Resume searching method and device, equipment and storage medium

Abstract

The invention provides a resume searching method and device, equipment and a storage medium. The method comprises the steps that search terms and search conditions, input by a user, of a specific industry are obtained; identifying a retrieval intention of the user based on the search term and the search condition; searching a resume matched with the retrieval intention from a pre-constructed resume library as a target resume, and pushing and displaying the target resume; word segmentation indexes of a plurality of resumes are stored in the resume library, and the word segmentation index of each resume is constructed based on a pre-constructed professional lexicon of a specific industry. According to the method, word segmentation is performed on the resume based on the professional lexicon of the specific industry, the vocabulary data of the specific industry can be accurately divided from the resume, and the word segmentation index of the resume is constructed based on the divided vocabulary data, so that when the resume is searched, the search efficiency is improved. And the identified retrieval intention can be compared with the accurate vocabulary data of the specific industry recorded in the word segmentation index of each resume, so that the accurate search of the resume of the specific industry is realized.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a resume search method, apparatus, device, and storage medium. Background Technology

[0002] Currently, recruitment websites widely utilize full-text search engines such as Lucene and Solr to assist companies in their recruitment processes, enabling rapid searches by comprehensively indexing job seekers' resumes. However, current search technologies struggle to accurately identify specialized terminology in resumes, especially in specific industries. This makes it difficult for companies to find resumes matching their recruitment needs when using recruitment platforms to search for resumes in specific industries, resulting in poor search quality. Therefore, how to achieve accurate resume searching in specific industries is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0003] In view of this, this disclosure proposes a resume search method, apparatus, device, and storage medium that can improve the accuracy of resume searches in specific industries.

[0004] According to a first aspect of this disclosure, a resume search method is provided, comprising:

[0005] Obtain the user's industry-specific search terms and search criteria;

[0006] Based on the search terms and search criteria, the user's search intent can be identified;

[0007] The system searches a pre-built resume database for resumes that match the search intent, selects target resumes as these, and then pushes and displays these target resumes.

[0008] The resume database stores word segmentation indexes for multiple resumes, and each resume's word segmentation index is constructed based on a pre-built professional thesaurus for the specific industry.

[0009] In one possible implementation, the construction of the industry-specific thesaurus includes:

[0010] Obtain literature and materials related to the specific industry;

[0011] Vocabulary data for the specific industry is selected from the literature, and a word segmentation lexicon is constructed based on the vocabulary data;

[0012] Semantic understanding is performed on the context in which the vocabulary data is located to obtain the synonyms of the vocabulary data, and a thesaurus is constructed based on the synonyms of the vocabulary data;

[0013] The word segmentation lexicon and the synonym lexicon are combined to obtain the professional lexicon for the specific industry.

[0014] In one possible implementation, when constructing the word segmentation index for the resume based on a professional thesaurus for the specific industry, the following is included:

[0015] The resume is segmented into words based on the professional thesaurus of the specific industry, and the word segmentation result of the resume is obtained.

[0016] The word segmentation results of the resume are matched with the relevant key fields defined in the pre-built word segmentation index pattern to obtain the word segmentation index of the resume.

[0017] In one possible implementation, after obtaining the word segmentation index of the resume, the method further includes: compressing the word segmentation index of the resume and storing the compressed word segmentation index in the resume database.

[0018] In one possible implementation, when identifying the user's search intent based on the search terms and search conditions, it is done based on a pre-built intent recognition model.

[0019] In one possible implementation, when pushing and displaying the target resume, the method further includes:

[0020] Obtain the activity level of the user to whom the target resume belongs;

[0021] The target resumes are reordered based on the user's activity level;

[0022] The rearranged results of the target resume will be pushed and displayed.

[0023] In one possible implementation, when obtaining industry-specific search terms input by the user, the method further includes optimizing the search terms.

[0024] According to a second aspect of this disclosure, a resume search device is provided, comprising:

[0025] The data input module is used to obtain the search terms and search criteria for a specific industry entered by the user;

[0026] The intent recognition module is used to identify the user's search intent based on the search terms and search conditions;

[0027] The resume search module is used to find resumes that match the search intent from a pre-built resume library as target resumes and push and display the target resumes.

