CN111461637A - Resume screening method and device, computer equipment and storage medium - Google Patents
Resume screening method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN111461637A CN111461637A CN202010128423.7A CN202010128423A CN111461637A CN 111461637 A CN111461637 A CN 111461637A CN 202010128423 A CN202010128423 A CN 202010128423A CN 111461637 A CN111461637 A CN 111461637A
- Authority
- CN
- China
- Prior art keywords
- resume
- text
- target
- detected
- similarity
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
-
- 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
-
- 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
- G06F16/35—Clustering; Classification
Abstract
The application relates to artificial intelligence and provides a resume screening method, a resume screening device, computer equipment and a storage medium. The method comprises the following steps: acquiring a plurality of resume texts to be detected; performing entity identification on each resume text to be detected to obtain resume keywords, and obtaining a resume keyword vector according to the resume keywords; inputting the resume keyword vectors into the established resume screening model to obtain the passing probability of each resume text to be detected, and determining at least one first target resume text according to the passing probability; acquiring a job description text, extracting job keywords in the job description text, and obtaining a job keyword vector according to the job keywords; acquiring a resume keyword vector corresponding to each first target resume text, and calculating text similarity according to the resume keyword vector and the position keyword vector; and determining a second target resume text according to the text similarity. By adopting the method, the accuracy of resume screening detection can be improved.
Description
Technical Field
The present application relates to the field of internet technologies, and in particular, to a resume screening method and apparatus, a computer device, and a storage medium.
Background
With the development of internet technology, job seekers usually post resumes through a recruitment website to seek work. Enterprises usually issue position related information on a recruitment website to obtain resume delivered by job seekers, and often a great amount of resume is delivered to one position. At present, enterprises filter and detect the resumes through a server according to preset rules, and filter a large number of unqualified resumes, so that the efficiency of resume screening is improved.
However, the server performs screening detection through a preset rule, and the accuracy of resume screening detection is low, which results in screen missing or screen error.
Disclosure of Invention
In view of the above, it is necessary to provide a resume screening method, apparatus, computer device and storage medium capable of improving accuracy of resume screening detection.
A resume screening method, the method comprising:
receiving a resume screening instruction, and acquiring a plurality of to-be-detected resume texts according to the resume screening instruction;
performing entity identification on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and vectorizing the resume keyword corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected;
inputting the resume keyword vector corresponding to each resume text to be detected into a pre-established resume classification model to obtain the passing probability corresponding to each resume text to be detected, and determining at least one first target resume text according to the passing probability;
acquiring a job description text, extracting job keywords in the job description text, and vectorizing the job keywords to obtain a job keyword vector;
acquiring a resume keyword vector corresponding to each first target resume text, and calculating the similarity of each text of each first target resume text and the job description text according to the resume keyword vector and the job keyword vector;
and determining a second target resume text from the first target resume text according to the text similarity.
In one embodiment, the entity recognition is performed on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and the method includes:
acquiring a basic field and a professional field in a resume text to be detected;
extracting basic keywords corresponding to the basic fields;
and inputting the professional field into the trained entity recognition model to obtain professional keywords corresponding to the professional field, and obtaining resume keywords corresponding to the resume text to be detected according to the basic keywords and the professional keywords.
In one embodiment, vectorizing the resume keywords corresponding to each resume text to be detected to obtain the resume keyword vector corresponding to each resume text to be detected includes:
inputting the resume keywords corresponding to each resume text to be detected into the trained neural network language model to obtain the resume keyword vector corresponding to each resume text to be detected, wherein the trained neural network language model is obtained by training through a deep neural network according to the existing corpus.
In one embodiment, calculating the respective text similarity of each first target resume text and the job description text according to the resume keyword vector and the job keyword vector comprises:
acquiring the number of resume keywords corresponding to the first target resume text and the number of position keywords corresponding to the position description text;
calculating the vector similarity between the resume keyword vector corresponding to the first target resume text and the position keyword vector corresponding to the position description text;
determining the similarity of the resume words of the resume keywords corresponding to the first target resume text from the vector similarity, and obtaining the first target similarity according to the similarity of the resume words of the resume keywords and the number of the resume keywords;
determining the job word similarity of a job keyword vector corresponding to the job description text from the vector similarity, and obtaining a second target similarity according to the job word similarity of the job keyword vector and the number of the job keywords;
and determining the text similarity between the first target resume text and the job description text according to the first target similarity and the second target similarity.
In one embodiment, determining each second target resume text from the first target resume text according to each text similarity includes:
and sequencing the text similarity to obtain a sequencing result, and selecting a preset number of first target resume texts according to the sequencing result to obtain a second target resume text.
In one embodiment, after determining the second target resume text from the first target resume text according to the text similarity, the method further includes:
acquiring a resume keyword vector corresponding to each second target resume text, and clustering the resume keyword vectors corresponding to the second target resume texts to obtain clustering results;
and determining a third target resume text from the second target resume text according to the clustering result.
In one embodiment, the method further comprises:
determining first text information corresponding to each resume text to be detected according to the passing probability;
calculating target similarity of the resume keyword vector corresponding to each resume text to be detected and the position keyword vector, and determining second text information corresponding to each resume text to be detected according to the target similarity;
acquiring a preset weight, and determining target text information corresponding to each resume text to be detected according to the preset weight, the first text information and the second text information;
and determining the resume texts to be detected which pass the screening according to the target text information.
