CN117592450A - Panoramic archive generation method and system based on employee information integration - Google Patents

Abstract

The invention belongs to the field of human resource management, and provides a panoramic archive generation method and a system based on employee information integration, wherein the method comprises the steps of receiving a data packet imported by an internal and external data source of an enterprise, and analyzing the data packet to obtain employee information resource text; establishing an employee information classification model, traversing the content of the input employee information resource text by text, identifying and classifying employee information data based on a keyword database of a file information list, determining the file category of the employee, and generating a file information item architecture tree corresponding to the file information list; loading a pre-configuration module database, screening corresponding templates based on file categories and file information item architecture trees, writing in according to a format, and automatically generating a panoramic file. The invention provides a more comprehensive and high-quality employee profile by automating and integrating employee information.

Description

Panoramic file generation method and system based on employee information integration

Technical field

The invention relates to the field of human resource management, and in particular to a method and system for generating panoramic files based on employee information integration.

Background technique

In business management, maintaining accurate and timely employee information is critical to support various human resources and payroll management activities. With the development of technology, some traditional employee information management methods and systems have emerged. However, these methods and systems usually have some limitations, including the following problems:

1. Information dispersion: Traditional enterprises use human resource information systems (HRIS) or similar systems to manage employee information. These systems usually contain basic employee information, such as name, contact information, work experience, training records and performance data. However, this information is usually scattered in different departments and databases, making it difficult to fully integrate and access it, resulting in information dispersion and inconsistency that is difficult to maintain.

2. Low timeliness: Traditional employee information management methods usually rely on manual maintenance, resulting in low timeliness of information. Employee information may change but is not updated in a timely manner, thus affecting the accuracy of human resources decision-making.

3. Data quality issues: Due to the data collection and maintenance process, employee information comes from multiple sources, including internal databases, external recruitment websites, employee handbooks, online forms, etc. This information exists in different formats, including text, spreadsheets and database records. There is a problem of diversity of data sources. Employee information is usually scattered in different databases and systems. Different departments may use different software and tools to manage employee information. There is the problem of data dispersion, and data quality issues caused by both are also a common challenge.

4. Manual integration and access issues: Traditional employee information management often involves manual integration and data access. Employee information usually needs to be manually integrated from multiple sources, and accessing employee information may require a long time of searching and data sorting. There is a lack of multi-angle query and it is difficult to view employee information from different angles, such as salary history, training records, performance evaluation, etc.

5. Difficulty supporting multi-scenario applications: Traditional employee information management systems usually cannot effectively support multi-scenario applications, such as salary adjustment, talent declaration, etc., and require multiple exports and processing of data.

Therefore, existing technologies are often based on single-purpose employee information systems, which are difficult to meet the diverse needs of enterprises and cannot solve the problem of data dispersion, manual processing, data inconsistency and inaccessibility, while ignoring the integration of information and multi-angle query. need. Based on the above problems, an all-round and multi-angle employee information management method and system is needed to respond to the challenges in modern enterprise management with a panoramic archive method and improve the efficiency and accuracy of employee information management.

Contents of the invention

The purpose of the embodiments of the present invention is to provide a panoramic file generation method and system based on employee information integration, to provide more comprehensive and high-quality employee files through automation and integration of employee information, and to solve various problems in traditional employee information management. problem, providing enterprises with better data access, information integration and compliance, helping to improve management efficiency and decision-making quality.

To achieve the above objects, embodiments of the present invention provide the following technical solutions.

In the first aspect, the present invention provides a panoramic profile generation method based on employee information integration, which includes the following steps:

Receive data packets imported from internal and external data sources within the enterprise, and parse the data packets to obtain employee information resource text;

Establish an employee information classification model, traverse the content of the input employee information resource text text by text, identify and classify employee information data based on the keyword database of the file information list, determine the employee's file category, and generate a corresponding file information list File information item structure tree;

Load the preconfigured module database, filter the corresponding templates based on the file category and file information item architecture tree, and write them according to the format to automatically generate panoramic files.

As a further solution of the present invention, the panoramic file generation method based on employee information integration also includes the invocation of the panoramic file; the invocation of the panoramic file includes the following steps:

Obtain the file retrieval request input by the user, and the file retrieval request includes the employee's unique identifier and the specific project information required for retrieval;

According to the employee's unique identifier, retrieve the corresponding employee's panoramic file data from the panoramic file resource database, and filter and organize it according to the retrieval requirements;

For the screened panoramic archive data, identify the keyword data related to the retrieval requirements, including but not limited to date and project name;

Based on the keyword data, screen out the panoramic archive resource data that meets the retrieval requirements, and generate the corresponding panoramic archive text;

Export the generated employee panoramic file text for users to review or further process.

