodiment, if Figure 2As shown, a job recommendation system based on archival document analysis includes the following modules: an information acquisition module 201, which is used to acquire archival documents of the personnel to be recommended, identify and extract information from the archival documents, and summarize to obtain preliminary information;

[0073] The format processing module 202 is used to perform format processing on the preliminary information to obtain archive information;

[0074] The matching module 203 is used to match the formatted archive information with the job information of the recommended job to obtain a matching result.

[0075] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A job recommendation method based on archival document analysis, characterized in that: The steps include: Obtain the archival documents of the persons to be recommended; Identify and extract information from archival documents and summarize them to obtain preliminary information; Format the preliminary information to obtain archival information; The formatted profile information is matched based on the job information of the recommended job to obtain a matching result.

2. A method for job recommendation based on archival document analysis as claimed in claim 1, characterized in that: The formatting process refers to setting all the required information content and then retrieving the archive information from the preliminary information.

3. A method for job recommendation based on archival document analysis as claimed in claim 2, characterized in that: The file information includes identity information, education information, and work experience.

4. A method for job recommendation based on archival document analysis as claimed in claim 3, characterized in that: The identity information includes comparable information and non-comparable information. The non-comparable information includes information used to determine the identity of a person but not related to the person's experience, including name, ID number, and identity information; the comparable information includes gender, age, and ethnicity; The educational information includes graduation year, graduation school, graduation school location, school category, school ranking information in China, graduation major, major ranking information in China, study subjects, academic performance, academic qualifications obtained, school-level awards received during school, provincial-level awards received during school, and national-level awards received during school; The work experience shall include the positions held, duration of service, location of service, work accomplished, awards received and specific reasons for the awards.

5. A method for job recommendation based on archival document analysis as claimed in claim 1, characterized in that: The matching is performed as follows: Obtain sample file information and on-the-job performance scores for all current personnel; Compare the information of the person to be recommended with the sample file information for similarity; Determine whether to recommend based on similarity.

6. A method for job recommendation based on archival document analysis as claimed in claim 5, characterized in that: The scoring is performed according to the following steps: Obtain sample file information and on-the-job performance scores of all existing personnel. The performance scores are divided into A, B, C, D, and E according to whether they meet the job requirements, where A represents the top 3% of personnel in terms of performance competence, B represents the top 10% of personnel in terms of performance competence, C represents the top 30% of personnel in terms of performance competence, D represents the top 50% of personnel in terms of performance competence, and E represents the rest of the personnel.

7. A method for job recommendation based on archival document analysis as claimed in claim 5, characterized in that: The similarity comparison is performed according to the following steps: For the job information of the recommended job, a subset of sample file information of existing job personnel who are the same or similar to the recommended job is obtained; Compare the information of the person to be recommended with the person assigned score A in the sample file information subset to obtain the final similarity, and calculate the percentage a of the similarity that exceeds the similarity threshold; Compare the information of the person to be recommended with the person assigned with score B in the sample file information subset to obtain the final similarity, and calculate the percentage b of the similarity that exceeds the similarity threshold; Compare the information of the person to be recommended with the person with a score of C in the sample file information subset to obtain the final similarity, and calculate the percentage c of the similarity that exceeds the similarity threshold; Compare the information of the person to be recommended with the person with a score of D in the sample file information subset to obtain the final similarity, and calculate the percentage d of the similarity that exceeds the similarity threshold; The information of the person to be recommended is compared with the person with a score of E in the sample file information subset to obtain the final similarity, and the percentage e of the similarity above the similarity threshold is calculated.

8. A method for job recommendation based on archival document analysis as claimed in claim 5, characterized in that: The decision on whether to recommend is made as follows: If (a+b) / 2>(c+d+e) / 3, the recommended person is considered as a basic recommendation; On the basis that (a+b) / 2>(c+d+e) / 3, a>b and a>2c and a>2d and a>2e, the recommended person is deemed to be highly recommended.

9. A method for job recommendation based on archival document analysis as claimed in claim 5, characterized in that: The similarity comparison is calculated using the Jaccard similarity coefficient algorithm; the similarity threshold is λ, 0.8≤λ≤0.9; The similarity threshold λ is determined as follows: Set λ to 0.

8. If the calculated values ​​of a, b, c, d, and e are all greater than 0.4, then increase λ by 0.01 each time until at least two of a, b, c, d, and e are less than 0.

4. Take the λ at this time as the similarity threshold. If λ is set to 0.8 and at least one of a, b, c, d, and e is not greater than 0.4, then set λ to 0.

8.

10. A system for implementing the job recommendation method based on archival document analysis according to any one of claims 1 to 9, characterized in that: Includes the following modules: The information acquisition module is used to obtain the archival documents of the person to be recommended, identify and extract information from the archival documents, and summarize the preliminary information; A format processing module is used to process the preliminary information in a format to obtain archival information; The matching module is used to match the formatted archive information with the job information of the recommended job to obtain a matching result.

Images