阅读说明：本技术 工作技能的确定方法和装置、电子设备和存储介质 (Method and device for determining work skill, electronic equipment and storage medium ) 是由 徐成国 王硕 周星杰 杨康 于 2021-08-19 设计创作，主要内容包括：本申请提供了一种工作技能的确定方法和装置、电子设备和存储介质,该方法包括：获取通讯软件中存储的聊天记录；利用第一方案对聊天记录中的词组进行实体的提取,确定实体所属的技能类别,其中,第一方案用于对词组中的词汇按照输入时序进行处理,并从处理后的词汇中提取出实体；利用第二方案对聊天记录中的词组进行文本类别的归类,得到目标帐号的文本标签,其中,第二方案用于根据预设文本类别标签对词组进行文本标签归类的处理,预设文本类别标签用于指示词组所属的文本类别；根据技能类别和文本标签,确定目标对象所具备的工作技能。通过本申请,解决了相关技术中存在的人工整理浪费较多人力资源,并且很容易出现细节遗漏或整理出错的问题。(The application provides a method and a device for determining work skills, an electronic device and a storage medium, wherein the method comprises the following steps: obtaining a chat record stored in communication software; extracting entities from phrases in the chat records by using a first scheme, and determining skill categories to which the entities belong, wherein the first scheme is used for processing vocabularies in the phrases according to an input time sequence and extracting the entities from the processed vocabularies; classifying the word groups in the chat records according to the text type labels by using a second scheme to obtain the text labels of the target account, wherein the second scheme is used for performing text label classification processing on the word groups according to the preset text type labels, and the preset text type labels are used for indicating the text types of the word groups; and determining the working skill of the target object according to the skill category and the text label. Through the method and the device, the problem that manual sorting wastes more human resources in the related technology and details are omitted or sorting errors easily occurs is solved.)

1. A method of determining a work skill, the method comprising:

the method comprises the steps of obtaining chat records stored in communication software, wherein the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating work skills of a target object corresponding to the target account;

extracting an entity from the phrases in the chat records by using a first scheme, and determining the skill category of the entity, wherein the first scheme is used for processing the words in the phrases according to an input time sequence and extracting the entity from the processed words;

classifying phrases in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the phrases according to preset text labels, and the preset text labels are used for indicating the text types to which the phrases belong;

and determining the work skill of the target object according to the skill category and the text label.

2. The method of claim 1, wherein the extracting the entity from the phrase in the chat log by using the first scheme, and the determining the skill category to which the entity belongs comprises:

acquiring preset keywords in the phrases, wherein the preset keywords are vocabularies with the association degree with the skill category larger than a preset threshold;

coding the preset keywords by using a first model to obtain coded data;

decoding the coded data by using a second model to obtain the entity, wherein the entity is data used for representing that the target object has the working skill in the phrase;

and determining the skill category to which the entity belongs according to preset conditions, wherein the preset conditions are used for representing the corresponding relation between the entity and the skill category.

3. The method according to claim 2, wherein the encoding the preset keyword by using the first model to obtain encoded data comprises:

processing the preset keywords by using a word vector model to obtain word vectors;

and inputting the word vector into a first target network for coding to obtain the coded data, wherein the first model comprises the word vector model and the first target network.

4. The method of claim 2, wherein the decoding the encoded data using the second model to obtain the entity comprises:

extracting the coded data by using a first target network to obtain hidden characterization data;

and decoding the hidden characterization data by using a conditional random field model to obtain the entity, wherein the second model comprises the first target network and the conditional random field model.

5. The method of claim 4, wherein the classifying the phrases in the chat log by the second scheme to obtain the text label of the target account comprises:

training the hidden representation data by using a second target network, and outputting a probability numerical value of the phrase belonging to the text category;

and taking the text category corresponding to the maximum probability numerical value as the text label of the phrase.

6. The method of claim 5, wherein the training the hidden representation data with the second target network and outputting the probability value that the phrase belongs to the text category comprises:

performing feature extraction on the hidden characterization data by using the second target network to obtain first feature data;

inputting the first feature data into a pooling layer to obtain a first average pooling feature;

inputting the first average pooling feature into the second target network for feature extraction to obtain second feature data;

inputting the second feature data into the pooling layer to obtain a second average pooled feature;

and decoding the second average pooling feature by using a third target network to obtain a probability value of the phrase belonging to the text category.

7. The method according to any one of claims 1 to 6, wherein said determining a work skill possessed by the target object based on the skill category and the text label comprises:

and determining that the target object has the work skill under the condition that the work skill corresponding to the skill category is consistent with the work skill corresponding to the text label.