As a further solution of the present invention, when parsing the data package to obtain the employee information resource text, it includes: identifying the format of the text file, table file or database file contained in the data package; based on the loaded text parser, table processing library and database The connection tool parses the data package, extracts employee information data and converts it into a structured text form; the extracted employee information data includes basic personal information, work experience and educational background, and converts it into a structured text form using XML or JSON. Format.

As a further solution of the present invention, establishing an employee information classification model includes the following steps:

Step 1. Data collection and processing:

Collect employee information data, remove duplicates and missing values, convert it into document data in a standardized data format, identify document category labels, and use it as sample data; where the employee information data includes personal information, work experience, educational background, and comes from internal Enterprise databases, external data sources or user-provided information;

Step 2. Keyword and phrase identification:

Based on the established keyword and phrase database, use natural language processing (NLP) technology to automatically identify and extract keywords and phrases in document data;

Step 3. Dataset partitioning:

Convert document data into a data set represented by computer-processed feature vectors, and divide the data set into a training set and a test set;

Step 4. Model training and testing:

Use the Naive Bayes algorithm to build a classification model, and use the training set data to train the classification model and learn to assign document data to different archive categories; use the test set to test the recall rate of the model, and evaluate the model based on the performance evaluation results. Parameter adjustment;

Step 5. Generate the file information item architecture tree:

Use the trained classification model to classify employee information texts and determine the profile category of each text; based on the classification results, generate an information item architecture tree for each profile category, including different information items: personal information, work experience, Education background;

Step 6. Deploy the model:

Deploy the trained model to the panoramic file generation system for automatic classification of employee information.

As a further solution of the present invention, when using the naive Bayes algorithm to build a classification model, it also includes calculating the category prior probability. The steps are as follows:

Prepare a training data set containing classified documents, where the training data set includes document data and category labels corresponding to the document data;

Make statistics on the training data set, calculate the number of documents under each category, and obtain the frequency of documents in each category;

Calculate the total number of documents in the training dataset and calculate the prior probability of the category by the formula:

P(C)＝N/N_total

Among them, P(C) is the prior probability of category C; N is the number of documents belonging to category C; N_total is the total number of documents;

The obtained prior probability of each category is used to predict the category of new documents.

As a further solution of the present invention, when predicting the category of a new document based on the obtained prior probability of each category, it also includes calculating the feature conditional probability (P(x|C)), that is, given the category C, observe To obtain the probability of feature x, the steps are as follows:

According to the training data set of classified documents, the documents in the training data set are counted, and the number of occurrences of each feature under each category is calculated; wherein the training data set includes document data, category labels corresponding to the document data, and the number of occurrences of each feature in the document. Characteristics that emerge;

In the case of a given category C, the frequency of feature x; if the document data has M different features, it is expressed as: N(x, C), the frequency of feature x under category C is for each category and each Each feature generates a frequency;

Calculate the relative frequency P(x|C) of feature x in each category C, and obtain the conditional probability of each feature under each category:

P(x|C)＝N(x,C)/N(C)

Among them, P(x|C) is the conditional probability of feature x under category C; N(x,C) is the frequency of feature x under category C; N(C) is the total number of documents under category C.

As a further solution of the present invention, when using the Naive Bayes classification model to predict the category of a new document, the following steps are included:

Prepare the trained Naive Bayes classification model, including the prior probability of the category (P(C)) and the conditional probability of the feature (P(x|C)) calculated from the training data;

Prepare new documents, perform the same text preprocessing on the new documents as the training data, and convert the new documents into feature vectors;

For each category C, use Bayes’ theorem to calculate the posterior probability P(C|X), where X represents the feature vector of the new document:

P(C|X)＝P(X|C)*P(C)/P(X)

Among them, P(C|X) is the posterior probability of category C given feature X; P(X|C) is the conditional probability of feature X under category C, obtained from the training data; P(C ) is the prior probability of category C, obtained from the training data; P(X) is the marginal probability of feature X, calculated by summing the posterior probabilities of each category. In classification, P(X) is for each The categories are the same;

After calculating the posterior probability for each category, the category with the highest posterior probability is selected as the predicted category for the new document; that is:

Prediction category=argmax P(C|X).

As a further solution of the present invention, the generation of the archive information item architecture tree corresponding to the archive information list includes:

Define the file information list for employee information integration, including information items for employee information integration;

Each information item in the archive information list is hierarchically structured to create an archive information item architecture tree, and a unique tree structure label is assigned to each information item in the information item architecture tree to establish an information item architecture tree.

As a further solution of the present invention, when the panoramic archive is generated, the employee information is mapped according to the corresponding information items in the information item architecture tree and stored in the panoramic archive resource database to create a complete panoramic archive for the user to use the employee's unique identifier. And retrieve the required specific project information to query and retrieve the corresponding employee's panoramic file data.