A resume screening apparatus, the apparatus comprising:
the text acquisition module is used for receiving the resume screening instruction and acquiring a plurality of to-be-detected resume texts according to the resume screening instruction;
the resume vector obtaining module is used for performing entity recognition on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, vectorizing the resume keyword corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected;
the first text determination module is used for inputting the resume keyword vector corresponding to each resume text to be detected into a pre-established resume classification model to obtain the passing probability corresponding to each resume text to be detected, and determining at least one first target resume text according to the passing probability;
the job vector obtaining module is used for obtaining a job description text, extracting job keywords in the job description text, and vectorizing the job keywords to obtain a job keyword vector;
the text similarity calculation module is used for acquiring a resume keyword vector corresponding to each first target resume text and calculating the text similarity between each first target resume text and the job description text according to the resume keyword vector and the job keyword vector;
and the second text determination module is used for determining a second target resume text from the first target resume text according to the text similarity.
A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
receiving a resume screening instruction, and acquiring a plurality of to-be-detected resume texts according to the resume screening instruction;
performing entity identification on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and vectorizing the resume keyword corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected;
inputting the resume keyword vector corresponding to each resume text to be detected into a pre-established resume classification model to obtain the passing probability corresponding to each resume text to be detected, and determining at least one first target resume text according to the passing probability;
acquiring a job description text, extracting job keywords in the job description text, and vectorizing the job keywords to obtain a job keyword vector;
acquiring a resume keyword vector corresponding to each first target resume text, and calculating the similarity of each text of each first target resume text and the job description text according to the resume keyword vector and the job keyword vector;
and determining a second target resume text from the first target resume text according to the text similarity.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
receiving a resume screening instruction, and acquiring a plurality of to-be-detected resume texts according to the resume screening instruction;
performing entity identification on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and vectorizing the resume keyword corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected;
inputting the resume keyword vector corresponding to each resume text to be detected into a pre-established resume classification model to obtain the passing probability corresponding to each resume text to be detected, and determining at least one first target resume text according to the passing probability;
acquiring a job description text, extracting job keywords in the job description text, and vectorizing the job keywords to obtain a job keyword vector;
acquiring a resume keyword vector corresponding to each first target resume text, and calculating the similarity of each text of each first target resume text and the job description text according to the resume keyword vector and the job keyword vector;
and determining a second target resume text from the first target resume text according to the text similarity.
According to the resume screening method, the resume screening device, the computer equipment and the storage medium, the plurality of resume texts to be detected are obtained, each resume text to be detected is firstly subjected to first screening detection through the pre-established resume classification model, and at least one first target resume text is determined from the plurality of resume texts to be detected. And then acquiring the job description text, determining a second target resume text from the first target resume text according to the similarity by calculating the similarity between the job description text and the first target resume text, and taking the second target resume text as the resume text passing the screening, so that the accuracy of screening and detecting the resume text is improved.
Drawings
FIG. 1 is a diagram illustrating an exemplary scenario for resume screening;
FIG. 2 is a schematic flow chart diagram illustrating a resume screening method in one embodiment;
FIG. 3 is a flowchart illustrating an embodiment of obtaining resume keywords;
FIG. 4 is a flow diagram that illustrates the determination of text similarity, according to one embodiment;
FIG. 5 is a flow diagram illustrating the determination of a third target resume text in one embodiment;
FIG. 6 is a schematic diagram of an embodiment of screening test based on text information;
FIG. 7 is a diagram illustrating a resume screening method in one embodiment;
FIG. 8 is a block diagram of a resume screening apparatus in one embodiment;
FIG. 9 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The resume screening method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network.
Receiving a resume screening instruction, and acquiring a plurality of to-be-detected resume texts according to the resume screening instruction;
performing entity identification on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and vectorizing the resume keyword corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected;
inputting the resume keyword vector corresponding to each resume text to be detected into a pre-established resume classification model to obtain the passing probability corresponding to each resume text to be detected, and determining at least one first target resume text according to the passing probability;
acquiring a job description text, extracting job keywords in the job description text, and vectorizing the job keywords to obtain a job keyword vector;
acquiring a resume keyword vector corresponding to each first target resume text, and calculating the similarity of each text of each first target resume text and the job description text according to the resume keyword vector and the job keyword vector;
and determining a second target resume text from the first target resume text according to the text similarity. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, a resume screening method is provided, which is described by taking the application of the method to the server in fig. 1 as an example, and includes the following steps:
s202, receiving a resume screening instruction, and acquiring a plurality of to-be-detected resume texts according to the resume screening instruction;
and S204, performing entity identification on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and vectorizing the resume keyword corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected.
The resume texts to be detected are obtained by vectorizing the resume keywords, and the resume keywords can be trained by using an existing neural network language model (NN L M, Neturularguagek L).
Specifically, the enterprise management terminal sends a resume screening instruction to the server, the server receives the resume screening instruction, acquires a plurality of to-be-detected resume texts needing to be screened from the resume database according to the resume screening instruction, and performs entity identification on each to-be-detected resume text to obtain a resume keyword corresponding to each to-be-detected resume text, wherein each to-be-detected resume text corresponds to a plurality of resume keywords, and different to-be-detected resume texts have different resume keywords. Vectorizing each resume keyword to obtain a resume keyword vector corresponding to each resume text to be detected.
S206, inputting the resume keyword vector corresponding to each resume text to be detected into the established resume classification model to obtain the passing probability corresponding to each resume text to be detected, and determining at least one first target resume text according to the passing probability.
