A resume-based entity determination method and apparatus

Abstract

The application discloses a resume-based entity determination method, comprising: obtaining a first entity set of a to-be-processed resume, wherein the first entity set comprises a plurality of entities in the to-be-processed resume, and the plurality of entities comprise a first entity; determining a first entity category corresponding to the first entity; determining whether the first entity satisfies a target entity filtering rule corresponding to the first entity category; and deleting the first entity from the first entity set in a case where the first entity satisfies the target entity filtering rule. According to the method, for any first entity in the first entity set, whether the first entity is an incorrectly-identified entity can be determined by using a target entity filtering rule corresponding to the first entity, and the first entity is deleted from the first entity set when the first entity is an incorrectly-identified entity. Therefore, the method can accurately obtain entities included in the to-be-processed resume.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method and apparatus for entity identification based on a resume. Background Technology

[0002] To efficiently manage and use resume data, resumes can be automatically parsed to obtain structured resume data. As an example, entities can be extracted from the resume, and structured resume data can be obtained based on the extracted entities.

[0003] Currently, the entities extracted from resumes are inaccurate, resulting in inaccurate structured resume data. Therefore, a solution is urgently needed to address these issues. Summary of the Invention

[0004] To solve, or at least partially solve, the above-mentioned technical problems, this application provides a resume-based entity recognition and determination method and apparatus.

[0005] In a first aspect, embodiments of this application provide a method for entity identification based on a resume, the method comprising:

[0006] Obtain a first entity set of resumes to be processed, the first entity set including multiple entities in the resumes to be processed, the multiple entities including the first entity;

[0007] Determine the first entity category corresponding to the first entity;

[0008] Determine whether the first entity satisfies the target entity filtering rule corresponding to the first entity category;

[0009] If the first entity satisfies the target entity filtering rule, the first entity will be removed from the first entity set.

[0010] Optionally, the target entity filtering rules are determined in the following manner:

[0011] Obtain the method for determining the first entity;

[0012] Based on the method of determining the first entity, determine the entity filtering mode of the first entity;

[0013] The target entity filtering rule is determined based on the entity filtering mode of the first entity.

[0014] Optionally, the method for determining the first entity includes any one of the following:

[0015] The resume to be processed is processed using an entity recognition model to obtain the first entity, wherein the entity recognition model is used to determine the entities in the resume; or...

[0016] The resume to be processed is matched with an entity dictionary to obtain the first entity, wherein the entity dictionary includes multiple resume entities; or...

[0017] The resume to be processed is processed according to the preset entity recognition rules to obtain the first entity.

[0018] Optionally, the entity filtering mode of the first entity includes:

[0019] Strict filtering mode, or non-strict filtering mode.

[0020] Optionally, determining the target entity filtering rule based on the entity filtering mode of the first entity includes:

[0021] The target entity filtering rule is determined based on the first entity category, the entity filtering mode of the first entity, and the pre-determined correspondence between the entity category, the entity filtering mode, and the entity filtering rule. The correspondence includes the correspondence between the first entity category, the entity filtering mode of the first entity, and the target entity filtering rule.

[0022] Optionally, the target entity filtering rules include any one or more of the following:

[0023] The number of characters included in the entity, the character structure of the entity, and the context information of the entity.

[0024] Optionally, the method further includes:

[0025] Obtain a second entity set, which includes entities in the first entity set that have not been deleted, but does not include the first entity;

[0026] Based on the second entity set, the resume summary of the resume to be processed is obtained.

[0027] Optionally, the first entity category is education level, and determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes:

[0028] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0029] The keyword immediately following the first entity is the first preset keyword; or,

[0030] The first entity is a second preset keyword, and the first entity is immediately followed by a third preset keyword; or,

[0031] The first entity is the fourth preset keyword; or,

[0032] The first entity has a length less than the first length and is immediately followed by the fifth preset keyword.

[0033] Optionally, the first entity category is a company, and determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes:

[0034] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0035] The first entity has a length less than the second length, or the first entity consists entirely of numbers; or the first entity is a sixth preset keyword; or the first entity is immediately followed by a seventh preset keyword; or the first entity ends with the first content; or the first entity begins with the second content; or the first entity appears in the URL.

[0036] Optionally, the first entity category is a job title, and determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes:

[0037] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0038] The length of the first entity is less than the third length, or the first entity is composed entirely of numbers; or the first entity is immediately preceded by the eighth preset keyword; or the sentence structure preceding the first entity is the first preset sentence structure; or the first entity is immediately followed by the ninth preset keyword; or the first entity is the tenth preset keyword; or the first entity ends with the third content; or the first entity begins with the fourth content or appears in the URL; or the first entity includes preset characters.

[0039] Optionally, the first entity category is a profession, and determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes:

[0040] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0041] If the length of the first entity is less than the fourth length, or the first entity is entirely composed of numbers; or, the sentence structure before the first entity is the second preset sentence structure; or, the first entity begins with the fifth content; or, the first entity is the eleventh preset keyword; or, the first entity is immediately followed by the sixth content; or, the text preceding the first entity ends with the seventh content; or, the first entity is preceded by the twelfth preset keyword.

[0042] Optionally, the first entity category is a school, and determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes:

[0043] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0044] If the length of the first entity is less than the fifth length, or the first entity consists entirely of numbers; or, the line containing the first entity ends with the eighth content; or, the first entity is preceded by the thirteenth preset keyword; or, the first entity is followed by the fourteenth preset keyword; or, the first entity ends with the ninth content; or, the first entity contains rich text information and the first entity's first letter is not an uppercase letter; or, the first letter of the entity is not an uppercase letter and the length of the first entity is less than the sixth length, and the sixth length is greater than the fifth length.