In a second aspect, the present invention also provides a panoramic file generation system based on employee information integration, including:

Data import module: used to receive data packages from internal and external data sources within the enterprise, and parse these data packages to obtain employee information resource text;

Employee information classification model: used to automatically identify and classify employee information data, and determine the employee's profile category; generate a profile information item architecture tree based on the information items defined in the profile information list;

Preconfigured module database: Contains predefined templates and formats, filters and automatically generates panoramic archives based on archive categories and archive information item architecture trees.

Panoramic archive resource database: used to store the generated panoramic archive data; employee information is stored according to the structure of the information item architecture tree;

Panoramic file call module: allows the user to retrieve and organize panoramic file data based on the employee's unique identifier and the specific project information required for retrieval, and generate panoramic file text that meets the retrieval requirements, and allows users to review it.

As a further solution of the present invention, the panoramic file generation system based on employee information integration also includes a text parser, a table processing library and a database connection tool; the text parser, table processing library and database connection tool are respectively used for parsing and importing The format of text files, table files or database files contained in the data package is used to extract and convert employee information data into structured text form.

As a further solution of the present invention, the panoramic profile generation system based on employee information integration also includes a naive Bayes classification model for automatic classification of employee information, using the training data set of classified documents to calculate the prior probability of the category and conditional probabilities of features to facilitate predicting the category of new documents.

In the third aspect, in yet another embodiment of the present invention, a panoramic profile generating device based on employee information integration is provided, including: a processor, a memory, a communication interface and a communication bus. The processor, the memory and the communication interface complete mutual communication through the communication bus; the memory is used to store at least one executable instruction, and the executable instruction causes the processor to execute the employee information-based method as described in the first aspect. Operations corresponding to the integrated panoramic archive generation method.

In a fourth aspect, in yet another embodiment of the present invention, a storage medium is provided, and at least one executable instruction is stored in the storage medium, and the executable instruction causes the processor to execute as described in the first aspect. The operations corresponding to the above-mentioned panoramic file generation method based on employee information integration.

Compared with the existing technology, the panoramic profile generation method and system based on employee information integration provided by embodiments of the present invention have the following beneficial effects:

1. Automated information integration: The panoramic file generation method and system based on employee information integration of the present invention can automatically parse and integrate employee information from different data sources without manual intervention, greatly improving the efficiency of information integration and reducing manual processing. time and labor costs.

2. Data accuracy: By automatically classifying and structuring employee information, the panoramic profile generation method and system based on employee information integration of the present invention helps reduce data input errors and information inconsistencies, and the generated panoramic profiles maintain high quality and accuracy sex.

3. Information traceability: This generated panoramic archive allows users to easily trace the history and changes of employee information, which is very important for auditing and compliance; it allows enterprises to define archive information lists and information item architecture trees according to their specific needs, Making it highly customizable allows the system to adapt to the requirements of different organizations, rather than mandatory specification of information.

4. Fast data retrieval: The panoramic file calling function in the system allows users to easily find and retrieve employee information based on employee unique identifiers and specific project information, improving the speed and efficiency of data access and retrieval.

5. Panoramic view: The generated panoramic profile provides a comprehensive view of employee information, including personal information, work experience, educational background, etc., allowing managers to better understand employees and make more informed decisions. The information in the panoramic profile Helps enterprises make more informed business decisions, such as recruitment, performance evaluation, employee training, etc.

In summary, the panoramic file generation method and system based on employee information integration of the present invention improve the efficiency and quality of employee information management, promote the consistency, traceability and security of information, and provide enterprises with better Data access and decision support tools.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings in the following description are only illustrative of the present invention. Some examples.

Figure 1 shows a flow chart of a panoramic profile generation method based on employee information integration provided by the present invention;

Figure 2 shows the implementation flow chart of panoramic file calling in a panoramic file generation method based on employee information integration provided by an embodiment of the present invention;

Figure 3 shows a system architecture diagram of a panoramic profile generation system based on employee information integration provided by an embodiment of the present invention.

Detailed ways

The following description and drawings illustrate specific embodiments herein sufficiently to enable those skilled in the art to practice them. Portions and features of some embodiments may be included in or substituted for those of other embodiments. The scope of the embodiments herein includes the full scope of the claims, and all available equivalents of the claims.

As used herein, the terms "first", "second", etc. are used only to distinguish one element from another element and do not require or imply any actual relationship or order between these elements. In fact a first element can also be called a second element and vice versa. Furthermore, the terms "comprises," "comprises," or any other variation thereof are intended to cover a non-exclusive inclusion, such that a structure, system or apparatus including a list of elements includes not only those elements, but also others not expressly listed elements, or also elements inherent to the structure, system or equipment.

To solve various problems in traditional employee information management and provide enterprises with better data access, information integration and compliance management solutions, the present invention provides a panoramic file generation method and system based on employee information integration. Automate and consolidate employee information to provide more comprehensive, high-quality employee profiles.

