CN112380421A - Resume searching method and device, electronic equipment and computer storage medium - Google Patents
Resume searching method and device, electronic equipment and computer storage medium Download PDFInfo
- Publication number
- CN112380421A CN112380421A CN202011254848.9A CN202011254848A CN112380421A CN 112380421 A CN112380421 A CN 112380421A CN 202011254848 A CN202011254848 A CN 202011254848A CN 112380421 A CN112380421 A CN 112380421A
- Authority
- CN
- China
- Prior art keywords
- resume
- search
- entities
- target
- searching
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000011218 segmentation Effects 0.000 claims abstract description 61
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000003062 neural network model Methods 0.000 claims description 51
- 230000015654 memory Effects 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 12
- 230000002596 correlated effect Effects 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 description 16
- 238000010586 diagram Methods 0.000 description 12
- 230000000875 corresponding effect Effects 0.000 description 10
- 238000013527 convolutional neural network Methods 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000005034 decoration Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 102100021723 Arginase-1 Human genes 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000007115 recruitment Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9532—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
Abstract
The embodiment of the application provides a resume searching method and device, electronic equipment and a computer storage medium, and relates to the field of information searching. The method comprises the following steps: acquiring a search sentence, and performing word segmentation processing on the search sentence to obtain a word segmentation result; determining a search intention, entities in a search sentence and a relation between the entities according to the word segmentation result; and determining the resume according to the search intention, the entities in the search statement and the relationship among the entities. According to the embodiment of the application, the search sentences input by the user are analyzed to determine the search intentions, the entities and the relation among the entities, so that the intentions of the user can be directly and accurately identified, more complete and comprehensive data can be obtained by using knowledge graph matching, and the user is assisted to make better decisions.
Description
Technical Field
The present application relates to the field of information search technologies, and in particular, to a resume search method and apparatus, an electronic device, and a computer storage medium.
Background
In the internet era, people look for work through the internet, enterprises also look for required talents on the internet, generally look for the required talents from a talent database, but as the data of the talent database of the enterprise is accumulated more and more, the enterprise is suitable for screening out the talents and faces a great challenge.
The existing human resource searching method mainly comprises keyword matching search, wherein indexes are built for all candidates in a talent database according to different fields, and when a user searches, the corresponding index is selected and the corresponding keyword is input for searching.
The searching mode needs to maintain the dictionary at regular time, the maintenance cost is high, the consumed resources are more, and the searching result is single, so that the requirement of an enterprise cannot be met.
Disclosure of Invention
Embodiments of the present application provide a resume searching method, apparatus, electronic device, and computer storage medium that overcome the above problems or at least partially solve the above problems.
In a first aspect, a method for searching resumes is provided, and the method includes:
acquiring a search sentence, and performing word segmentation processing on the search sentence to obtain a word segmentation result;
determining a search intention, entities in a search sentence and a relation between the entities according to the word segmentation result;
and obtaining a search result from a talent library and a pre-constructed knowledge graph according to the search intention, the entities in the search sentence and the relationship among the entities, wherein the talent library is used for storing the resume, and the knowledge graph is used for storing the entities in the resume and the relationship among the entities.
In one possible implementation, determining the search intention, the entities and the relationship among the entities according to the word segmentation result comprises:
inputting the word segmentation result into a pre-trained intention recognition neural network model to obtain a search intention output by the intention recognition neural network model;
inputting the word segmentation result into a pre-trained entity recognition neural network model to obtain an entity in the output search sentence of the entity recognition neural network model;
and inputting the entities into a preset semantic relation model to obtain the relation between the entities output by the preset semantic relation model.
In another possible implementation manner, obtaining search results from the talent base and the pre-constructed knowledge graph according to the search intention, the entities in the search sentence and the relationship among the entities comprises:
searching the resume containing the entities in the search sentence in a preset talent library, and taking the resume as a first target resume;
determining a first target entity in the first target resume, wherein the first target entity and the entity meet the relationship between the entities;
searching an entity with the same semantic as the first target entity in a preset knowledge graph as a second target entity according to the searching intention;
and searching the resume containing the second target entity in the talent library to serve as a second target resume, and taking the second target resume as a search result.
In another possible implementation manner, searching the talent base for a resume containing a second target entity as a second target resume, using the second target resume as a search result, and then ranking the search result, including:
the score of the second target resume is determined according to the time difference between the last update of the second target resume and the current moment and the similarity between the search statement and the second target resume;
sorting the second target resumes according to the sequence of scores from large to small, and returning the sorted search results;
the score is negatively correlated with the time difference between the last update of the second target resume and the current moment, and positively correlated with the similarity between the search statement and the second target resume.
In another possible implementation manner, the talent base is searched for a resume containing a second target entity as a second target resume, the second target resume is used as a search result, and then the method further includes a step of expanding the search result, including:
searching attribute information of the entity contained in the second target resume in the knowledge graph;
adding the attribute information into a second target resume to obtain an expanded search result; the attribute information of the entity refers to the characteristics of the entity.