[0045] Optionally, the first entity category is telephone, and determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes:

[0046] If the number corresponding to the first entity falls between two dates, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category. Secondly, embodiments of this application provide a resume-based entity determination device, the device comprising:

[0047] The first acquisition unit is used to acquire a first entity set of the resume to be processed, the first entity set including multiple entities in the resume to be processed, the multiple entities including the first entity;

[0048] The first determining unit is used to determine the first entity category corresponding to the first entity;

[0049] The second determining unit is used to determine whether the first entity satisfies the target entity filtering rule corresponding to the first entity category;

[0050] The deletion unit is used to delete the first entity from the first entity set if the first entity satisfies the target entity filtering rule.

[0051] Optionally, the target entity filtering rules are determined in the following manner:

[0052] Obtain the method for determining the first entity;

[0053] Based on the method of determining the first entity, determine the entity filtering mode of the first entity;

[0054] The target entity filtering rule is determined based on the entity filtering mode of the first entity.

[0055] Optionally, the method for determining the first entity includes any one of the following:

[0056] The resume to be processed is processed using an entity recognition model to obtain the first entity, wherein the entity recognition model is used to determine the entities in the resume; or...

[0057] The resume to be processed is matched with an entity dictionary to obtain the first entity, wherein the entity dictionary includes multiple resume entities; or...

[0058] The resume to be processed is processed according to the preset entity recognition rules to obtain the first entity.

[0059] Optionally, the entity filtering mode of the first entity includes:

[0060] Strict filtering mode, or non-strict filtering mode.

[0061] Optionally, determining the target entity filtering rule based on the entity filtering mode of the first entity includes:

[0062] The target entity filtering rule is determined based on the first entity category, the entity filtering mode of the first entity, and the pre-determined correspondence between the entity category, the entity filtering mode, and the entity filtering rule. The correspondence includes the correspondence between the first entity category, the entity filtering mode of the first entity, and the target entity filtering rule.

[0063] Optionally, the target entity filtering rules include any one or more of the following:

[0064] The number of characters included in the entity, the character structure of the entity, and the context information of the entity.

[0065] Optionally, the device further includes:

[0066] The second acquisition unit is used to acquire a second entity set, the second entity set including entities in the first entity set that have not been deleted, and the second entity set not including the first entity;

[0067] The obtaining unit is used to obtain the resume summary of the resume to be processed based on the second entity set.

[0068] Optionally, the first entity category is academic qualifications, and the second determining unit is used for:

[0069] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0070] The keyword immediately following the first entity is the first preset keyword; or,

[0071] The first entity is a second preset keyword, and the first entity is immediately followed by a third preset keyword; or,

[0072] The first entity is the fourth preset keyword; or,

[0073] The first entity has a length less than the first length and is immediately followed by the fifth preset keyword.

[0074] Optionally, the first entity category is a company, and the second determining unit is used for:

[0075] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0076] The first entity has a length less than the second length, or the first entity consists entirely of numbers; or the first entity is a sixth preset keyword; or the first entity is immediately followed by a seventh preset keyword; or the first entity ends with the first content; or the first entity begins with the second content; or the first entity appears in the URL.

[0077] Optionally, the first entity category is a job title, and the second determining unit is used for:

[0078] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0079] The length of the first entity is less than the third length, or the first entity is composed entirely of numbers; or the first entity is immediately preceded by the eighth preset keyword; or the sentence structure preceding the first entity is the first preset sentence structure; or the first entity is immediately followed by the ninth preset keyword; or the first entity is the tenth preset keyword; or the first entity ends with the third content; or the first entity begins with the fourth content or appears in the URL; or the first entity includes preset characters.

[0080] Optionally, the first entity category is a profession, and the second determining unit is used for:

[0081] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0082] If the length of the first entity is less than the fourth length, or the first entity is entirely composed of numbers; or, the sentence structure before the first entity is the second preset sentence structure; or, the first entity begins with the fifth content; or, the first entity is the eleventh preset keyword; or, the first entity is immediately followed by the sixth content; or, the text preceding the first entity ends with the seventh content; or, the first entity is preceded by the twelfth preset keyword.

[0083] Optionally, the first entity category is a school, and the second determining unit is used for:

[0084] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0085] If the length of the first entity is less than the fifth length, or the first entity consists entirely of numbers; or, the line containing the first entity ends with the eighth content; or, the first entity is preceded by the thirteenth preset keyword; or, the first entity is followed by the fourteenth preset keyword; or, the first entity ends with the ninth content; or, the first entity contains rich text information and the first entity's first letter is not an uppercase letter; or, the first letter of the entity is not an uppercase letter and the length of the first entity is less than the sixth length, and the sixth length is greater than the fifth length.

[0086] Optionally, the first entity category is telephone, and the second determining unit is used for:

[0087] If the number corresponding to the first entity falls between two dates, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category. Thirdly, embodiments of this application provide a device, the device including a processor and a memory;

[0088] The processor is configured to execute instructions stored in the memory to cause the device to perform the method as described in any one of the first aspects above.

[0089] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that instruct a device to perform the method as described in any one of the first aspects above.

[0090] Fifthly, embodiments of this application provide a computer program product that, when run on a computer, causes the computer to perform the method described in any of the first aspects above.

[0091] Compared with the prior art, the embodiments of this application have the following advantages:

[0092] This application provides a method for entity identification based on a resume. In one example, the method includes: obtaining a first entity set of resumes to be processed, the first entity set including multiple entities in the resumes to be processed, the multiple entities including a first entity. The first entity can be any entity in the first entity set. Then, determining the first entity category corresponding to the first entity. After determining the first entity category, a target entity filtering rule corresponding to the first entity category can be used to determine whether to filter out the first entity. Specifically, it can be determined whether the first entity satisfies the target entity filtering rule. When the first entity satisfies the target entity filtering rule, it indicates that the first entity is an incorrectly identified entity, and therefore the first entity can be filtered out. In other words, in one example, if the first entity satisfies the target entity filtering rule, the first entity can be deleted from the first entity set. Therefore, using this solution, for any first entity in the first entity set, the target entity filtering rule corresponding to the first entity can be used to determine whether the first entity is an incorrectly identified entity, and if the first entity is an incorrectly identified entity, it can be deleted from the first entity set. Thus, using this solution, the entities included in the resumes to be processed can be accurately obtained. Attached Figure Description

[0093] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0094] Figure 1A flowchart illustrating a resume-based entity determination method provided in this application embodiment;

[0095] Figure 2 A flowchart illustrating a method for determining target entity filtering rules provided in an embodiment of this application;

[0096] Figure 3 This is a schematic diagram of a resume-based entity determination device provided in an embodiment of this application. Detailed Implementation

[0097] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0098] The various non-limiting embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0099] Exemplary methods

[0100] See Figure 1 This figure is a flowchart illustrating a resume-based entity identification method provided in an embodiment of this application. In this embodiment, the method can be executed by the client or the server; this embodiment does not impose any specific limitations.

[0101] In one example Figure 1 The method shown may include, for example, the following steps: S101-S104.

[0102] S101: Obtain a first entity set of resumes to be processed, the first entity set including multiple entities in the resumes to be processed, the multiple entities including the first entity.

[0103] In one example, the first entity set may be the first entity set obtained after preliminary entity recognition of the resume to be processed. The first entity set may include multiple entities, which may include correctly identified entities and incorrectly identified entities. The first entity may be any entity in the first entity set.

[0104] In this embodiment of the application, the first entity set may include various types of entities. For example, it may include entities such as name, school name, company name, major, position, date, telephone number, education level, and email address.

[0105] This application does not specifically limit the specific implementation of the preliminary identification of the resume to be processed to obtain the first entity set.

[0106] In one example, a preset resume entity recognition algorithm can be used to process the resume to obtain the first entity type.

[0107] In another example, the resume to be processed can be processed in various ways to obtain the first entity set. Next, the method for determining the entities in the first entity set will be illustrated using the first entity in the first entity set as an example.

[0108] As an example, an entity recognition model can be used to process the resume to obtain the first entity, wherein the entity recognition model is used to determine the entities in the resume. That is, after the resume to be processed is input into the entity recognition model, the entity recognition model can output the entities that may exist in the resume. This application embodiment does not specifically limit the entity recognition model, and the entity recognition model may be, for example, a NER model.

[0109] As another example, the resume to be processed can be matched with an entity dictionary to obtain the first entity. The entity dictionary may include multiple resume entities. Each resume entity in the entity dictionary can be matched one by one with the resume to be processed. If the entity dictionary includes the first entity, and the resume to be processed also includes keywords that match the first entity, then it can be determined that the resume to be processed includes the first entity. The keywords matching the first entity can be the same keywords as the first entity, or keywords with the same semantic meaning as the first entity.

[0110] As another example, the resume to be processed can be processed according to preset entity recognition rules to obtain the first entity. For example, it can be determined whether the word segmentation in the resume to be processed meets the entity recognition rules, and the keywords that meet the entity recognition rules are identified as entities included in the resume to be processed.

[0111] S102: Determine the first entity category corresponding to the first entity.

[0112] After determining the first entity, the first entity category corresponding to the first entity can be further determined. This application embodiment does not specifically limit the first entity category; the first entity category can be any one of the aforementioned name, school name, company name, major, position, date, telephone number, education level, and email address.

[0113] In one example, the first entity category corresponding to the first entity can be determined based on the first entity and a pre-built correspondence between entities and entity categories. For example, the first entity can be matched with entities in the correspondence, and when the first entity matches a second entity in the correspondence, the entity category corresponding to the second entity is determined to be the first entity category. Here, the first entity matching the second entity in the correspondence can be, for example, the first entity and the second entity being the same, or the first entity and the second entity having the same semantics.

[0114] In yet another example, the first entity category can be determined based on an entity category determination model. For instance, the first entity can be input into the entity category determination model to obtain the first entity category.

[0115] S103: Determine whether the first entity satisfies the target entity filtering rule corresponding to the first entity category.

[0116] S104: If the first entity satisfies the target entity filtering rule, the first entity is deleted from the first entity set.

[0117] In one example, after determining the first entity category, the target entity filtering rules corresponding to the first entity category can be further determined.

[0118] In one example, one entity category can correspond to one entity filtering rule. In this case, after determining the first entity category, the unique entity rule corresponding to that first entity type can be determined as the aforementioned target entity filtering rule.

[0119] Next, we will briefly introduce the specific implementation of S103 for each entity category.

[0120] In one example, the first entity category is education level. In this case, in the specific implementation of S103, for example, the first entity can be determined to satisfy the target entity filtering rule corresponding to the first entity category if the first entity satisfies one or more of the following:

[0121] 1. The keyword immediately following the first entity is a first preset keyword. This application does not specifically limit the first preset keyword; the first preset keyword can be determined according to actual circumstances. For example, the first preset keyword could be: mentor, intern, or professors.

[0122] 2. The first entity is a second preset keyword, and the first entity is immediately followed by a third preset keyword. This application does not specifically limit the second and third preset keywords; the second and third preset keywords can be determined according to actual circumstances. For example, the second preset keyword could be "master," and the third preset keyword could be "mentored by," "scrum," or "and."

[0123] 3. The first entity is a fourth preset keyword. This application embodiment does not specifically limit the fourth preset keyword; the fourth preset keyword may be, for example, ms, ma, bs, ba, or aa.

[0124] 4. The length of the first entity is less than the first length, and the first entity is immediately followed by the fifth preset keyword. This application embodiment does not specifically limit the first length; the first length may be, for example, 3. This application embodiment does not specifically limit the fifth preset keyword; the fifth preset keyword may be, for example, Excel, Word, or PowerPoint.

[0125] In one example, the first entity category is "company". In this case, in the specific implementation of S103, for example, the first entity can be determined to satisfy the target entity filtering rule corresponding to the first entity category if the first entity satisfies one or more of the following conditions:

[0126] 1. The length of the first entity is less than the second length, or the first entity is entirely composed of numbers. This application does not specifically limit the second length; for example, the second length could be 2.

[0127] 2. The first entity is the sixth preset keyword. This application does not specifically limit the sixth preset keyword; for example, the sixth preset keyword may be student, member, various, wfh, task, fullstack, or workexprence.

[0128] 3. The first entity is immediately followed by a seventh preset keyword. This application does not specifically limit the seventh preset keyword; for example, it may be responsible, contribute, duration, processes, platforms, or workflow.

[0129] 4. The first entity ends with the first content. This application does not specifically limit the first content; the first content may be, for example, "project" or "programme".

[0130] 5. Alternatively, the first entity may begin with the second content. This application does not specifically limit the second content; the second content may be, for example, "publish" or "extra".

[0131] 6. Alternatively, the first entity may appear in a URL. A URL mentioned in this embodiment may be, for example, www.xx.com. The "xx" in the URL can be any character that can appear in a URL.

[0132] In one example, the first entity category is a job title. In this case, in the specific implementation of S103, for example, the first entity can be determined to satisfy the target entity filtering rule corresponding to the first entity category if the first entity satisfies one or more of the following conditions:

[0133] 1. The length of the first entity is less than the third length, or the first entity is entirely composed of numbers. This application does not specifically limit the third length; for example, the third length could be 2.

[0134] 2. The first entity is immediately followed by an eighth preset keyword. This application embodiment does not specifically limit the eighth preset keyword; the eighth preset keyword may be, for example, "report to," "directly to," "industry," or "task."

[0135] 3. The sentence structure preceding the first entity is a first preset sentence structure. This application does not specifically limit the first preset sentence structure; for example, the first preset sentence structure could be: at / by / how / when + "to" + the / a / an; or: major / minor + in / of / for.

[0136] 4. The first entity is immediately followed by a ninth preset keyword. This application does not specifically limit the ninth preset keyword; for example, it may be a degree, diploma, certificates, club, or institute.

[0137] 5. The first entity is the tenth preset keyword. This application embodiment does not specifically limit the tenth preset keyword; the tenth preset keyword may be, for example, remote, corner, present, experience, exchange, or aliexpress.

[0138] 6. Alternatively, the first entity ends with a third content. The embodiments of the present application do not specifically limit the third content, and the third content may be, for example: experiences or in or at followed by a non-numeric / non-alphabetic character, or project, experience, or tasks.

[0139] 7. The first entity starts with a fourth content or appears in a web address. The embodiments of the present application do not specifically limit the fourth content, and the fourth content may be, for example: publish or extra. For the web address, reference may be made to the relevant description part above, and no repeated description is given here.

[0140] 8. The first entity includes a preset character. The embodiments of the present application do not specifically limit the preset character, and the preset character may be, for example, the character "to".

[0141] In an example, the first entity category is a major. In this case, when S103 is implemented specifically, for example, it may be determined that the first entity satisfies the target entity filtering rule corresponding to the first entity category when the first entity meets one or more of the following conditions:

[0142] 1. The entity length of the first entity is less than a fourth length, or the first entity is entirely composed of numbers. The embodiments of the present application do not specifically limit the fourth length, and the fourth length may be 2, for example.

[0143] 2. The sentence pattern before the first entity is a second preset sentence pattern. The embodiments of the present application do not specifically limit the second preset sentence pattern, and the second preset sentence pattern may be, for example:

[0144] Sentence pattern 1: language / course / coursework / subject / interest / objective + ":"; or

[0145] Sentence pattern 2: planning on + specialized / concentrating.

[0146] 3. The first entity starts with a fifth content. The embodiments of the present application do not specifically limit the fifth content, and the fifth content may be, for example: degree in, or "English" + "4 / 6 / 8 / four / six / eight".

[0147] 4. The first entity is an eleventh preset keyword. The embodiments of the present application do not specifically limit the eleventh preset keyword, and the eleventh preset keyword may be, for example: art, science or technology.

[0148] 5. The first entity is immediately followed by the sixth content. This application embodiment does not specifically limit the sixth content; the sixth content may, for example, be: based.

[0149] 6. The preceding line of text for the first entity ends with the seventh content. This application embodiment does not specifically limit the seventh content; the seventh content may, for example, be: courses or responsibilities.

[0150] 7. The first entity is preceded by a twelfth preset keyword. This application embodiment does not specifically limit the twelfth preset keyword; the twelfth preset keyword may be, for example, "course" or "department".

[0151] In one example, the first entity category is "school". In this case, in the specific implementation of S103, for example, the first entity can be determined to satisfy the target entity filtering rule corresponding to the first entity category if the first entity satisfies one or more of the following conditions:

[0152] 1. The length of the first entity is less than the fifth length, or the first entity is entirely composed of numbers. This application does not specifically limit the fifth length; for example, the fifth length could be 2.

[0153] 2. The line containing the first entity ends with the eighth content. This application embodiment does not specifically limit the eighth content; the eighth content may, for example, be: education.

[0154] 3. The first entity is preceded by a thirteenth preset keyword. This application embodiment does not specifically limit the thirteenth preset keyword; the thirteenth preset keyword may be, for example, "by," "for," or "member of."

[0155] 4. The first entity is followed by a fourteenth preset keyword. This application embodiment does not specifically limit the fourteenth preset keyword; the fourteenth preset keyword may be, for example, "is" or "was".

[0156] 5. The first entity ends with the ninth content. This application embodiment does not specifically limit the ninth content; the ninth content may, for example, be "'s".

[0157] 6. The first entity contains rich text information and the first entity does not start with a capital letter.

[0158] 7. The first letter of the entity is not an uppercase letter, and the entity length of the first entity is less than the sixth length, wherein the sixth length is greater than the fifth length. This application does not specifically limit the sixth length; for example, the sixth length can be 6.

[0159] In one example, the first entity category is "telephone". In this case, in the specific implementation of S103, for example, if the number corresponding to the first entity is between two dates, it can be determined that the first entity meets the target entity filtering rule corresponding to the first entity category. Here, the fact that the number corresponding to the first entity is between two dates indicates that the first entity may be a date, rather than a telephone number.

[0160] In another example, one entity category can correspond to multiple entity filtering rules. As an example, one entity filtering rule can correspond to one entity filtering mode. This application does not specifically limit the entity filtering mode; in one example, the entity filtering mode may include a strict mode and a non-strict mode. The strict filtering mode has more filtering conditions than the non-strict filtering mode. For an entity, such as the first entity, it may satisfy the strict filtering mode even if it does not meet the filtering conditions of the non-strict filtering mode.

[0161] In this situation, in one example, the target entity filtering rule can be determined through [method / method] in its specific implementation. Figure 2 The implementations of S1031-S1033 are shown. Figure 2 This is a flowchart illustrating a method for determining target entity filtering rules, provided in an embodiment of this application.

[0162] S1031: Obtain the determination method for the first entity.

[0163] In one example, during the initial entity identification process, the identification method for each identified entity can be recorded. Therefore, when acquiring the first entity, the pre-recorded identification method for the first entity can also be acquired.

[0164] The method for determining the first entity can be any one of the following: "processing the resume to be processed using an entity recognition model to obtain the first entity", "matching the resume to be processed with an entity dictionary to obtain the first entity", and "processing the resume to be processed according to a preset entity recognition rule to obtain the first entity".

[0165] S1032: Determine the entity filtering mode of the first entity according to the determination method of the first entity.

[0166] In one example, considering that different methods of entity identification result in different accuracy rates, the entity filtering mode for the first entity can be determined based on its identification method. For instance, if the accuracy rate of identifying the entity using the first entity's identification method is high, then the entity filtering mode for the first entity can be determined to be a non-strict mode. Conversely, if the accuracy rate of identifying the entity using the first entity's identification method is low, then the entity filtering mode for the first entity can be determined to be a strict mode. Examples are provided below.

[0167] Entity recognition models have low accuracy in identifying entities. Therefore, if the first entity is identified by an entity recognition model, its entity filtering mode can be determined to be strict mode. Entity dictionaries or entity recognition rules have higher accuracy in identifying entities. Therefore, if the first entity is identified by an entity dictionary or entity recognition rules, its entity filtering mode can be determined to be non-strict mode.

[0168] As an example, entity filtering rules in strict mode include more stringent entity filtering conditions than entity filtering rules in non-strict mode.

[0169] S1033: Determine the target entity filtering rule based on the entity filtering mode of the first entity.

[0170] After determining the entity filtering mode of the first entity, the target entity filtering rules can be further determined based on the entity filtering mode of the first entity.

[0171] As an example, multiple entity filtering rules for a first entity category can be determined first, and then the target entity filtering rule can be determined from the multiple entity filtering rules for the first entity category based on the entity filtering pattern of the first entity.

[0172] As another example, a correspondence between entity categories, entity filtering modes, and entity rule patterns can be established in advance. After determining the first entity category and the first entity's entity filtering mode, the first entity category and the first entity's entity filtering mode can be matched with this correspondence to obtain the target entity filtering rule. The correspondence between entity categories, entity filtering modes, and entity rule patterns includes at least the correspondence between the first entity category, the first entity's entity filtering mode, and the target entity filtering rule.

[0173] After determining the target entity filtering rules, it can be further determined whether the first entity satisfies the target entity filtering rules. If the first entity satisfies the target entity filtering rules, it indicates that the first entity is likely to be an incorrectly identified entity; therefore, the first entity can be deleted from the first entity set. If the first entity does not satisfy the target entity filtering rules, it indicates that the first entity is likely to be a correctly identified entity; therefore, the first entity can be retained.

[0174] Regarding the target entity filtering rules, it should be noted that, in one example, the target entity filtering rules may include at least one entity filtering condition. This application embodiment does not specifically limit the at least one filtering condition. In one example, considering that an entity may be identified as incorrectly, its number of characters, character structure, and contextual information could all indicate that the entity is incorrectly identified. Therefore, the at least one entity filtering condition may, for example, include one or more of the following: the number of characters in the entity, the character structure of the entity, and the contextual information of the entity. The number of characters in the entity can also be referred to as the "entity length." In one example, the character structure of the entity may indicate the type of characters included in the entity, such as all numbers, numbers plus other non-numeric characters, etc. Next, S104 will be described with reference to specific examples.

[0175] In one example, when the first entity is an academic qualification entity, the first entity is deleted if it meets one or more of the following three entity filtering conditions.

[0176] 1. The first entity is followed by keywords such as mentor, intern, or professors.

[0177] 2. The first entity is master, and the entity is immediately preceded by keywords such as mentored by, scrum, or and.

[0178] 3. The first entity is a keyword such as ms, ma, bs, ba, or aa; or the entity length is less than 3 and the entity is immediately followed by a keyword such as excel, word, or powerpoint.

[0179] In one example, when the first entity is a telephone entity, if the telephone entity appears in more than two date entities, or in other words, the number corresponding to the telephone entity is between two dates, then the first entity is deleted.

[0180] In one example, when the first entity is a company entity:

[0181] If the length of the first entity is less than 2 or the first entity consists entirely of numbers, then the first entity can be deleted.

[0182] If the length of the first entity is greater than 2 and the first entity is not entirely composed of numbers, then the first entity shall be deleted when the first entity meets one or more of the following three conditions.

[0183] 1. The entity content consists of words such as student, member, various, wfh, task, fullstack, or workexprence.

[0184] 2. The entity is preceded by words such as responsible, contribute, duration, processes, platforms, or workflow.

[0185] 3. The entity content ends with "project" or "programme", begins with "publish" or "extra", and the entity appears in the URL www.xx.com.

[0186] In one example, when the first entity is a job entity:

[0187] If the length of the first entity is less than 2 or the first entity consists entirely of numbers, then the first entity can be deleted.

[0188] If the length of the first entity is greater than 2 and the first entity is not entirely composed of numbers, then in the non-strict filtering mode, the first entity is deleted when it meets one or more of the following 5 conditions.

[0189] 1. The entity is immediately preceded by keywords such as report to, directly to, industry, or task.

[0190] 2. The sentence structure before the entity is: at / by / how / when + "to" + the / a / an; or: major / minor + in / of / for. 3. The entity is immediately followed by keywords such as degree, diploma, certificates, club, or institute.

[0191] 4. The entity content is remote, corner, present, experience, intership, or aliexpress; or the entity content ends with keywords such as project, experience, or tasks.

[0192] 5. The entity content must begin with "publish" or "extra" or appear in the URL www.xx.com.

[0193] In strict filtering mode, even if the first entity does not meet the above 5 conditions, the first entity will still be deleted if it meets one or more of the following two conditions.

[0194] 1. Entity content ends with "experiences", "in", or "at" followed by a non-numeric / letter ending.

[0195] 2. The entity content includes the character "to".

[0196] In one example, when the first entity is a professional entity:

[0197] If the length of the first entity is less than 2 or the first entity consists entirely of numbers, then the first entity can be deleted.

[0198] If the length of the first entity is greater than 2 and the first entity is not entirely composed of numbers, then in the non-strict filtering mode, the first entity is deleted when it meets one or more of the following three conditions.

[0199] 1. The sentence structure before the entity is:

[0200] Sentence pattern 1: language / course / coursework / subject / interest / objective+“:”

[0201] Sentence pattern 2: planning on+specialized / concentrating

[0202] 2. Entity content begins with "degree in".

[0203] 3. The entity content is art, science, or technology.

[0204] In strict filtering mode, even if the first entity does not meet the above three conditions, the first entity will still be deleted if it meets one or more of the following five conditions.

[0205] 1. The entity content is immediately followed by "based".

[0206] 2. The text preceding the entity content ends with "courses" or "responsibilities".

[0207] 3. The content of the entity contains the keyword "course".

[0208] 4. The word before the entity content is "department".

[0209] 5. The entity content starts with "English" + "4 / 6 / 8 / four / six / eight".

[0210] In an example, when the first entity is a school entity:

[0211] If the entity length of the first entity is less than 2 or the first entity is entirely composed of numbers, the first entity may be deleted.

[0212] If the entity length of the first entity is greater than 2 and the first entity is not entirely composed of numbers, then in the non-strict filtering mode, when the first entity meets one or more of the following 4 conditions, the first entity is deleted.

[0213] 1. The line where the entity content is located starts with education.

[0214] 2. Keywords such as by, for or member of are included before the entity content.

[0215] 3. Keywords such as is or was are included after the entity content.

[0216] 4. The entity content ends with "'s".

[0217] In the strict filtering mode, even if the first entity does not meet one or more of the above 4 conditions, if the first entity meets the following 2 conditions, the first entity is also deleted.

[0218] 1. Rich text information is included and the first letter of the entity is not a capital letter.

[0219] 2. The first letter of the entity is not a capital letter and the entity length is less than 6.

[0220] It can be seen from the above description that with the solution of the present application, for any first entity in the first entity set, the target entity filtering rule corresponding to the first entity can be used to determine whether the first entity is an entity with recognition error, and when the first entity is an entity with recognition error, the first entity is deleted from the first entity set. It can be seen that the solution of the present application can accurately obtain the entities included in the resume to be processed.

[0221] In this embodiment, S102-S104 can be executed for any entity in the first entity set. After executing S102-S104 for any entity in the first entity set, a second entity set is obtained, which includes the entities in the first entity set that were not deleted. It is easy to understand that the second entity set does not include the aforementioned first entities. After obtaining the second entity set, a resume summary of the resume to be processed can be obtained based on the second entity set. For example, multiple entities in the second entity set can be processed to obtain an education experience summary, a work experience summary, and a personal basic information summary corresponding to the resume to be processed.

[0222] Since the entities in the second entity set are relatively accurate, the resume summary obtained based on the second entity set is also relatively accurate. Therefore, this solution can yield a more accurate resume summary. In one example, the resume summary can also be understood as a concise version of the resume.

[0223] Exemplary device

[0224] Based on the methods provided in the above embodiments, this application also provides an apparatus, which will be described below with reference to the accompanying drawings.

[0225] See Figure 3 The figure is a schematic diagram of the structure of a resume-based entity determination device provided in an embodiment of this application. The device 300 may specifically include, for example, a first acquisition unit 301, a first determination unit 302, a second determination unit 303, and a deletion unit 304.

[0226] The first acquisition unit 301 is used to acquire a first entity set of the resume to be processed, the first entity set including multiple entities in the resume to be processed, the multiple entities including the first entity;

[0227] The first determining unit 302 is used to determine the first entity category corresponding to the first entity;

[0228] The second determining unit 303 is used to determine whether the first entity satisfies the target entity filtering rule corresponding to the first entity category;

[0229] The deletion unit 304 is used to delete the first entity from the first entity set if the first entity satisfies the target entity filtering rule.

[0230] Optionally, the target entity filtering rules are determined in the following manner:

[0231] Obtain the method for determining the first entity;

[0232] Based on the method of determining the first entity, determine the entity filtering mode of the first entity;

[0233] The target entity filtering rule is determined based on the entity filtering mode of the first entity.

[0234] Optionally, the method for determining the first entity includes any one of the following:

[0235] The resume to be processed is processed using an entity recognition model to obtain the first entity, wherein the entity recognition model is used to determine the entities in the resume; or...

[0236] The resume to be processed is matched with an entity dictionary to obtain the first entity, wherein the entity dictionary includes multiple resume entities; or...

[0237] The resume to be processed is processed according to the preset entity recognition rules to obtain the first entity.

[0238] Optionally, the entity filtering mode of the first entity includes:

[0239] Strict filtering mode, or non-strict filtering mode.

[0240] Optionally, determining the target entity filtering rule based on the entity filtering mode of the first entity includes:

[0241] The target entity filtering rule is determined based on the first entity category, the entity filtering mode of the first entity, and the pre-determined correspondence between the entity category, the entity filtering mode, and the entity filtering rule. The correspondence includes the correspondence between the first entity category, the entity filtering mode of the first entity, and the target entity filtering rule.

[0242] Optionally, the target entity filtering rules include any one or more of the following:

[0243] The number of characters included in the entity, the character structure of the entity, and the context information of the entity.

[0244] Optionally, the device further includes:

[0245] The second acquisition unit is used to acquire a second entity set, the second entity set including entities in the first entity set that have not been deleted, and the second entity set not including the first entity;

[0246] The obtaining unit is used to obtain the resume summary of the resume to be processed based on the second entity set.

[0247] Optionally, the first entity category is educational background, and the second determining unit 303 is used for:

[0248] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0249] The keyword immediately following the first entity is the first preset keyword; or,

[0250] The first entity is a second preset keyword, and the first entity is immediately followed by a third preset keyword; or,

[0251] The first entity is the fourth preset keyword; or,

[0252] The first entity has a length less than the first length and is immediately followed by the fifth preset keyword.

[0253] Optionally, the first entity category is a company, and the second determining unit 303 is used for:

[0254] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0255] The first entity has a length less than the second length, or the first entity consists entirely of numbers; or the first entity is a sixth preset keyword; or the first entity is immediately followed by a seventh preset keyword; or the first entity ends with the first content; or the first entity begins with the second content; or the first entity appears in the URL.

[0256] Optionally, the first entity category is a job title, and the second determining unit 303 is used for:

[0257] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0258] The length of the first entity is less than the third length, or the first entity is composed entirely of numbers; or the first entity is immediately preceded by the eighth preset keyword; or the sentence structure preceding the first entity is the first preset sentence structure; or the first entity is immediately followed by the ninth preset keyword; or the first entity is the tenth preset keyword; or the first entity ends with the third content; or the first entity begins with the fourth content or appears in the URL; or the first entity includes preset characters.

[0259] Optionally, the first entity category is a profession, and the second determining unit 303 is used for:

[0260] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0261] If the length of the first entity is less than the fourth length, or the first entity is entirely composed of numbers; or, the sentence structure before the first entity is the second preset sentence structure; or, the first entity begins with the fifth content; or, the first entity is the eleventh preset keyword; or, the first entity is immediately followed by the sixth content; or, the text preceding the first entity ends with the seventh content; or, the first entity is preceded by the twelfth preset keyword.

[0262] Optionally, the first entity category is a school, and the second determining unit 303 is used for:

[0263] If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category:

[0264] If the length of the first entity is less than the fifth length, or the first entity consists entirely of numbers; or, the line containing the first entity ends with the eighth content; or, the first entity is preceded by the thirteenth preset keyword; or, the first entity is followed by the fourteenth preset keyword; or, the first entity ends with the ninth content; or, the first entity contains rich text information and the first entity's first letter is not an uppercase letter; or, the first letter of the entity is not an uppercase letter and the length of the first entity is less than the sixth length, and the sixth length is greater than the fifth length.

[0265] Optionally, the first entity category is telephone, and the second determining unit 303 is used for:

[0266] If the number corresponding to the first entity falls between two dates, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category.

[0267] Since the device 300 is a device corresponding to the method provided in the above method embodiments, the specific implementation of each unit of the device 300 is based on the same concept as the above method embodiments. Therefore, for the specific implementation of each unit of the device 300, please refer to the relevant descriptions in the above method embodiments, which will not be repeated here.

[0268] This application also provides an apparatus, which includes a processor and a memory;

[0269] The processor is used to execute instructions stored in the memory so that the device performs the method provided in the above method embodiments.

[0270] This application provides a computer-readable storage medium including instructions that instruct a device to perform the method provided in the above method embodiments.

[0271] This application provides a computer program product that, when run on a computer, causes the computer to execute the method provided in the above method embodiments.

[0272] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0273] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

[0274] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for entity identification based on a resume, characterized in that, The method includes: Obtain a first entity set of resumes to be processed, the first entity set including multiple entities in the resumes to be processed, the multiple entities including the first entity; Determine the first entity category corresponding to the first entity, where the first entity category includes: education level, company, job title, major, school, or telephone number; Determine whether the first entity satisfies the target entity filtering rule corresponding to the first entity category; If the first entity satisfies the target entity filtering rule, the first entity is deleted from the first entity set; The target entity filtering rules are determined in the following manner: Obtain the method for determining the first entity; Based on the method of determining the first entity, determine the entity filtering mode of the first entity; The target entity filtering rule is determined based on the entity filtering mode of the first entity.

2. The method according to claim 1, characterized in that, The method for determining the first entity includes any one of the following: The resume to be processed is processed using an entity recognition model to obtain the first entity; the entity recognition model is used to determine the entities in the resume; or... The resume to be processed is matched with an entity dictionary to obtain the first entity, wherein the entity dictionary includes multiple resume entities; or... The resume to be processed is processed according to the preset entity recognition rules to obtain the first entity.

3. The method according to claim 1 or 2, characterized in that, The entity filtering modes of the first entity include: Strict filtering mode, or non-strict filtering mode.

4. The method according to claim 1, characterized in that, The step of determining the target entity filtering rule based on the entity filtering pattern of the first entity includes: The target entity filtering rule is determined based on the first entity category, the entity filtering mode of the first entity, and the pre-determined correspondence between the entity category, the entity filtering mode, and the entity filtering rule. The correspondence includes the correspondence between the first entity category, the entity filtering mode of the first entity, and the target entity filtering rule.

5. The method according to claim 1, characterized in that, The target entity filtering rules include any one or more of the following: The number of characters included in the entity, the character structure of the entity, and the context information of the entity.

6. The method according to claim 1, characterized in that, The method further includes: Obtain a second entity set, which includes entities in the first entity set that have not been deleted, but does not include the first entity; Based on the second entity set, the resume summary of the resume to be processed is obtained.

7. The method according to claim 1, characterized in that, The first entity category is education level. Determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes: If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category: The keyword immediately following the first entity is the first preset keyword; or, The first entity is a second preset keyword, and the first entity is immediately followed by a third preset keyword; or, The first entity is the fourth preset keyword; or, The first entity has a length less than the first length and is immediately followed by the fifth preset keyword.

8. The method according to claim 1, characterized in that, The first entity category is "company". Determining whether the first entity meets the target entity filtering rules corresponding to the first entity category includes: If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category: The first entity has a length less than the second length, or the first entity consists entirely of numbers; or the first entity is a sixth preset keyword; or the first entity is immediately followed by a seventh preset keyword; or the first entity ends with the first content; or the first entity begins with the second content; or the first entity appears in the URL.

9. The method according to claim 1, characterized in that, The first entity category is a job title. Determining whether the first entity meets the target entity filtering rules corresponding to the first entity category includes: If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category: The length of the first entity is less than the third length, or the first entity is composed entirely of numbers; or the first entity is immediately preceded by the eighth preset keyword; or the sentence structure preceding the first entity is the first preset sentence structure; or the first entity is immediately followed by the ninth preset keyword; or the first entity is the tenth preset keyword; or the first entity ends with the third content; or the first entity begins with the fourth content or appears in the URL; or the first entity includes preset characters.

10. The method according to claim 1, characterized in that, The first entity category is a profession. Determining whether the first entity meets the target entity filtering rules corresponding to the first entity category includes: If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category: If the length of the first entity is less than the fourth length, or the first entity is entirely composed of numbers; or, the sentence structure before the first entity is the second preset sentence structure; or, the first entity begins with the fifth content; or, the first entity is the eleventh preset keyword; or, the first entity is immediately followed by the sixth content; or, the text preceding the first entity ends with the seventh content; or, the first entity is preceded by the twelfth preset keyword.

11. The method according to claim 1, characterized in that, The first entity category is "school". Determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes: If the first entity satisfies one or more of the following conditions, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category: If the length of the first entity is less than the fifth length, or the first entity consists entirely of numbers; or, the line containing the first entity ends with the eighth content; or, the first entity is preceded by the thirteenth preset keyword; or, the first entity is followed by the fourteenth preset keyword; or, the first entity ends with the ninth content; or, the first entity contains rich text information and the first entity's first letter is not an uppercase letter; or, the first letter of the entity is not an uppercase letter and the length of the first entity is less than the sixth length, and the sixth length is greater than the fifth length.

12. The method according to claim 1, characterized in that, The first entity category is telephone. Determining whether the first entity satisfies the target entity filtering rule corresponding to the first entity category includes: If the number corresponding to the first entity falls between two dates, then the first entity is determined to satisfy the target entity filtering rule corresponding to the first entity category.

13. A resume-based entity identification device, characterized in that, The device includes: The first acquisition unit is used to acquire a first entity set of the resume to be processed, the first entity set including multiple entities in the resume to be processed, the multiple entities including the first entity; The first determining unit is used to determine the first entity category corresponding to the first entity, wherein the first entity category includes: education level, company, job position, major, school, or telephone number; The second determining unit is used to determine whether the first entity satisfies the target entity filtering rule corresponding to the first entity category; The deletion unit is used to delete the first entity from the first entity set if the first entity satisfies the target entity filtering rule; The target entity filtering rules are determined in the following manner: Obtain the method for determining the first entity; Based on the method of determining the first entity, determine the entity filtering mode of the first entity; The target entity filtering rule is determined based on the entity filtering mode of the first entity.

14. A device, characterized in that, The device includes a processor and a memory; The processor is configured to execute instructions stored in the memory to cause the device to perform the method as described in any one of claims 1 to 12.

15. A computer-readable storage medium, characterized in that, Includes instructions that instruct the device to perform the method as described in any one of claims 1 to 12.

Images