As shown in Figure 1, an embodiment of the present invention provides a panoramic profile generation method based on employee information integration, which includes the following steps:

Step S10: Receive data packets imported from internal and external data sources of the enterprise, and parse the data packets to obtain employee information resource text;

Among them, when parsing the data package to obtain the employee information resource text, it includes: identifying the format of the text file, table file or database file contained in the data package; analyzing the data package based on the loaded text parser, table processing library and database connection tool Analyze, extract employee information data and convert it into a structured text form; the extracted employee information data includes basic personal information, work experience and educational background, and convert it into a structured text form using XML or JSON format.

Step S20: Establish an employee information classification model, traverse the content of the input employee information resource text text by text, identify and classify employee information data based on the keyword database of the file information list, determine the employee's file category, and generate a corresponding file The file information item architecture tree of the information list;

Step S30: Load the preconfigured module database, filter the corresponding template based on the file category and the file information item architecture tree and write it according to the format, and automatically generate the panoramic file.

In step S10 of this embodiment, the system receives data packages imported from internal and external data sources of the enterprise. The data packages may include various file formats, such as text files, table files, or database files. The data package contains employee information resource text. Use text parsers, table processing libraries and database connection tools to parse data packets. This process can identify different file formats, then extract employee information data and convert it into structured text form, usually in XML or JSON format. The data package contains employee information such as basic personal information, work experience and educational background.

In step S20, during the construction of the employee information classification model, during the data collection and processing stages, employee information data is collected, duplicates and missing values in the data are removed, the above data is converted into a standardized document data format, and each document is Data identifies document category tags, and employee information data comes from the company's internal database, external data sources, or is provided by users. Then perform keyword and phrase recognition, using natural language processing (NLP) technology, the system automatically identifies and extracts keywords and phrases in the document data based on the established keyword and phrase database, which helps to distinguish employee information in more detail. different parts. Then the data set is divided to convert the document data into a data set represented by a computer-processable feature vector. Subsequently, the data set was divided into training and test sets for training and performance evaluation of the classification model. During model training and testing, the naive Bayes algorithm is used to build a classification model. The goal of the classification model is to learn how to assign document data to different archive categories. When training the model, the training set data is used for training. When testing the model, use the test set data to evaluate the performance of the model, and the system can adjust the model parameters based on the performance evaluation results. Finally, a profile information item architecture tree is generated, and the trained classification model is used to classify the employee information text. Based on the classification results, an information item architecture tree for each profile category is generated. The information item architecture tree includes different information items, such as personal Information, work experience, educational background, etc. Different sections used to organize and store employee information.

In step S30, file generation and calling include loading a preconfigured module database and generating a panoramic file. In the step of loading the preconfigured module database, the system loads the preconfigured module database. The module database contains the information required for generating employee information files. Various templates are used to generate panoramic files of different file categories; when generating panoramic files, based on the file category and file information item architecture tree, the system filters applicable templates and writes the data in the specified format to automatically generate panoramic files. ,The panoramic archive includes data of different information ,items, organized according to the pre-defined information item ,architecture tree structure.

The panoramic file generation method based on employee information integration of the present invention provides high-quality panoramic file generation and access through automation and integration of employee information. This helps improve data management efficiency and decision support within the enterprise.

In some embodiments, as shown in Figure 2, the panoramic profile generation method based on employee information integration also includes calling the panoramic profile; the calling of the panoramic profile includes the following steps:

Step S101: Obtain the file retrieval request input by the client. The file retrieval request includes two main parts:

Employee Unique Identifier: The user provides the employee's unique identifier, usually a unique identification number or employee ID, used to accurately identify the employee;

Retrieve requested project-specific information: Users specify project-specific information they want to retrieve, such as an employee's education, work experience, or project information within a specific date range.

Step S102: Retrieve the panoramic file data of the corresponding employee from the panoramic file resource database according to the employee's unique identifier, and filter and organize according to the retrieval requirements;

Step S103: Identify keyword data related to the retrieval requirements for the screened panoramic archive data, including but not limited to date and project name;

Step S104: Based on the keyword data, screen out the panoramic archive resource data that meets the retrieval requirements, and generate the corresponding panoramic archive text;

Step S105: Export the generated employee panoramic file text for user review or further processing.

In this embodiment, according to the user's request, relevant information is retrieved and organized from the panoramic archive resources to meet the specific needs of the user, which helps the user to quickly access the required employee information and improves the usability and accessibility of the information.

In the embodiment of the present invention, establishing an employee information classification model includes the following steps:

Step 1. Data collection and processing:

Collect employee information data, remove duplicates and missing values, convert it into document data in a standardized data format, identify document category labels, and use it as sample data; where the employee information data includes personal information, work experience, educational background, and comes from internal Enterprise databases, external data sources or user-provided information;

Step 2. Keyword and phrase identification:

Based on the established keyword and phrase database, use natural language processing (NLP) technology to automatically identify and extract keywords and phrases in document data;

Step 3. Dataset partitioning:

Convert document data into a data set represented by computer-processed feature vectors, and divide the data set into a training set and a test set;

Step 4. Model training and testing:

Use the Naive Bayes algorithm to build a classification model, and use the training set data to train the classification model and learn to assign document data to different archive categories; use the test set to test the recall rate of the model, and evaluate the model based on the performance evaluation results. Parameter adjustment;

Step 5. Generate the file information item architecture tree:

Use the trained classification model to classify employee information texts and determine the profile category of each text; based on the classification results, generate an information item architecture tree for each profile category, including different information items: personal information, work experience, Education background;

Step 6. Deploy the model:

Deploy the trained model to the panoramic file generation system for automatic classification of employee information.