The established resume classification model is obtained by training through a machine learning classification algorithm according to the historical resume text and the corresponding historical screening result. The machine learning classification algorithm may be a decision tree or a random forest or a tree classification prediction algorithm such as GBDT (gradient boosting decision tree). The main purpose of the established resume classification model is to guarantee the recall rate, and the characteristics used by the established resume classification model in training are basic information (such as sex, age, academic history, education background and the like) in the historical resume text, namely non-professional information. The passing probability refers to the screening passing probability of the resume text to be detected, which is obtained after the resume classification model is established for prediction. The first target resume text is the resume text to be screened, which is obtained after the established resume screening model is subjected to preliminary screening.
Specifically, the server inputs the resume keyword vector corresponding to each resume text to be detected into the established resume classification model for screening, so as to obtain the passing probability corresponding to each resume text to be detected, and determines at least one first target resume text according to the passing probability. The resume text to be detected with the passing probability exceeding the preset threshold value can also be used as the first target resume text.
And S208, acquiring the job description text, extracting job keywords in the job description text, and vectorizing the job keywords to obtain a job keyword vector.
The job keywords refer to keywords in the job description text and can be extracted by using a keyword extraction algorithm, wherein the keyword extraction algorithm can be TextRank, TF-IDF (Term Frequency-Inverse Document Frequency), L DA (L event digital Document Allocation) and the like, and can also be extracted by a trained neural network language model.
Specifically, the server acquires a job description text, which is a delivery job description text corresponding to each resume text to be detected. And extracting the position keywords in the position description text by using a keyword extraction algorithm, and vectorizing the position keywords by using a bag-of-words model to obtain a position keyword vector, wherein the bag-of-words model refers to a one-hot model, a TF-IDF model, a Huffman coding model and the like. The position keywords may also be vectorized using a Bag-of-Words Model using a Skip-gram (Skip-gram Model) or cbow (Continuous Bag-of-Words Model) Model to obtain a position keyword vector.
S210, obtaining a resume keyword vector corresponding to each first target resume text, and calculating text similarity between each first target resume text and the position description text according to the resume keyword vector and the position keyword vector.
Specifically, the server obtains the resume keyword vectors corresponding to the first target resume texts, and may calculate the similarity between the resume keyword vectors and the position keyword vectors using a distance similarity algorithm, where the distance similarity algorithm may use an euclidean distance algorithm, a cosine similarity algorithm, or the like. And taking the obtained similarity as the text similarity of the corresponding first target resume text and the corresponding position description text, and calculating the text similarity of each first target resume text and each position description text.
S212, determining a second target resume text from the first target resume text according to the text similarity.
The second target resume text refers to the first target resume text obtained after similarity screening detection is carried out on the second target resume text and the job description text.
Specifically, the server may select a text similarity exceeding a text similarity threshold from the text similarities according to a preset text similarity threshold, determine a first target resume text corresponding to the text similarity exceeding the text similarity threshold, and use the first target resume text as a second target resume text. The second target resume text can be used as the resume text to be detected which passes through the resume screening, and then the resume text to be detected which passes through the resume screening is returned to the management terminal for displaying.
In the resume screening method, each resume text to be detected is obtained, first screening detection is carried out on each resume text to be detected through an established resume screening model, and at least one first target resume text is determined from each resume text to be detected. And at the moment, acquiring the job description text, determining a second target resume text from the first target resume text according to the similarity by calculating the similarity between the job description text and each first target resume text, and taking the second target resume text as the screened resume text.
In one embodiment, as shown in fig. 3, step S204 is to perform entity recognition on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and includes the steps of:
s302, acquiring basic fields and professional fields in the resume text to be detected.
The basic field refers to a field in which the resume text to be detected describes basic information of the job seeker, such as age, gender, graduation school, academic history, past work company, working age, working place and the like. The professional field refers to a professional information field describing the relation between the job seeker and the job position in the resume text to be detected, such as: project experience information, work experience information, procurement certification information, programming language information, and work skill information, among others.
Specifically, the server acquires a basic field and a professional field in the resume text to be detected.
S304, extracting the basic key words corresponding to the basic fields.
Specifically, the basic keyword refers to a keyword corresponding to the basic field, and since the basic field describes basic information of the job seeker, the server can directly use the content of the basic field as the basic keyword.
S306, inputting the professional fields into the trained entity recognition model to obtain professional keywords corresponding to the professional fields, and obtaining resume keywords corresponding to the resume text to be detected according to the basic keywords and the professional keywords.
The trained entity recognition model is obtained by using CRF (conditional random field) and Bi L STM (Bi-directional L one Short-term-Termmemory neural network) to train in advance according to the existing professional information of positions, wherein the Bi-L STM and the CRF are optimized on the basis of the original Bi-L STM and the maximum entropy, a layer of conditional random field model is hung on the Bi-L STM to serve as a decoding layer of the model, the rationality of prediction results is considered in the conditional random field model, and the recognition accuracy is improved.
Specifically, the server inputs the professional field into the trained entity recognition model to obtain the professional keyword corresponding to the professional field, and the basic keyword and the professional keyword are used as the resume keyword corresponding to the resume text to be detected.
In the above example, the basic fields are extracted, and the professional fields are identified by using the entity identification model, so that the identification efficiency of obtaining the resume keywords corresponding to the resume text to be detected is improved.
In one embodiment, step S204, vectorizing the resume keywords corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected includes the steps of:
inputting the resume keywords corresponding to each resume text to be detected into the trained neural network language model to obtain the resume keyword vector corresponding to each resume text to be detected, wherein the trained neural network language model is obtained by training through a deep neural network according to the existing corpus.
