CN114330333A - Method for processing skill information, model training method and device - Google Patents

Abstract

The present disclosure provides a method, a model training method, and an apparatus for processing skill information, and relates to the technical field of artificial intelligence, in particular to the technical field of machine learning. The specific implementation scheme is as follows: acquiring information to be investigated; determining at least one skill word based on the information to be investigated and a skill word generation model completed in advance; and outputting at least one skill word. In this implementation manner, skill information can be inspected based on skill words, which can improve the accuracy of skill inspection.

Description

Method, model training method and device for processing skill information

technical field

The present disclosure relates to the field of artificial intelligence technology, in particular to the field of machine learning technology.

Background technique

At present, for different positions, it is necessary to examine the skill information corresponding to each position, so as to select candidates who are suitable for each position.

However, the inspection of skill information now relies on human experience, which makes the inspection of skills more subjective and less accurate.

SUMMARY OF THE INVENTION

The present disclosure provides a method, model training method and apparatus for processing skill information.

According to an aspect of the present disclosure, a method for processing skill information is provided, including: acquiring information to be investigated; determining at least one skill word based on the information to be investigated and a skill word generation model completed in advance; outputting at least one skill word.

According to another aspect of the present disclosure, a model training method is provided, including: obtaining sample information to be inspected and inspection result labeling data; determining at least one sample based on a preset skill word map, sample information to be inspected, and a model to be trained Skill word; based on at least one sample skill word and the test result labeling data, train the model to be trained, and obtain the trained skill word generation model.

According to another aspect of the present disclosure, there is provided an apparatus for processing skill information, comprising: an information acquisition unit configured to acquire information to be investigated; a skill word determination unit configured to complete based on the information to be investigated and pre-training The skill word generation model determines at least one skill word; the skill word output unit is configured to output at least one skill word.

According to another aspect of the present disclosure, there is provided a model training device, comprising: a sample acquisition unit configured to acquire sample information to be investigated and inspection result labeling data; a sample skill word determination unit configured to be based on preset skills The word map, the sample information to be inspected, and the model to be trained determine at least one sample skill word; the model training unit is configured to label data based on the at least one sample skill word and the inspection result, train the model to be trained, and obtain the trained skill word Generate the model.

According to another aspect of the present disclosure, there is provided an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, One or more processors are caused to implement any of the above methods for processing skill information or model training methods.

According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute any one of the above methods for processing skill information or model training methods.

According to another aspect of the present disclosure, a computer program product is provided, including a computer program, which when executed by a processor implements any one of the above methods for processing skill information or a model training method.

According to the technology of the present disclosure, a method for processing skill information is provided, which can inspect skill information based on skill words, and can improve the accuracy of skill inspection.

It should be understood that what is described in this section is not intended to identify key or critical features of embodiments of the disclosure, nor is it intended to limit the scope of the disclosure. Other features of the present disclosure will become readily understood from the following description.

Description of drawings

The accompanying drawings are used for better understanding of the present solution, and do not constitute a limitation to the present disclosure. in:

FIG. 1 is an exemplary system architecture diagram to which an embodiment of the present disclosure may be applied;

2 is a flowchart of one embodiment of a method for processing skill information according to the present disclosure;

3 is a schematic diagram of an application scenario of the method for processing skill information according to the present disclosure;

4 is a flowchart of one embodiment of a model training method according to the present disclosure;

5 is a flowchart of another embodiment of a model training method according to the present disclosure;

6 is a schematic structural diagram of an embodiment of an apparatus for processing skill information according to the present disclosure;

7 is a schematic structural diagram of an embodiment of a model training apparatus according to the present disclosure;

8 is a block diagram of an electronic device used to implement the method for processing skill information or the model training method according to an embodiment of the present disclosure.

Detailed ways

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding and should be considered as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted from the following description for clarity and conciseness.