In this embodiment, when using the Naive Bayes algorithm to build a classification model, it also includes calculating the category prior probability. The steps are as follows:

Prepare a training data set containing classified documents, where the training data set includes document data and category labels corresponding to the document data;

Make statistics on the training data set, calculate the number of documents under each category, and obtain the frequency of documents in each category;

Calculate the total number of documents in the training dataset and calculate the prior probability of the category by the formula:

P(C)＝N/N_total

Among them, P(C) is the prior probability of category C; N is the number of documents belonging to category C; N_total is the total number of documents;

The obtained prior probability of each category is used to predict the category of new documents.

Among them, the working process of calculating the category prior probability when using the Naive Bayes algorithm to build a classification model is:

(1) Prepare training data set: First, collect a training data set containing classified documents. Each document should be associated with its corresponding category label, and the training data set includes document data and the category labels corresponding to the document data.

(2) Count the number of documents: Count the training data set and calculate the number of documents under each category. This is to determine the frequency of documents in each category; the number of documents indicates how many documents there are in each category.

(3) Calculate the total number of documents: Calculate the total number of documents in the entire training data set, that is, the sum of the number of documents in all categories.

(4) Calculate the prior probability of the category: Calculate the prior probability (P(C)) for each category C. Among them, the prior probability represents the probability that a document belongs to a specific category C; the following formula is used to calculate the prior probability of a category:

P(C)＝N/N_total

Among them, P(C) is the prior probability of category C, N is the number of documents belonging to category C, and N_total is the total number of documents.

Illustrative: Assume that an employee profile classification model is being built, in which the categories of documents are different types of employees (for example, full-time employees, part-time employees, interns, etc.); among them, a training data set has been prepared, which includes the following information :

100 documents are labeled "Full-time Employees".

50 documents are labeled "Part-time Employees."

30 documents are marked "Intern".

Now, we need to calculate the prior probability for each class.

1. For the “full-time employee” category:

N ("full-time employees") = 100 (100 documents belong to the full-time employee category)

N_total (total number of documents) = 100 + 50 + 30 = 180

Calculate the prior probability: P("full-time employee")=100/180≈0.5556

2. For the “part-time employee” category:

N ("part-time employees") = 50

N_total (total number of documents) = 180

Calculate the prior probability: P ("part-time employee") = 50/180≈0.2778

3. For the “Intern” category:

N("Intern")=30

N_total (total number of documents) = 180

Calculate the prior probability: P("Intern")=30/180≈0.1667.

Therefore, the prior probability of each category has been calculated above, which will be used to predict the category of the new document. The posterior probability is calculated based on the feature vector of the new document and Bayes' theorem. Based on the category with the highest posterior probability, the category of the new document can be predicted, which helps the model automatically classify the new document.

In this embodiment, when predicting the category of a new document based on the obtained prior probability of each category, it also includes calculating the feature conditional probability (P(x|C)), that is, given the category C, it is observed that The probability of feature x, the steps are as follows:

According to the training data set of classified documents, the documents in the training data set are counted, and the number of occurrences of each feature under each category is calculated; wherein the training data set includes document data, category labels corresponding to the document data, and the number of occurrences of each feature in the document. Characteristics that emerge;

In the case of a given category C, the frequency of feature x; if the document data has M different features, it is expressed as: N(x, C), the frequency of feature x under category C is for each category and each Each feature generates a frequency;

Calculate the relative frequency P(x|C) of feature x in each category C, and obtain the conditional probability of each feature under each category:

P(x|C)＝N(x,C)/N(C)

Among them, P(x|C) is the conditional probability of feature x under category C; N(x,C) is the frequency of feature x under category C; N(C) is the total number of documents under category C.

For example, assuming there is an employee file classification model, the present invention has constructed a frequency matrix, which contains the number of occurrences of each feature under each category. The following is a partial example:

Suppose the invention has three categories: A, B, C; and four features: X, Y, Z, W. An example of a partial frequency matrix is shown below:

A B C X 20 15 10 Y 12 18 9 Z 8 7 5 W 6 4 3

Then, calculate the feature conditional probability of the present invention, where the present invention calculates the conditional probability of feature X under category A:

P(X|A)＝N(X,A)/N(A)＝20/(20+12+8+6)≈0.4651

This represents the probability of observing feature X under category A. Similarly, the present invention can calculate the conditional probabilities of other features under different categories. The conditional probabilities will be used in the naive Bayes classification model to help the model predict the category of new documents, and calculate the posterior of each category based on the observation of features in the new document. empirical probability to determine the most likely category.