The existing corpus is a corpus database built according to historical resume texts.
Specifically, the server is trained by NN L M in advance according to an existing corpus, and when a preset completion threshold or the maximum iteration number is reached, a trained neural network language model is obtained, wherein an activation function can use a tanh (hyperbolic tangent) function.
In one embodiment, as shown in fig. 4, the step S210 of calculating the respective text similarity of each first target resume text and the job description text according to the resume keyword vector and the job keyword vector includes the steps of:
s402, obtaining the number of resume keywords corresponding to the first target resume text and the number of position keywords corresponding to the position description text.
S404, calculating the vector similarity between the resume keyword vector corresponding to the first target resume text and the position keyword vector corresponding to the position description text.
The vector similarity refers to the similarity between the resume keyword vector and the position keyword vector.
Specifically, the server calculates the number of resume keywords in the first target resume text and the number of position keywords in the position description text. A similarity between each resume keyword vector and each position keyword vector is calculated using a distance similarity algorithm.
S406, determining the similarity of the resume words of the resume keywords corresponding to the first target resume text from the vector similarity, and obtaining the first target similarity according to the similarity of the resume words of the resume keywords and the number of the resume keywords.
The resume word similarity refers to the maximum similarity among the similarities between the resume keyword vector and each position keyword vector. The first target similarity refers to the similarity between the first target resume text and the job description text.
Specifically, the server determines each similarity between the resume keyword vector and each position keyword vector from each vector similarity, compares each similarity to obtain a maximum similarity, and takes the maximum similarity as the resume word similarity corresponding to the resume keyword. For example, the server determines each age similarity between the resume age vector and all position keyword vectors, compares the obtained each age similarity, determines the maximum age similarity according to the comparison result, the maximum age similarity is the age similarity between the resume age vector and the position age vector, and the maximum age similarity is used as the resume word similarity of the age keyword. The server determines the similarity of the resume words of each resume keyword. And summing the similarity of the resume words of each resume keyword, calculating the ratio of the sum result to the number of the resume keywords by the server, and taking the ratio result as the first target similarity.
And S408, determining the job word similarity of the job keyword vector corresponding to the job description text from the vector similarity, and obtaining a second target similarity according to the job word similarity of the job keyword vector and the number of the job keywords.
The term similarity refers to the maximum similarity among the similarities between the term keyword vector and each resume keyword vector. The second similarity refers to the similarity between the position description text and the first target resume text.
Specifically, the server determines the similarity between the position keyword vector and each resume keyword vector from the vector similarities, compares the similarities to obtain the maximum similarity, uses the maximum similarity as the position word similarity corresponding to the position keyword, and determines the position word similarity of each position keyword. And summing the job keyword similarity of each job keyword, calculating the ratio of the summation result to the number of the job keywords by the server, and taking the ratio result as the second target similarity.
And S410, determining the text similarity between the first target resume text and the job description text according to the first target similarity and the second target similarity.
Specifically, the server calculates the similarity and the average value of the first target and the second target to obtain the text similarity between the first target resume text and the job description text.
In a specific embodiment, the server may calculate the text similarity using the following formula (1). Equation (1) is as follows:
wherein S1 refers to the first target resume text, and S2 refers to the job description text. The SIM (S1, S2) represents a text similarity between the first target resume text and the job description text. nums (S1) represents the number of resume keywords, and nums (S1) represents the number of position keywords. w1i refers to resume keyword vectors and w2j refers to position keyword vectors. sim (w1i, w2j) refers to the similarity between the resume keyword vector and the position keyword vector. max [ sim (w1i, w2j)]The maximum similarity among the similarities between the resume keyword vector and each position keyword vector, that is, the resume word similarity, is referred to. sim (w2j, w1i) refers to the similarity between the position keyword vector and the resume keyword vector. max [ sim (w2j, w1i)]The maximum similarity among the similarities between the role keyword vector and each resume keyword vector, i.e., the role word similarity ∑w1imax[sim(w1i,w2j)]∑ showing the sum of the similarity of the various resume wordsw2jmax[sim(w2j,w1i)]Refers to the sum of similarity of the various terms.
It is referred to the first object similarity that,
it is referred to the second object similarity that,
in the embodiment, the text similarity between the first target resume text and the job description text is determined according to the first target similarity and the second target similarity, so that the accuracy of the obtained text similarity is improved.
In one embodiment, the step S212 of determining the second target resume text from the first target resume text according to the text similarity includes the steps of:
and sequencing the text similarity to obtain a sequencing result, and selecting a preset number of first target resume texts according to the sequencing result to obtain a second target resume text.
Specifically, the server sequences the text similarities to obtain a sequencing result, sequentially selects the first target resume texts with the largest text similarity according to the size of the sequencing result until a preset number of first target resume texts are selected, and uses the selected preset number of first target resume texts as each second target resume text, so that the efficiency of obtaining the second target resume texts is improved.
In one embodiment, as shown in fig. 5, after step S212, that is, after determining the second target resume text from the first target resume text according to the text similarity, the method further includes the steps of:
s502, obtaining the resume keyword vector corresponding to the second target resume text, and clustering the resume keyword vector corresponding to the second target resume text to obtain a clustering result.
S504, determining a third target resume text from the second target resume text according to the clustering result.
Clustering, among other things, refers to the process of dividing a collection of physical or abstract objects into classes composed of similar objects. Clustering can be performed using a clustering algorithm, such as: k-means clustering algorithm, DBscan density clustering algorithm, agglomerative hierarchical clustering, and the like.
