• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    Computer-implemented job matching is revealed. Based on a combination of pattern matching and concept extraction from natural language information, a template is automatically adapted with an abstract concept of a concept network. Concept extraction is based on associating a natural language utterance with an abstract concept of a concept network. Pattern matching is used to distinguish between relevant and irrelevant natural language utterances based on the context of a natural language utterance. For each of the different combinations of a candidate and a job, a numerical match value is automatically calculated based on the corresponding candidate and job templates. An ordered list associated with the multiple different combinations is displayed via a visualizer, or sent to a user device for display via a visualizer. The list contains an order based on the numerical match values.
    公开号:BE1027696B1
    申请号:E20205783
    申请日:2020-11-02
    公开日:2021-06-14
    发明作者:Frank Platteau；Geert Devos
    申请人:Nalantis Nv；Nalantis Holding Ltd；
    IPC主号:
    专利说明:

    ANALYSIS AND COMPARISON OF CHARACTER-CODED DIGITAL DATA, PARTICULARLY FOR JOB MATCHING
    FIELD OF THE INVENTION The present invention relates to computer-implemented analysis and comparison of character-encoded digital data, in particular lexical, phrase and semantic analysis (GO6F 17/27), as well as natural language processing and generation (G06F 17/28). .
    BACKGROUND WO 2019/106 437 A2 describes a computer-implemented method for matching job offers (e.g., job openings) with job offers (e.g., resumes).
    Natural language processing techniques are used to interpret job specific terminology from job offers and/or job offers. The job bids are matched with the job offers based on a predetermined distance measure between the job bids and the job offers.
    WO 2019/106 437 A2 discloses a compartmentalized set of vocabulary terms (Jobzi Ontology), which are contextualized, hierarchical, synonym-enriched and relationship-oriented. Relationships between terms can take various forms, such as "is a synonym for", "part of", "is a", etc.
    WO 2019/106 437 A2 discloses a job title matching (Meta Work) with associated information such as knowledge, skills, courses and professional requirements including experience and salary. Skills are weighted, indicating the importance of a skill to the job title. The matching is based on weighted skills.
    WO 2019/106 437 A2 discloses penalties for lapse of experience, e.g. having worked as a 'truck driver' 10 years ago versus currently working as a 'truck driver'.
    However, WO 2019/106 437 A2 is not suitable for parsing documents containing narrative information, such as "I am a marketing director, reporting to the general manager," as often found in a cover letter. With the methodology of WO 2019/106 437 A2, it would be unclear whether 'marketing director', 'general manager', none or both are relevant.
    WO 2019/106 437 A2 discloses resume and job parsing automatically, but is silent on verification of the parsed information.
    WO 2019/106 437 A2 describes an ontology, but is silent about the creation of the ontology. In a rapidly evolving job market, with new job titles and skills emerging and declining, maintaining ontology is labor-intensive.
    WO 2019/106 437 A2 describes an ontology, but is silent about the use of multiple languages.
    Speer, Chin and Havasi, entitled “ConceptNet 5.5: “An Open Multilingual Graph of General Knowledge,” in Proceedings of the Thirty-First AAA! Conference on Artificial Intelligence (AAAI-17), pages 4444-4451 (2017) reveals a multilingual concept network. A concept network is a knowledge graph that connects words and phrases with labeled edges (e.g. "Synonym", "FormOf"). Multilingual functionality is realized in ConceptNet by labeling terms with a language (e.g. 'en', 'fr', 'it') and linking words in different languages, e.g. via a 'synonymous' relationship, e.g. 'polyglotte (fr)' being a 'synonym' of 'multilingual(s)'.
    A problem with Speer's (2017) concept network is that a multilingual diversity of natural language utterances that can be termed synonyms can each include one or more typical common meanings in a given language, details or partial overlaps of which can be lost in translation.
    The present invention aims to find a solution to at least some of the above-mentioned problems.
    SUMMARY OF THE INVENTION In a first aspect, the present invention provides a computer-implemented method for job matching, according to claim 1.
    In a second aspect, the present invention provides a computer system for job matching, wherein the computer system, such as a server, comprises means, such as a processor, configured to perform the method according to the first aspect.
    In a third aspect, the present invention provides a computer program for job matching, the computer program comprising instructions which, when the computer program is executed by a computer, such as a computer system according to the second aspect, causes the computer to perform the method according to the perform the first aspect.
    The present invention may further provide a physical, durable computer-readable data carrier, such as a compact disc (CD), a hard disk (HDD), or a solid state drive (SSD), comprising the computer program.
    The three aspects of the present invention are interrelated.
    Therefore, any feature described above or below may relate to any of the aspects of the present invention, even when described in connection with a particular aspect.
    The pattern matching makes it possible to process natural language information in narrative form, distinguishing between relevant and irrelevant natural language utterances.
    For example, in the utterance "I am a marketing director, reporting to the general manager and working closely with a graphic designer," three natural language utterances can be identified (e.g., "marketing director (nl)"; "general manager (nl)"; and " graphic designer (nl)') related to abstract professional concepts (eg.
    Marketing Director_OCC; General_ Director OCC; and Graphic_Designer_OCC). Through pattern matching, it can be determined that the occupation of the person associated with the narrative text, usually a candidate, is marketing director and not general manager or graphic designer.
    This should be contrasted with a template completed in a natural language, where a priori connections are clear through the fields of the template, i.e. a natural language utterance in the field for the current profession being natural language information of the candidate's current profession.
    In particular, the present invention provides for automatically at least partially populating a template with abstract concepts of a concept network based on natural language information in narrative form, such as a cover letter or job vacancy, based on a combination of pattern matching and concept extraction from the natural language information. in narrative form.
    The natural language information in narrative form is provided as character-encoded digital data. This automation reduces the need to manually fill in different templates. Template change can be performed for a candidate and/or a job, such as a candidate's cover letter or a job opening.
    Further advantages of the invention, and in particular of preferred embodiments, are described in the detailed description below.
    DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a computer-implemented method, a computer system and a computer program for job matching. The present invention is summarized in the corresponding section above. In what follows, the present invention is described in detail, preferred embodiments are discussed, and the present invention is illustrated by way of non-limiting examples.
    Unless otherwise defined, all terms used in disclosing the invention, including technical and scientific terms, have the meaning commonly understood by one of ordinary skill in the art to which this invention belongs. By way of further guidance, term definitions are included to better appreciate the teachings of the present invention.
    "A," "the," and "the," as used herein, include both singular and plural referents unless the context clearly dictates otherwise. By way of example, "a compartment" refers to one or more compartments.
    'Include', 'comprising' and 'includes' and 'consisting of' as used herein are synonymous with 'contain', 'containing' or 'contains' and are inclusive or open terms specifying the presence of what follows (e.g. a component) and do not exclude the presence of additional, not mentioned components, features, elements, parts, steps, which are well known in the art or described therein.
    "Based on" as used herein is synonymous with "based at least in part on" and is an inclusive or open term indicating the presence of what follows and excludes the presence of additional, unnamed components, features, parts, steps, which are well known in the art or described therein.
    5 For each candidate of one or more candidates, preferably several candidates, a candidate template is automatically adapted based on character-encoded digital data that includes candidate information expressed in a natural language. For each job of one or more jobs, preferably multiple jobs, a job template is automatically adapted based on character-encoded digital data that includes job information expressed in a natural language. For each of several different combinations of a candidate and a job, a numerical match value is automatically calculated based on corresponding candidate and job templates, preferably based on predetermined heuristic rules, more preferably based on a predefined distance measure between the candidate and job templates. An ordered list associated with the multiple different combinations is displayed via a visualizer, or sent to a user device for display via a visualizer. The list contains an order based on the numerical values.
    A template is automatically adapted by retrieving an abstract concept from a concept network based on a combination of pattern matching and concept extraction from the natural language information in the character-encoded digital data and inserting the abstract concept into the template. A candidate template can be automatically modified by retrieving an abstract concept from a concept network based on a combination of pattern matching and concept extraction from candidate information expressed in a natural language, particularly in narrative form, contained in character-encoded digital data and inserting the abstract concept in the candidate template. A job template can be automatically modified by retrieving an abstract concept from a concept network based on a combination of pattern matching and concept extraction from job information expressed in a natural language, especially in narrative form, contained in character-encoded digital data and inserting of the abstract concept in the job template. Concept extraction is based on associating a natural language utterance with an abstract concept of a concept network. Pattern matching is used to distinguish between relevant and irrelevant natural language utterances based on the context of a natural language utterance, in particular based on phrase analysis and/or semantic analysis of a phrase and/or paragraph containing a natural language utterance.
    A "concept network" as used herein is synonymous with "ontology" and refers to a knowledge graph that includes nodes and labeled edges. A 'node' of the concept network represents an abstract concept or a natural language expression. Preferably, an abstract concept is a natural language independent concept, in the sense that it is not used as such in a natural language. Marketing Director OCC, for example, is an abstract professional concept for 'marketing director (nl)'. The former is not used as such in any natural language, while the latter can be used in the natural language Dutch. A non-limiting list of examples of abstract concept types includes occupational concepts, competency concepts, work experience concepts, and qualification (or education) concepts. An "edge" or "connection" of the concept network connects two nodes and includes a relationship type. A relationship can be symmetric or asymmetric. A non-limiting list of examples of symmetric relationship types includes antonym', 'different from', 'etymologically related to', 'nearby', 'related to', 'similar to' and 'synonymous'. A non-limiting list of examples of asymmetric relationship types includes "by location," "capable of," "causes," "desire caused," "created by," "defined as," "derived from," "desires," 'implies', 'external URL', 'form of', 'has a', 'has context', 'has first subevent', 'has last subevent', 'has condition', 'has property', 'example of' , 'is a', 'made of', 'way of', 'motivated by purpose', 'impeded by', 'part of', 'receives action', 'purpose of', 'symbol of' and 'used for '. For example, 'concept network' is a 'synonym' for 'ontology': 'copper' is a 'metal'; and 'wheel' is a 'part of' a 'car'. Particular relationships used in the concept network of the present invention are "connotation", "parent", "child" and "has domain" relationships.
    Preferably, the concept network comprises a plurality of interconnected abstract concepts. Preferably, the concept network comprises a natural language utterance for each concept. Preferably, the concept network comprises for each concept an utterance in each natural language of several natural languages. Preferably, the multiple natural languages comprise at least two of English, German, French, Chinese, Japanese, Spanish, Portuguese, Swedish, Danish, Italian and Dutch. Preferably, the multiple natural languages include at least English, German, French, Chinese, Japanese, Spanish, Portuguese, Swedish, Danish, Italian and Dutch.
    This is beneficial because it allows to match a resume or cover letter in a first natural language with a vacancy in a second natural language that is different from the first natural language. In multilingual countries or regions, such as Belgium or Switzerland, or in multilingual environments, such as English being a preferred scientific communication language, regardless of the natural language of the place, this avoids the need to manually or automatically translate a CV, cover letter or vacancy before the matching can be performed. Preferably, a natural language utterance is linked to an abstract concept. Preferably, a natural language utterance is connected to an abstract concept by a synonym connection. Preferably, a natural language utterance is associated with one or more abstract concepts. Preferably, a natural language utterance is only associated with abstract concepts. Preferably, a natural language utterance is not connected to another natural language utterance.
    This is advantageous because it mitigates mistranslation due to only partially overlapping meanings for different natural language utterances, both for the same and for different natural languages. Where a natural language utterance can have multiple meanings, for example a 'handyman (nl) who is a Zeeman_OCC or a Repairer_OCC, the natural language utterance is associated with each of the abstract professional concepts, but most preferably not with other natural language utterances. The appropriate abstract occupational concept can be identified through pattern matching and/or contextual information. By traversing the concept network through abstract concepts, mistranslations or synonyms can be limited. Preferably, the concept network comprises a hierarchy of abstract occupational concepts. Preferably, abstract concepts of a hierarchy are connected via parent-child connections. Preferably, the hierarchy of occupational concepts is based on the Standard Occupational Classification (SOC) System. Preferably, the concept network comprises a plurality of abstract competency concepts. Preferably, the concept network comprises a hierarchy of abstract work experience concepts. Preferably, the multitude of competency concepts and/or the hierarchy of work experience concepts is based on the International Standard Classification of Occupations (ISCO). Preferably, a professional concept is linked to one or more competency and/or work experience concepts. Preferably, a professional concept is linked to one or more competency and/or work experience concepts via a connotation connection. Preferably, the concept network comprises a hierarchy of abstract qualification concepts. Preferably, a qualification concept is linked to one or more competency concepts. Preferably, a qualification concept is linked to one or more competence concepts via a connotation connection.
    In a preferred embodiment, a job batch comprising a plurality of digital job documents is provided and processed. Each digital job record contains character-encoded digital data that includes job information expressed in a natural language. A job template is automatically adapted for each digital job document of the job batch, as described above.
    In a preferred embodiment, a candidate batch comprising a plurality of candidate digital documents is provided and processed. Each candidate digital document contains character-encoded digital data that includes candidate information expressed in a natural language. A candidate template is automatically adapted for each digital candidate document of the candidate batch, as described above. In one embodiment, a modified template is manually verified. Preferably, verification information is automatically generated from the modified template. Preferably, the verification information is expressed in a natural language. Preferably, each abstract concept is converted into a natural language expression. Preferably, the verification information is displayed via a visualization means, or sent to a user device for display via a visualization means. Preferably, confirmation and/or correction data based on displayed verification information is obtained via a user input device, or received from a user device. In a preferred embodiment, filter data is obtained. Preferably, the filter data is obtained through a graphical user interface or from a user device based on input through a graphical user interface. A custom ordered list is displayed through the visualization assets, or sent to the user device for display through a visualization asset. The custom ordered list is based on the filter data. Preferably, the filter data is based on one or more abstract concepts of the concept network. Preferably, a user can filter for items in the ordered list based on one or more abstract concepts of the concept network. Preferably, the graphical user interface is associated with the visualization means. Preferably, the filter data is based on spatial reconfiguration, such as drag and drop, preferably drag and drop with a cursor device, from a displayed abstract concept to a filter area displayed through the visualization means.
    In a preferred embodiment, a new occupational concept is automatically detected on the basis of a corpus of documents, preferably via machine learning, by detecting clusters, preferably recurring clusters, of competence concepts in the corpus that do not sufficiently correspond to a common occupational concept present in the corpus. the concept of network.
    A candidate can be associated with competency concepts. A candidate may be directly associated with competency concepts, e.g. when competencies are explicitly stated in natural language utterances in the sign-coded digital data (e.g. a cover letter or CV), or indirectly, e.g. as implied by a qualification concept (e.g. education) or an occupational concept from a previous employment.
    A job can be associated with competency concepts. A job can be directly associated with competency concepts, e.g. when competencies are explicitly stated in natural language utterances in the sign-coded digital data (e.g. vacancy), or indirectly, e.g. as implied by a qualification concept (e.g. education) as required or a vocational concept as required work experience.
    In a preferred embodiment, a gap in competency concepts between a candidate and a job is automatically identified. A qualification concept is automatically determined to at least partially fill the gap in competence concepts. A natural language suggestion of the particular qualification concept associated with the job is displayed via a visualization resource, or sent to a user device for display via a visualization resource.
    In a preferred embodiment, the calculation of a numerical match value for a candidate and a job depends at least in part on the competency concepts and/or work experience concepts associated with the candidate and the competency concepts and/or work experience concepts associated with the job.
    In a preferred embodiment, a first natural language utterance is automatically added to the draft network, preferably via machine learning, based on translation pairs of documents from a corpus. Preferably, the first natural language is a 'new' natural language, for which the inclusion in the concept network is incomplete. A translation pair of documents refers to the first and a second natural language, i.e. a document in the first natural language and a document in the second natural language, both documents being translations of each other. In a first document in the first natural language, a first group of utterances is detected. An 'unknown' expression of the first group is not included in the concept network. At least two 'known' expressions of the first group are included in the concept network. In a corresponding second document in the second natural language, a corresponding second group of utterances is detected, based on locations within the (first and second) documents and/or the abstract concepts associated with the known utterances of the first group. A target utterance in the second group that matches the unknown utterance of the first group is determined. The unknown utterance of the first group is connected with the abstract concept associated with the target utterance of the second group.
    This is advantageous as it allows bootstrap acquisition of a new language in the concept network. Based on an initial set of natural language utterances in the new language added based on automatic translation and/or manual addition, other natural language utterances in the new language can be added automatically as described above.
    The invention is further described by the following non-limiting examples which further illustrate the invention, and are not intended, nor should be interpreted, to limit the scope of the invention.
    EXAMPLE A candidate uploads a document (eg resume or cover letter) to a server according to the second aspect of the present invention via his user device. The document includes character-encoded digital data, which is analyzed on the server, and based on which a candidate template associated with the candidate is modified. The document and candidate template are stored in a server database.
    For example, consider that the document has the sentence "I have been closely monitoring the metabolomics project." in natural language English. Through pattern matching and concept extraction, the concepts 'Initiative' and 'Biomedical Engineering WE' are retrieved from the document and added to the candidate template. Concept extraction identifies the natural language utterances 'follow (nl)' and 'metabolomics- project (nl)' in the sentence, as well as their respective associated concepts 'Initiative' and 'Biomedical Engineering WE' from the concept network. Through pattern recognition, a degree of certainty is obtained that the identified concepts relate to the candidate, and if the level of certainty is sufficiently high, the identified concepts are added to the candidate template, in this way the document is fully analyzed.
    The candidate template consists of a detail part and a concept part. The detail section contains fields related to candidate details, such as, for example, name, address, references to uploaded files, language, seniority, work type, academic level, work domain, address and/or telephone number. Most preferably, these details are also obtained automatically based on pattern matching. In particular, the detail portion may include geographic coordinates, such as latitude and longitude, to allow distance filtering for job matching. Geographic coordinates can be automatically retrieved based on a candidate's address. In particular, the detail portion may include a candidate ID (for database retrieval), which is preferably a string containing alphanumeric characters and/or dashes. The concept section is divided into a multitude of topics, including, for example, Experience, Language, Education and Competencies. The concepts extracted from the document have been added to the corresponding topics. For example, 'Biomedical Technology WE' stands for work experience in biomedical technology and is therefore added to the subject Experience. A concept can include a weight. Each concept can include a weight. A work experience concept may include a weight based on the length of the experience, preferably corrected for penalizing experiential decay. A competency concept may include a weight based on weights of related concepts, eg work experience concepts, and/or prevalence in the document.
    Preferably, the template also includes in conjunction with each concept the corresponding natural language utterance of the document and/or a corresponding reference to a particular corresponding portion of the document.
    A web page is generated with authentication information. The verification information is expressed in natural language and derived from the candidate template. Each draft of the template is therefore converted into a natural language utterance that corresponds to the natural language indicated in the detail section of the candidate template. The web page is sent to the candidate's user device and displayed through a user device screen. The web page is configured to modify natural language utterances at the candidate's option via a user input device of the user device. The web page further includes a confirmation button to confirm the original displayed and/or modified natural language expressions of the web page. This makes it possible to automatically generate an at least partially completed template and only edit it manually where necessary. Preferably, corrected information is retained for automatic improvement, preferably through machine learning, or document analysis.
    The candidate can then search for jobs based on the candidate template. A numerical matching value is obtained for a combination of the candidate template and a job template. An ordered list is generated with job identifiers and/or job prompts, arranged according to corresponding numerical match values. A web page containing the ordered list is generated, sent to the user device and presented to the candidate through the user device screen. The web page further includes functionality to enable or disable filters and/or reorder the list, such as, for example, based on distance or numerical match value. Selecting a job identifier and/or referral address from the ordered list on the web page will display information specific to the job, such as numeric match value for each concept category, a general numeric match value, a job vacancy or a reference address to it, and information related to the job template. The job template may also include a detail section and a draft section. The detail section contains fields related to job details, such as, for example, company name, address, references to uploaded files, language, seniority, work type, academic level, work domain, address and/or telephone number. Most preferably, these details are also obtained automatically based on pattern matching. In particular, the detail portion may include geographic coordinates, such as latitude and longitude, to allow distance filtering for job matching. Geographic coordinates can be automatically retrieved based on an address of the job. In particular, the detail portion may include a job ID (for database retrieval), which is preferably a string containing alphanumeric characters and/or dashes. The concept section is divided into a multitude of topics, including, for example, Experience,
    Language, Education and Competencies.
    A concept can include a weight.
    Each concept can include a weight.
    If certain competencies or training are missing to make the candidate fully comply with the full overview of the vacancy, suggestions for further qualification and/or training can also be presented to the candidate.
    权利要求:
    Claims (25)
    [1]
    A computer-implemented job matching method comprising the steps of: e automatically matching for each candidate one or more candidates of a candidate template based on character-encoded digital data comprising candidate information expressed in a natural language; e automatically adapting for each job of one or more jobs a job template based on character-encoded digital data comprising job information expressed in a natural language; e automatically calculating for each of a plurality of different combinations of a candidate and a job a numerically corresponding value based on the corresponding candidate and job templates; and e displaying via a visualization means, or sending to a user device for display via a visualization means, an ordered list associated with the plurality of different combinations, said list comprising an order based on the numerical match values, wherein a template is automatically modified by retrieving an abstract concept from a concept network based on a combination of pattern matching and concept extraction from the natural language information and inserting the abstract concept into the template, where concept extraction is based on an association of a natural language utterance with an abstract concept of a concept network, which uses pattern matching to distinguish between relevant and irrelevant natural language utterances based on the context of a natural language utterance.
    [2]
    A computer-implemented method according to any preceding claim, comprising the steps of: providing a job batch comprising a plurality of digital job records, each digital job record comprising character-encoded digital data comprising job information expressed in a natural language; and automatically adapting a job template for each digital job document of the job batch.
    [3]
    A computer-implemented method according to any one of the preceding claims, comprising the steps of: providing a candidate batch comprising a plurality of candidate digital documents, wherein each candidate digital document comprises character-encoded digital data comprising candidate information expressed in a natural language; and automatically adapting a candidate template for each candidate digital document of the candidate batch.
    [4]
    A computer-implemented method according to any preceding claim, comprising verifying a custom template via: e automatically generating from the custom template verification information expressed in a natural language, thereby converting each abstract concept into a natural language utterance; e displaying via a visualization means, or transmitting to a user device for display via a visualization means, the verification information; e obtaining, via a user input device, or receiving a user device, confirmation and/or correction data based on the displayed verification information.
    [5]
    A computer-implemented method according to any one of the preceding claims, comprising the steps of: obtaining filter data, preferably via a graphical user interface or from a user device based on input via a graphical user interface; and displaying via the visualization means, or sending to the user device for display via a visualization means, a custom ordered list, the custom ordered list being based on the filter data.
    [6]
    A computer-implemented method according to preceding claim 5, wherein the filter data is based on one or more abstract concepts of the concept network.
    [7]
    A computer-implemented method according to preceding claim 6, wherein the filter data is obtained via a graphical user interface or from a user device based on input via a graphical user interface, the graphical user interface being associated with the visualization means, the filtering data being based on spatial reconfiguration, such as drag-and-drop, of a rendered abstract concept to a filter area rendered through the visualization tool.
    [8]
    A computer-implemented method according to any preceding claim, wherein pattern matching is used to distinguish between relevant and irrelevant natural language utterances based on the context of a natural language utterance, in particular based on phrase analysis and/or semantic analysis of a phrase and/or paragraph containing a natural language utterance.
    [9]
    A computer-implemented method according to any preceding claim, wherein the concept network comprises a plurality of interconnected abstract concepts, and for each concept an utterance in each of multiple natural languages.
    [10]
    A computer-implemented method according to preceding claim 9, wherein a natural language utterance is connected to an abstract concept, preferably by a synonym connection.
    [11]
    A computer-implemented method according to any one of the preceding claims 9 and 10, wherein a natural language utterance is not directly associated with another natural language utterance.
    [12]
    A computer-implemented method according to any one of the preceding claims 9 to 11, wherein a first natural language utterance is automatically added to the concept network, preferably via machine learning, based on translation pairs of documents of a corpus, wherein a translation pair of documents relates to the first and a second natural language, wherein the automatic addition of an utterance comprises the following steps: e detecting in a first document in the first natural language a first group of utterances, whereby an unknown utterance of the first group is not included in the draft network, wherein at least two known utterances of the first group are included in the draft network; e determining in a corresponding second document in the second natural language a corresponding second group of utterances based on locations in the documents and/or the abstract concepts associated with the known utterances of the first group; e determining a target utterance in the second group that corresponds to the unknown utterance of the first group; and e associating the unknown utterance of the first group with the abstract concept associated with the target utterance of the second group.
    [13]
    A computer-implemented method according to any one of the preceding claims 9 to 12, wherein said multiple natural languages comprise at least English, German, French, Chinese, Japanese, Spanish, Portuguese, Swedish, Danish, Italian and Dutch.
    [14]
    A computer-implemented method according to any preceding claim, wherein the concept network comprises a hierarchy of abstract occupational concepts.
    [15]
    A computer-implemented method according to preceding claim 14, wherein the hierarchy of occupational concepts is based on the Standard Occupational Classification (SOC) System.
    [16]
    A computer-implemented method according to any preceding claim, wherein the concept network comprises a hierarchy of abstract competency concepts.
    [17]
    The computer-implemented method of the preceding claim 16, wherein the hierarchy of competency concepts is based on the International Standard Classification of Occupations (ISCO).
    [18]
    A computer-implemented method according to claim 14 or 15 and according to claim 16 or 17, wherein a professional concept is associated with one or more competency concepts, preferably through a connotation connection.
    [19]
    A computer-implemented method according to the preceding claim 18, comprising the step of automatically detecting a new occupational concept based on a corpus of documents, preferably recurring clusters, of competence concepts in the corpus that do not sufficiently correspond to a current occupational concept.
    [20]
    A computer-implemented method according to any preceding claim, wherein the concept network comprises a hierarchy of abstract qualification concepts.
    [21]
    A computer-implemented method according to preceding claim 20, wherein a qualification concept is associated with one or more competence concepts.
    [22]
    The computer-implemented method of claim 21, wherein a candidate is associated with competency concepts, a job is associated with competency concepts, automatically determining a gap in competency concepts between the candidate and the job, automatically establishing a qualification concept to at least partially filling the gap in competency concepts, wherein a natural language suggestion of the established qualification concept associated with the job is displayed via a visualization resource, or sent to a user device for display via a visualization resource.
    [23]
    A computer-implemented method according to any one of the preceding claims 21 and 22, wherein a candidate is associated with competency concepts, wherein a job is associated with competency concepts, wherein the calculation of a numerical match value for a candidate and a job depends at least on the competency concepts associated with the candidate and the competency concepts associated with the job.
    [24]
    A computer system for job matching, wherein the computer system comprises means configured to perform a method according to any one of the preceding claims 1 to 23.
    [25]
    A computer program for job matching, the computer program comprising instructions which, when the program is executed by a computer, cause the computer to perform a method according to any one of claims 1 to 23.
    类似技术:
    公开号 | 公开日 | 专利标题
    US20160306846A1|2016-10-20|Visual representation of question quality
    Bonta et al.2019|A comprehensive study on lexicon based approaches for sentiment analysis
    US9881037B2|2018-01-30|Method for systematic mass normalization of titles
    US9053418B2|2015-06-09|System and method for identifying one or more resumes based on a search query using weighted formal concept analysis
    US9639522B2|2017-05-02|Methods and apparatus related to determining edit rules for rewriting phrases
    US20100114789A1|2010-05-06|System and method for guiding users to candidate resumes and current in-demand job specification matches using predictive tag clouds of common, normalized elements for navigation
    US20180366013A1|2018-12-20|System and method for providing an interactive visual learning environment for creation, presentation, sharing, organizing and analysis of knowledge on subject matter
    US20170103439A1|2017-04-13|Searching Evidence to Recommend Organizations
    US20160103920A1|2016-04-14|System for, and method of, searching data records
    CN106294466A|2017-01-04|Disaggregated model construction method, disaggregated model build equipment and sorting technique
    Modaresi et al.2016|Exploring the Effects of Cross-Genre Machine Learning for Author Profiling in PAN 2016.
    US10783877B2|2020-09-22|Word clustering and categorization
    Khaouja et al.2019|Building a soft skill taxonomy from job openings
    JP4600045B2|2010-12-15|Opinion extraction learning device and opinion extraction classification device
    JP5218409B2|2013-06-26|Related information search system and related information search method
    Phan et al.2019|A sentiment analysis method of objects by integrating sentiments from tweets
    BE1027696B1|2021-06-14|ANALYSIS AND COMPARISON OF CHARACTER-CODED DIGITAL DATA, PARTICULARLY FOR JOB MATCHING
    KR102099364B1|2020-04-09|Artificial intelligence book writing guiding system and thereof
    US20200394724A1|2020-12-17|Persona based intervention
    KR20200064490A|2020-06-08|Server and method for automatically generating profile
    JP5877775B2|2016-03-08|Content management apparatus, content management system, content management method, program, and storage medium
    US20190303437A1|2019-10-03|Status reporting with natural language processing risk assessment
    US20160103836A1|2016-04-14|System for, and method of, ranking search results
    Chathuranga et al.2018|Annotating opinions and opinion targets in student course feedback
    Jiménez et al.2021|Opinion Mining for Instructor Evaluations at the Autonomous University of Ciudad Juarez
    同族专利:
    公开号 | 公开日
    WO2021089129A1|2021-05-14|
    BE1027696A1|2021-05-17|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20090276415A1|2008-05-01|2009-11-05|Myperfectgig|System and method for automatically processing candidate resumes and job specifications expressed in natural language into a common, normalized, validated form|
    US20150317610A1|2014-05-05|2015-11-05|Zlemma, Inc.|Methods and system for automatically obtaining information from a resume to update an online profile|
    US20160232160A1|2014-11-26|2016-08-11|Vobis, Inc.|Systems and methods to determine and utilize conceptual relatedness between natural language sources|
    WO2019106437A2|2017-11-30|2019-06-06|Jobzi Inteligencia De Dados Na Internet, Ltda.|Matching bids for work with offers for work|
    法律状态:
    2021-06-28| FG| Patent granted|Effective date: 20210614 |
    优先权:
    申请号 | 申请日 | 专利标题
    PCT/EP2019/080286|WO2021089129A1|2019-11-05|2019-11-05|Analysis and comparison of character-coded digital data, in particular for job matching|
    [返回顶部]