It should be noted that the embodiments of the present disclosure and the features of the embodiments may be combined with each other under the condition of no conflict. The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

As shown in FIG. 1 , the system architecture 100 may include terminal devices 101 , 102 , and 103 , a network 104 and a server 105 . The network 104 is a medium used to provide a communication link between the terminal devices 101 , 102 , 103 and the server 105 . The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The terminal devices 101, 102, and 103 interact with the server 105 through the network 104 to receive or send messages and the like. Wherein, the terminal devices 101, 102, 103 can obtain the object information that needs to be investigated, the post information waiting for investigation information, and send the information to be investigated to the server 105 through the network 104, so that the server 105 returns the corresponding at least one skill word, At least one skill word is output, so that the user conducts a skill survey based on the at least one skill word. Moreover, in the model training stage, the terminal devices 101, 102, and 103 can obtain the sample information to be investigated and the annotation data of the inspection results, and send the sample information to be investigated and the annotation data of the inspection results to the server 105 through the network 104, so that the server 105 is based on These data are used for model training, and the trained skill word generation model is returned.

The terminal devices 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, and 103 are hardware, they may be various electronic devices, including but not limited to mobile phones, computers, tablets, and so on. When the terminal devices 101, 102, and 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (eg, to provide distributed services), or as a single software or software module. There is no specific limitation here.

The server 105 may be a server that provides various services. For example, the server 105 may acquire the information to be investigated sent by the terminal devices 101, 102, and 103, and determine at least one item corresponding to the information to be investigated based on the pre-trained skill word generation model. Label words, and return at least one label word to the terminal devices 101 , 102 , and 103 through the network 104 . Or, in the model training stage, the server 105 may also receive the sample information to be investigated and the data marked on the inspection results sent by the terminal devices 101, 102 and 103, and based on the preset skill word map, the information to be inspected and the data marked on the inspection results. , train the to-be-trained model, and get the trained skill word generation model.

It should be noted that the server 105 may be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or can be implemented as a single server. When the server 105 is software, it can be implemented as a plurality of software or software modules (for example, for providing distributed services), or can be implemented as a single software or software module. There is no specific limitation here.

It should be noted that the method for processing skill information or the model training method provided by the embodiments of the present disclosure may be executed by the terminal devices 101, 102, 103, or may be executed by the server 105, an apparatus or model for processing skill information The training device may be installed in the terminal devices 101 , 102 , and 103 , or may be installed in the server 105 .

It should be understood that the numbers of terminal devices, networks and servers in FIG. 1 are merely illustrative. There can be any number of terminal devices, networks and servers according to implementation needs.

With continued reference to FIG. 2, a flow 200 of one embodiment of a method for processing skill information in accordance with the present disclosure is shown. The method for processing skill information of this embodiment includes the following steps:

Step 201, obtaining information to be investigated.

In this embodiment, the execution subject (such as the terminal devices 101, 102, 103 or the server 105 in FIG. 1 ) can obtain the information to be investigated related to the skill inspection when the skills inspection is required. Wherein, the information to be investigated may include but not limited to the information of the post to be investigated, the information of the object to be investigated, the historical investigation result information of the object to be investigated, etc., which is not limited in this embodiment. In addition, the execution subject may acquire the information to be investigated from pre-stored local data, or may acquire the information to be investigated based on an electronic device that has established a connection in advance.

In some optional implementation manners of this embodiment, the information to be inspected includes information of a position to be inspected and information of an object to be inspected. Among them, the information of the position to be inspected is the information of the position to be inspected, such as the job requirement information. The information of the object to be inspected is the information of the applicant to be inspected, such as the resume information of the applicant.

Step 202: Determine at least one skill word based on the information to be investigated and the skill word generation model that has been trained in advance.

In this embodiment, the pre-trained skill word generation model is used to determine at least one skill word matching the information to be investigated, so that the recruiting object can conduct a skill examination on the recruiting object based on the at least one skill word.

The pre-trained skill word generation model may be determined based on historical survey data and a skill word map corresponding to the historical survey data, and the historical survey data may include historical information to be investigated and inspection results corresponding to the historical information to be investigated.

Step 203, output at least one skill word.

In this embodiment, the execution body may output at least one skill word, so that the recruiting object conducts a recruitment investigation on the recruiting object according to the at least one skill word. Optionally, the execution subject may output at least one skill word to the binding electronic device of the recruiting object.

In some optional implementations of this embodiment, outputting at least one skill word includes: determining a target recruiting object based on the at least one skill word and recruiting object information; sending at least one skill word to the target recruiting object, so that the target recruiting object The subject is subject to a skill test based on at least one skill word.

In this implementation manner, after determining that at least one skill word is obtained, the execution subject may acquire the recruitment target information. The recruitment object information may be object information corresponding to each recruitment object. Optionally, the object information may include object tags. The executive body can match at least one skill word with the object information of each recruiting object, obtain the matching degree of each recruiting object and at least one skill word, and select the recruiting object with the highest matching degree as the target recruiting object. And at least one skill word is sent to the binding electronic device of the target recruiting object, so that the target recruiting object conducts a skill inspection on the object to be investigated based on the at least one skill word.

And, for each skill word, the executive body can determine the target recruiting object with the highest matching degree with the skill word based on the skill word and the recruiting object information, and send the skill word to the target recruiting object, so that the target recruiting object can inspect the target recruiting object. The skill corresponding to the skill word of the object to be investigated. That is, the executive body can select different recruiting objects for recruitment inspection for different skill words, so as to improve the inspection effect of multi-skill words.

Continue to refer to FIG. 3 , which shows a schematic diagram of an application scenario of the method for processing skill information according to the present disclosure. In the application scenario of FIG. 3 , the execution subject can obtain the information to be investigated 301 , where the information to be investigated 301 includes information of the position to be investigated and information of the object to be investigated. The execution subject inputs the information to be investigated 301 into the skill word generation model 302, and obtains the skill word 303 output by the skill word generation model 302, wherein the skill word 303 may include but is not limited to the skill word A, the skill word B, and the skill word C. After that, the executive body can send the skill word 303 to the target recruitment object 304 . The target recruitment object 304 here may be one object or multiple objects, which is not limited in this embodiment.

The method for processing skill information provided by the above-mentioned embodiments of the present disclosure can determine the skill word corresponding to the information to be investigated based on the pre-trained skill word generation model, and inspect the skill information based on the skill word, which can improve the skill inspection precision.

Continuing to refer to FIG. 4 , a flow 400 of one embodiment of a model training method according to the present disclosure is shown. As shown in FIG. 4 , the model training method of this embodiment may include the following steps:

Step 401: Obtain the sample information to be inspected and the inspection result labeling data.

In this embodiment, the sample information to be investigated may be the information of the post to be investigated and the information of the object to be investigated in the historical investigation data, and the marked data of the investigation result may be the information of the post to be investigated and the information of the object to be investigated in the historical investigation data that match the information of the post to be investigated and the information of the object to be investigated Survey result information, such as historical interview result information.

Step 402: Determine at least one sample skill word based on a preset skill word graph, sample information to be investigated, and a model to be trained.

In this embodiment, the preset skill word graph includes a plurality of skill words and an association relationship between each skill word. The execution subject may determine at least one sample skill word output by the model to be trained based on the preset skill word graph and the sample information to be investigated.

The execution subject may input both the preset skill word map and the sample information to be investigated into the model to be trained, so that the model to be trained determines at least one sample skill word that matches the sample information to be investigated from the preset skill word map .

Step 403 , based on the at least one sample skill word and the marked data of the inspection result, train the model to be trained, and obtain a trained skill word generation model.

In this embodiment, the execution subject may mark the degree of matching between the data and the above-mentioned at least one sample skill word based on the inspection result, and in response to determining that the degree of matching is low, adjust the model parameters of the model to be trained, and repeat iterations until the training is to be performed. The matching degree between at least one sample word output by the model and the labeled data of the inspection result is relatively high, and the preset convergence condition is satisfied, and the trained skill word generation model is obtained.

The trained skill word generation model here can be applied to the method for processing skill information described in FIG. 1 , and a matching skill word can be determined based on the information to be investigated.

The model training method provided by the above-mentioned embodiments of the present disclosure can also use a preset skill word graph, as well as the sample information to be investigated and the inspection result labeling data in the historical investigation data, to train the model to be trained, and obtain the skill words that have been trained. The generation model realizes the generation of skill words that match the information to be investigated.

Continuing to refer to FIG. 5, a flow 500 of another embodiment of a model training method according to the present disclosure is shown. As shown in FIG. 5 , the model training method of this embodiment may include the following steps:

Step 501: Obtain the sample information to be inspected and the inspection result labeling data.

In this embodiment, the sample information to be inspected includes sample information of a post to be inspected and sample information of an object to be inspected. Among them, the sample information of the post to be investigated is the post information of the historical investigation, such as the historical post requirement information. The sample information of the object to be investigated is the information of the historically investigated candidate, such as the resume information of the historical candidate. The inspection result labeling data may be the interview result information of the historical inspection, corresponding to the corresponding position information and corresponding object information of the historical inspection.

For the detailed description of step 501, please refer to the detailed description of step 401, which is not repeated here.

Step 502: Determine the skill word graph based on the sample information to be investigated and the labeling data of the inspection result.

In this embodiment, the execution subject can extract each skill word from the sample information to be investigated and the labeling data of the inspection result, and establish a list of each skill word based on the appearance position of each skill word in the sample information to be inspected and the labeling data of the inspection result. The relationship between them is obtained, and the skill word graph is obtained.

In some optional implementations of this embodiment, determining the skill word graph based on the sample information to be investigated includes: determining each candidate skill word from the sample to be investigated information and the labeling data of the inspection result; based on each candidate skill word, The sample information to be inspected and the inspection result labeling data are used to determine the connection information between each candidate skill word; based on each candidate skill word and the connection information between each candidate skill word, the skill word map is determined.

In this implementation manner, the executing subject may perform text analysis on the sample inspection information and the inspection result labeling data, and extract each candidate skill word from the text analysis. Optionally, the execution subject may use a combination of a recurrent neural network, an attention mechanism, and a replication mechanism to extract candidate skill words from sample inspection information and inspection result labeling data. Among them, the copy mechanism is used to directly copy words beyond the vocabulary, and the attention mechanism is used to pay more attention to key texts for attention resource scheduling. By combining the recurrent neural network, attention mechanism and replication mechanism, the determination accuracy of candidate skill words can be improved.

Afterwards, the execution subject may determine the connection information between each candidate skill word based on each candidate skill word, the sample information to be examined, and the examination result labeling data. Optionally, the execution subject may determine connection information between each candidate skill word based on the position of each candidate skill word in each piece of information and the association relationship between each piece of information. Among them, the information refers to the information of the position to be inspected, the information of the object to be inspected, and the annotation data of the inspection result of the historical data in the sample information to be inspected and the inspection result labeling data.

In some other optional implementation manners of this embodiment, the connection information between each candidate skill word is determined based on each candidate skill word, sample information to be investigated, and inspection result labeling data, including: based on sample information to be investigated and investigation The result is marked with data to determine at least one group of sample data tuples; for each group of sample data tuples, the connection information between candidate skill words in the sample data tuples is determined; based on the number of candidate skill words in each group of sample data tuples The connection information between each candidate skill word is determined.

In this implementation manner, the sample information to be inspected may include information of positions to be inspected and information of objects to be inspected in a plurality of historical inspection data, and the inspection result labeling data may include historical interview result information in a plurality of historical inspection data. The execution body can determine at least one set of sample data tuples from the sample information to be investigated and the annotation data of the inspection results. For each group of sample data tuples, the sample data tuples include the information of the position to be investigated and the information of the object to be investigated that have a corresponding relationship. and historical interview result information.

For each group of sample data tuples, determining the connection information between the candidate skill words in the sample data tuple may include: for each information in each group of sample data tuples, using a preset sliding window to traverse the information, Based on the skill words in the information that are also in the preset sliding window, the connection information between the candidate skill words in the information is established; and, for each information in the same sample data tuple, the candidate skill words in each information are , to establish the connection relationship between candidate skill words across informative texts. Based on the connection relationship between candidate skill words within the information in each sample data tuple, and the connection relationship between candidate skill words across information texts, a preset skill word graph is generated. The information is the information of the position to be inspected, the information of the object to be inspected, and the historical interview result information.

Step 503: Input the preset skill word graph and sample information to be investigated into the model to be trained.

In this embodiment, the execution subject may input the above-mentioned skill word graph and sample information to be investigated into the model to be trained, so that the model to be trained outputs at least one sample skill word.

Step 504 , based on the preset skill word graph, determine the label representation information.

In this embodiment, the execution subject can obtain the label representation information by using a preset skill word graph. Among them, the label representation information is used to represent each skill word.

Step 505: Determine topic representation information based on a preset skill word graph and sample information to be investigated.

In this embodiment, the execution subject may map the preset skill word graph to a preset multi-dimensional space to obtain mapping data corresponding to the skill word graph. and mapping the sample information to be investigated to a preset multi-dimensional space to obtain mapping data corresponding to the sample to be investigated information. Afterwards, the execution subject can integrate the mapping data corresponding to the skill word graph and the mapping data corresponding to the sample information to be investigated to obtain the topic representation information. The topic representation information is used to represent the preset skill word graph and the text topic corresponding to the sample information to be investigated.

In some optional implementations of this embodiment, the following steps may also be performed: using a multi-layer perceptron to perform data processing on the mapping data corresponding to the skill word graph and the mapping data corresponding to the sample information to be investigated, to obtain integrated data, The integrated data is output for subsequent data analysis and processing based on the integrated data.

Step 506: Determine at least one sample skill word based on the tag representation information and the topic representation information.

In this embodiment, the execution of the protective body may determine at least one sample skill word based on matching the label representation information and the topic representation information. The sample skill words here can be skill words in the label representation information.

Step 507 , based on the at least one sample skill word and the inspection result labeling data, train the model to be trained, and obtain a trained skill word generation model.

In this embodiment, the skill word generation model may be a Bayesian latent variable model, or may be another machine learning model in the prior art, which is not limited in this embodiment. For the detailed description of step 507, please refer to the detailed description of step 403, which is not repeated here.

The model training method provided by the above-mentioned embodiments of the present disclosure can also extract skill words from the sample information to be investigated and the annotation data of the inspection result, and determine each sample data corresponding to the skill words extracted from the sample information to be inspected and the annotation data of the inspection result. Tuple, based on the relationship between each information in each sample data tuple and between each information, determine the relationship between skill words, get skill word graph, and use skill word graph to assist model training, which can improve the accuracy of the model. And, using the topic representation information and the label representation information to determine the sample skill words can improve the determination accuracy of the sample skill words.

Further referring to FIG. 6 , as an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an apparatus for processing skill information, and the apparatus embodiment corresponds to the method embodiment shown in FIG. 2 , The apparatus can be specifically applied to electronic equipment such as terminal equipment and servers.

As shown in FIG. 6 , the apparatus 600 for processing skill information in this embodiment includes: an information acquisition unit 601 , a skill word determination unit 602 and a skill word output unit 603 .

The information acquisition unit 601 is configured to acquire the information to be investigated.

The skill word determining unit 602 is configured to determine at least one skill word based on the information to be investigated and the skill word generation model that has been trained in advance.

The skill word output unit 603 is configured to output at least one skill word.

In some optional implementation manners of this embodiment, the information to be inspected includes information of a position to be inspected and information of an object to be inspected.

In some optional implementations of this embodiment, the skill word output unit 603 is further configured to: determine a target recruiting object based on at least one skill word and recruiting object information; send at least one skill word to the target recruiting object, so that The target recruiting object conducts a skill inspection of the object to be inspected based on at least one skill word.

It should be understood that the units 601 to 603 recorded in the apparatus 600 for processing skill information correspond to respective steps in the method described with reference to FIG. 2 . Therefore, the operations and features described above with respect to the method for processing skill information are also applicable to the apparatus 600 and the units included therein, and will not be repeated here.

Referring further to FIG. 7 , as an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a model training apparatus. The apparatus embodiment corresponds to the method embodiment shown in FIG. 4 . Specifically, the apparatus may It is used in electronic equipment such as terminal equipment and servers.

As shown in FIG. 7 , the model training apparatus 700 in this embodiment includes: a sample acquisition unit 701 , a sample skill word determination unit 702 , and a model training unit 703 .

The state obtaining unit 701 is configured to obtain sample information to be investigated and investigation result labeling data.

The sample skill word determining unit 702 is configured to determine at least one sample skill word based on a preset skill word graph, sample information to be investigated, and a model to be trained.

The model training unit 703 is configured to train the model to be trained based on the at least one sample skill word and the labeling data of the inspection result, and obtain a trained skill word generation model.

In some optional implementations of this embodiment, the method further includes: a skill word graph determining unit, configured to determine the skill word graph based on the sample information to be inspected and the labeling data of the inspection result.

In some optional implementations of this embodiment, the skill word map determining unit is further configured to: determine each candidate skill word from the sample information to be investigated and the labeling data of the inspection result; The information and inspection results are marked with data, and the connection information between each candidate skill word is determined; based on each candidate skill word and the connection information between each candidate skill word, a skill word map is determined.

In some optional implementations of this embodiment, the skill word graph determining unit is further configured to: determine at least one group of sample data tuples based on the sample information to be investigated and the inspection result labeling data; for each group of sample data tuples , determine the connection information between the candidate skill words in the sample data tuple; determine the connection information between each candidate skill word based on the connection information between the candidate skill words in each group of sample data tuples.

In some optional implementations of this embodiment, the sample information to be investigated includes sample information of positions to be investigated and sample information of objects to be investigated.

In some optional implementations of this embodiment, the sample skill word determination unit 702 is further configured to: input a preset skill word graph and sample information to be investigated into the model to be trained; based on the preset skill word graph, determine Label representation information; determine topic representation information based on a preset skill word graph and sample information to be investigated; determine at least one sample skill word based on label representation information and topic representation information.

It should be understood that the units 701 to 703 described in the model training apparatus 700 correspond to respective steps in the method described with reference to FIG. 4 . Therefore, the operations and features described above with respect to the model training method are also applicable to the apparatus 700 and the units included therein, and details are not described herein again.

In the technical solution of the present disclosure, the acquisition, storage and application of the user's personal information involved are all in compliance with the provisions of relevant laws and regulations, and do not violate public order and good customs.

According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

FIG. 8 shows a schematic block diagram of an example electronic device 800 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are by way of example only, and are not intended to limit implementations of the disclosure described and/or claimed herein.

As shown in FIG. 8 , the device 800 includes a computing unit 801 that can be executed according to a computer program stored in a read only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803 Various appropriate actions and handling. In the RAM 803, various programs and data necessary for the operation of the device 800 can also be stored. The computing unit 801 , the ROM 802 , and the RAM 803 are connected to each other through a bus 804 . An input/output (I/O) interface 805 is also connected to bus 804 .

Various components in the device 800 are connected to the I/O interface 805, including: an input unit 806, such as a keyboard, mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, an optical disk, etc. ; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 809 allows the device 800 to exchange information/data with other devices through a computer network such as the Internet and/or various telecommunication networks.

Computing unit 801 may be various general-purpose and/or special-purpose processing components with processing and computing capabilities. Some examples of computing units 801 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various specialized artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as a method for processing skill information or a model training method. For example, in some embodiments, a method for processing skill information or a model training method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 808 . In some embodiments, part or all of the computer program may be loaded and/or installed on device 800 via ROM 802 and/or communication unit 809 . When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the method for processing skill information or the model training method described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured by any other suitable means (eg, by means of firmware) to perform a method for processing skill information or a model training method.

Various implementations of the systems and techniques described herein above may be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips system (SOC), load programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpretable on a programmable system including at least one programmable processor that The processor, which may be a special purpose or general-purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device an output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer or other programmable data processing apparatus, such that the program code, when executed by the processor or controller, performs the functions/functions specified in the flowcharts and/or block diagrams. Action is implemented. The program code may execute entirely on the machine, partly on the machine, partly on the machine and partly on a remote machine as a stand-alone software package or entirely on the remote machine or server.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in connection with the instruction execution system, apparatus or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media may include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), fiber optics, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having a display device (eg, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user ); and a keyboard and pointing device (eg, a mouse or trackball) through which a user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (eg, visual feedback, auditory feedback, or tactile feedback); and can be in any form (including acoustic input, voice input, or tactile input) to receive input from the user.

The systems and techniques described herein may be implemented on a computing system that includes back-end components (eg, as a data server), or a computing system that includes middleware components (eg, an application server), or a computing system that includes front-end components (eg, a user's computer having a graphical user interface or web browser through which a user may interact with implementations of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system may be interconnected by any form or medium of digital data communication (eg, a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

A computer system can include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, a distributed system server, or a server combined with blockchain.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, the steps described in the present disclosure can be executed in parallel, sequentially, or in different orders. As long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, there is no limitation herein.

The above-mentioned specific embodiments do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure should be included within the protection scope of the present disclosure.

Claims (21)

1. A method for processing skill information, comprising:

acquiring information to be examined;

determining at least one skill word based on the information to be examined and a skill word generation model trained in advance;

outputting the at least one skill word.

2. The method of claim 1, wherein the information to be investigated comprises position information to be investigated and object information to be investigated.

3. The method of claim 1, wherein the outputting the at least one skill word comprises:

determining a target recruitment object based on the at least one skill word and the recruitment object information;

and sending the at least one skill word to the target recruitment object so that the target recruitment object performs skill investigation on the object to be investigated based on the at least one skill word.

4. A model training method, comprising:

obtaining sample to-be-examined information and investigation result marking data;

determining at least one sample skill word based on a preset skill word graph, the sample to-be-examined information and a to-be-trained model;

and training the model to be trained based on the at least one sample skill word and the survey result labeling data to obtain a trained skill word generation model.

5. The method of claim 4, further comprising:

and determining the skill word graph based on the sample to-be-examined information and the investigation result labeling data.

6. The method of claim 5, wherein the determining the skill word graph based on the sample to-be-examined information comprises:

determining each candidate skill word from the sample to-be-examined information and the investigation result labeling data;

determining connection information among the candidate skill words based on the candidate skill words, the sample to-be-examined information and the investigation result labeling data;

determining the skill word graph based on the candidate skill words and connection information between the candidate skill words.

7. The method of claim 6, wherein the determining connection information between the candidate skill words based on the candidate skill words, the sample review information, and the survey result annotation data comprises:

determining at least one group of sample data tuples based on the sample information to be examined and the survey result marking data;

for each group of sample data tuples, determining connection information among candidate skill words in the sample data tuples;

and determining the connection information among the candidate skill words based on the connection information among the candidate skill words in each group of sample data tuples.

8. The method according to claim 4, wherein the sample information to be examined comprises position sample information to be examined and object sample information to be examined.

9. The method of claim 4, wherein the determining at least one sample skill word based on a preset skill word graph, the sample scout information, and a model to be trained comprises:

inputting the preset skill word graph and the sample to-be-examined information into the model to be trained;

determining label representation information based on the preset skill word graph;

determining topic representation information based on the preset skill word graph and the sample to-be-examined information;

determining the at least one sample skill word based on the tag characterization information and the topic characterization information.

10. An apparatus for processing skill information, comprising:

an information acquisition unit configured to acquire information to be examined;

a skill word determination unit configured to determine at least one skill word based on the information to be examined and a pre-trained skill word generation model;

a skill word output unit configured to output the at least one skill word.

11. The apparatus of claim 10, wherein the information to be investigated comprises position information to be investigated and object information to be investigated.

12. The apparatus of claim 10, wherein the skill word output unit is further configured to:

determining a target recruitment object based on the at least one skill word and the recruitment object information;

and sending the at least one skill word to the target recruitment object so that the target recruitment object performs skill investigation on the object to be investigated based on the at least one skill word.

13. A model training apparatus comprising:

the sample acquisition unit is configured to acquire sample to-be-inspected information and inspection result marking data;

the sample skill word determining unit is configured to determine at least one sample skill word based on a preset skill word graph, the sample to-be-examined information and a to-be-trained model;

and the model training unit is configured to train the model to be trained on the basis of the at least one sample skill word and the investigation result marking data to obtain a trained skill word generation model.

14. The apparatus of claim 13, further comprising:

and the skill word graph determining unit is configured to determine the skill word graph based on the sample to-be-examined information and the investigation result labeling data.

15. The apparatus of claim 14, wherein the skill word graph determination unit is further configured to:

determining each candidate skill word from the sample to-be-examined information and the investigation result labeling data;

determining connection information among the candidate skill words based on the candidate skill words, the sample to-be-examined information and the investigation result labeling data;

determining the skill word graph based on the candidate skill words and connection information between the candidate skill words.

16. The apparatus of claim 15, wherein the skill word graph determination unit is further configured to:

determining at least one group of sample data tuples based on the sample information to be examined and the survey result marking data;

for each group of sample data tuples, determining connection information among candidate skill words in the sample data tuples;

and determining the connection information among the candidate skill words based on the connection information among the candidate skill words in each group of sample data tuples.

17. The apparatus of claim 13, wherein the sample to-be-examined information includes position to-be-examined sample information and object to-be-examined sample information.

18. The apparatus of claim 13, wherein the sample skill word determination unit is further configured to:

inputting the preset skill word graph and the sample to-be-examined information into the model to be trained;

determining label representation information based on the preset skill word graph;

determining topic representation information based on the preset skill word graph and the sample to-be-examined information;

determining the at least one sample skill word based on the tag characterization information and the topic characterization information.

19. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-9.

20. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-9.

21. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-9.

Images