Entity noun linking method, device, computer equipment and storage medium

Abstract

The present application relates to the field of big data technology, and provides an entity noun linking method, device, computer equipment and storage medium. The method includes: obtaining a text to be processed, and determining a domain dictionary of the vertical field to which the text to be processed belongs, performing word segmentation processing on the text to be processed based on a word segmentation tool associated with the domain dictionary, obtaining a word segmentation result, marking the nouns in the word segmentation result that exist in the domain dictionary as candidate entity nouns, and establishing a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary. For vertical fields, lightweight entity links can be quickly constructed through domain dictionaries and word segmentation tools associated with the domain dictionaries, effectively improving the efficiency of entity noun links. The present application can be applied to smart government affairs, thereby promoting the construction of smart cities.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to an entity noun linking method, apparatus, computer equipment and storage medium. Background Art

[0002] With the development of natural language processing technology, entity linking has been applied in many fields of natural language processing, such as question answering, semantic search, information extraction, etc.

[0003] Existing entity linking requires a large knowledge base as support, and also requires a large amount of entity recognition corpus to train an entity recognition model and a disambiguation model. The requirements for data infrastructure and computing power infrastructure are relatively high. Building a huge knowledge base and a large amount of entity recognition corpus and disambiguation corpus requires a lot of computing resources. Moreover, a large base of corpus will lead to a decrease in the efficiency of the entity noun link establishment process in the text. Summary of the invention

[0004] Based on this, it is necessary to provide an entity noun linking method, device, computer equipment and storage medium that can improve the efficiency of entity noun linking in response to the above technical problems.

[0005] A method for linking entity nouns, comprising:

[0006] Obtain the text to be processed, and determine the domain dictionary of the vertical field according to the vertical field to which the text to be processed belongs;

[0007] Based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain the word segmentation result;

[0008] Mark the nouns in the domain dictionary in the word segmentation results as candidate entity nouns;

[0009] Establish a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary.

[0010] In one embodiment, the entity noun categories in the domain dictionary include entity nouns to be displayed and entity nouns that do not need to be displayed;

[0011] Marking nouns in the domain dictionary in the word segmentation results as candidate entity nouns includes:

[0012] According to the word segmentation results, the target entity nouns existing in the domain dictionary are screened out;

[0013] Determine the entity noun category of the target entity noun according to the entity noun category identifier carried by the entity noun in the domain dictionary;

[0014] The target entity noun whose entity noun category is the entity noun to be displayed is marked as a candidate entity noun.

[0015] In one embodiment, based on a word segmentation tool associated with a domain dictionary, the text to be processed is segmented, and the segmentation results obtained include:

[0016] Based on the word segmentation tool associated with the domain dictionary, the domain dictionary is traversed to match the text to be processed with the entity nouns in the domain dictionary to obtain a matching result;

[0017] According to the matching results, the text to be processed is segmented to obtain the segmentation results.

[0018] In one embodiment, establishing a link relationship between a candidate entity noun and a corresponding entity noun in a domain dictionary includes:

[0019] When a candidate entity noun meets a preset link display condition, a link relationship is established between the candidate entity noun and the corresponding entity noun in the domain dictionary, wherein the preset link display condition includes at least one of the following conditions: the text to be processed does not contain an interpretation text of the candidate entity noun, the candidate entity noun appears for the first time in the text to be processed, the text to be processed does not contain a link to the candidate entity noun, and the ratio of the number of links in the text to the number of segmentation results does not reach a preset ratio.

[0020] In one embodiment, establishing a link relationship between a candidate entity noun and a corresponding entity noun in a domain dictionary includes:

[0021] Obtain the ratio data of the number of links and the number of word segmentation results in the text to be processed;

[0022] When the proportion data is less than the preset proportion, obtaining link information of the candidate entity noun in the text to be processed;

[0023] When the link information indicates that the candidate entity noun has no corresponding link, the number of occurrences of the candidate entity noun in the text to be processed is obtained;

[0024] When the occurrence is the first time, find the definition text of the candidate entity noun from the text to be processed;

[0025] When the paraphrase text search fails, a link is established between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0026] In one embodiment, establishing a link relationship between a candidate entity noun and a corresponding entity noun in a domain dictionary includes:

[0027] Obtaining data on the ratio of the number of links to the number of word segmentation results in the text to be processed, link information of the candidate entity nouns in the text to be processed, the number of occurrences of the candidate entity nouns in the text to be processed, and the interpretation text search results of the candidate entity nouns in the text to be processed;

[0028] When the proportion data is less than the preset proportion, the link information is that the candidate entity noun has no corresponding link, the number of occurrences is the first time, and the interpretation text search result is a search failure, a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary is established.

[0029] In one embodiment, after obtaining the ratio data of the number of links to the number of word segmentation results in the text to be processed, the link information of the candidate entity noun in the text to be processed, the number of occurrences of the candidate entity noun in the text to be processed, and the interpretation text search results of the candidate entity noun in the text to be processed, the method further includes:

[0030] When the proportion data is not less than the preset proportion, or the link information indicates that the candidate entity noun has a corresponding link, or the number of occurrences is not the first time, or the interpretation text search result is a successful search, the candidate entity noun is marked as an entity noun that does not need to be linked.

[0031] An entity noun linking device, the device comprising:

[0032] A text acquisition module is used to acquire the text to be processed and determine the domain dictionary of the vertical domain according to the vertical domain to which the text to be processed belongs;

[0033] The word segmentation module is used to perform word segmentation processing on the text to be processed based on the word segmentation tool associated with the domain dictionary to obtain the word segmentation result;

[0034] A tagging module is used to tag nouns in the word segmentation results that exist in the domain dictionary as candidate entity nouns;

[0035] The link relationship establishment module is used to establish a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0036] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0037] Obtain the text to be processed, and determine the domain dictionary of the vertical field according to the vertical field to which the text to be processed belongs;

[0038] Based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain the word segmentation result;

[0039] Mark the nouns in the domain dictionary in the word segmentation results as candidate entity nouns;

[0040] Establish a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary.

[0041] A computer storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0042] Obtain the text to be processed, and determine the domain dictionary of the vertical field according to the vertical field to which the text to be processed belongs;

[0043] Based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain the word segmentation result;

[0044] Mark the nouns in the domain dictionary in the word segmentation results as candidate entity nouns;

[0045] Establish a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary.

[0046] The above-mentioned entity noun linking method, device, computer equipment and storage medium obtain the text to be processed, and determine the domain dictionary of the vertical field to which the text to be processed belongs, and perform word segmentation on the text to be processed based on the word segmentation tool associated with the domain dictionary to obtain the word segmentation result, mark the nouns existing in the domain dictionary in the word segmentation result as candidate entity nouns, establish a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary, and for the vertical field, quickly build a lightweight entity link through the domain dictionary and the word segmentation tool associated with the domain dictionary, effectively improving the efficiency of entity noun linking. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is an application environment diagram of an entity noun linking method in an embodiment;

[0048] Figure 2 Schematic diagram of a flow chart of an entity noun linking method in one embodiment;

[0049] Figure 3 is a flowchart of an entity noun linking method in another embodiment;

[0050] Figure 4 is a flowchart of an entity noun linking method in yet another embodiment;

[0051] Figure 5 A schematic diagram of a flow chart of an entity noun linking method in yet another embodiment;

[0052] Figure 6 is a structural block diagram of an entity noun linking device in one embodiment;

[0053] Figure 7FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] The entity noun linking method provided in this application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The server 104 receives the text to be processed uploaded by the terminal 102, and determines the domain dictionary of the vertical field according to the vertical field to which the text to be processed belongs. The word segmentation tool associated with the domain dictionary performs word segmentation on the text to be processed to obtain a word segmentation result. The nouns in the domain dictionary in the word segmentation result are marked as candidate entity nouns. A link relationship is established between the candidate entity noun and the corresponding entity noun in the domain dictionary, and finally the text with the link relationship is fed back to the terminal. Among them, the terminal 102 can be but is not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices, and the server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0056] In one embodiment, Figure 2 As shown, a method for linking entity nouns is provided, which is applied to Figure 1 The server in is taken as an example to illustrate, including the following steps 202 to 208.

[0057] Step 202: Acquire the text to be processed, and determine the domain dictionary of the vertical domain according to the vertical domain to which the text to be processed belongs.

[0058] The text to be processed refers to the text that needs to link some words in the text to improve readability. The text to be processed contains several entities, each of which is a noun that needs to be interpreted through linking. The text to be processed can be selected by the user and submitted to the server through the terminal, or it can be obtained by the server through detection and recognition of the stored text.

[0059] The words that need to be linked can be specific domain proper nouns, custom words, etc. A link refers to a connection relationship from an interface to a target. The target pointed to can be a web page interface, or a different location on the same interface, or a picture, an email address, a file, or even an application. In an embodiment, the link of the entity noun in the text to be processed points to the definition of the noun in the domain dictionary of the vertical field of the text to be processed. The definition can be in text format or in image format.

[0060] A vertical field refers to a field that focuses on a certain part of an industry, and the audience attributes are limited to a certain specific group. In the field dictionary of a vertical field, various field-specific nouns, custom words, and various words that need to be explained through interpretation are included in the vertical field. For each word included in the field dictionary, there is a corresponding interpretation, and the interpretation can be used as a link to display content of the word. In the embodiment, the specific vertical field can be the insurance vertical field, the medical vertical field, etc.

[0061] For example, the text to be processed can be a document related to the insurance field. By linking the entities in the document, the interpretation of the relevant insurance terms in the document can be given to improve the user experience. For another example, in the question-answering system in the insurance field, the system's answers often contain a large number of insurance terms. Many of these terms may be rarely used by users, or even the first time they are used. Therefore, the need for further explanation of insurance terms can be achieved through entity linking.

[0062] Step 204 , based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain a word segmentation result.

[0063] A word segmentation tool refers to a tool used to perform word segmentation on a text. A word segmentation tool associated with a domain dictionary refers to a tool that performs word segmentation on a text based on the nouns that have been determined in the domain dictionary. Based on the word segmentation tool associated with the domain dictionary, it is possible to avoid incorrect segmentation of specific nouns and ensure the validity of each word in the word segmentation result. Furthermore, when performing word segmentation, the word segmentation tool can first use the domain dictionary to filter out specific words that have been included in the domain dictionary from the text to be processed, and then perform word segmentation on the text to be processed based on the specific words.

[0064] In the embodiment, the word segmentation results can be divided into two categories: words that exist in the domain dictionary and words that do not exist in the domain dictionary, so as to distinguish each word from the traditional word segmentation method, so as to improve the efficiency of word segmentation processing.

[0065] Step 206: Mark the nouns in the word segmentation results that exist in the domain dictionary as candidate entity nouns.

[0066] In the process of word segmentation processing by the word segmentation tool, by matching with the nouns included in the domain dictionary, two matching results will be obtained: successful matching and failed matching. Among them, the successfully matched nouns are the nouns that exist in the domain dictionary, and the failed matching words are the words that do not exist in the domain dictionary. Combining the two matching results of successful matching and failed matching, the word segmentation result of the text to be processed is obtained. By marking the nouns that exist in the domain dictionary, that is, the successfully matched words, candidate entity nouns can be obtained, so that corresponding links can be established for the candidate entity nouns in the subsequent processing process.

[0067] Step 208: Establish a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0068] In an embodiment, corresponding links can be directly established for each marked candidate entity noun, or the candidate entity nouns can be screened according to set conditions, some entity nouns that do not need to be linked are discarded, and corresponding links are established for the filtered candidate entity nouns.

[0069] Furthermore, for each candidate entity noun that needs to establish a link relationship, first determine the target entity noun corresponding to the candidate entity noun in the domain dictionary, then obtain a link object containing the corresponding interpretation of the target entity noun and the link address of the link object, and then insert the link address into the corresponding preset area where the candidate entity noun is located in the text to be processed, thereby establishing a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0070] Specifically, the corresponding preset area where the candidate entity noun is inserted can be a newly added triggerable link behind the candidate entity noun, or the candidate entity noun can be set as a clickable triggerable link, where the clickable triggerable link can be marked prominently by changing the font color, underlining, etc. By inserting a link, the definition of the entity noun in the text can be inserted without modifying the original text, which facilitates the reading of the text content and improves the user experience.

[0071] The above-mentioned entity noun linking method obtains the text to be processed, and determines the domain dictionary of the vertical field to which the text to be processed belongs. Based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain the word segmentation result, and the nouns existing in the domain dictionary in the word segmentation result are marked as candidate entity nouns, and a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary is established. For the vertical field, a lightweight entity link is quickly constructed through the domain dictionary and the word segmentation tool associated with the domain dictionary, thereby effectively improving the efficiency of entity noun linking.

[0072] In one embodiment, if Figure 3 As shown, based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain the word segmentation result, namely step 204, including steps 302 to 304.

[0073] Step 302: Based on the word segmentation tool associated with the domain dictionary, the domain dictionary is traversed to match the text to be processed with the entity nouns in the domain dictionary to obtain a matching result.

[0074] Step 304: perform word segmentation on the text to be processed according to the matching result to obtain a word segmentation result.

[0075] The matching results of the text to be processed include words that are successfully matched and words that fail to match. Among them, the words that are successfully matched are entity nouns, and the words that fail to match include non-entity nouns or entity nouns that are not included in the domain dictionary.

[0076] The text to be processed is segmented according to the words that are successfully matched and the words that are unmatched in the matching results. The segmentation results obtained can accurately identify which words in the text to be processed are successfully matched and which words are unmatched.

[0077] For example, in a certain insurance text to be processed, the word segmentation result of one sentence is: "The basic matters prepared in advance by the insurer and stipulated in the insurance policy include two parts: statutory terms and optional terms." Among them, the insurer, insurance policy, statutory terms and optional terms are entity nouns that are matched successfully, and the rest are words that fail to match.

[0078] In one embodiment, the entity noun categories in the domain dictionary include entity nouns to be displayed and entity nouns that do not need to be displayed.

[0079] like Figure 4 As shown, the nouns in the word segmentation results that exist in the domain dictionary are marked as candidate entity nouns, that is, step 206 includes steps 402 to 406.

[0080] Step 402: Filter out target entity nouns existing in the domain dictionary based on the word segmentation results.

[0081] Step 404 , determining the entity noun category of the target entity noun according to the entity noun category identifier carried by the entity noun in the domain dictionary.

[0082] Step 406 , marking the target entity noun whose entity noun category is the entity noun to be displayed as a candidate entity noun.

[0083] The entity nouns to be displayed refer to nouns that need to be interpreted when they appear in the text to be processed, and the entity nouns that do not need to be displayed refer to nouns that do not need to be interpreted when they appear in the text to be processed.

[0084] The word segmentation results include nouns that successfully match the domain dictionary and words that fail to match the domain dictionary. By comparing the successfully matched nouns, i.e., the target entity nouns that exist in the domain dictionary, with the entity noun category identifiers carried by each entity noun included in the domain dictionary, based on the entity noun category identifier, it can be determined whether the entity noun category of the target entity noun is an entity noun to be displayed or an entity noun that does not need to be displayed. Since only the links of the entity nouns to be displayed need to be displayed in the text to be processed, the target entity nouns whose entity noun category is an entity noun to be displayed are marked as candidate entity nouns. By distinguishing between entity nouns to be displayed and entity nouns that do not need to be displayed in the domain dictionary, the entity nouns in the text to be processed can be effectively filtered, unnecessary data processing processes can be reduced, and data processing speed can be improved.

[0085] In one embodiment, establishing a link relationship between a candidate entity noun and a corresponding entity noun in a domain dictionary includes:

[0086] When the candidate entity noun meets the preset link display conditions, a link relationship is established between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0087] Among them, the preset link display conditions include at least one of the following conditions: the text to be processed does not contain the interpretation text of the candidate entity noun, the candidate entity noun appears for the first time in the text to be processed, the text to be processed does not contain the link of the candidate entity noun, and the ratio of the number of links to the number of segmentation results in the text to be processed does not reach a preset ratio.

[0088] Furthermore, a link display strategy is obtained when an entity noun appears multiple times in the answer to the same FAQ. When the link display strategy is to display only the first entity noun, the preset link display condition includes the first appearance of the candidate entity noun in the text to be processed.

[0089] By performing interpretation text detection on whether the candidate entity noun is contained in the text to be processed, it is determined whether the text to be processed contains the interpretation text of the candidate entity noun, specifically including: identifying whether the candidate entity noun contains interpretation keywords in the sentence, and judging whether the text to be processed contains the interpretation text of the candidate entity. For example, "pre-health letter" refers to the customer..., where "refers to" or "refers to" are interpretation keywords, that is, interpretation keywords can include interpretation words or a combination of pre-punctuation and interpretation words.

[0090] Whether the text to be processed contains links to candidate entity nouns, specifically including: detecting whether the candidate entity nouns are inIn the Tag tag, determine whether the text to be processed contains a link to the candidate entity.

[0091] Whether the ratio of the number of links in the text to be processed to the number of word segmentation results reaches a preset ratio, specifically includes: counting the number or density of entity links in the text, and based on the number of word segmentation results, judging whether the ratio of the number of links in the text to be processed to the number of word segmentation results reaches a preset ratio.

[0092] In one of the embodiments, establishing a link relationship between a candidate entity noun and a corresponding entity noun in a domain dictionary includes: obtaining data on the ratio of the number of links to the number of segmentation results in the text to be processed. When the ratio data is less than a preset ratio, obtaining link information of the candidate entity noun in the text to be processed. When the link information indicates that the candidate entity noun has no corresponding link, obtaining the number of occurrences of the candidate entity noun in the text to be processed. When the number of occurrences is the first time, searching for an interpretation text of the candidate entity noun from the text to be processed. When the interpretation text search fails, establishing a link between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0093] like Figure 5 As shown, there are multiple preset link display conditions. The server first obtains the ratio data of the number of links and the number of word segmentation results in the text to be processed, and determines whether the ratio data is less than the preset ratio. If so, the link information of the candidate entity noun in the text to be processed is obtained. If not, the candidate entity noun is marked as an entity noun that does not need to be linked.

[0094] After obtaining the link information of the candidate entity noun in the text to be processed, determine whether the link information has a corresponding link for the candidate entity noun. If not, obtain the number of occurrences of the candidate entity noun in the text to be processed. If so, mark the candidate entity noun as an entity noun that does not need to be linked.

[0095] After obtaining the number of occurrences of the candidate entity noun in the text to be processed, determine whether the number of occurrences is multiple. If not, find the interpretation text of the candidate entity noun from the text to be processed. If so, mark the candidate entity noun as an entity noun that does not need to be linked.

[0096] Search for the definition text of the candidate entity noun from the text to be processed, and determine whether the search result is successful. If not, establish a link between the candidate entity noun and the corresponding entity noun in the domain dictionary. If so, mark the candidate entity noun as an entity noun that does not need to be linked.

[0097] By sorting multiple link display conditions by priority and processing them in sequence, unnecessary consumption of data processing resources can be reduced. By configuring the priority of each link display condition, the link display conditions are judged in sequence. When a certain condition is not met, the candidate entity noun can be marked as an entity noun that does not need to be linked, the processing flow of the entity noun is ended, and the process jumps to the next entity noun.

[0098] In one embodiment, establishing a link relationship between a candidate entity noun and a corresponding entity noun in a domain dictionary includes:

[0099] Obtain the ratio data of the number of links and the number of word segmentation results in the text to be processed, the link information of the candidate entity nouns in the text to be processed, the number of occurrences of the candidate entity nouns in the text to be processed, and the interpretation text search results of the candidate entity nouns in the text to be processed.

[0100] When the proportion data is less than the preset proportion, the link information is that the candidate entity noun has no corresponding link, the number of occurrences is the first time, and the interpretation text search result is a search failure, a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary is established.

[0101] In the embodiment, by judging multiple conditions in parallel, it is possible to process multiple conditions simultaneously, thereby improving data processing efficiency. Only when all conditions are met, a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary is established.

[0102] In one embodiment, after obtaining the ratio data of the number of links to the number of word segmentation results in the text to be processed, the link information of the candidate entity noun in the text to be processed, the number of occurrences of the candidate entity noun in the text to be processed, and the interpretation text search results of the candidate entity noun in the text to be processed, the method further includes:

[0103] When the proportion data is not less than the preset proportion, or the link information indicates that the candidate entity noun has a corresponding link, or the number of occurrences is not the first time, or the interpretation text search result is a successful search, the candidate entity noun is marked as an entity noun that does not need to be linked.

[0104] In the case where multiple link display conditions are configured, as long as there is a piece of data information that does not meet the link display conditions, the candidate entity noun will be marked as an entity noun that does not need to be linked, so that when the candidate entity noun appears multiple times in the text to be processed, the candidate entity noun can be directly skipped and the processing of the next candidate entity noun can be executed, thereby improving data processing efficiency.

[0105] It should be understood that, although each step in each flow chart involved in the above-described embodiment is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each flow chart involved in the above-described embodiment may include a plurality of steps or a plurality of stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of the steps or stages in other steps or other steps.

[0106] In one embodiment, Figure 6 As shown, an entity noun linking device is provided, including: a text acquisition module 602, a word segmentation module 604, a marking module 606 and a link relationship establishment module 608, wherein:

[0107] The text acquisition module 602 is used to acquire the text to be processed and determine the domain dictionary of the vertical domain according to the vertical domain to which the text to be processed belongs.

[0108] The word segmentation module 604 is used to perform word segmentation processing on the text to be processed based on the word segmentation tool associated with the domain dictionary to obtain a word segmentation result.

[0109] The marking module 606 is used to mark the nouns in the word segmentation results that exist in the domain dictionary as candidate entity nouns.

[0110] The link relationship establishment module 608 is used to establish a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0111] In one of the embodiments, entity noun categories in the domain dictionary include entity nouns to be displayed and entity nouns that do not need to be displayed; the marking module is also used to filter out target entity nouns existing in the domain dictionary based on the word segmentation results; determine the entity noun category of the target entity noun based on the entity noun category identifier carried by the entity noun in the domain dictionary; and mark the target entity noun whose entity noun category is entity noun to be displayed as a candidate entity noun.

[0112] In one of the embodiments, the word segmentation module is also used to traverse the domain dictionary based on a word segmentation tool associated with the domain dictionary, match the text to be processed with the entity nouns in the domain dictionary, and obtain a matching result; based on the matching result, the text to be processed is segmented to obtain a word segmentation result.

[0113] In one embodiment, the link relationship establishment module is also used to establish a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary when the candidate entity noun meets the preset link display conditions, wherein the preset link display conditions include at least one of the following conditions: the text to be processed does not contain an interpretation text of the candidate entity noun, the candidate entity noun appears for the first time in the text to be processed, the text to be processed does not contain a link to the candidate entity noun, and the ratio of the number of links to the number of segmentation results in the text to be processed does not reach a preset ratio.

[0114] In one of the embodiments, the link relationship establishment module is also used to obtain the ratio data of the number of links and the number of word segmentation results in the text to be processed; when the ratio data is less than the preset ratio, the link information of the candidate entity noun in the text to be processed is obtained; when the link information is that the candidate entity noun has no corresponding link, the number of occurrences of the candidate entity noun in the text to be processed is obtained; when the number of occurrences is the first time, the interpretation text of the candidate entity noun is searched from the text to be processed; when the interpretation text search fails, a link is established between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0115] In one of the embodiments, the link relationship establishment module is also used to obtain the ratio data of the number of links and the number of segmentation results in the text to be processed, the link information of the candidate entity noun in the text to be processed, the number of occurrences of the candidate entity noun in the text to be processed, and the interpretation text search result of the candidate entity noun in the text to be processed; when the ratio data is less than the preset ratio, the link information is that the candidate entity noun has no corresponding link, the number of occurrences is the first time and the interpretation text search result is a search failure, a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary is established.

[0116] In one of the embodiments, the marking module is also used to mark the candidate entity noun as an entity noun that does not need to be linked when the proportion data is not less than a preset proportion, or the link information indicates that there is a corresponding link for the candidate entity noun, or the number of occurrences is not the first time, or the interpretation text search result is a successful search.

[0117] The above-mentioned entity noun linking device obtains the text to be processed, and determines the domain dictionary of the vertical field according to the vertical field to which the text to be processed belongs. Based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain the word segmentation result, and the nouns existing in the domain dictionary in the word segmentation result are marked as candidate entity nouns, and a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary is established. For the vertical field, a lightweight entity link is quickly constructed through the domain dictionary and the word segmentation tool associated with the domain dictionary, thereby effectively improving the efficiency of entity noun linking.

[0118] For the specific definition of the entity noun linking device, please refer to the definition of the entity noun linking method above, which will not be repeated here. The various modules in the above-mentioned entity noun linking device can be implemented in whole or in part by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0119] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data in a domain dictionary. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for linking entity nouns is implemented.

[0120] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0121] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0122] The text to be processed is obtained, and according to the vertical field to which the text to be processed belongs, a domain dictionary of the vertical field is determined; based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain a word segmentation result; the nouns existing in the domain dictionary in the word segmentation result are marked as candidate entity nouns; and a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary is established.

[0123] In one embodiment, the entity noun category in the domain dictionary includes entity nouns to be displayed and entity nouns that do not need to be displayed; when the processor executes the computer program, the following steps are also implemented:

[0124] According to the word segmentation results, the target entity nouns existing in the domain dictionary are screened out; according to the entity noun category identifier carried by the entity noun in the domain dictionary, the entity noun category of the target entity noun is determined; and the target entity noun whose entity noun category is the entity noun to be displayed is marked as a candidate entity noun.

[0125] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0126] Based on the word segmentation tool associated with the domain dictionary, the domain dictionary is traversed to match the text to be processed with the entity nouns in the domain dictionary to obtain a matching result; based on the matching result, the text to be processed is segmented to obtain a segmentation result.

[0127] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0128] When a candidate entity noun meets a preset link display condition, a link relationship is established between the candidate entity noun and the corresponding entity noun in the domain dictionary, wherein the preset link display condition includes at least one of the following conditions: the text to be processed does not contain an interpretation text of the candidate entity noun, the candidate entity noun appears for the first time in the text to be processed, the text to be processed does not contain a link to the candidate entity noun, and the ratio of the number of links in the text to the number of segmentation results does not reach a preset ratio.

[0129] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0130] Obtain the ratio data of the number of links and the number of word segmentation results in the text to be processed; when the ratio data is less than the preset ratio, obtain the link information of the candidate entity noun in the text to be processed; when the link information indicates that the candidate entity noun has no corresponding link, obtain the number of occurrences of the candidate entity noun in the text to be processed; when the number of occurrences is the first time, search for the interpretation text of the candidate entity noun from the text to be processed; when the interpretation text search fails, establish a link between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0131] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0132] Obtain data on the ratio of the number of links to the number of segmentation results in the text to be processed, link information of the candidate entity nouns in the text to be processed, the number of occurrences of the candidate entity nouns in the text to be processed, and the interpretation text search results of the candidate entity nouns in the text to be processed; when the ratio data is less than a preset ratio, the link information is that the candidate entity noun has no corresponding link, the number of occurrences is the first time, and the interpretation text search result is a search failure, establish a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0133] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0134] When the proportion data is not less than the preset proportion, or the link information indicates that the candidate entity noun has a corresponding link, or the number of occurrences is not the first time, or the interpretation text search result is a successful search, the candidate entity noun is marked as an entity noun that does not need to be linked.

[0135] The computer device for implementing the entity noun linking method obtains the text to be processed, and determines the domain dictionary of the vertical field to which the text to be processed belongs, performs word segmentation on the text to be processed based on the word segmentation tool associated with the domain dictionary to obtain the word segmentation result, marks the nouns in the word segmentation result that exist in the domain dictionary as candidate entity nouns, establishes a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary, and for the vertical field, quickly constructs lightweight entity links through the domain dictionary and the word segmentation tool associated with the domain dictionary, effectively improving the efficiency of entity noun linking.

[0136] In one embodiment, a computer storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0137] The text to be processed is obtained, and according to the vertical field to which the text to be processed belongs, a domain dictionary of the vertical field is determined; based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain a word segmentation result; the nouns existing in the domain dictionary in the word segmentation result are marked as candidate entity nouns; and a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary is established.

[0138] In one embodiment, the entity noun category in the domain dictionary includes entity nouns to be displayed and entity nouns that do not need to be displayed; when the computer program is executed by the processor, the following steps are also implemented:

[0139] According to the word segmentation results, the target entity nouns existing in the domain dictionary are screened out; according to the entity noun category identifier carried by the entity noun in the domain dictionary, the entity noun category of the target entity noun is determined; and the target entity noun whose entity noun category is the entity noun to be displayed is marked as a candidate entity noun.

[0140] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0141] Based on the word segmentation tool associated with the domain dictionary, the domain dictionary is traversed to match the text to be processed with the entity nouns in the domain dictionary to obtain a matching result; based on the matching result, the text to be processed is segmented to obtain a segmentation result.

[0142] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0143] When a candidate entity noun meets a preset link display condition, a link relationship is established between the candidate entity noun and the corresponding entity noun in the domain dictionary, wherein the preset link display condition includes at least one of the following conditions: the text to be processed does not contain an interpretation text of the candidate entity noun, the candidate entity noun appears for the first time in the text to be processed, the text to be processed does not contain a link to the candidate entity noun, and the ratio of the number of links in the text to the number of segmentation results does not reach a preset ratio.

[0144] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0145] Obtain the ratio data of the number of links and the number of word segmentation results in the text to be processed; when the ratio data is less than the preset ratio, obtain the link information of the candidate entity noun in the text to be processed; when the link information indicates that the candidate entity noun has no corresponding link, obtain the number of occurrences of the candidate entity noun in the text to be processed; when the number of occurrences is the first time, search for the interpretation text of the candidate entity noun from the text to be processed; when the interpretation text search fails, establish a link between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0146] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0147] Obtain data on the ratio of the number of links to the number of segmentation results in the text to be processed, link information of the candidate entity nouns in the text to be processed, the number of occurrences of the candidate entity nouns in the text to be processed, and the interpretation text search results of the candidate entity nouns in the text to be processed; when the ratio data is less than a preset ratio, the link information is that the candidate entity noun has no corresponding link, the number of occurrences is the first time, and the interpretation text search result is a search failure, establish a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary.

[0148] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0149] When the proportion data is not less than the preset proportion, or the link information indicates that the candidate entity noun has a corresponding link, or the number of occurrences is not the first time, or the interpretation text search result is a successful search, the candidate entity noun is marked as an entity noun that does not need to be linked.

[0150] The computer storage medium for implementing the entity noun linking method obtains the text to be processed, and determines the domain dictionary of the vertical domain according to the vertical domain to which the text to be processed belongs, performs word segmentation on the text to be processed based on the word segmentation tool associated with the domain dictionary, obtains the word segmentation result, marks the nouns in the word segmentation result that exist in the domain dictionary as candidate entity nouns, establishes a link relationship between the candidate entity nouns and the corresponding entity nouns in the domain dictionary, and for the vertical domain, quickly constructs lightweight entity links through the domain dictionary and the word segmentation tool associated with the domain dictionary, effectively improving the efficiency of entity noun linking.

[0151] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0152] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for linking entity nouns, characterized in that: The method comprises: Acquire a text to be processed, and determine a domain dictionary of the vertical field according to the vertical field to which the text to be processed belongs; Based on the word segmentation tool associated with the domain dictionary, the text to be processed is segmented to obtain a word segmentation result; the word segmentation tool refers to a tool that segments the text according to the nouns that have been determined in the domain dictionary; Marking the nouns in the word segmentation result that exist in the domain dictionary as candidate entity nouns; Establishing a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary; The establishing of the link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary includes: when the candidate entity noun meets a preset link display condition, establishing the link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary, wherein the preset link display condition includes at least one of the following conditions: the text to be processed does not contain a definition text of the candidate entity noun, the candidate entity noun appears for the first time in the text to be processed, the text to be processed does not contain a link to the candidate entity noun, and the ratio of the number of links to the number of segmentation results in the text to be processed does not reach a preset ratio; The establishing of the link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary includes: obtaining the ratio data of the number of links and the number of word segmentation results in the text to be processed; when the ratio data is less than a preset ratio, obtaining the link information of the candidate entity noun in the text to be processed; when the link information indicates that the candidate entity noun has no corresponding link, obtaining the number of occurrences of the candidate entity noun in the text to be processed; when the number of occurrences is the first time, searching the text to be processed for the definition text of the candidate entity noun; when the definition text search fails, establishing a link between the candidate entity noun and the corresponding entity noun in the domain dictionary.

2. The method according to claim 1, characterized in that The entity noun categories in the domain dictionary include entity nouns to be displayed and entity nouns that do not need to be displayed; The step of marking the nouns in the word segmentation result that are in the domain dictionary as candidate entity nouns comprises: According to the word segmentation results, the target entity nouns existing in the domain dictionary are screened out; Determining the entity noun category of the target entity noun according to the entity noun category identifier carried by the entity noun in the domain dictionary; The target entity noun whose entity noun category is the entity noun to be displayed is marked as a candidate entity noun.

3. The method according to claim 1, characterized in that: The word segmentation tool associated with the domain dictionary performs word segmentation processing on the text to be processed, and the word segmentation results obtained include: Based on a word segmentation tool associated with the domain dictionary, traverse the domain dictionary, match the text to be processed with entity nouns in the domain dictionary, and obtain a matching result; According to the matching result, the text to be processed is segmented to obtain a segmentation result.

4. The method according to claim 1, characterized in that: The step of establishing a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary includes: Obtaining data on the ratio of the number of links to the number of segmentation results in the text to be processed, link information of the candidate entity noun in the text to be processed, the number of occurrences of the candidate entity noun in the text to be processed, and a text search result of the interpretation of the candidate entity noun in the text to be processed; When the ratio data is less than a preset ratio, the link information indicates that the candidate entity noun has no corresponding link, the number of occurrences is the first time, and the interpretation text search result is a search failure, a link relationship is established between the candidate entity noun and the corresponding entity noun in the domain dictionary.

5. The method according to claim 4, characterized in that After obtaining the ratio data of the number of links and the number of word segmentation results in the text to be processed, the link information of the candidate entity noun in the text to be processed, the number of occurrences of the candidate entity noun in the text to be processed, and the interpretation text search result of the candidate entity noun in the text to be processed, the method further includes: When the ratio data is not less than a preset ratio, or the link information indicates that there is a corresponding link for the candidate entity noun, or the number of occurrences is not the first time, or the interpretation text search result is a successful search, the candidate entity noun is marked as an entity noun that does not require a link.

6. An entity noun linking device, characterized in that: The device comprises: A text acquisition module, used to acquire the text to be processed, and determine the domain dictionary of the vertical field according to the vertical field to which the text to be processed belongs; A word segmentation module, used to perform word segmentation processing on the text to be processed based on a word segmentation tool associated with the domain dictionary to obtain a word segmentation result; the word segmentation tool refers to a tool that performs word segmentation on the text based on the nouns that have been determined in the domain dictionary; A marking module, used for marking nouns in the word segmentation result that exist in the domain dictionary as candidate entity nouns; A link relationship establishment module, used to establish a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary; The link relationship establishment module is also used for: when the candidate entity noun meets the preset link display condition, establishing a link relationship between the candidate entity noun and the corresponding entity noun in the domain dictionary, wherein the preset link display condition includes at least one of the following conditions: the text to be processed does not contain the interpretation text of the candidate entity noun, the candidate entity noun appears for the first time in the text to be processed, the text to be processed does not contain the link of the candidate entity noun, and the ratio of the number of links to the number of segmentation results in the text to be processed does not reach a preset ratio; obtaining the ratio data of the number of links to the number of segmentation results in the text to be processed; when the ratio data is less than the preset ratio, obtaining the link information of the candidate entity noun in the text to be processed; when the link information is that the candidate entity noun has no corresponding link, obtaining the number of occurrences of the candidate entity noun in the text to be processed; when the number of occurrences is the first time, searching for the interpretation text of the candidate entity noun from the text to be processed; when the interpretation text search fails, establishing a link between the candidate entity noun and the corresponding entity noun in the domain dictionary.

7. The device according to claim 6, characterized in that The entity noun categories in the domain dictionary include entity nouns to be displayed and entity nouns that do not need to be displayed; the marking module is also used to: filter out target entity nouns existing in the domain dictionary according to the word segmentation results; determine the entity noun category of the target entity noun according to the entity noun category identifier carried by the entity nouns in the domain dictionary; mark the target entity noun whose entity noun category is entity noun to be displayed as a candidate entity noun.

8. The device according to claim 6, characterized in that The word segmentation module is also used to: traverse the domain dictionary based on the word segmentation tool associated with the domain dictionary, match the text to be processed with the entity nouns in the domain dictionary to obtain a matching result; and perform word segmentation processing on the text to be processed according to the matching result to obtain a word segmentation result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images