In this embodiment, using the Naive Bayes classification model to predict the category of a new document includes the following steps:

Prepare the trained Naive Bayes classification model, including the prior probability of the category (P(C)) and the conditional probability of the feature (P(x|C)) calculated from the training data;

Prepare new documents, perform the same text preprocessing on the new documents as the training data, and convert the new documents into feature vectors;

For each category C, use Bayes’ theorem to calculate the posterior probability P(C|X), where X represents the feature vector of the new document:

P(C|X)＝P(X|C)*P(C)/P(X)

Among them, P(C|X) is the posterior probability of category C given feature X; P(X|C) is the conditional probability of feature X under category C, obtained from the training data; P(C ) is the prior probability of category C, obtained from the training data; P(X) is the marginal probability of feature X, calculated by summing the posterior probabilities of each category. In classification, P(X) is for each The categories are the same;

After calculating the posterior probability for each category, the category with the highest posterior probability is selected as the predicted category for the new document; that is:

Prediction category=argmax P(C|X).

In this embodiment, for example, it is assumed that the present invention has a trained Naive Bayes classification model for classifying news articles, in which there are two categories: sports and technology; now, the present invention has a new sports News articles need to be classified:

1. This invention first performs preprocessing on this news article, including word segmentation, removal of stop words, etc.

2. For this article, the present invention extracts features from the model, which may include word frequency, bag-of-words model, etc.

3. Next, the present invention uses the Naive Bayes model to calculate the posterior probability that the article belongs to the sports and technology categories according to Bayes theorem.

4. Finally, the present invention selects the category with the highest posterior probability as the classification result of the article. If P(Sports|X)>P(Technology|X), it is classified as sports news.

This process assigns new documents to the most likely category so that the new document can be classified correctly.

In some embodiments, generating an archive information item architecture tree corresponding to the archive information list includes:

Define the file information list for employee information integration, including information items for employee information integration;

Each information item in the archive information list is hierarchically structured to create an archive information item architecture tree, and a unique tree structure label is assigned to each information item in the information item architecture tree to establish an information item architecture tree.

Among them, when the panoramic file is generated, employee information is mapped according to the corresponding information items in the information item architecture tree and stored in the panoramic file resource database. A complete panoramic file is created for the user to use the employee's unique identifier and retrieve the required information. Specific project information queries and retrieves the corresponding employee's panoramic profile data.

The invention proposes a panoramic file generation method based on employee information integration to effectively manage and utilize employee information in internal and external data sources of the enterprise, and has the following advantages:

1. Panoramic profile generation: This method can transform employee information from internal and external data sources into panoramic profiles, including personal information, work experience, educational background, etc. This panoramic archive is organized in a structured manner, making the information easily accessible and manageable.

2. Automatic classification and file generation: This invention uses an employee information classification model and based on a keyword database to automatically identify and classify employee information. Then, according to the structure of the file information list, the file information item architecture tree is automatically generated, thereby realizing automatic file generation and reducing the workload of manual processing.

3. Automatic generation of templates: The method also supports the automatic generation of templates. According to the file category and information item architecture tree, select the appropriate template and fill the data into the template. This ensures that the resulting panorama archives comply with specific standards and formats.

4. Intelligent file calling: In some embodiments, this method also provides a panoramic file calling function, allowing users to query and retrieve panoramic file data based on employee unique identifiers and specific project information. This adds intelligence and interactivity to the system, making employee information more accessible.

5. Naive Bayes classification model: Use the Naive Bayes algorithm to build a classification model to achieve automatic classification of new documents and ensure the flexibility and performance of the system.

6. Efficient employee information integration: The present invention improves the integration and management efficiency of employee information and is suitable for various enterprises, especially in human resources and employee file management.

In summary, the panoramic file generation method based on employee information integration of the present invention will greatly improve the information management efficiency of the enterprise, reduce manual work, ensure the consistency and integrity of data, and provide highly accessible employees information resource. This is of great value to the operation and management of the enterprise.

It should be understood that although the above is described in a certain order, these steps are not necessarily performed in the order described above. Unless explicitly stated in this article, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, some steps of this embodiment may include multiple steps or stages. These steps or stages are not necessarily executed at the same time, but may be executed at different times. The order of execution of these steps or stages does not necessarily change. It must be performed sequentially, but may be performed in turn or alternately with other steps or at least part of steps or stages in other steps.

In one embodiment, as shown in Figure 3, a panoramic profile generation system based on employee information integration is provided, including:

Data import module: used to receive data packets from internal and external data sources within the enterprise, and parse these data packets to obtain employee information resource text; it can extract employee information resource text, providing raw data for subsequent processing.