In a second aspect, an apparatus for searching for a resume is provided, the apparatus comprising:
the acquisition module is used for acquiring a search sentence and performing word segmentation processing on the search sentence to obtain a word segmentation result;
the analysis module is used for determining the search intention, the entities in the search sentence and the relationship among the entities according to the word segmentation result;
and the searching module is used for obtaining a searching result from the talent base and a pre-constructed knowledge graph according to the searching intention, the entities in the searching sentence and the relation among the entities, wherein the talent base is used for storing the resume, and the knowledge graph is used for storing the entities in the resume and the relation among the entities.
Further, the analysis module includes:
the intention analysis module is used for inputting the word segmentation result into a pre-trained intention recognition neural network model to obtain a search intention output by the intention recognition neural network model;
the entity analysis module is used for inputting the word segmentation result into a pre-trained entity recognition neural network model to obtain an entity in a search sentence output by the entity recognition neural network model;
and the relationship analysis module is used for inputting the entities into the preset semantic relationship model and obtaining the relationship between the entities output by the preset semantic relationship model.
Further, the search module includes:
the first resume searching module searches resumes containing entities in the search sentences in a preset talent library and takes the resumes as a first target resume;
the first entity searching module is used for determining a first target entity in the first target resume, and the first target entity and the entity meet the relationship between the entities;
the second entity searching module is used for searching an entity with the same semantic as the first target entity in a preset knowledge graph as a second target entity according to the searching intention;
and the second resume searching module is used for searching the resume containing the second target entity in the talent library to be used as the second target resume, and the second target resume is used as a searching result.
Further, the second resume search module includes:
the ranking module is used for determining the score of the second target resume according to the time difference between the last update of the second target resume and the current moment and the similarity between the search statement and the second target resume;
sorting the second target resumes according to the sequence of scores from large to small, and returning the sorted search results;
the score is negatively correlated with the time difference between the last update of the second target resume and the current moment, and positively correlated with the similarity between the search statement and the second target resume.
Further, the second resume search module further comprises:
the expansion module is used for searching attribute information of the entity contained in the second target resume in the knowledge graph;
adding the attribute information into a second target resume, and returning the expanded search result; the attribute information of the entity refers to the characteristics of the entity.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the method provided in the first aspect are implemented.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of the method as provided in the first aspect.
According to the resume searching method and device, the electronic device and the storage medium, the searching intention, the entities and the relation among the entities are determined by analyzing the searching sentences input by the user, the intention of the user can be directly and accurately identified, more complete and comprehensive data can be obtained by using knowledge graph matching, and the user is assisted in making better decisions.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the description of the embodiments of the present application will be briefly described below.
Fig. 1 is a network architecture diagram illustrating operation of a resume search system according to an embodiment of the present disclosure;
fig. 2 is a flowchart of a resume searching method according to an embodiment of the present application;
FIG. 3 is a schematic diagram of an intent-to-recognize neural network model provided by an embodiment of the present application;
FIG. 4 is a schematic diagram of an entity recognition neural network model provided by an embodiment of the present application;
FIG. 5 is a flowchart for obtaining search results according to an embodiment of the present application;
FIG. 6 is a schematic diagram of a resume search process according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a resume searching apparatus according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to the embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary only for the purpose of explaining the present application and are not to be construed as limiting the present invention.
As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. As used herein, the term "and/or" includes all or any element and all combinations of one or more of the associated listed items.
The resume searching method provided by the application aims to solve the technical problems in the prior art.
To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
The following describes the technical solutions of the present application and how to solve the above technical problems with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
First, the application can be applied to various resume search scenes, for example, a scene of HR (human resources) of an enterprise when searching for job seekers meeting the requirements of the enterprise, a resume is a brief introduction sent by a job seeker to a recruitment unit, and contains various basic information, such as name, gender, age, name family, native place, political aspect, academic history, contact information and the like, and information of self-evaluation, work experience, learning experience, glory and achievement, job hunting desire, understanding of the work and the like. The resume is generally stored in the talent library, and when the resume needs to be searched, the user can go to the talent library to search for the corresponding resume.
It should be understood that the method for searching for a resume provided by the present application may be applied to any computer device or system having resume search, specifically, the system for searching for a resume may operate in the network architecture diagram as shown in fig. 1, where fig. 1 is a network architecture diagram in which the system for searching for a resume provided by the embodiment of the present application operates, where the server 11 performs corresponding processing and search by receiving a search statement uploaded by the terminal 12, and sends an obtained resume back to the terminal 12.