Specifically, the server obtains the resume keyword vectors corresponding to the second target resume texts, and performs clustering calculation on the resume keyword vectors corresponding to the second target resume texts by using a clustering algorithm to obtain a clustering result. And then screening second target resume texts with obvious abnormality in the clustering result, wherein the obvious abnormality refers to the second target resume texts exceeding a preset clustering threshold value in the clustering result. And taking the second target resume text without obvious exception as a third target resume text. In this embodiment, each third target resume text is obtained by performing cluster screening detection on the second target resume text, so that the accuracy of obtaining the third target resume text that passes the screening detection is further improved.
In one embodiment, as shown in fig. 6, the resume screening method further includes the steps of:
s602, determining first text information corresponding to each resume text to be detected according to the passing probability.
The first text information refers to model evaluation information of the resume text to be detected, and the model evaluation information is obtained according to the passing probability. The model evaluation information may specifically be an evaluation score. For example, if the passing probability of the resume text to be detected is 0.8, the first text information may be 80 points.
Specifically, the server determines first text information of each resume text to be detected according to the passing probability.
S604, calculating the target similarity between the resume keyword vector corresponding to each resume text to be detected and the position keyword vector, and determining second text information corresponding to each resume text to be detected according to the target similarity.
The target similarity refers to the similarity between the resume keyword vector and the position keyword vector, the second text information is similar evaluation information of the resume text to be detected, and the similar evaluation information is determined according to the target similarity.
Specifically, the server respectively calculates the target similarity between the resume keyword vector corresponding to each resume text to be detected and the position keyword vector, and determines second text information corresponding to each resume text to be detected according to the target similarity. For example, the value of the target similarity corresponding to all the resume keywords in each resume text to be detected can be calculated, and the second text information of the resume text to be detected can be determined according to the average similarity. For example, if the average similarity is 75%, the obtained second text information may be 75 points.
In one embodiment, the formula (1) may be used for calculation according to the target similarity to obtain the text similarity corresponding to each resume text to be detected, and the final second text information is determined according to the text similarity.
And S606, acquiring a preset weight, and determining target text information corresponding to each resume text to be detected according to the preset weight, the first text information and the second text information.
And S608, determining the resume texts to be detected which pass the screening according to the target text information.
The preset weight refers to the preset weight corresponding to the first text information and the second text information. For example, the preset weight of the first text information may be 0.4, and the preset weight of the second text information may be 0.6. The target text information refers to the final evaluation information of the resume text to be detected.
Specifically, the server acquires a preset weight, and then calculates target text information corresponding to each resume text to be detected according to the preset weight, the first text information and the second text information. For example, the second text information may be 75 based on the first text information 80. The preset weight of the first text information may be 0.4, and the preset weight of the second text information may be 0.6. The calculated target text information may be (80 × 0.4+75 × 0.6)/2 ═ 54.5. And then the server calculates target text information corresponding to each resume text to be detected. Then, each target text message can be screened according to preset text messages, and the resume text to be detected, which exceeds the preset text messages, in each target text message is used as the resume text to be detected, which passes the detection. The weights are set for the model detection result and the similarity detection result, and the final detection result is obtained through calculation according to the set weights, so that the accuracy of the resume text to be detected passing the detection is improved, and the accuracy of the resume screening is improved.
In one embodiment, as shown in FIG. 7, a schematic diagram of resume screening is shown. Specifically;
the service area obtains a named entity model and a neural network language model through resume database training in advance. And then obtaining each resume text and position description text to be detected, performing text analysis and post-structure processing on each resume text and position description text to be detected, namely performing entity identification through a named entity model to obtain an entity, and performing vectorization processing on the entity through a neural network language model to obtain a resume keyword vector and a position keyword vector. Inputting the resume keyword vectors into an established resume screening model for model detection, primarily screening resumes to be detected according to a model detection result to obtain first target resume texts, performing text matching according to the resume keyword vectors and position keyword vectors corresponding to the first target resume texts, performing text similarity calculation to obtain text similarity calculation results, performing secondary fine screening according to the text similarity calculation results, sequencing the text similarity calculation results, sequentially selecting a preset number of first target resume texts from large to small according to the sequencing result to obtain second target resume texts, and obtaining second target resume texts which are resume screening results.
It should be understood that although the various steps in the flowcharts of fig. 2-6 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 described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-6 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 8, there is provided a resume screening apparatus 800 comprising: a text obtaining module 802, a resume vector obtaining module 804, a first text determining module 806, a position vector obtaining module 808, a text similarity calculating module 810 and a second text determining module 812, wherein:
the text acquisition module 802 is configured to receive a resume screening instruction and acquire a plurality of resume texts to be detected according to the resume screening instruction;
a resume vector obtaining module 804, configured to perform entity identification on each resume text to be detected, obtain a resume keyword corresponding to each resume text to be detected, and vectorize the resume keyword corresponding to each resume text to be detected, to obtain a resume keyword vector corresponding to each resume text to be detected;
a first text determining module 806, configured to input the resume keyword vector corresponding to each resume text to be detected into a pre-established resume classification model, obtain a passing probability corresponding to each resume text to be detected, and determine at least one first target resume text according to the passing probability;
a job vector obtaining module 808, configured to obtain a job description text, extract job keywords in the job description text, and vectorize the job keywords to obtain a job keyword vector;
the text similarity calculation module 810 is configured to obtain a resume keyword vector corresponding to each first target resume text, and calculate each text similarity between each first target resume text and the job description text according to the resume keyword vector and the job keyword vector;
and a second text determining module 812, configured to determine a second target resume text from the first target resume text according to the text similarity.
In one embodiment, the resume vector derivation module 804 includes:
the field acquisition unit is used for acquiring basic fields and professional fields in the resume text to be detected;
the extraction unit is used for extracting basic keywords corresponding to the basic fields;
and the recognition unit is used for inputting the professional fields into the trained entity recognition model to obtain the professional keywords corresponding to the professional fields, and obtaining the resume keywords corresponding to the resume text to be detected according to the basic keywords and the professional keywords.
In one embodiment, the resume vector derivation module 804 includes:
and the model vectorization unit is used for inputting the resume keywords corresponding to each resume text to be detected into the trained neural network language model to obtain the resume keyword vector corresponding to each resume text to be detected, and the trained neural network language model is obtained by training through a deep neural network according to the existing corpus.
In one embodiment, the text similarity calculation module 810 includes:
the quantity acquiring unit is used for acquiring the quantity of resume keywords corresponding to the first target resume text and the quantity of position keywords corresponding to the position description text;
the vector calculation unit is used for calculating the vector similarity between the resume keyword vector corresponding to the first target resume text and the position keyword vector corresponding to the position description text;
the first target obtaining unit is used for determining the resume word similarity of the resume keywords corresponding to the first target resume text from the vector similarity and obtaining the first target similarity according to the resume word similarity of the resume keywords and the number of the resume keywords;
the second target obtaining unit is used for determining the job word similarity of the job keyword vector corresponding to the job description text from the vector similarity and obtaining second target similarity according to the job word similarity of the job keyword vector and the number of the job keywords;
and the text similarity determining unit is used for determining the text similarity between the first target resume text and the job description text according to the first target similarity and the second target similarity.
In one embodiment, the second text determination module 812 includes:
and the sorting unit is used for sorting the text similarity to obtain a sorting result, and selecting a preset number of first target resume texts according to the sorting result to obtain a second target resume text.
In one embodiment, the resume screening apparatus 800 further includes:
the clustering detection module is used for acquiring the resume keyword vector corresponding to each second target resume text and clustering the resume keyword vector corresponding to each second target resume text to obtain a clustering result; and determining a third target resume text from the second target resume text according to the clustering result.
In one embodiment, the resume screening apparatus 800 further includes:
the text information determining module is used for determining first text information corresponding to each resume text to be detected according to the passing probability; calculating target similarity of the resume keyword vector corresponding to each resume text to be detected and the position keyword vector, and determining second text information corresponding to each resume text to be detected according to the target similarity;
the target information determining module is used for acquiring a preset weight and determining target text information corresponding to each resume text to be detected according to the preset weight, the first text information and the second text information;
and the text determination module is used for determining the resume texts to be detected which pass the screening according to the target text information.
For the specific definition of the resume screening device, reference may be made to the above definition of the resume screening method, which is not described herein again. The modules in the resume screening apparatus may be implemented in whole or in part by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 9. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store resume text data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a resume screening method.
Those skilled in the art will appreciate that the architecture shown in fig. 9 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, there is provided a computer device comprising a memory storing a computer program and a processor implementing the steps of the resume screening method in any of the above embodiments when the processor executes the computer program.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the steps of the resume screening method in any of the embodiments described above.
It will be understood by those of ordinary skill in the art that all or a portion of the processes of the methods of the embodiments described above may be implemented by a computer program that may be stored on a non-volatile computer-readable storage medium, which when executed, may include the processes of the embodiments of the methods described above, wherein any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A resume screening method, the method comprising:
receiving a resume screening instruction, and acquiring a plurality of to-be-detected resume texts according to the resume screening instruction;
performing entity identification on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and vectorizing the resume keyword corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected;
inputting the resume keyword vector corresponding to each resume text to be detected into a pre-established resume classification model to obtain the passing probability corresponding to each resume text to be detected, and determining at least one first target resume text according to the passing probability;
acquiring a job description text, extracting job keywords in the job description text, and vectorizing the job keywords to obtain a job keyword vector;
acquiring a resume keyword vector corresponding to each first target resume text, and calculating text similarity between each first target resume text and the position description text according to the resume keyword vector and the position keyword vector;
and determining a second target resume text from the first target resume text according to the text similarity.
2. The method according to claim 1, wherein the performing entity recognition on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected comprises:
acquiring a basic field and a professional field in a resume text to be detected;
extracting basic keywords corresponding to the basic fields;
and inputting the professional field into a trained entity recognition model to obtain professional keywords corresponding to the professional field, and obtaining resume keywords corresponding to the resume text to be detected according to the basic keywords and the professional keywords.
3. The method according to claim 1, wherein vectorizing the resume keyword corresponding to each resume text to be detected to obtain the resume keyword vector corresponding to each resume text to be detected comprises:
inputting the resume keywords corresponding to each resume text to be detected into a trained neural network language model to obtain the resume keyword vector corresponding to each resume text to be detected, wherein the trained neural network language model is obtained by training through a deep neural network according to an existing corpus.
4. The method of claim 1, wherein computing respective text similarities for each first target resume text and the job description text based on the resume keyword vector and the job keyword vector comprises:
acquiring the number of resume keywords corresponding to a first target resume text and the number of position keywords corresponding to the position description text;
calculating the vector similarity between the resume keyword vector corresponding to the first target resume text and the position keyword vector corresponding to the position description text;
determining the similarity of the resume words of the resume keywords corresponding to the first target resume text from the vector similarity, and obtaining the first target similarity according to the similarity of the resume words of the resume keywords and the number of the resume keywords;
determining the job word similarity of a job keyword vector corresponding to the job description text from the vector similarity, and obtaining a second target similarity according to the job word similarity of the job keyword vector and the number of the job keywords;
and determining the text similarity between the first target resume text and the job description text according to the first target similarity and the second target similarity.
5. The method of claim 1, wherein determining a second target resume text from the first target resume text based on the text similarity comprises:
and sequencing the text similarity to obtain a sequencing result, and selecting a preset number of first target resume texts according to the sequencing result to obtain a second target resume text.
6. The method of claim 1, further comprising, after determining a second target resume text from the first target resume text according to the text similarity:
acquiring a resume keyword vector corresponding to a second target resume text, and clustering the resume keyword vector corresponding to the second target resume text to obtain a clustering result;
and determining a third target resume text from the second target resume text according to the clustering result.
7. The method of claim 1, further comprising:
determining first text information corresponding to each resume text to be detected according to the passing probability;
calculating target similarity between the resume keyword vector corresponding to each resume text to be detected and the position keyword vector, and determining second text information corresponding to each resume text to be detected according to the target similarity;
acquiring a preset weight, and determining target text information corresponding to each resume text to be detected according to the preset weight, the first text information and the second text information;
and determining the resume texts to be detected which pass the screening according to the target text information.
8. A resume screening apparatus, the apparatus comprising:
the text acquisition module is used for receiving the resume screening instruction and acquiring a plurality of to-be-detected resume texts according to the resume screening instruction;
the resume vector obtaining module is used for performing entity recognition on each resume text to be detected to obtain a resume keyword corresponding to each resume text to be detected, and vectorizing the resume keyword corresponding to each resume text to be detected to obtain a resume keyword vector corresponding to each resume text to be detected;
the first text determination module is used for inputting the resume keyword vector corresponding to each resume text to be detected into a pre-established resume classification model to obtain the passing probability corresponding to each resume text to be detected, and determining at least one first target resume text according to the passing probability;
the job description module is used for acquiring a job description text, extracting job keywords in the job description text, and vectorizing the job keywords to obtain a job keyword vector;
the text similarity calculation module is used for acquiring a resume keyword vector corresponding to each first target resume text and calculating the text similarity between each first target resume text and the job description text according to the resume keyword vector and the job keyword vector;
and the second text determination module is used for determining at least one second target resume text from the first target resume text according to the text similarity.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010128423.7A CN111461637A (en) | 2020-02-28 | 2020-02-28 | Resume screening method and device, computer equipment and storage medium |
| PCT/CN2020/098678 WO2021169111A1 (en) | 2020-02-28 | 2020-06-29 | Resume screening method and apparatus, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010128423.7A CN111461637A (en) | 2020-02-28 | 2020-02-28 | Resume screening method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111461637A true CN111461637A (en) | 2020-07-28 |
Family
ID=71682467
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010128423.7A Pending CN111461637A (en) | 2020-02-28 | 2020-02-28 | Resume screening method and device, computer equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111461637A (en) |
| WO (1) | WO2021169111A1 (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112100492A (en) * | 2020-09-11 | 2020-12-18 | 河北冀联人力资源服务集团有限公司 | Batch delivery method and system for resumes of different versions |
| CN112417154A (en) * | 2020-11-25 | 2021-02-26 | 上海创米科技有限公司 | Method and device for determining similarity of documents |
| CN113298495A (en) * | 2021-05-27 | 2021-08-24 | 中国科学院深圳先进技术研究院 | Resume screening method, resume screening device, terminal device and storage medium |
| CN113672705A (en) * | 2021-08-27 | 2021-11-19 | 工银科技有限公司 | Resume screening method, apparatus, device, medium and program product |
| CN113806544A (en) * | 2021-09-23 | 2021-12-17 | 湖北天天数链技术有限公司 | Resume information processing method and device, electronic equipment and storage medium |
| US20220215196A1 (en) * | 2021-01-04 | 2022-07-07 | Atos It Solutions And Services, Inc. | Devices, systems and methods for assessing a match between job descriptions and resumes |
| CN116644184A (en) * | 2023-07-27 | 2023-08-25 | 浙江厚雪网络科技有限公司 | Human Resource Information Management System Based on Data Clustering |
| CN117390173A (en) * | 2023-11-02 | 2024-01-12 | 江苏优丞信息科技有限公司 | Massive resume screening method for semantic similarity matching |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113780996A (en) * | 2021-09-16 | 2021-12-10 | 湖北天天数链技术有限公司 | Post data detection method, model training method and device and electronic equipment |
| CN113886562A (en) * | 2021-10-02 | 2022-01-04 | 智联(无锡)信息技术有限公司 | AI resume screening method, system, equipment and storage medium |
| CN114282106B (en) * | 2021-12-22 | 2023-07-25 | 北京网聘咨询有限公司 | Quick delivering method for position information |
| CN113988045B (en) * | 2021-12-28 | 2022-04-12 | 浙江口碑网络技术有限公司 | Text similarity determining method, text processing method, corresponding device and equipment |
| CN115879901B (en) * | 2023-02-22 | 2023-07-28 | 陕西湘秦衡兴科技集团股份有限公司 | Intelligent personnel self-service platform |
| CN116562837A (en) * | 2023-07-12 | 2023-08-08 | 深圳须弥云图空间科技有限公司 | Person post matching method, device, electronic equipment and computer readable storage medium |
| CN117540917A (en) * | 2023-11-14 | 2024-02-09 | 大能手教育科技(北京)有限公司 | Training platform aided training method, device, equipment and medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117863A (en) * | 2015-09-28 | 2015-12-02 | 北京橙鑫数据科技有限公司 | Resume position matching method and device |
| US10657498B2 (en) * | 2017-02-17 | 2020-05-19 | Walmart Apollo, Llc | Automated resume screening |
| CN108550019B (en) * | 2018-03-22 | 2022-03-25 | 创新先进技术有限公司 | Resume screening method and device |
| CN110399475A (en) * | 2019-06-18 | 2019-11-01 | 平安科技（深圳）有限公司 | Resume matching process, device, equipment and storage medium based on artificial intelligence |
| CN110263148A (en) * | 2019-06-27 | 2019-09-20 | 中国工商银行股份有限公司 | Intelligent resume selection method and device |
-
2020
- 2020-02-28 CN CN202010128423.7A patent/CN111461637A/en active Pending
- 2020-06-29 WO PCT/CN2020/098678 patent/WO2021169111A1/en active Application Filing
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112100492A (en) * | 2020-09-11 | 2020-12-18 | 河北冀联人力资源服务集团有限公司 | Batch delivery method and system for resumes of different versions |
| CN112417154A (en) * | 2020-11-25 | 2021-02-26 | 上海创米科技有限公司 | Method and device for determining similarity of documents |
| US20220215196A1 (en) * | 2021-01-04 | 2022-07-07 | Atos It Solutions And Services, Inc. | Devices, systems and methods for assessing a match between job descriptions and resumes |
| CN113298495A (en) * | 2021-05-27 | 2021-08-24 | 中国科学院深圳先进技术研究院 | Resume screening method, resume screening device, terminal device and storage medium |
| CN113672705A (en) * | 2021-08-27 | 2021-11-19 | 工银科技有限公司 | Resume screening method, apparatus, device, medium and program product |
| CN113806544A (en) * | 2021-09-23 | 2021-12-17 | 湖北天天数链技术有限公司 | Resume information processing method and device, electronic equipment and storage medium |
| CN116644184A (en) * | 2023-07-27 | 2023-08-25 | 浙江厚雪网络科技有限公司 | Human Resource Information Management System Based on Data Clustering |
| CN116644184B (en) * | 2023-07-27 | 2023-10-20 | 浙江厚雪网络科技有限公司 | Human resource information management system based on data clustering |
| CN117390173A (en) * | 2023-11-02 | 2024-01-12 | 江苏优丞信息科技有限公司 | Massive resume screening method for semantic similarity matching |
| CN117390173B (en) * | 2023-11-02 | 2024-03-29 | 江苏优丞信息科技有限公司 | Massive resume screening method for semantic similarity matching |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021169111A1 (en) | 2021-09-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111461637A (en) | Resume screening method and device, computer equipment and storage medium | |
| CN109190023B (en) | Collaborative recommendation method and device and terminal equipment | |
| US11074434B2 (en) | Detection of near-duplicate images in profiles for detection of fake-profile accounts | |
| US11521372B2 (en) | Utilizing machine learning models, position based extraction, and automated data labeling to process image-based documents | |
| CN110598206A (en) | Text semantic recognition method and device, computer equipment and storage medium | |
| CN110401545B (en) | Chat group creation method, chat group creation device, computer equipment and storage medium | |
| CN110825894A (en) | Data index establishing method, data index retrieving method, data index establishing device, data index retrieving device, data index establishing equipment and storage medium | |
| CN112395875A (en) | Keyword extraction method, device, terminal and storage medium | |
| Zhang et al. | Semisupervised particle swarm optimization for classification | |
| CN110765292A (en) | Image retrieval method, training method and related device | |
| CN112070506A (en) | Risk user identification method, device, server and storage medium | |
| CN111159481B (en) | Edge prediction method and device for graph data and terminal equipment | |
| Pristyanto et al. | The effect of feature selection on classification algorithms in credit approval | |
| CN113254687B (en) | Image retrieval and image quantification model training method, device and storage medium | |
| CN115034315A (en) | Business processing method and device based on artificial intelligence, computer equipment and medium | |
| CN114329004A (en) | Digital fingerprint generation method, digital fingerprint generation device, data push method, data push device and storage medium | |
| Faure et al. | Estimating Kolmogorov entropy from recurrence plots | |
| CN111611395B (en) | Entity relationship identification method and device | |
| Le et al. | Ent-Boost: Boosting using entropy measures for robust object detection | |
| Pohl et al. | Active online learning for social media analysis to support crisis management | |
| Wang et al. | Averaging versus voting: A comparative study of strategies for distributed classification. | |
| CN112733645B (en) | Handwritten signature verification method, handwritten signature verification device, computer equipment and storage medium | |
| CN115310606A (en) | Deep learning model depolarization method and device based on data set sensitive attribute reconstruction | |
| CN109918583B (en) | Task information processing method and device | |
| Patel | Large high dimensional data handling using data reduction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210129 Address after: 518066 Room 201, building A, No. 1, Qian Wan Road, Qianhai Shenzhen Hong Kong cooperation zone, Shenzhen, Guangdong (Shenzhen Qianhai business secretary Co., Ltd.) Applicant after: Shenzhen saiante Technology Service Co.,Ltd. Address before: 1-34 / F, Qianhai free trade building, 3048 Xinghai Avenue, Mawan, Qianhai Shenzhen Hong Kong cooperation zone, Shenzhen, Guangdong 518000 Applicant before: Ping An International Smart City Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |