Enterprise external migration prediction method and device based on news public opinions, electronic equipment and storage medium

Abstract

The invention discloses an enterprise external migration prediction method and device based on news public opinions, electronic equipment and a storage medium. The method comprises the steps of obtaining a target enterprise list; public opinion information is collected according to the target enterprise list; preprocessing the public opinion information; in combination with the preprocessed public opinion information, establishing a plurality of enterprise external migration prediction models through different algorithms; evaluating all the enterprise ex-migration prediction models to obtain an optimal enterprise ex-migration prediction model; and obtaining new public opinion information of a new target enterprise list, preprocessing the new public opinion information, and inputting the preprocessed new public opinion information into the optimal enterprise ex-migration prediction model to obtain the ex-migration tendency of the target enterprise. According to the method and the device, the technical problem of hysteresis of ex-migration early warning of enterprises in related technologies is solved.

Description

Technical Field

[0001] The present invention relates to the field of text classification technology, and in particular to a method, device, electronic device and storage medium for predicting enterprise relocation based on news and public opinion. Background Art

[0002] As society continues to develop, businesses often relocate or even relocate parts of their operations based on their development needs. As the main actors in market economic activity, businesses, especially key tax-generating enterprises, are a crucial source of local fiscal revenue. Relocation can significantly reduce local tax revenue, impacting local government revenue and, in turn, local urban development. Currently, local governments often receive information about relocations late, necessitating the use of technology to provide early warning of relocations.

[0003] Invention patent CN114511250A discloses a method, device, and computer-readable storage medium for predicting enterprise relocation risk. This method obtains an enterprise relocation information dataset, uses basic enterprise information and operational information as feature variables, vectorizes the dataset, and constructs an enterprise relocation decision tree model based on the sample data. When assessing whether a target enterprise faces relocation risk, the target enterprise's characteristics are input into the enterprise relocation decision tree to obtain a relocation risk prediction result. Machine learning based on big data can improve the accuracy of the prediction results, while reducing the workload of manual identification and avoiding human misjudgment. Invention patent CN109377058A discloses an enterprise relocation risk assessment method based on a logistic regression model. This method collects enterprise relocation case information and desensitized telecommunications data provided by operators. Using data binning and the Pearson correlation coefficient, it eliminates low-predictive power and high-correlation indicators. Using logistic regression, it constructs an enterprise relocation prediction model, ultimately outputting the enterprise's relocation probability. Both invention patents utilize quantitative models to predict enterprise relocation.

[0004] Both of these inventions utilize only relatively lagging data as characteristic variables for enterprises. Even if a company's relocation risk is predicted, these behaviors may have already occurred for some time, and relevant data may have accumulated to a certain extent. This indicates that such predictions are based on signs after the company has already demonstrated relocation intentions or behaviors, and thus still have a lag. Therefore, early warning of relocation still requires more advanced information.

[0005] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0006] The embodiments of the present invention provide a method, device, electronic device and storage medium for predicting enterprise relocation based on news and public opinion, so as to at least solve the technical problem of lag in enterprise relocation warning in related technologies.

[0007] According to one aspect of an embodiment of the present invention, a method for predicting enterprise relocation based on news public opinion is provided, including: obtaining a list of target enterprises; collecting public opinion information based on the target enterprise list; preprocessing the public opinion information; combining the preprocessed public opinion information to establish multiple enterprise relocation prediction models through different algorithms; evaluating all enterprise relocation prediction models to obtain an optimal enterprise relocation prediction model; obtaining new public opinion information of a new target enterprise list, preprocessing the new public opinion information and inputting it into the optimal enterprise relocation prediction model to obtain the relocation tendency of the target enterprises.

[0008] Optionally, the target enterprises in the target enterprise list are the top 20% of enterprises in terms of tax contribution in the target area.

[0009] Optionally, preprocessing the public opinion information includes: marking the public opinion data in the public opinion information and constructing a data set.

[0010] Optionally, labeling the public opinion data of the public opinion information includes: dividing the public opinion information into information with relocation risk and information without relocation risk, wherein the information with relocation risk includes: information about enterprises going to other places for inspection, signing cooperation agreements with foreign governments, projects being put into production in other places, and foreign government leaders visiting the enterprise for exchanges.

[0011] Optionally, constructing the data set includes the following steps: deduplicating public opinion information marked as having no risk of relocation, randomly extracting the same amount of data from the deduplicated public opinion information having no risk of relocation and the public opinion information having risk of relocation, and constructing the data set.

[0012] Optionally, before establishing multiple enterprise relocation prediction models through different algorithms, it is also necessary to segment the public opinion information and delete words that have no practical meaning; and perform digital vector conversion.

[0013] Optionally, the model with the highest evaluation accuracy adopts the xgboost model.

[0014] According to another aspect of an embodiment of the present invention, a device for predicting enterprise relocation based on news public opinion is also provided, including: a list acquisition module for obtaining a list of target enterprises; a public opinion information acquisition module for collecting public opinion information based on the target enterprise list; a preprocessing module for preprocessing the public opinion information; a model building module for combining the preprocessed public opinion information to establish multiple enterprise relocation prediction models through different algorithms; an evaluation module for evaluating all enterprise relocation prediction models to obtain an optimal enterprise relocation prediction model; an application module for obtaining new public opinion information of a new target enterprise list, and inputting the new public opinion information obtained by the public opinion information acquisition module into the optimal enterprise relocation prediction model after processing through the preprocessing module to obtain the relocation tendency of the target enterprise.

[0015] According to another aspect of an embodiment of the present invention, an electronic device is provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods of various embodiments of the present invention when running.

[0016] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0017] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0018] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0019] According to another aspect of the embodiments of the present invention, a computer program is provided. When the computer program is executed by a processor, the methods in various embodiments of the present invention are implemented.

[0020] In an embodiment of the present invention, a method for predicting enterprise relocation based on news and public opinion is provided, comprising: obtaining a list of target enterprises; collecting public opinion information based on the list of target enterprises; preprocessing the public opinion information; establishing multiple enterprise relocation prediction models using different algorithms based on the preprocessed public opinion information; evaluating all enterprise relocation prediction models to obtain an optimal enterprise relocation prediction model; obtaining new public opinion information for a new list of target enterprises, preprocessing the new public opinion information, and inputting the optimal enterprise relocation prediction model to obtain the relocation tendency of the target enterprises. Existing technologies mainly rely on basic enterprise information and operating characteristics, such as enterprise type, industry, paid-in capital, length of operation, tax change rate, number of employees applying for social security, number of outbound investments, and whether they are Class A taxpayers. However, these characteristics are generally not reflected until the enterprise decides to relocate. Even if the enterprise relocation is predicted, the government's efforts to retain the enterprise are unlikely to be successful. The present invention, however, uses public opinion information to understand public opinion information such as enterprises going on outbound inspections, which is more timely. Local governments can learn of such information and visit the enterprises in a timely manner to inquire about their demands, thereby retaining the enterprises by addressing their demands. This solves the technical problem of delayed early warning of enterprise relocation in related technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0022] Figure 1 This is a flowchart of a method for predicting enterprise relocation based on news and public opinion according to an embodiment of the present invention;

[0023] Figure 2 2 is a schematic diagram of a device for predicting enterprise relocation based on news and public opinion according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] According to one aspect of an embodiment of the present invention, a method for predicting enterprise relocation based on news and public opinion is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0027] Figure 1 FIG. 1 is a flow chart of a method for predicting enterprise relocation based on news and public opinion according to an embodiment of the present invention. Figure 1 As shown, the method includes the following steps:

[0028] Step S101: Obtain a list of target companies.

[0029] In terms of tax concentration, the top 20% of enterprises in a region contribute 80% of taxes, and the major taxpayers contribute the most to fiscal revenue mainly in the top 20% of enterprises. Therefore, the target enterprises in the target enterprise list are the top 20% of enterprises in terms of tax contribution in the target region.

[0030] Step S102: Collect public opinion information based on the target enterprise list.

[0031] Step S103: pre-process the public opinion information.

[0032] The above-mentioned pre-processing of public opinion information may include: marking public opinion data in the public opinion information and constructing a data set.

[0033] Among them, the labeling of public opinion data includes: dividing public opinion information into information with relocation risk and information without relocation risk. Information with relocation risk includes: enterprises going to other places for inspection, signing cooperation agreements with foreign governments, projects being put into production in other places, and information about foreign government leaders visiting enterprises for exchanges, etc. Information without relocation risk is information other than relocation risk information.

[0034] The above-mentioned construction of the dataset may include the following steps: deduplicate the public opinion information labeled as having no risk of relocation. Since the number of public opinion information with the risk of relocation is much less than that without the risk of relocation, to balance the two labels, randomly select the same number of data from the deduplicated public opinion information without the risk of relocation and the public opinion information with the risk of relocation to construct the dataset.

[0035] Step S104: Combine the preprocessed public opinion information and establish multiple enterprise relocation prediction models through different algorithms.

[0036] In this step, before establishing multiple enterprise relocation prediction models through different algorithms, it is also necessary to segment the public opinion information and delete the words without practical significance; and perform digital vector conversion.

[0037] Among them, the words without practical significance include: "of", "in", and some numbers, etc.

[0038] The above-mentioned digital vector conversion can adopt the tfidf algorithm to convert the dataset into digital vectors recognizable by the computer.

[0039] The above-mentioned establishment of multiple enterprise relocation prediction models through different algorithms may specifically include: selecting multiple classifier models for modeling. The classifier models include: logistic regression, random forest, naive Bayes, support vector machine, and xgboost model, etc.

[0040] Step S105: Evaluate all the enterprise relocation prediction models to obtain the optimal enterprise relocation prediction model.

[0041] In an optional embodiment, evaluating all the enterprise relocation prediction models mainly uses the f1-score as a measurement index. This index is the harmonic mean of precision and recall, and precision and recall are equally important. The expression of the f1-score is: <>

[0042] F1 = 2 * precision * recall / (precision + recall)

[0043] Among them, precision is the precision rate, which refers to the proportion of positive samples among the positive examples determined by the classifier model. Recall is the recall rate, which refers to the proportion of the predicted positive examples among the total positive examples. The higher the f1-score value, the more accurate the model. That is, the model with the highest accuracy rate obtained through evaluation is used as the enterprise relocation prediction model to improve the accuracy rate of new public opinion data prediction.

[0044] Table 1 shows the f1-score values ​​obtained through cross-validation of methods such as logistic regression, random forest, naive Bayes, support vector machine, and XGBoost models. The XGBoost model performed best, with an f1-score of 0.726. Furthermore, XGBoost gradient boosting tree is an ensemble learning algorithm that iteratively trains a series of weak classifiers (decision trees), with each iteration attempting to fit the residuals of the previous iteration. Ultimately, these weak classifiers are combined to form a more powerful model. XGBoost excels at processing large amounts of data, is capable of parallel computing, and has high efficiency. Furthermore, by iteratively optimizing the model, it generally outperforms other algorithms in terms of prediction accuracy. Therefore, the model with the highest evaluation accuracy uses the XGBoost model, which can also improve prediction accuracy to a certain extent.

[0045] Table 1 Evaluation results

[0046] method f1-score value Logistic regression 0.672 Random Forest 0.705 Naive Bayes 0.6705 Support Vector Machine 0.4583 Xgboost model 0.726

[0047] Step S106: Obtain new public opinion information of a new list of target enterprises, pre-process the new public opinion information, and input it into the optimal enterprise relocation prediction model to obtain the relocation tendency of the target enterprises.

[0048] In addition, the data of enterprises with relocation tendencies can be sent to local government departments, so that local governments can communicate with enterprises in a timely manner to retain enterprises to the greatest extent possible and avoid tax source loss.

[0049] An embodiment of the present invention provides a method for predicting enterprise relocation based on news and public opinion. The method comprises the following steps: obtaining a list of target enterprises; collecting public opinion information based on the list of target enterprises; preprocessing the public opinion information; establishing multiple enterprise relocation prediction models using different algorithms based on the preprocessed public opinion information; evaluating all enterprise relocation prediction models to obtain an optimal enterprise relocation prediction model; obtaining new public opinion information for a new list of target enterprises, preprocessing the new public opinion information, and inputting it into the optimal enterprise relocation prediction model to determine the relocation tendency of the target enterprises. Existing technologies primarily rely on basic enterprise information and operating characteristics, such as enterprise type, industry, paid-in capital, length of operation, tax change rate, number of employees applying for social security, number of outbound investments, and whether the target enterprises have an A-level tax rate. However, these characteristics are generally not apparent until the enterprise decides to relocate. Even if the enterprise relocation is predicted, government efforts to retain the enterprise are unlikely to succeed. The present invention, however, uses public opinion information to understand public opinion information such as enterprises' outbound inspections, which is more timely. Upon learning of this information, local governments can promptly visit the enterprises, inquire about their demands, and address their demands to retain them.

[0050] According to another aspect of an embodiment of the present invention, a device for predicting enterprise relocation based on news and public opinion is also provided. The system can execute the enterprise relocation prediction method based on news and public opinion of the above embodiment. The specific implementation method and preferred application scenario are the same as the above embodiment and will not be repeated here.

[0051] Figure 2 is a schematic diagram of a device for predicting enterprise relocation based on news and public opinion according to an embodiment of the present application. Figure 2 As shown, the device includes the following: a list acquisition module 201, a public opinion information acquisition module 202, a preprocessing module 203, a model building module 204, an evaluation module 205 and an application module 206.

[0052] The list acquisition module 201 is used to acquire a list of target enterprises.

[0053] Among them, the public opinion information acquisition module 202 is used to collect public opinion information based on the target enterprise list.

[0054] Among them, the preprocessing module 203 is used to preprocess the public opinion information.

[0055] Among them, the model building module 204 is used to combine the pre-processed public opinion information and establish multiple enterprise relocation prediction models through different algorithms.

[0056] The evaluation module 205 is used to evaluate all enterprise relocation prediction models to obtain the optimal enterprise relocation prediction model.

[0057] Among them, the application module 206 is used to obtain new public opinion information of the new target enterprise list, and inputs the new public opinion information obtained by the public opinion information acquisition module 202 into the optimal enterprise relocation prediction model after being processed by the preprocessing module 203 to obtain the relocation tendency of the target enterprise.

[0058] An embodiment of the present application further provides an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods of various embodiments of the present invention when running.

[0059] The above-mentioned memory may refer to a device inside a computer for storing data and programs, and may include memory, hard disk, etc., wherein the memory may be used to temporarily store running programs and data, the hard disk may be used to store programs and data for a long time, and the memory may be used to enable the computer to read and write data, as well as execute programs; the above-mentioned processor may be responsible for executing instructions in computer programs and performing data processing, and may be responsible for controlling and executing various operations, including arithmetic operations, logical operations, data transmission, etc.

[0060] An embodiment of the present application further provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0061] The above-mentioned computer storage medium may refer to a medium in a computer memory used to store certain discontinuous physical quantities. Computer storage media mainly include semiconductors, magnetic cores, magnetic drums, magnetic tapes, laser disks, etc. The stored program included in the computer-readable storage medium may be a set of instructions that can be recognized and executed by a computer, running on an electronic computer, and serving as an information tool to meet certain needs of people.

[0062] An embodiment of the present application further provides a computer program product, including a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0063] The above-mentioned computer program product may refer to a software program that has been written, tested and released, which can be run on a computer or other device. The computer program product may include an application, an operating system, tool software, etc., which is used to implement specific functions or solve specific problems.

[0064] An embodiment of the present application further provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0065] The above-mentioned non-volatile computer-readable storage medium may refer to a medium for storing data. The non-volatile computer-readable storage medium can keep the data from being lost when the power is off, and can be used to store long-term data, such as operating systems, applications and user files. The non-volatile storage medium may include hard disk drives, solid-state drives, optical disks and flash memory storage devices, etc.

[0066] The embodiments of the present application further provide a computer program, which implements the methods in the above-mentioned embodiments of the present invention when executed by a processor.

[0067] The above-mentioned computer program may refer to a collection of instructions used to tell a computer to perform a specific task or operation. A computer program may be written by a programmer using a specific programming language and may include algorithms, data structures, logic, and control flows. Computer programs may be used for a variety of purposes, including application software, operating systems, and the like.

[0068] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0069] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0070] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0071] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0072] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc. Various media that can store program codes.

[0073] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for predicting enterprise relocation based on news and public opinion, characterized in that: include: Obtain a list of target companies; Collect public opinion information based on the target enterprise list; Preprocessing the public opinion information; Combined with the pre-processed public opinion information, various enterprise relocation prediction models are established through different algorithms; Evaluate all enterprise relocation prediction models to obtain the optimal enterprise relocation prediction model; Obtain new public opinion information of the new list of target enterprises, pre-process the new public opinion information and input it into the optimal enterprise relocation prediction model to obtain the relocation tendency of the target enterprises.

2. The enterprise relocation prediction method based on news public opinion according to claim 1 is characterized in that: The target enterprises in the target enterprise list are the top 20% of enterprises in terms of tax contribution in the target area.

3. The enterprise relocation prediction method based on news public opinion according to claim 1 is characterized in that: Preprocessing the public opinion information includes: marking the public opinion data in the public opinion information and constructing a data set.

4. The enterprise relocation prediction method based on news public opinion according to claim 3 is characterized in that: The labeling of the public opinion data of the public opinion information includes: dividing the public opinion information into information with relocation risk and information without relocation risk, wherein the information with relocation risk includes: information about enterprises going to other places for inspection, signing cooperation agreements with foreign governments, projects being put into production in other places, and foreign government leaders visiting enterprises for exchanges.

5. The enterprise relocation prediction method based on news public opinion according to claim 1 is characterized in that: Constructing the data set includes the following steps: deduplicating the public opinion information marked as having no risk of relocation, randomly extracting the same amount of data from the public opinion information with no risk of relocation and the public opinion information with risk of relocation after deduplication, and constructing the data set.

6. The enterprise relocation prediction method based on news and public opinion according to claim 1 is characterized in that: Before establishing multiple enterprise relocation prediction models through different algorithms, it is also necessary to segment the public opinion information and delete words that have no practical meaning; as well as perform digital vector conversion.

7. The enterprise relocation prediction method based on news and public opinion according to claim 5 is characterized in that: The model with the highest evaluation accuracy adopts the xgboost model.

8. A device for predicting enterprise relocation based on news and public opinion, characterized in that: include: List acquisition module, used to obtain the target enterprise list; A public opinion information acquisition module, used to collect public opinion information according to the target enterprise list; A preprocessing module, used for preprocessing the public opinion information; The model building module is used to combine the pre-processed public opinion information and establish a variety of enterprise relocation prediction models through different algorithms; Evaluation module, used to evaluate all enterprise relocation prediction models and obtain the optimal enterprise relocation prediction model; The application module is used to obtain new public opinion information of the new target enterprise list. The new public opinion information obtained by the public opinion information acquisition module is processed by the preprocessing module and then input into the optimal enterprise relocation prediction model to obtain the relocation tendency of the target enterprise.

9. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein when the program is running, the enterprise relocation prediction method based on news public opinion described in any one of claims 1 to 7 is executed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the enterprise relocation prediction method based on news public opinion as described in any one of claims 1 to 7.

Images