In order to explain the technical scheme of the application more clearly, the application is explained in the search scene of the resume in the following, the existing search method of the resume generally directly searches in a talent library, indexes such as inverted indexes are established in the existing talent library, keyword matching is carried out in the talent library, and the corresponding resume is searched according to the corresponding indexes, so that the search mode is single, the searched resumes are single and small in quantity, the maintenance cost of the talent library is high, the consumed resources are large, the requirements of enterprises cannot be completely met, search sentences input by users cannot be completely matched with the corresponding indexes in some cases when searching the resumes, the resumes which the users want to search cannot be completely understood, namely the intention of the users cannot be clearly determined, and the decision of the users cannot be assisted.
In an embodiment of the present application, a method for searching a resume is provided, as shown in fig. 2, where fig. 2 is a flowchart of a method for searching a resume provided in an embodiment of the present application, and the method includes:
s101, obtaining a search sentence, and performing word segmentation processing on the search sentence to obtain a word segmentation result.
The search sentence is input by a user and can be input on a terminal such as a computer, a mobile phone and the like, and the search sentence can be characters and numbers, such as 'Zhang three graduation school'; the mobile phone number is 12345678. The word segmentation processing means that a search sentence is simply segmented to obtain a word segmentation result, for example, the word segmentation processing is performed on the graduation school of Zhang III to obtain Zhang III and graduation school.
After the search sentence is obtained, the search sentence is subjected to word segmentation to obtain a word segmentation result, it should be understood that the search sentence input by the user may not be standard and meets the search requirement, and the search sentence input by the user needs to be subjected to certain processing before the subsequent search step is performed.
And S102, determining the search intention, entities in the search sentence and the relation among the entities according to the word segmentation result.
The search intention is the content that the user wants to search and is expressed by the search sentence input by the user in the thinking of the user, and can be obtained by utilizing a pre-trained intention recognition neural network model, for example, the search sentence input by the user is 'Zhang three alumni', and the search intention is a person who graduates with Zhang three in the same school. An entity refers to something that exists objectively, for example, "zhang san" is an entity. The relationship between the entities refers to the relationship between the entities, and the relationship between the two entities, such as Zhang III and Beijing university, may be graduation.
After the word segmentation result is obtained, the search intention, the entities in the search sentence and the relation among the entities can be determined according to the search result, and the step can be obtained through a pre-trained neural network model, for example, the search intention can be obtained by utilizing the pre-trained intention to identify the neural network model; entities in the search sentences can be obtained by utilizing a pre-trained entity recognition neural network model; the relationship between the entities can be obtained by using a preset semantic relationship model.
S103, obtaining a search result from a talent library and a pre-constructed knowledge graph according to the search intention, the entities in the search sentence and the relationship among the entities, wherein the talent library is used for storing the resume, and the knowledge graph is used for storing the entities in the resume and the relationship among the entities.
After the search intention, the entities in the search sentence and the relationship among the entities are obtained, the embodiment of the application can search in a talent base which is established in advance, and determine the resume thought by the user by combining a knowledge graph, and a specific search method is described in the embodiment later in the application.
According to the resume searching method and device, the electronic device and the storage medium, the searching intention, the entities and the relation among the entities are determined by analyzing the searching sentences input by the user, the intention of the user can be directly and accurately identified, more complete and comprehensive data can be obtained by using knowledge graph matching, and the user is assisted in making better decisions.
The embodiment of the present application further provides a possible implementation manner, and determining a search intention, an entity, and a relationship between entities according to the word segmentation result includes:
and S1021, inputting the word segmentation result into a pre-trained intention recognition neural network model to obtain a search intention output by the intention recognition neural network model.
The embodiment of the application obtains the search intention by way of machine learning, specifically, the word segmentation result is input into a pre-trained intention recognition neural network model, and the search intention of the user output by the intention recognition neural network model can be obtained, wherein a SoftMax (flexible maximum value) activation function is used in an output layer of the intention recognition neural network model, the SoftMax activation function can map outputs of a plurality of neural networks to a probability in a (0,1) interval, and a result with the maximum probability is selected as the search intention, and a formula is shown as follows:
wherein, f (Z)i) Is the ith sample search intention ZiThe probability of (a) of (b) being,
is the ith sample search intention ZiIs a mathematical constant, is the base of the natural logarithm, also known as the euler number, and n is the total number of sample search intents.
It should be understood that, before performing step S1021, the intention-to-recognize neural network model may also be trained in advance, and may be obtained by the following method: firstly, collecting a certain number of sample search sentences, performing word segmentation processing to obtain word segmentation results of each sample search sentence, and then determining search intentions represented by the word segmentation results of each sample search sentence, wherein the search intentions can include names, birth dates, mobile phone numbers, schools, companies, positions, skills and the like, and can be used for searching resumes according to names or resumes according to mobile phone numbers. And randomly training based on the word segmentation result of the sample search sentence and the search intention represented by the word segmentation result of the sample search sentence so as to obtain an intention recognition neural network model. The initial model may be a single neural network model or a combination of multiple neural network models.
Fig. 3 is a schematic diagram of an intention recognition Neural network model provided in an embodiment of the present application, where w0, w1, w2, w3, and w4 are individual words in a segmentation result, and the intention recognition Neural network model in the embodiment of the present application may adopt a CNN + MLP Neural network model, which includes a CNN (Convolutional Neural network) Convolutional Neural network and an MLP (multi layer Perceptron) fully-connected Neural network.
The training method for the intention recognition neural network model comprises the following steps:
s201, initializing a CNN layer and an MLP layer;
s202, taking word segmentation results of a certain number of sample search sentences as training samples, taking search intents represented by the word segmentation results of the sample search sentences as sample labels, and inputting the training samples and the sample labels to a CNN layer to obtain feature vectors of the training samples output by the CNN layer;
s203, inputting the feature vector of the training sample into an MLP layer to obtain a search intention prediction result of the training sample output by the MLP layer;
and S204, calculating the deviation between the prediction result and the sample label corresponding to the training sample, and adjusting the parameters of the CNN layer and the MLP layer through reverse feedback adjustment until the deviation reaches a convergence condition to obtain the trained intention recognition neural network model.
And S1022, inputting the word segmentation result into a pre-trained entity recognition neural network model to obtain an entity in a search sentence output by the entity recognition neural network model.
The entity is obtained through a machine learning mode, specifically, the word segmentation result is input into a pre-trained entity recognition neural network model, and the entity output by the entity recognition neural network model can be obtained.
It should be understood that, before performing step S1022, the entity recognition neural network model may also be trained in advance, and specifically, obtained by the following method: firstly, a certain number of sample search sentences are collected, word segmentation processing is carried out to obtain word segmentation results of each sample search sentence, and then entities contained in the word segmentation results of each sample search sentence are determined, for example, the word segmentation results are 'Zhangsan' and 'Beijing university', the name entity Zhangsan and the school entity Beijing university can be obtained. And randomly training based on the word segmentation result of the sample search statement and the entity contained in the word segmentation result of the sample search statement, thereby obtaining the entity recognition neural network model. The initial model may be a single neural network model or a combination of multiple neural network models.
Fig. 4 is a schematic diagram of an entity recognition neural network model provided in the embodiment of the present application, where w0, w1, w2, w3, and w4 are words in the word segmentation result, and the entity recognition neural network model in the embodiment of the present application may adopt a BiLSTM + CRF neural network model, which includes a Bi-directional Long Short-Term Memory (Bi-directional Short-Term Memory) neural network and a CRF (conditional random field) neural network.
The training method of the entity recognition neural network model comprises the following steps:
s301, initializing a BilSTM layer and a CRF layer;
s302, taking the word segmentation results of a certain number of sample search sentences as training samples, taking entities contained in the word segmentation results of the sample search sentences as sample labels, and inputting the training samples and the sample labels into a BilSTM layer to obtain the classification probability of the training samples output by the BilSTM layer;
s303, inputting the classification probability of the training sample into a CRF layer to obtain an entity prediction result of the training sample output by the CRF layer;
s304, calculating the deviation of the prediction result and the sample label corresponding to the training sample, and adjusting parameters of a BilSTM layer and a CRF layer in a reverse feedback mode until the deviation reaches a convergence condition to obtain the trained entity recognition neural network model.
And S1023, inputting the entities into a preset semantic relation model, and obtaining the relation between the entities output by the preset semantic relation model.
Inputting an entity into a preset semantic relationship model, obtaining a relationship between entities output by the preset semantic relationship model, using an open source LTP (Language Technology Platform) model, where various semantic relationships are defined in the LTP model, and after inputting an entity into the LTP model, determining a relationship between the entity and the entity by using the defined semantic relationships, such as name entity zhang and school entity, to obtain a relationship between the entities as graduation, specifically, table 1 is a semantic role relationship definition table, as shown in table 1:
| type of relationship | Description of the invention | Examples of the invention |
| ARG0 | Main body | [ Zhangsan ARG0]Graduate to Beijing university |
| ARG1 | The person who is in charge | Zhang Sanbei university ARG1] |
| FEAT | Decoration | Zhang three (graduation from FEAT)]Peking University |
Table 1: semantic role relationship definition table
According to the embodiment of the application, the pre-trained intention recognition neural network model, the entity recognition neural network model and the preset semantic relation model are utilized, the search intention of the user expressed by the search statement, the included entities and the relation among the entities can be accurately recognized, and more accurate search can be carried out according to the requirements of the user.
Fig. 5 is a schematic flow chart of obtaining a search result provided in the embodiment of the present application, and obtaining the search result from a talent base and a pre-constructed knowledge graph according to a search intention, an entity and a relationship between the entities includes:
and S1031, searching the resume containing the entity in a preset talent library, and taking the resume as a first target resume.
The talent base is established in advance, the resume is stored in the talent base, the entity obtained according to the steps can be searched in the talent base to obtain the resume containing the entity, and the resume obtained by searching at the moment is taken as a first target resume, for example: if the search statement input by the user is 'Zhang III alumni', Zhang III, an entity can be obtained through the steps, and the first target resume, namely the resume with the name of Zhang III, is obtained through searching in the talent base.
S1032, determining a first target entity in the first target resume, wherein the first target entity and the entity meet the relationship among the entities.
After the first target resume is determined, an entity having a relationship with the entity may be determined as a first target entity according to the relationship between the entities obtained in the above steps, for example, a search sentence input by a user is "zhang san schoolmate", and after the relationship between the entities is obtained as graduation, the first target entity may be determined as "beijing university".
And S1033, searching an entity with the same semantic meaning as that of the first target entity in a preset knowledge graph according to the searching intention to serve as a second target entity.
After the first entity is obtained, the entity having the same actual meaning as the first target entity can be obtained by searching in a preset knowledge graph according to the search intention of the user, and the second target entity is obtained, it should be understood that the knowledge graph is established in advance, the knowledge graph stores various entities, relationships among the entities and attributes of the entities, the attributes of the entities refer to various characteristics of the entities, the content written in the resume may not be all the attributes represented by the entities, and the second target entity can be determined according to the attributes of the entities stored in the knowledge graph, such as the school entity "beijing university", which is called "the school of academic garden occupational technology" and "beida", respectively.
S1034, searching the talent library for the resume containing the second target entity as a second target resume, and taking the second target resume as a search result.
After the second target entity is obtained, the search may be continued in the predetermined talent base to obtain all resumes including the second target entity as the second target resume, where the second target resume is the resume that the user wants to search for, for example, all resumes including "beijing university", "beida", "clengyun academy of professional technology" are searched as the second target resume.
According to the embodiment of the application, the knowledge graph is utilized to carry out more extensive and accurate search according to the search intention expressed by the search sentence input by the user, the entity and the relation among the entities, and all the requirements of the user for searching are met.
A possible implementation manner is also provided in the embodiment of the present application, as shown in fig. 6, fig. 6 is a schematic diagram of a resume search process provided in the embodiment of the present application:
the server F searches the resume containing the entity in the search statement in the talent bank R, obtains the resume containing the entity in the search statement in the talent bank R and takes the resume as a first target resume; then the server F takes the entity which satisfies the relation between the entities in the search statement in the first target resume as the first target entity.
And the server F searches in the knowledge graph T according to the search intention of the user to obtain an entity with the same semantic as the first target entity in the knowledge graph T as a second target entity.
Z represents a terminal, the server F searches in the talent library R to obtain the resume containing the second target entity in the talent library R as a search result, and the resume is returned to the terminal Z.
According to the embodiment of the application, the knowledge graph and the talent library are organically combined for searching, so that the problem that accurate searching is difficult to perform due to the fact that a large number of resumes are stored in the talent library is solved, and the searching result is wide and accurate.
The embodiment of the present application further provides a possible implementation manner, the talent base is searched for a resume containing a second target entity, the resume is used as a second target resume, the second target resume is used as a search result, and then the method further includes a step of ranking the search result, including:
the score of the second target resume is determined according to the time difference between the last update of the second target resume and the current moment and the similarity between the search statement and the second target resume;
sorting the second target resumes according to the sequence of scores from large to small, and returning the sorted search results;
the score is negatively correlated with the time difference between the last update of the second target resume and the current moment, and positively correlated with the similarity between the search statement and the second target resume.
The talent base stores a large number of resumes, and according to the time lapse, the experience of the job seeker can continuously update the content of the resumes, so that the difference between the last update time and the current search time can be used as a judgment standard to allow the user to select the required resumes, the similarity between the search sentence of the user and the resume content can also be a condition closely related to the requirement of the user, and the smaller the difference between the last update time and the current search time, the greater the similarity between the search sentence of the user and the resume content, the higher the score, the algorithm formula shown below can calculate the score of the resumes under the two conditions:
formula (1):
score＝bm25+0.5*e-0.05*t
formula (2):
equation (1) is a calculation equation of score, where bm25 is the similarity between the search statement and the resume, t is the difference between the last update time of the resume and the current search time, and e is a natural constant in mathematics, also called euler number.
Equation (2) is a calculation equation of the similarity bm25 between the search sentence and the resume, where riIs the number of related resumes containing the word segmentation i, R is the total number of related resumes containing the word segmentation i, niIs the total number of resumes, N is the total number of resumes, K, k1、k2Is a constant number fiIs the frequency of occurrence of the participle i in the resume, qfiIs the frequency with which the participle i appears in the user's search sentence.
According to the embodiment of the application, the resumes are sequenced according to the sequence from large to small through two judgment conditions, namely the difference between the last updated time of the resumes and the current searching time and the similarity between the searching sentences and the resumes, so that a user can have a very visual view angle on the searched resumes, and the requirements of the user can be better met.
The embodiment of the present application further provides a possible implementation manner, the talent base is searched for a resume containing a second target entity, the resume is used as a second target resume, the second target resume is used as a search result, and then the method further includes a step of expanding the search result, including:
searching attribute information of the entity contained in the second target resume in the knowledge graph;
adding the attribute information to a second target resume; the attribute information of the entity refers to the characteristics of the entity.
After the user searches for the resume, whether the job seeker with the resume meets the requirements of the user may be judged according to the attribute information of one or some entities in the resume, or comparison may be performed to allow the user to select the job seeker, and the attribute information of the entities, that is, the characteristics of the entities, is stored in the knowledge graph, and the attribute information of the entities in the knowledge graph may be added to the resume to allow the user to select the entity.
For example, if a user wants to search for a resume with graduation in Beijing university or Qinghua university, the attribute information of the entity Beijing university and the attribute information of the entity Qinghua university stored in the knowledge map may be added to the resume for comparison, such as the direction of the strong term specialty of the Beijing university, teaching resources, etc., the direction of the strong term specialty of the Qinghua university, teaching resources, etc.
According to the embodiment of the application, the attribute information of the entity in the knowledge graph is added into the resume, so that the user can fully know the attribute information of different entities in the resumes of different job seekers, and the user can make a better decision according to the requirement to select the job seekers more suitable for user enterprises.
An embodiment of the present application provides a resume searching apparatus, as shown in fig. 7, fig. 7 is a schematic diagram of the resume searching apparatus provided in the embodiment of the present application, and the apparatus may include: an acquisition module 101, an analysis module 102, and a search module 103, wherein,
an obtaining module 101, configured to obtain a search statement, perform word segmentation processing on the search statement, and obtain a word segmentation result;
the analysis module 102 is configured to determine a search intention, entities in a search sentence, and relationships between the entities according to the word segmentation result;
and the searching module 103 is used for obtaining a searching result from a talent base and a pre-constructed knowledge graph according to the searching intention, the entities in the searching sentence and the relationship among the entities, wherein the talent base is used for storing the resume, and the knowledge graph is used for storing the entities in the resume and the relationship among the entities.
The apparatus for searching for a resume provided in this embodiment of the application specifically executes the process of the method embodiment, and please refer to the content of the method embodiment for searching for a resume in detail, which is not described herein again. According to the resume searching device provided by the embodiment of the application, the searching intention, the entities and the relation among the entities are determined by analyzing the searching sentences input by the user, the intention of the user can be directly and accurately identified, more complete and comprehensive data can be obtained by utilizing knowledge graph matching, and the user is assisted to make better decisions.
Further, the analysis module 102 includes:
the intention analysis module is used for inputting the word segmentation result into a pre-trained intention recognition neural network model to obtain a search intention output by the intention recognition neural network model;
the entity analysis module is used for inputting the word segmentation result into a pre-trained entity recognition neural network model to obtain an entity in a search sentence output by the entity recognition neural network model;
and the relationship analysis module is used for inputting the entities into the preset semantic relationship model and obtaining the relationship between the entities output by the preset semantic relationship model.
Further, the search module 103 includes:
the first resume searching module searches resumes containing entities in the search sentences in a preset talent library and takes the resumes as a first target resume;
the first entity searching module is used for determining a first target entity in the first target resume, and the first target entity and the entity meet the relationship between the entities;
the second entity searching module is used for searching an entity with the same semantic as the first target entity in a preset knowledge graph as a second target entity according to the searching intention;
and the second resume searching module is used for searching the resume containing the second target entity in the talent library to be used as the second target resume, and the second target resume is used as a searching result.
Further, the second resume search module includes:
the ranking module is used for determining the score of the second target resume according to the time difference between the last update of the second target resume and the current moment and the similarity between the search statement and the second target resume;
sorting the second target resumes according to the sequence of scores from large to small, and returning the sorted search results;
the score is negatively correlated with the time difference between the last update of the second target resume and the current moment, and positively correlated with the similarity between the search statement and the second target resume.
Further, the second resume search module further comprises:
the expansion module is used for searching attribute information of the entity contained in the second target resume in the knowledge graph;
adding the attribute information into a second target resume, and returning the expanded search result; the attribute information of the entity refers to the characteristics of the entity.
An embodiment of the present application provides an electronic device, including: a memory and a processor; at least one program stored in the memory for execution by the processor, which when executed by the processor, implements: the method has the advantages that search sentences input by the user are analyzed to determine the search intentions, the entities and the relation among the entities, so that the intentions of the user can be identified more directly and accurately, more complete and comprehensive data can be obtained by using knowledge graph matching, and the user is assisted to make better decisions.
In an alternative embodiment, an electronic device is provided, as shown in fig. 8, the electronic device 4000 shown in fig. 8 comprising: a processor 4001 and a memory 4003. Processor 4001 is coupled to memory 4003, such as via bus 4002. Optionally, the electronic device 4000 may further comprise a transceiver 4004. In addition, the transceiver 4004 is not limited to one in practical applications, and the structure of the electronic device 4000 is not limited to the embodiment of the present application.
The Processor 4001 may be a CPU (Central Processing Unit), a general-purpose Processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other Programmable logic device, a transistor logic device, a hardware component, or any combination thereof. Which may implement or perform the various illustrative logical blocks, modules, and circuits described in connection with the disclosure. The processor 4001 may also be a combination that performs a computational function, including, for example, a combination of one or more microprocessors, a combination of a DSP and a microprocessor, or the like.
The Memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical Disc storage, optical Disc storage (including Compact Disc, laser Disc, optical Disc, digital versatile Disc, blu-ray Disc, etc.), a magnetic Disc storage medium or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to these.
The memory 4003 is used for storing application codes for executing the scheme of the present application, and the execution is controlled by the processor 4001. Processor 4001 is configured to execute application code stored in memory 4003 to implement what is shown in the foregoing method embodiments.
The present application provides a computer-readable storage medium, on which a computer program is stored, which, when running on a computer, enables the computer to execute the corresponding content in the foregoing method embodiments. Compared with the prior art, the method has the advantages that the search intention, the entities and the relation among the entities are determined by analyzing the search sentences input by the user, the intention of the user can be directly and accurately identified, more complete and comprehensive data can be obtained by utilizing knowledge graph matching, and the user is assisted to make better decisions.
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and may be performed in other orders unless explicitly stated herein. Moreover, at least a portion of the steps in the flow chart of the figure may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
The foregoing is only a partial embodiment of the present application, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present application, and these modifications and decorations should also be regarded as the protection scope of the present application.
Claims (8)
1. A resume searching method is characterized by comprising the following steps:
acquiring a search sentence, and performing word segmentation processing on the search sentence to obtain a word segmentation result;
determining a search intention, entities in a search sentence and a relation among the entities according to the word segmentation result;
and obtaining a search result from a talent library and a pre-constructed knowledge graph according to the search intention, the entities in the search sentence and the relationship among the entities, wherein the talent library is used for storing the resume, and the knowledge graph is used for storing the entities in the resume and the relationship among the entities.
2. The method for searching resumes according to claim 1, wherein the determining of the search intention, the entities in the search sentence and the relationship among the entities according to the word segmentation result comprises:
inputting the word segmentation result into a pre-trained intention recognition neural network model to obtain a search intention output by the intention recognition neural network model;
inputting the word segmentation result into a pre-trained entity recognition neural network model to obtain an entity in a search sentence output by the entity recognition neural network model;
and inputting the entities into a preset semantic relation model to obtain the relation between the entities output by the preset semantic relation model.
3. The resume searching method of claim 1, wherein the obtaining of the search result from the talent base and the pre-constructed knowledge graph according to the search intention, the entities in the search sentence and the relationship between the entities comprises:
searching the resume containing the entity in the search sentence in a preset talent library, and taking the resume as a first target resume;
determining a first target entity in the first target resume, wherein the first target entity and the entity meet the relationship between the entities;
searching an entity with the same semantic as the first target entity in a preset knowledge graph as a second target entity according to the searching intention;
and searching the talent base for the resume containing the second target entity to serve as a second target resume, and taking the second target resume as a search result.
4. The resume searching method of claim 3, wherein the step of searching the talent bank for the resume containing the second target entity as the second target resume and using the second target resume as the search result further comprises the step of ranking the search result, comprising:
the score of the second target resume is determined according to the time difference between the last update of the second target resume and the current moment and the similarity between the search statement and the second target resume;
sorting the second target resumes according to the sequence of scores from large to small, and returning the sorted search results;
the score is negatively correlated with the time difference between the last update of the second target resume and the current moment, and positively correlated with the similarity between the search statement and the second target resume.
5. The resume searching method of claim 3, wherein the step of searching the talent bank for the resume containing the second target entity as the second target resume, and using the second target resume as the search result, and then expanding the search result further comprises:
searching the knowledge graph for attribute information of entities contained in the second target resume;
adding the attribute information to the second target resume to obtain an expanded search result; the attribute information of the entity refers to the characteristics of the entity.
6. A resume searching apparatus, comprising:
the acquisition module is used for acquiring a search sentence and performing word segmentation processing on the search sentence to obtain a word segmentation result;
the analysis module is used for determining a search intention, entities in a search sentence and a relation among the entities according to the word segmentation result;
and the searching module is used for obtaining a searching result from a talent base and a pre-constructed knowledge graph according to the searching intention, the entities in the searching sentence and the relation among the entities.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method for searching for a resume of any of claims 1 to 5 are implemented by the processor when executing the program.
8. A computer-readable storage medium, characterized in that it stores computer instructions that cause the computer to perform the steps of the resume searching method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011254848.9A CN112380421A (en) | 2020-11-11 | 2020-11-11 | Resume searching method and device, electronic equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011254848.9A CN112380421A (en) | 2020-11-11 | 2020-11-11 | Resume searching method and device, electronic equipment and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112380421A true CN112380421A (en) | 2021-02-19 |
Family
ID=74582624
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011254848.9A Pending CN112380421A (en) | 2020-11-11 | 2020-11-11 | Resume searching method and device, electronic equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112380421A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113434687A (en) * | 2021-07-22 | 2021-09-24 | 高向咨询(深圳)有限公司 | Automatic resume finding method, automatic recruitment system and computer storage medium |
| CN113742455A (en) * | 2021-09-09 | 2021-12-03 | 平安科技（深圳）有限公司 | Resume searching method, device and equipment based on artificial intelligence and storage medium |
| CN113761206A (en) * | 2021-09-10 | 2021-12-07 | 平安科技（深圳）有限公司 | Intelligent information query method, device, equipment and medium based on intention recognition |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017024884A1 (en) * | 2015-08-07 | 2017-02-16 | 广州神马移动信息科技有限公司 | Search intention identification method and device |
| US20190065507A1 (en) * | 2017-08-22 | 2019-02-28 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for information processing |
| CN110516260A (en) * | 2019-08-30 | 2019-11-29 | 腾讯科技（深圳）有限公司 | Entity recommended method, device, storage medium and equipment |
-
2020
- 2020-11-11 CN CN202011254848.9A patent/CN112380421A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017024884A1 (en) * | 2015-08-07 | 2017-02-16 | 广州神马移动信息科技有限公司 | Search intention identification method and device |
| US20190065507A1 (en) * | 2017-08-22 | 2019-02-28 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for information processing |
| CN110516260A (en) * | 2019-08-30 | 2019-11-29 | 腾讯科技（深圳）有限公司 | Entity recommended method, device, storage medium and equipment |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113434687A (en) * | 2021-07-22 | 2021-09-24 | 高向咨询(深圳)有限公司 | Automatic resume finding method, automatic recruitment system and computer storage medium |
| CN113742455A (en) * | 2021-09-09 | 2021-12-03 | 平安科技（深圳）有限公司 | Resume searching method, device and equipment based on artificial intelligence and storage medium |
| CN113742455B (en) * | 2021-09-09 | 2023-11-10 | 平安科技（深圳）有限公司 | Resume searching method, device, equipment and storage medium based on artificial intelligence |
| CN113761206A (en) * | 2021-09-10 | 2021-12-07 | 平安科技（深圳）有限公司 | Intelligent information query method, device, equipment and medium based on intention recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112632385B (en) | Course recommendation method, course recommendation device, computer equipment and medium | |
| CN110147551B (en) | Multi-category entity recognition model training, entity recognition method, server and terminal | |
| CN110222160B (en) | Intelligent semantic document recommendation method and device and computer readable storage medium | |
| CN106649818B (en) | Application search intention identification method and device, application search method and server | |
| WO2020177282A1 (en) | Machine dialogue method and apparatus, computer device, and storage medium | |
| CN109376222B (en) | Question-answer matching degree calculation method, question-answer automatic matching method and device | |
| CN111753060A (en) | Information retrieval method, device, equipment and computer readable storage medium | |
| CN111858859A (en) | Automatic question-answering processing method, device, computer equipment and storage medium | |
| CN107609185B (en) | Method, device, equipment and computer-readable storage medium for similarity calculation of POI | |
| CN112380421A (en) | Resume searching method and device, electronic equipment and computer storage medium | |
| CN111444320A (en) | Text retrieval method and device, computer equipment and storage medium | |
| CN108304373B (en) | Semantic dictionary construction method and device, storage medium and electronic device | |
| US20170103337A1 (en) | System and method to discover meaningful paths from linked open data | |
| CN110795527B (en) | Candidate entity ordering method, training method and related device | |
| CN110929524A (en) | Data screening method, device, equipment and computer readable storage medium | |
| CN111221936B (en) | Information matching method and device, electronic equipment and storage medium | |
| CN112084435A (en) | Search ranking model training method and device and search ranking method and device | |
| CN113761868B (en) | Text processing method, text processing device, electronic equipment and readable storage medium | |
| CN114036322A (en) | Training method for search system, electronic device, and storage medium | |
| CN111291187B (en) | Emotion analysis method and device, electronic equipment and storage medium | |
| CN112463944A (en) | Retrieval type intelligent question-answering method and device based on multi-model fusion | |
| CN113987161A (en) | Text sorting method and device | |
| CN113569118A (en) | Self-media pushing method and device, computer equipment and storage medium | |
| CN112508177A (en) | Network structure searching method and device, electronic equipment and storage medium | |
| CN113569018A (en) | Question and answer pair mining method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |