A method and device for identifying enterprise name

Abstract

The present invention provides a method and device for identifying company names, wherein the method comprises: obtaining company text data to be identified; performing event identification on the company text data to obtain a target event; identifying the target event according to a part-of-speech recognition algorithm to obtain a company name, which can solve the problem of low efficiency in manually screening company names, classify the text into events based on event identification, clarify the relationship between event categories and companies, and extract the company name based on the event identification results, thereby achieving the purpose of quickly extracting company names.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method and device for identifying enterprise names. Background Art

[0002] The existing method is to screen manually, so the efficiency is relatively low, the speed is very slow, and the labor cost is high.

[0003] There is a low efficiency problem in manually screening company names through related technologies, and no solution has been proposed yet. Summary of the invention

[0004] The embodiment of the present invention provides a method and device for identifying company names, so as to at least solve the problem of low efficiency in manually screening company names.

[0005] According to an embodiment of the present invention, a method for identifying a company name is provided, comprising:

[0006] Acquire enterprise text data to be identified;

[0007] Performing event recognition on the enterprise text data to obtain target events;

[0008] The target event is identified according to a part-of-speech recognition algorithm to obtain the company name.

[0009] Optionally, before performing event recognition on the enterprise text data to obtain the target event, the method further includes:

[0010] The enterprise text data is preprocessed to obtain a plurality of sentence units.

[0011] Optionally, preprocessing the enterprise text data includes:

[0012] Performing word segmentation processing on the enterprise text data;

[0013] Stop words in the enterprise text data are removed.

[0014] Optionally, performing event recognition on the enterprise text data to obtain target events includes:

[0015] The multiple sentence units are input into a pre-trained target event recognition model to obtain the probabilities of multiple events corresponding to the multiple sentence units output by the target event recognition model, wherein the event whose probability is greater than a preset threshold is the target event.

[0016] Optionally, before performing event recognition on the enterprise text data to obtain a target event, the method further includes:

[0017] Obtaining a predetermined number of training samples;

[0018] Based on the part-of-speech recognition algorithm, the initial event recognition model is trained according to the training samples to obtain the target event recognition model.

[0019] Optionally, based on a part-of-speech recognition algorithm, the initial event recognition model is trained according to the training samples to obtain the target event recognition model, including:

[0020] Dividing each of the training samples into multiple event categories, and respectively labeling the multiple event categories with different event labels;

[0021] Filtering texts outside the event category from the training samples to obtain target training samples;

[0022] Based on the part-of-speech recognition algorithm, the target training sample is input into the initial event recognition model for training to obtain the target event recognition model.

[0023] According to another embodiment of the present invention, there is also provided a device for identifying a company name, comprising:

[0024] A first acquisition module is used to acquire enterprise text data to be identified;

[0025] An event recognition module, used to perform event recognition on the enterprise text data to obtain a target event;

[0026] The name recognition module is used to recognize the target event according to the part-of-speech recognition algorithm to obtain the company name.

[0027] Optionally, the device further comprises:

[0028] The preprocessing module is used to preprocess the enterprise text data to obtain multiple sentence units.

[0029] Optionally, the preprocessing module includes:

[0030] A word segmentation submodule, used for performing word segmentation processing on the enterprise text data;

[0031] The removal submodule is used to remove stop words in the enterprise text data.

[0032] Optionally, the event identification module is also used to

[0033] The multiple sentence units are input into a pre-trained target event recognition model to obtain the probabilities of multiple events corresponding to the multiple sentence units output by the target event recognition model, wherein the event whose probability is greater than a preset threshold is the target event.

[0034] Optionally, the device further comprises:

[0035] A second acquisition module, used to acquire a predetermined number of training samples;

[0036] The training module is used to train the initial event recognition model based on the training samples based on the part-of-speech recognition algorithm to obtain the target event recognition model.

[0037] Optionally, the training module includes:

[0038] A division submodule, used for dividing each of the training samples into a plurality of event categories, and marking different event labels for the plurality of event categories respectively;

[0039] A screening submodule, used for screening out texts outside the event category from the training samples to obtain target training samples;

[0040] The training submodule is used to input the target training sample into the initial event recognition model for training based on a part-of-speech recognition algorithm to obtain the target event recognition model.

[0041] According to yet another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when run.

[0042] According to yet another embodiment of the present invention, there is provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0043] Through the present invention, enterprise text data to be identified is obtained; event identification is performed on the enterprise text data to obtain a target event; the target event is identified according to a part-of-speech recognition algorithm to obtain a company name, which can solve the problem of low efficiency in manually screening company names. Based on event identification, the text is classified into event categories, the relationship between event categories and companies is clarified, and the company name is extracted based on the event identification results, thereby achieving the purpose of quickly extracting the company name. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] 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:

[0045] Figure 1 is a hardware structure block diagram of a mobile terminal of the enterprise name recognition method according to an embodiment of the present invention;

[0046] Figure 2 is a flow chart of a method for identifying a company name according to an embodiment of the present invention;

[0047] Figure 3 is a schematic diagram of enterprise name extraction based on event recognition according to an embodiment of the present invention;

[0048] Figure 4 4 is a block diagram of a company name recognition device according to this embodiment. DETAILED DESCRIPTION

[0049] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0050] It should be noted that the terms "first", "second", etc. in the specification 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.

[0051] Example 1

[0052] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a method for identifying a corporate name according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) processor 102 (processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Optionally, the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0053] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the enterprise name recognition method in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0054] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0055] In this embodiment, a method for identifying an enterprise name running on the above mobile terminal or network architecture is provided. Figure 2 FIG. 1 is a flow chart of a method for identifying a digital enterprise name according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0056] Step S202, obtaining enterprise text data to be identified;

[0057] Step S204, performing event recognition on the enterprise text data to obtain a target event;

[0058] Step S206: Identify the target event according to a part-of-speech recognition algorithm to obtain the company name.

[0059] Through the above steps S202 to S206, the enterprise text data to be identified is obtained; event identification is performed on the enterprise text data to obtain the target event; the target event is identified according to the part-of-speech recognition algorithm to obtain the enterprise name, which can solve the problem of low efficiency in manually screening enterprise names. Based on event recognition, the text is classified into events, the relationship between the event category and the enterprise is clarified, and the enterprise name is extracted based on the event recognition result, thereby achieving the purpose of quickly extracting the enterprise name.

[0060] In an optional embodiment, before performing event recognition on the enterprise text data to obtain the target event, the enterprise text data is preprocessed to obtain multiple sentence units. Specifically, the enterprise text data is segmented to remove stop words in the enterprise text data to obtain multiple sentence units.

[0061] In an embodiment of the present invention, the above-mentioned step S204 may specifically include: inputting the multiple sentence units into a pre-trained target event recognition model, obtaining the probabilities of multiple events corresponding to the multiple sentence units output by the target event recognition model, wherein the event whose probability is greater than a preset threshold is the target event.

[0062] In another optional embodiment, before performing event recognition on the enterprise text data to obtain the target event, a predetermined number of training samples are obtained; based on a part-of-speech recognition algorithm, an initial event recognition model is trained according to the training samples to obtain the target event recognition model. Specifically, each of the training samples is divided into multiple event categories, and different event labels are respectively assigned to the multiple event categories; texts outside the event categories are filtered out from the training samples to obtain target training samples; based on a part-of-speech recognition algorithm, the target training samples are input into the initial event recognition model for training to obtain the target event recognition model.

[0063] The embodiment of the present invention uses event-based recognition to first classify the text into events, clarify the relationship between the event category and the target company organization, and then extract the company name based on the event recognition result, so as to achieve the purpose of quickly extracting and analyzing the company organization name. Figure 3 is a schematic diagram of enterprise name extraction based on event recognition according to an embodiment of the present invention, such as Figure 3 As shown, including:

[0064] Preprocess the input enterprise text data to obtain the input samples of the event recognition module. Preprocessing includes word segmentation and stop word removal. Word segmentation refers to dividing a Chinese character sequence into individual words. The high-frequency common words counted in the financial corpus can be used to improve the accuracy of word segmentation. Stop words are some high-frequency pronouns, conjunctions, prepositions and other words in the text that are meaningless to text recognition. Removing stop words through the stop word list can improve the efficiency of text analysis.

[0065] Based on the massive financial text corpus, the event recognition model is obtained by using text classification algorithms (including but not limited to fastText, transformer and other algorithms) and the event keyword thesaurus accumulated through experience, and the input samples are divided into different event categories, thereby filtering out irrelevant texts. Keywords in the titles of financial corpus (including chapter titles, subtitles, etc.) can be analyzed, and different event labels can be given to this part of the text to train the text classification algorithm and obtain the event recognition model. The trained event recognition model is then used to recognize the input text of the event recognition module to obtain the output text of the event recognition module.

[0066] For example, event categories can be divided into the following categories: main business, R&D investment, corporate cooperation, share reduction, rating adjustment, and examples of various events are shown in Table 1.

[0067] Table 1

[0068] Event Name example Main Business Company A focuses on financial and management services for small and micro enterprises, and provides professional accounting and inventory management software for small and micro enterprises at different growth stages. R&D Investment Company B focuses on accounting and innovative entrepreneurship education, increases investment in the research and development of virtual reality / augmented reality teaching products, and strives to expand the overseas practical training education market. Corporate Cooperation Company X has been deeply involved in the field of customized data centers, strengthening the stickiness of existing customers and deepening strategic cooperation relationships. During the reporting period, Company X once again cooperated with Company C to build data center projects such as ZH13, with a total investment of approximately RMB 1.859 billion. Shareholding reduction In 2018, Company Y reduced its holdings of Caitong Securities' unrestricted tradable shares by a total of 18 million shares through the securities centralized trading system, and the company obtained an investment income of approximately RMB 105 million from the above reduction. Rating Adjustment The credit rating of a certain automobile group was downgraded again by a certain company, from CCC to C.

[0069] Based on the target event category text, the name of the company organization is obtained by using the part-of-speech recognition algorithm (including but not limited to the hidden Markov model, etc.). The part-of-speech recognition algorithm can give each word in the output text of the event recognition module its own part of speech, and the required words can be filtered out through the part of speech. For example, in "Mu* said that he was considering further downgrading the bank", "Mu*" and "the bank (X Bank)" are nouns / pronouns, and "the bank (X Bank)" is in the object position, which is the word that needs to be filtered out, and the final output result is X Bank.

[0070] The following explanation is given using the "rating adjustment" event category as an example.

[0071] Corporate text data: The credit rating of the second largest bank in a certain region, Bank X, is currently B3, and Muhammad said that he is considering further downgrading the bank. A certain region must raise 1.8 billion euros by the 30th of this month to inject capital into Bank X, which is affected by the debt crisis, otherwise it will have to apply for rescue.

[0072] The event recognition module input text includes:

[0073] Text 1: The second largest bank in a region, Bank X, currently has a credit rating of B3. Mu* said it is considering further downgrading the bank.

[0074] Text 2: A certain region must raise 1.8 billion euros to inject capital into Bank X, which is affected by the debt crisis, before the 30th of this month, otherwise it will have to apply for bailout.

[0075] Both text 1 and text 2 include multiple sentence units.

[0076] The event recognition module outputs the text: The second largest bank in a region, Bank X, currently has a credit rating of B3, and Mu* said it is considering further downgrading the bank;

[0077] The company name is output through the part-of-speech recognition algorithm: Bank X.

[0078] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0079] Example 2

[0080] According to another embodiment of the present invention, a device for identifying a company name is provided. Figure 4 is a block diagram of a device for identifying a company name according to this embodiment. Figure 4 As shown, including:

[0081] A first acquisition module 42, used to acquire enterprise text data to be identified;

[0082] An event recognition module 44 is used to perform event recognition on the enterprise text data to obtain a target event;

[0083] The name recognition module 46 is used to recognize the target event according to the part-of-speech recognition algorithm to obtain the company name.

[0084] Optionally, the device further comprises:

[0085] The preprocessing module is used to preprocess the enterprise text data to obtain multiple sentence units.

[0086] Optionally, the preprocessing module includes:

[0087] A word segmentation submodule, used for performing word segmentation processing on the enterprise text data;

[0088] The removal submodule is used to remove stop words in the enterprise text data.

[0089] Optionally, the event identification module 44 is also used to

[0090] The multiple sentence units are input into a pre-trained target event recognition model to obtain the probabilities of multiple events corresponding to the multiple sentence units output by the target event recognition model, wherein the event whose probability is greater than a preset threshold is the target event.

[0091] Optionally, the device further comprises:

[0092] A second acquisition module, used to acquire a predetermined number of training samples;

[0093] The training module is used to train the initial event recognition model based on the training samples based on the part-of-speech recognition algorithm to obtain the target event recognition model.

[0094] Optionally, the training module includes:

[0095] A division submodule, used for dividing each of the training samples into a plurality of event categories, and marking different event labels for the plurality of event categories respectively;

[0096] A screening submodule, used for screening out texts outside the event category from the training samples to obtain target training samples;

[0097] The training submodule is used to input the target training sample into the initial event recognition model for training based on a part-of-speech recognition algorithm to obtain the target event recognition model.

[0098] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0099] Example 3

[0100] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0101] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0102] S1, obtaining the enterprise text data to be identified;

[0103] S2, performing event recognition on the enterprise text data to obtain a target event;

[0104] S3, identifying the target event according to a part-of-speech recognition algorithm to obtain a company name.

[0105] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0106] Example 4

[0107] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0108] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0109] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0110] S1, obtaining the enterprise text data to be identified;

[0111] S2, performing event recognition on the enterprise text data to obtain a target event;

[0112] S3, identifying the target event according to a part-of-speech recognition algorithm to obtain a company name.

[0113] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0114] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for identifying a company name, characterized in that: include: Acquire enterprise text data to be identified; Performing event recognition on the enterprise text data to obtain a target event, including: inputting a plurality of sentence units into a pre-trained target event recognition model, obtaining the probabilities of a plurality of events corresponding to the plurality of sentence units output by the target event recognition model, wherein the event with a probability greater than a preset threshold is the target event; Identify the target event according to a part-of-speech recognition algorithm to obtain a company name; Before performing event recognition on the enterprise text data to obtain the target event, the method further includes: Obtaining a predetermined number of training samples; Based on the part-of-speech recognition algorithm, the initial event recognition model is trained according to the training samples to obtain the target event recognition model, including: dividing each of the training samples into multiple event categories, and marking the multiple event categories with different event labels respectively; filtering out texts outside the event categories from the training samples to obtain target training samples; based on the part-of-speech recognition algorithm, inputting the target training samples into the initial event recognition model for training to obtain the target event recognition model.

2. The method according to claim 1, characterized in that Before performing event recognition on the enterprise text data to obtain the target event, the method further includes: The enterprise text data is preprocessed to obtain a plurality of sentence units.

3. The method according to claim 2, characterized in that Preprocessing the enterprise text data includes: Performing word segmentation processing on the enterprise text data; Stop words in the enterprise text data are removed.

4. A device for identifying a company name, characterized in that: include: A first acquisition module is used to acquire enterprise text data to be identified; An event recognition module, used to perform event recognition on the enterprise text data to obtain a target event; A name recognition module, used to recognize the target event according to a part-of-speech recognition algorithm to obtain the company name; Wherein, the event recognition module is also used to Inputting a plurality of sentence units into a pre-trained target event recognition model, obtaining the probabilities of a plurality of events corresponding to the plurality of sentence units output by the target event recognition model, wherein the event whose probability is greater than a preset threshold is the target event; The device also includes: A second acquisition module, used to acquire a predetermined number of training samples; A training module, used for training the initial event recognition model according to the training samples based on a part-of-speech recognition algorithm to obtain the target event recognition model; Wherein, the training module includes: A division submodule, used for dividing each of the training samples into a plurality of event categories, and marking different event labels for the plurality of event categories respectively; A screening submodule, used for screening out texts outside the event category from the training samples to obtain target training samples; The training submodule is used to input the target training sample into the initial event recognition model for training based on a part-of-speech recognition algorithm to obtain the target event recognition model.

5. The device according to claim 4, characterized in that The device also includes: The preprocessing module is used to preprocess the enterprise text data to obtain multiple sentence units.

6. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 3 when executed.

7. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 3.

Images