[0028] The resume database stores word segmentation indexes for multiple resumes, and each resume's word segmentation index is constructed based on a pre-built professional thesaurus for the specific industry.

[0029] According to a third aspect of this disclosure, a resume search device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to perform the method described in the first aspect of this disclosure.

[0030] According to a fourth aspect of this disclosure, a non-volatile computer-readable storage medium is provided that stores computer program instructions thereon, wherein the computer program instructions, when executed by a processor, implement the method described in the first aspect of this disclosure.

[0031] This disclosure provides a resume search method, apparatus, device, and storage medium. The method includes: acquiring search terms and search conditions for a specific industry input by a user; identifying the user's search intent based on the search terms and search conditions; searching a pre-built resume database for resumes matching the search intent as target resumes, and pushing and displaying the target resumes; wherein the resume database stores word segmentation indexes for multiple resumes, and each resume's word segmentation index is constructed based on a pre-built industry-specific professional thesaurus. In this disclosure, resumes are segmented based on an industry-specific professional thesaurus, thereby accurately extracting industry-specific vocabulary data from the resumes, and constructing a resume's word segmentation index based on the segmented vocabulary data. Thus, during resume search, the identified search intent can be compared with the accurate industry-specific vocabulary data recorded in the word segmentation indexes of each resume, thereby achieving accurate search for industry-specific resumes.

[0032] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0033] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this disclosure together with the specification and serve to explain the principles of this disclosure.

[0034] Figure 1 A flowchart illustrating a resume search method according to an embodiment of the present disclosure is shown;

[0035] Figure 2 A schematic block diagram of a resume search device according to an embodiment of the present disclosure is shown;

[0036] Figure 3 A schematic block diagram of a resume search device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0037] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0038] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0039] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0040] <Method Implementation>

[0041] Figure 1 A flowchart illustrating a resume search method according to an embodiment of the present disclosure is shown. Figure 1 As shown, the method includes steps S1100-S1300.

[0042] S1100 retrieves the user-inputted industry-specific search terms and search criteria. The specific industry refers to highly specialized industries, typically containing numerous technical terms, such as the power industry, telecommunications industry, and computer industry. Search criteria include key fields from the resume, such as work experience, salary, and educational background. Search terms are specific descriptions of these key fields.

[0043] In one possible implementation, to guide users in entering accurate search terms, the process of acquiring industry-specific search terms input by the user also includes optimizing the search terms. This optimization can be achieved through features such as providing search suggestions, auto-completion, and auto-correction.

[0044] Optimizing search terms using search suggestions may include the following steps: First, collect search logs from all users who used a search engine (i.e., the search engine executing the resume search method described in this disclosure) within a set time period. These search logs record information such as the search terms entered by the users, the search results clicked, and the search time. Second, identify popular search terms from all collected search logs and recommend these popular searches as search suggestions. Users can then select a popular search term from the search suggestions that matches their search needs as their final input search term.

[0045] When optimizing search terms using autocomplete, the following steps can be included: First, obtain the text currently entered by the user. Second, obtain the popular search terms and frequently used search terms identified above, and use a prefix matching algorithm to filter out search terms that match the currently entered text from the popular search terms and frequently used search terms. Then, use the search terms that match the currently entered text to autocomplete the text, and use the autocompleted search terms as the final input search terms. The prefix matching algorithm can be a Trie prefix tree matching algorithm or other prefix matching algorithms; no specific limitation is made here. In one possible implementation, the parts of the popular search terms and frequently used search terms that match the currently entered text can also be highlighted to make them visually easier to identify.

[0046] When optimizing search terms through automatic error correction, a pre-built professional thesaurus is used. This professional thesaurus includes a word segmentation thesaurus and a synonym thesaurus. The professional thesaurus records vocabulary data specific to certain industries, including standard professional terms and common words. The synonym thesaurus records synonyms for each vocabulary item in the word segmentation thesaurus.

[0047] In one possible implementation, when automatically correcting user-input search terms based on a pre-built professional thesaurus, it is necessary to first determine whether there is vocabulary data in the word segmentation thesaurus that is the same as the search term: if it exists, it is determined that the user-input search term is standard, and then the user-input search term is directly used as the final input search term; if it does not exist, it is determined that the user-input search term is non-standard, and then the operation of correcting the user-input search term is performed.

[0048] In one possible implementation, correcting user-input search terms may include the following steps: First, convert the user-input search term into Pinyin and determine if there is any word data in the word segmentation dictionary with the same Pinyin as the search term. If there is, push the word data with the same Pinyin as the search term so that the user can select the correct search term from it, thus correcting the search term. If no word data with the same Pinyin is found, continue to determine if there is a synonym in the thesaurus with the same Pinyin as the user-input search term. If a synonym exists, directly read the corresponding word data and use it as the corrected search term. If no synonym exists, select a synonym with the same Pinyin as the search term from the thesaurus and push the word data of the synonym with the same Pinyin as the search term for the user to select and correct the search term.

[0049] After obtaining the user's input search terms and search conditions, step S1200 can be executed to identify the user's search intent based on the search terms and search conditions. This search intent includes the job position the user needs to recruit for and the job requirements. In one possible implementation, the user's search intent is identified based on a pre-built intent recognition model. Specifically, the obtained search terms and search conditions are input into the pre-built intent recognition model, which automatically outputs the user's search intent. When building the intent recognition model, a set amount of intent recognition samples needs to be constructed. Each intent recognition sample includes the user-defined search conditions and search terms, and the corresponding search intent is labeled for each user-defined search condition and search term. Then, the parameters of the neural network model are trained using a set number of intent recognition samples. After training, the intent recognition model is obtained. This neural network model can be CNN, RNN, etc., without specific limitations.

[0050] After identifying the user's search intent, step S1300 can be executed, which involves searching a pre-built resume database for resumes that match the search intent as target resumes and then pushing and displaying these target resumes. The resume database stores word segmentation indexes for multiple resumes, each index being built based on a pre-built industry-specific thesaurus.

[0051] It should be noted here that, to ensure the accuracy of the word segmentation index construction for each resume, it is necessary to first build a professional thesaurus for a specific industry. One possible implementation involves the following steps when building this industry-specific thesaurus:

[0052] First, obtain industry-specific literature. Specifically, you can retrieve industry-specific literature from the internet. This literature can include industry reports, conference materials, technical standards, dissertations, patent documents, etc., without further specific limitations.

[0053] Second, industry-specific vocabulary data is selected from industry-specific literature, and a word segmentation lexicon is constructed based on this selected vocabulary data. This industry-specific vocabulary data includes industry-specific technical terms and common words. Storing all the selected technical terms and common words in the word segmentation lexicon completes the construction of the lexicon.

[0054] Third, semantic understanding is performed on the context of the vocabulary data to obtain synonyms, and a thesaurus is constructed based on these synonyms. Specifically, the literature and the vocabulary data identified within it are input into a Natural Language Processing (NLP) model. NLP performs contextual understanding on the vocabulary data appearing in the literature and then identifies synonyms based on the contextual understanding results. The identified vocabulary data and its synonyms are then associated and stored in the thesaurus, thus completing the construction of the thesaurus. In other words, the thesaurus includes synonyms for each vocabulary data item in the word segmentation lexicon.

[0055] Fourth, the word segmentation lexicon and the synonym lexicon are combined to obtain a professional lexicon for a specific industry. In other words, the professional lexicon is composed of two lexicons: the word segmentation lexicon and the synonym lexicon.

[0056] After constructing a specialized thesaurus for a specific industry, it will be automatically updated. Specifically, new literature from the specific industry will be retrieved periodically, and then vocabulary data and synonyms for each vocabulary data will be extracted from the new literature data following the steps described above. The newly extracted vocabulary data and synonyms will then be added to the specialized thesaurus to achieve automatic updates.

[0057] After building a professional thesaurus, word segmentation indexes can be constructed for each resume based on the built professional thesaurus. It should be noted that the search engine includes personal clients and enterprise clients. Each job seeker needs to register an account during the job search process, and then enter their personal resume through the registered account. After the personal resume is submitted, the system will obtain the submitted resume and build a word segmentation index for the newly submitted resume.

[0058] In one possible implementation, when constructing a word segmentation index for a resume based on a specialized thesaurus for a specific industry, the following steps may be included:

[0059] First, the resume is segmented based on a specialized industry-specific thesaurus, yielding segmentation results. Specifically, all vocabulary data and synonyms recorded in the specialized thesaurus are added to the default dictionary of the IK Chinese word segmenter, thus enriching the default dictionary with industry-specific terminology, common words, and synonyms. Then, the IK_SMART_WORD tool in the IK Chinese word segmenter uses the enriched default dictionary to segment the resume, resulting in multiple segmentation results.

[0060] Second, the word segmentation results of the resume are matched with the relevant key fields defined in the pre-built word segmentation index pattern to obtain the resume's word segmentation index. Specifically, a pre-configured word segmentation index pattern is retrieved from the Elasticsearch engine. This word segmentation index pattern records all the key fields that may be included in the resume. Then, the word segmentation results of the resume are matched with the relevant key fields defined in the word segmentation index pattern, thus completing the construction of the resume's Elasticsearch word segmentation index. For ease of description, the resume's Elasticsearch word segmentation index is referred to as the resume's word segmentation index in this publication.

[0061] In one possible implementation, to improve the standardization of resume word segmentation index construction, when matching the word segmentation results of the resume to the relevant key fields defined in the pre-built word segmentation index pattern, the word segmentation results will also be cleaned by removing irrelevant characters and unifying the vocabulary expression form (such as converting traditional Chinese to simplified Chinese, unifying capitalization, etc.). Then, the cleaned word segmentation results will be matched to the relevant key fields defined in the pre-built word segmentation index pattern.

[0062] In one possible implementation, the word segmentation index for a resume is constructed using an inverted sorting method. Constructing the word segmentation index using an inverted sorting method allows for faster document retrieval, making the information retrieval process more efficient.

[0063] In one possible implementation, the search engine also includes a monitoring system that monitors in real time whether resumes are being updated and whether new resumes are being added. When a new resume is detected, a synchronization mechanism is triggered. This mechanism builds a word segmentation index for the new resume using the same process as described above and synchronizes it to the resume database. Similarly, when a resume is updated, a synchronization mechanism is triggered again. This mechanism reconstructs the resume's word segmentation index using the same process, thus achieving synchronized updates to the resume's word segmentation index. By using a monitoring system, the word segmentation indexes for each resume in the database can be generated and updated synchronously, allowing enterprise users to access the latest resume data in real time and ensuring the timeliness of search results.

[0064] In one possible implementation, the monitoring system uses RabbitMQ message middleware to monitor resume updates and new resume additions in real time. When a resume update or new resume is detected, RabbitMQ triggers an XXL-job to perform data synchronization. This process includes periodic data synchronization between the database and Elasticsearch index, as well as necessary data integrity checks. The system also includes a retry mechanism to prevent unexpected packet loss or communication failures during data synchronization, ensuring data security and accuracy. The retry mechanism works as follows: First, using aspect-oriented programming, when an anomaly occurs during data synchronization, the operation is captured and recorded in the database. Second, a scheduled task reads the anomaly operation every 5 minutes and re-executes it. Then, after a successful retry, the anomaly is cleared. If the number of retries exceeds the system-defined limit, an email alert is sent for manual intervention. Ultimately, this ensures that user resume data is not lost.

[0065] In one possible implementation, after obtaining the word segmentation index of the resume, the process further includes: compressing the word segmentation index and storing the compressed index in the resume database. It should be noted that, to improve the subsequent retrieval efficiency, the resume database is divided into at least two storage areas. When storing the compressed word segmentation index in the resume database, the resume ID is first obtained, then the hash value of the resume ID is calculated, and resumes are randomly and relatively evenly distributed to the corresponding storage areas based on the hash value of the resume ID.

[0066] For each resume entered by an individual client, a corresponding word segmentation index is constructed using the method described above. The word segmentation indexes of numerous resumes constitute the resume library. After the resume library is constructed, and the identified user intent is obtained, resumes matching the user's search intent can be found in the resume library as target resumes, which are then pushed and displayed.

[0067] In one possible implementation, when searching for resumes that match the user's search intent, it is necessary to segment the user's search intent using a professional thesaurus and filter out entity words with practical meaning. Then, the entity words with practical meaning are matched with the segmented words in the segmentation index of each resume, and resumes that include entity words with practical meaning in the segmentation index are filtered out as target resumes.

[0068] After selecting target resumes, they will be scored, and the top N resumes with the highest scores will be retrieved and displayed according to their scores. The scoring of target resumes can be implemented using similarity evaluation algorithms such as BM25 or PageRank, which will not be elaborated upon here.

[0069] In one possible implementation, the following steps are also included when pushing and displaying the target resume:

[0070] First, the activity level of the users associated with the target resumes is obtained. Specifically, after recalling the top N target resumes, the behavioral data of the users associated with those resumes is also obtained, and the activity level of each user is calculated based on this behavioral data. In one possible implementation, user behaviors include logging into the search engine app using a personal client, participating in IM conversations, viewing jobs, applying for jobs, and refreshing resumes. For each behavior, the number of times it is executed and the time are first obtained. Then, based on the number of times and the time, an activity score for each behavior is determined. Finally, the activity scores for each behavior are weighted and summed, and the result is the user's final activity level. It should be noted that the system sets scoring rules for each behavior based on its number of executions and the time. Thus, after obtaining the number of times and the time for an behavior, the score can be determined according to the scoring rules. For example, if a user applies for 10 jobs within 3 days, and the system stipulates that applying for 10 jobs within 3 days earns 6 points, then the activity score for the behavior of applying for jobs is 6 points.

[0071] Second, the target resumes are reordered based on user activity. Specifically, the retrieved target resumes are reordered according to user activity, and the result of this reordering is the final ranking of the retrieved target resumes.

[0072] Third, push and display the results of the target resume's reordering.

[0073] Reordering the retrieved resumes based on activity levels makes it easier for companies to discover active and proactive job seekers. This means that companies can accurately retrieve resumes that meet their search intent and understand the activity level of the users whose resumes they belong to based on the final ranking of the resumes.

[0074] In one possible implementation, for each target resume ultimately pushed to the system, status data for that resume will also be displayed. This status data includes whether the resume has been viewed, whether it has been submitted to a specific job posting, and whether it has been downloaded. This status data helps enterprise users quickly understand the activity level and application progress of each candidate's resume.

[0075] In one possible implementation, the search engine also provides two search modes: fuzzy search and precise search, to meet the search needs of different users. Specifically, after a user enters a search term, it will determine whether the search term includes the user's unique identifier. If the user's unique identifier is included, it is determined that the user needs to perform a precise search for resumes. In this case, resumes containing the user's unique identifier will be directly retrieved from the resume database and pushed to the enterprise user as the final target resumes, thereby achieving precise search for candidate resumes. This unique identifier can be a resume number, mobile phone number, email address, etc., and is not specifically limited here. If the search term does not include the user's unique identifier, it is determined that the user needs to perform a fuzzy search for resumes. The specific process of fuzzy search is described in steps S1200-S1300, and will not be repeated here.

[0076] This disclosure provides a resume search method, including: obtaining search terms and search conditions for a specific industry input by a user; identifying the user's search intent based on the search terms and search conditions; searching a pre-built resume database for resumes matching the search intent as target resumes, and pushing and displaying the target resumes; wherein, the resume database stores word segmentation indexes for multiple resumes, and each resume's word segmentation index is constructed based on a pre-built professional thesaurus for a specific industry. In this disclosure, resumes are segmented based on a professional thesaurus for a specific industry, thereby accurately extracting specific industry vocabulary data from the resumes, and constructing word segmentation indexes for the resumes based on the segmented vocabulary data. Thus, during resume search, the identified search intent can be compared with the accurate specific industry vocabulary data recorded in the word segmentation indexes of each resume, thereby achieving accurate search for resumes in a specific industry.

[0077] <Device Embodiment>

[0078] Figure 2 A schematic block diagram of a resume search device according to an embodiment of the present disclosure is shown. Figure 2 As shown, the device 100 includes:

[0079] Data input module 110 is used to obtain the search terms and search conditions for a specific industry entered by the user;

[0080] The intent recognition module 120 is used to identify the user's search intent based on search terms and search conditions;

[0081] The resume search module 130 is used to find resumes that match the search intent from a pre-built resume library as target resumes and push and display the target resumes.

[0082] The resume database stores word segmentation indexes for multiple resumes, each index being built based on a pre-constructed industry-specific thesaurus.

[0083] <Equipment Example>

[0084] Figure 3 A schematic block diagram of a resume search device according to an embodiment of the present disclosure is shown. Figure 3 As shown, the resume search device 200 includes a processor 210 and a memory 220 for storing executable instructions of the processor 210. The processor 210 is configured to implement any of the resume search methods described above when executing the executable instructions.

[0085] It should be noted here that the number of processors 210 can be one or more. Furthermore, the resume search device 200 of this embodiment may also include an input device 230 and an output device 240. The processors 210, memory 220, input device 230, and output device 240 can be connected via a bus or other means, which are not specifically limited here.

[0086] The memory 220, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and various modules, such as the program or module corresponding to the resume search method of this disclosure embodiment. The processor 210 executes various functional applications and data processing of the resume search device 200 by running the software program or module stored in the memory 220.

[0087] Input device 230 can be used to receive input digital numbers or signals. These signals may include key signals related to user settings and function control of the device / terminal / server. Output device 240 may include a display device such as a screen.

[0088] <Storage Medium Examples>

[0089] According to a fourth aspect of this disclosure, a non-volatile computer-readable storage medium is also provided, on which computer program instructions are stored, which, when executed by processor 210, implement any of the resume search methods described above.

[0090] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical applications, or technical improvements to the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A resume search method, characterized in that, include: Obtain the user's industry-specific search terms and search criteria; Based on the search terms and search criteria, the user's search intent can be identified; The system searches a pre-built resume database for resumes that match the search intent, selects target resumes as these, and then pushes and displays these target resumes. The resume database stores word segmentation indexes for multiple resumes, and each resume's word segmentation index is constructed based on a pre-built professional thesaurus for the specific industry.

2. The method according to claim 1, characterized in that, When constructing a specialized thesaurus for the specific industry, the following are included: Obtain literature and materials related to the specific industry; Vocabulary data for the specific industry is selected from the literature, and a word segmentation lexicon is constructed based on the vocabulary data; Semantic understanding is performed on the context in which the vocabulary data is located to obtain the synonyms of the vocabulary data, and a thesaurus is constructed based on the synonyms of the vocabulary data; The word segmentation lexicon and the synonym lexicon are combined to obtain the professional lexicon for the specific industry.

3. The method according to claim 1, characterized in that, When constructing a word segmentation index for resumes based on a professional thesaurus for the specific industry, the following is included: The resume is segmented into words based on the professional thesaurus of the specific industry, and the word segmentation result of the resume is obtained. The word segmentation results of the resume are matched with the relevant key fields defined in the pre-built word segmentation index pattern to obtain the word segmentation index of the resume.

4. The method according to claim 3, characterized in that, After obtaining the word segmentation index of the resume, the method further includes: compressing the word segmentation index of the resume and storing the compressed word segmentation index in the resume database.

5. The method according to claim 1, characterized in that, When identifying a user's search intent based on the search terms and search conditions, it is done based on a pre-built intent recognition model.

6. The method according to claim 1, characterized in that, When pushing and displaying the target resume, the following is also included: Obtain the activity level of the user to whom the target resume belongs; The target resumes are reordered based on the user's activity level; The rearranged results of the target resume will be pushed and displayed.

7. The method according to claim 1, characterized in that, When obtaining search terms for a specific industry input by the user, the process also includes optimizing the search terms.

8. A resume search device, characterized in that, include: The data input module is used to obtain the search terms and search criteria for a specific industry entered by the user; The intent recognition module is used to identify the user's search intent based on the search terms and search conditions; The resume search module is used to find resumes that match the search intent from a pre-built resume library as target resumes and push and display the target resumes. The resume database stores word segmentation indexes for multiple resumes, and each resume's word segmentation index is constructed based on a pre-built professional thesaurus for the specific industry.

9. A resume search device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method of any one of claims 1 to 7 when executing the executable instructions.

10. A non-volatile computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images