Employee information classification model: used to automatically identify and classify employee information data and determine the employee's profile category; generate a profile information item architecture tree based on the information items defined in the profile information list; using the naive Bayes classification model, it can effectively Employee information is assigned to different profile categories. This model also utilizes a predefined keyword database to identify and tag key information, making subsequent profile generation more targeted.

Preconfigured module database: Contains predefined templates and formats, filters and automatically generates panoramic archives based on archive categories and archive information item architecture trees.

Panoramic archive resource database: used to store the generated panoramic archive data; employee information is stored according to the structure of the information item architecture tree;

Panoramic file call module: allows the user to retrieve and organize panoramic file data based on the employee's unique identifier and the specific project information required for retrieval, and generate panoramic file text that meets the retrieval requirements, and allows users to review it.

Among them, the panoramic file generation system based on employee information integration also includes a text parser, a table processing library and a database connection tool; the text parser, table processing library and database connection tool are respectively used to parse the data contained in the imported data package. Text file, table file or database file format to extract and convert employee information data into a structured text form.

Among them, the panoramic profile generation system based on employee information integration also includes a naive Bayes classification model for automatic classification of employee information, using the training data set of classified documents to calculate the prior probability of categories and the conditional probabilities of features. , which facilitates predicting the category of new documents.

In this embodiment, the panoramic file generation system based on employee information integration of the present invention provides a powerful employee information management solution for enterprises and organizations. The system receives, integrates, categorizes and stores employee information in an efficient manner and then automatically generates panoramic profiles based on a pre-defined information architecture. The panoramic file generation system based on employee information integration adopts the steps of the panoramic file generation method based on employee information integration as mentioned above during execution. Therefore, in this embodiment, the operation process of the panoramic file generation system based on employee information integration does not change. More details.

In one embodiment, an embodiment of the present invention also provides a computer device, including at least one processor, and a memory communicatively connected to the at least one processor, and the memory stores instructions that can be executed by the at least one processor. , the instruction is executed by at least one processor, so that at least one processor executes the panoramic profile generation method based on employee information integration. When the processor executes the instruction, the steps of the above panoramic profile generation method based on employee information integration are implemented.

In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium stores computer instructions, and the computer instructions are used to cause the computer to execute the steps of a panoramic profile generating method based on employee information integration.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through computer programs represented by computer instructions. The computer program can be stored in a non-volatile computer readable device. In the storage medium, when executed, the computer program may include the processes of the above method embodiments. Any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory.

Non-volatile memory may include read-only memory, magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory or external cache memory. By way of illustration and not limitation, RAM can be in many forms, such as static random access memory or dynamic random access memory.

The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the protection scope of the present invention. Inside.

Claims (10)

1. A panoramic profile generation method based on employee information integration, which is characterized by including the following steps: Receive data packets imported from internal and external data sources within the enterprise, and parse the data packets to obtain employee information resource text; Establish an employee information classification model, traverse the content of the input employee information resource text text by text, identify and classify employee information data based on the keyword database of the file information list, determine the employee's file category, and generate a corresponding file information list File information item structure tree; Load the preconfigured module database, filter the corresponding templates based on the file category and file information item architecture tree, and write them according to the format to automatically generate panoramic files. 2. The panoramic file generation method based on employee information integration according to claim 1, characterized in that the panoramic file generation method based on employee information integration also includes the calling of the panoramic file; the calling of the panoramic file includes the following step: Obtain the file retrieval request input by the user, and the file retrieval request includes the employee's unique identifier and the specific project information required for retrieval; According to the employee's unique identifier, retrieve the corresponding employee's panoramic file data from the panoramic file resource database, and filter and organize it according to the retrieval requirements; For the screened panoramic archive data, identify the keyword data related to the retrieval requirements, including but not limited to date and project name; Based on the keyword data, screen out the panoramic archive resource data that meets the retrieval requirements, and generate the corresponding panoramic archive text; Export the generated employee panoramic file text for users to review or further process. 3. The panoramic file generation method based on employee information integration according to claim 2, characterized in that when parsing the data package to obtain the employee information resource text, it includes: identifying text files, table files or databases contained in the data package. The format of the file; parses the data package based on the loaded text parser, table processing library and database connection tool, extracts employee information data and converts it into a structured text form. 4. The panoramic profile generation method based on employee information integration according to claim 1, characterized in that establishing an employee information classification model includes the following steps: Step 1) Data collection and processing: Collect employee information data, remove duplicates and missing values, convert it into document data in a standardized data format, identify document category labels, and use it as sample data; where the employee information data includes personal information, work experience, educational background, and comes from internal Enterprise databases, external data sources or user-provided information; Step 2) Keyword and phrase identification: Based on the established keyword and phrase database, use natural language processing technology to automatically identify and extract keywords and phrases in document data; Step 3) Dataset partitioning: Convert document data into a data set represented by computer-processed feature vectors, and divide the data set into a training set and a test set; Step 4) Model training and testing: Use the Naive Bayes algorithm to build a classification model, and use the training set data to train the classification model and learn to assign document data to different archive categories; use the test set to test the recall rate of the model, and evaluate the model based on the performance evaluation results. Parameter adjustment; Step 5) Generate the file information item architecture tree: Use the trained classification model to classify employee information texts and determine the profile category of each text; based on the classification results, generate an information item architecture tree for each profile category, including different information items: personal information, work experience, Education background; Step 6) Deploy the model: Deploy the trained model to the panoramic file generation system for automatic classification of employee information. 5. The panoramic file generation method based on employee information integration according to claim 4, characterized in that when using the Naive Bayes algorithm to build the classification model, it also includes calculating the category prior probability, and the steps are as follows: Prepare a training data set containing classified documents, where the training data set includes document data and category labels corresponding to the document data; Make statistics on the training data set, calculate the number of documents under each category, and obtain the frequency of documents in each category; Calculate the total number of documents in the training dataset and calculate the prior probability of the category by the formula: P(C)＝N/N_total Among them, P(C) is the prior probability of category C; N is the number of documents belonging to category C; N_total is the total number of documents; The obtained prior probability of each category is used to predict the category of new documents. 6. The panoramic profile generation method based on employee information integration according to claim 5, characterized in that when predicting the category of a new document based on the obtained prior probability of each category, it also includes calculating the characteristic conditional probability P(x| C), that is, the probability of observing feature x given category C, the steps are as follows: According to the training data set of classified documents, the documents in the training data set are counted, and the number of occurrences of each feature under each category is calculated; wherein the training data set includes document data, category labels corresponding to the document data, and the number of occurrences of each feature in the document. Characteristics that emerge; In the case of a given category C, the frequency of feature x; if the document data has M different features, it is expressed as: N(x, C), the frequency of feature x under category C is for each category and each Each feature generates a frequency; Calculate the relative frequency P(x|C) of feature x in each category C, and obtain the conditional probability of each feature under each category: P(x|C)＝N(x,C)/N(C) Among them, P(x|C) is the conditional probability of feature x under category C; N(x,C) is the frequency of feature x under category C; N(C) is the total number of documents under category C. 7. The panoramic profile generation method based on employee information integration according to claim 6, characterized in that when using a naive Bayes classification model to predict the category of a new document, it includes the following steps: Prepare the trained Naive Bayes classification model, including the prior probability of the category (P(C)) and the conditional probability of the feature (P(x|C)) calculated from the training data; Prepare new documents, perform the same text preprocessing on the new documents as the training data, and convert the new documents into feature vectors; For each category C, use Bayes’ theorem to calculate the posterior probability P(C|X), where X represents the feature vector of the new document: P(C|X)＝P(X|C)*P(C)/P(X) Among them, P(C|X) is the posterior probability of category C given feature X; P(X|C) is the conditional probability of feature X under category C, obtained from the training data; P(C ) is the prior probability of category C, obtained from the training data; P(X) is the marginal probability of feature X, calculated by summing the posterior probabilities of each category. In classification, P(X) is for each The categories are the same; After calculating the posterior probability for each category, the category with the highest posterior probability is selected as the predicted category for the new document; that is: Prediction category=argmax P(C|X). 8. The panoramic profile generation method based on employee information integration according to claim 7, characterized in that the generated profile information item architecture tree corresponding to the profile information list includes: Define the file information list for employee information integration, including information items for employee information integration; Each information item in the archive information list is hierarchically structured to create an archive information item architecture tree, and a unique tree structure label is assigned to each information item in the information item architecture tree to establish an information item architecture tree. 9. The panoramic archive generation method based on employee information integration according to claim 8, characterized in that when the panoramic archive is generated, the employee information is mapped according to the corresponding information items in the information item architecture tree and stored in the panoramic archive resource database. , create a complete panoramic file for the user to query and retrieve the corresponding employee's panoramic file data using the employee's unique identifier and the specific project information requested. 10. A panoramic profile generation system based on employee information integration, used to execute the panoramic profile generation method based on employee information integration according to any one of claims 1 to 9, characterized in that the system includes: Data import module: used to receive data packages from internal and external data sources within the enterprise, and parse these data packages to obtain employee information resource text; Employee information classification model: used to automatically identify and classify employee information data, and determine the employee's profile category; generate a profile information item architecture tree based on the information items defined in the profile information list; Preconfigured module database: Contains predefined templates and formats, filters and automatically generates panoramic archives based on archive categories and archive information item architecture trees; Panoramic archive resource database: used to store the generated panoramic archive data; employee information is stored according to the structure of the information item architecture tree; Panoramic file call module: allows the user to retrieve and organize panoramic file data based on the employee's unique identifier and the specific project information required for retrieval, and generate panoramic file text that meets the retrieval requirements, and allows users to review it.