8. An apparatus for determining a work skill, the apparatus comprising:

the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring chat records stored in communication software, the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating work skills of a target object corresponding to the target account;

the extracting unit is used for extracting an entity from the phrase in the chat record by using a first scheme and determining the skill category to which the entity belongs, wherein the first scheme is used for processing the vocabulary in the phrase according to an input time sequence and extracting the entity from the processed vocabulary;

the classification unit is used for classifying the word groups in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the word groups according to preset text labels, and the preset text labels are used for indicating the text categories to which the word groups belong;

and the determining unit is used for determining the work skill of the target object according to the skill category and the text label.

9. An electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein said processor, said communication interface and said memory communicate with each other via said communication bus,

the memory for storing a computer program;

the processor for performing the method steps of work skill determination of any of claims 1 to 7 by running the computer program stored on the memory.

10. A computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to carry out the method steps of determining a work skill as claimed in any one of claims 1 to 7 when executed.

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for determining work skills, an electronic device, and a storage medium.

Background

With the popularization of communication tools in enterprises, the communication between the inside and the outside of employees increasingly depends on-line communication; moreover, because the scales of all companies are developed and increased, the cross-department contact becomes close after the number of the employees reaches a certain scale; however, the working skill characteristics of each employee have certain differences, and it becomes important how to accurately position the working skills of the employees and further contact corresponding colleagues according to the needs of the employees.

The mode that sums up staff's work skill at present is manual arrangement usually, can appear arranging time overlength like this, extravagant more manpower resources to the condition that the detail is omitted or the arrangement is made mistakes appears very easily.

Therefore, the related art has the problems that manual sorting wastes more human resources, and details are easy to miss or sorting errors occur.

Disclosure of Invention

The application provides a method and a device for determining work skills, electronic equipment and a storage medium, which are used for at least solving the problems that manual sorting wastes more human resources and details are easy to miss or sorting errors in the related technology.

According to an aspect of an embodiment of the present application, there is provided a work skill determination method, including: the method comprises the steps of obtaining chat records stored in communication software, wherein the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating work skills of a target object corresponding to the target account; extracting an entity from the phrases in the chat records by using a first scheme, and determining the skill category of the entity, wherein the first scheme is used for processing the words in the phrases according to an input time sequence and extracting the entity from the processed words; classifying phrases in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the phrases according to preset text labels, and the preset text labels are used for indicating the text types to which the phrases belong; and determining the work skill of the target object according to the skill category and the text label.

According to another aspect of the embodiments of the present application, there is also provided an apparatus for determining a work skill, including: the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring chat records stored in communication software, the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating work skills of a target object corresponding to the target account; the extracting unit is used for extracting an entity from the phrase in the chat record by using a first scheme and determining the skill category to which the entity belongs, wherein the first scheme is used for processing the vocabulary in the phrase according to an input time sequence and extracting the entity from the processed vocabulary; the classification unit is used for classifying the word groups in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the word groups according to preset text labels, and the preset text labels are used for indicating the text categories to which the word groups belong; and the determining unit is used for determining the work skill of the target object according to the skill category and the text label.

Optionally, the extraction unit comprises: the acquiring module is used for acquiring preset keywords in the phrases, wherein the preset keywords are vocabularies with the association degree between the preset keywords and the skill category larger than a preset threshold; the coding module is used for coding the preset keyword by using a first model to obtain coded data; the decoding module is used for decoding the coded data by using a second model to obtain the entity, wherein the entity is data used for representing that the target object has the working skill in the phrase; the first determining module is used for determining the skill category to which the entity belongs according to preset conditions, wherein the preset conditions are used for representing the corresponding relation between the entity and the skill category.

Optionally, the encoding module comprises: the processing subunit is used for processing the preset keywords by using a word vector model to obtain word vectors; and the coding subunit is used for inputting the word vector into a first target network for coding to obtain the coded data, wherein the first model comprises the word vector model and the first target network.

Optionally, the decoding module comprises: the first extraction subunit is used for extracting the coded data by using a first target network to obtain hidden representation data; a first decoding subunit, configured to decode the hidden characterization data by using a conditional random field model to obtain the entity, where the second model includes the first target network and the conditional random field model.

Optionally, the classifying unit includes: the training module is used for training the hidden representation data by utilizing a second target network and outputting a probability value of the phrase belonging to the text category; and the setting module is used for taking the text category corresponding to the maximum probability numerical value as the text label of the phrase.

Optionally, the training module comprises: the second extraction subunit is configured to perform feature extraction on the hidden characterization data by using the second target network to obtain first feature data; the first input subunit is used for inputting the first characteristic data into the pooling layer to obtain a first average pooling characteristic; the second input subunit is used for inputting the first average pooling feature into the second target network for feature extraction to obtain second feature data; a third input subunit, configured to input the second feature data into the pooling layer to obtain a second average pooling feature; and the second decoding subunit is configured to decode the second average pooling feature by using a third target network to obtain a probability value that the phrase belongs to the text category.

Optionally, the determining unit includes: and the second determination module is used for determining that the target object has the work skill under the condition that the work skill corresponding to the skill category is consistent with the work skill corresponding to the text label.

According to another aspect of the embodiments of the present application, there is also provided an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory communicate with each other through the communication bus; wherein the memory is used for storing the computer program; a processor for performing the method steps of the work skill determination in any of the above embodiments by running the computer program stored on the memory.

According to a further aspect of an embodiment of the present application, there is also provided a computer-readable storage medium having a computer program stored therein, wherein the computer program is configured to perform the method steps of determining a work skill in any of the above embodiments when the computer program is run.

The method can be applied to natural language processing in the technical field of deep learning, and in the embodiment of the method, chat records stored in communication software are acquired, wherein the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating the working skill of a target object corresponding to the target account; extracting entities from phrases in the chat records by using a first scheme, and determining skill categories to which the entities belong, wherein the first scheme is used for processing vocabularies in the phrases according to an input time sequence and extracting the entities from the processed vocabularies; classifying the word groups in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the word groups according to preset text labels, and the preset text labels are used for indicating the text types to which the word groups belong; and determining the working skill of the target object according to the skill category and the text label. The method and the device for determining the working skills of the employees automatically generate the working skills of the employees by combining entity recognition and classification of phrase text categories, so that the working skills of each employee can be quickly and accurately obtained, the technical effect of promoting cross-department communication to be more efficient and smooth is achieved, and the problems that more manpower resources are wasted in manual arrangement and details are easily omitted or arrangement errors occur in the related technology are solved.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.

FIG. 1 is a schematic diagram of a hardware environment for an alternative method of determining a work skill according to an embodiment of the present invention;

FIG. 2 is a schematic flow diagram of an alternative method of determining work skills according to an embodiment of the application;

FIG. 3 is a schematic overall flow diagram of an alternative method of determining a work skill according to an embodiment of the present application;

fig. 4 is a block diagram of an alternative apparatus for determining an operational skill according to an embodiment of the present disclosure;

fig. 5 is a block diagram of an alternative electronic device according to an embodiment of the present application.

Detailed Description

In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

According to one aspect of an embodiment of the present application, a method of determining a work skill is provided. Alternatively, in the present embodiment, the above-described work skill determination method may be applied to a hardware environment as shown in fig. 1. As shown in fig. 1, the terminal 102 may include a memory 104, a processor 106, and a display 108 (optional components). The terminal 102 may be communicatively coupled to a server 112 via a network 110, the server 112 may be configured to provide services (e.g., gaming services, application services, etc.) to the terminal or to clients installed on the terminal, and a database 114 may be provided on the server 112 or separate from the server 112 to provide data storage services to the server 112. Additionally, a processing engine 116 may be run in the server 112, and the processing engine 116 may be used to perform the steps performed by the server 112.

Alternatively, the terminal 102 may be, but is not limited to, a terminal capable of calculating data, such as a mobile terminal (e.g., a mobile phone, a tablet Computer), a notebook Computer, a PC (Personal Computer) Computer, and the like, and the network may include, but is not limited to, a wireless network or a wired network. Wherein, this wireless network includes: bluetooth, WIFI (Wireless Fidelity), and other networks that enable Wireless communication. Such wired networks may include, but are not limited to: wide area networks, metropolitan area networks, and local area networks. The server 112 may include, but is not limited to, any hardware device capable of performing computations.

In addition, in the present embodiment, the method for determining the work skill can also be applied to, but not limited to, a stand-alone processing device with a relatively high processing capability without data interaction. For example, the processing device may be, but is not limited to, a terminal device with a relatively high processing capability, that is, the operations in the above-mentioned work skill determination method may be integrated into a single processing device. The above is merely an example, and this is not limited in this embodiment.

Optionally, in this embodiment, the method for determining the work skill may be executed by the server 112, the terminal 102, or both the server 112 and the terminal 102. The method for determining the work skill of the terminal 102 according to the embodiment of the present application may be executed by a client installed thereon.

Taking the example of being operated in a server, fig. 2 is a schematic flowchart of an alternative work skill determination method according to an embodiment of the present application, and as shown in fig. 2, the flowchart of the method may include the following steps:

step S201, obtaining a chat record stored in the communication software, where the chat record is used to record a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used to indicate a work skill of a target object corresponding to the target account.

Optionally, with the popularization of communication tools inside enterprises, communication between the inside and outside of employees increasingly depends on online communication, and the server in the embodiment of the present application acquires chat records among the employees by using some commonly used communication software, such as wx software, and the like, where the chat records have some skill problem lists generated for the current target account, such as "is you good at xxx colleagues asking you to ask you to use a tensrflow framework? "," xxx colleagues hello, ask you for your office software office of office? "etc., wherein the skill problem list is to know which work skills a target object (i.e. a user using a target account) using a target account has itself, where the target account is an account recording a chat record in current communication software, and for example, the target account is: zhang San, etc.

Step S202, utilizing a first scheme to extract an entity from a phrase in the chat record, and determining the skill category to which the entity belongs, wherein the first scheme is used for processing the vocabulary in the phrase according to the input time sequence and extracting the entity from the processed vocabulary.

Optionally, the word in the word group in the first scheme may be processed according to a processing mode of an input time sequence, and then the processed word is extracted to complete the extraction of the entity of the word group in the chat record, and then the skill category to which the entity belongs is determined according to the extracted entity.

Before the entity extraction is performed by using the first scheme, the server may define in advance a scenario in which a skill list of a conversation employee is generated based on a chat record, the work skills possessed by the employee may be divided into N technical categories, for example, N may be 7, then the 7 categories may be "AI technology", "background technology", "front-end technology", "operation and maintenance technology", "business technology", "financial technology", and "operation technology", and then corresponding entity names are set for the 7 categories, and multiple specific entities may be included in each technical category, for example, "AI technology" includes "NLP", "CV", "machine learning", and "business technology" includes "maintenance", "welding", and the like. Thus, after the entity is extracted by the first scheme, the skill category corresponding to the entity can be directly found.

Step S203, classifying the word groups in the chat records according to a second scheme to obtain the text labels of the target account, wherein the number of the text types is at least one, the second scheme is used for performing text label classification processing on the word groups according to preset text type labels, and the preset text type labels are used for indicating the text types to which the word groups belong.

Optionally, the present application embodiment may adopt a second scheme to classify the word group in the chat record according to the text type, where the present application embodiment sets a preset text type label for indicating the text type to which the word group belongs in advance, and the second scheme is to perform text label classification on the word group according to the preset text type label, and more specifically, the second scheme mainly functions to perform text intent classification, and therefore, the second scheme may be a text type determination scheme based on a convolutional neural network.

And then, based on the acquired text type of the phrase, acquiring a text label type corresponding to the target account. It can be understood that the target account usually corresponds to the target object using the target account, so that the text type of the phrase in the target account is obtained, that is, the text label of the target account is determined. The number of the text categories is greater than or equal to one, and if there are 7 technology categories, the text categories are respectively "AI technology", "background technology", "front-end technology", "operation and maintenance technology", "business technology", "financial technology" and "operation technology".

And step S204, determining the work skill of the target object according to the skill category and the text label.

Optionally, after the server obtains the skill category determined by the first scheme and the text label determined by the second scheme, the skill category and the text label are compared to determine the work skill of the target object.

In the embodiment of the application, chat records stored in communication software are obtained, wherein the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating work skills of a target object corresponding to the target account; extracting entities from phrases in the chat records by using a first scheme, and determining skill categories to which the entities belong, wherein the first scheme is used for processing vocabularies in the phrases according to an input time sequence and extracting the entities from the processed vocabularies; classifying the word groups in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the word groups according to preset text labels, and the preset text labels are used for indicating the text types to which the word groups belong; and determining the working skill of the target object according to the skill category and the text label. The method and the device for determining the working skills of the employees automatically generate the working skills of the employees by combining entity recognition and classification of phrase text categories, so that the working skills of each employee can be quickly and accurately obtained, the technical effect of promoting cross-department communication to be more efficient and smooth is achieved, and the problems that more manpower resources are wasted in manual arrangement and details are easily omitted or arrangement errors occur in the related technology are solved.

As an alternative embodiment, the extracting the entity from the phrase in the chat record by using the first scheme, and the determining the skill category to which the entity belongs includes:

acquiring preset keywords in the phrases, wherein the preset keywords are vocabularies with the association degree with the skill category larger than a preset threshold;

coding preset keywords by using a first model to obtain coded data;

decoding the coded data by using a second model to obtain an entity, wherein the entity is data used for representing that the target object has work skills in the phrase;

and determining the skill category to which the entity belongs according to preset conditions, wherein the preset conditions are used for representing the corresponding relation between the entity and the skill category.

Optionally, the server in the embodiment of the present application needs to first obtain or extract preset keywords in the phrase, where the association between the preset keywords and each skill category is greater than a preset threshold, and entity information to be extracted may be found according to the preset keywords.

Further, the preset keywords are encoded by using a first model to obtain encoded data, where the first model may be a model or a tool, such as an encoder, that can process text data into encoded data.

And then decoding the encoded data by using a second model to obtain a corresponding entity, wherein the second model can be a model or a tool which can decode the encoded data into required entity data, such as a decoder.

And according to preset conditions preset by the server, obtaining the corresponding skill category according to the determined entity data. The preset conditions are recorded with the corresponding relations between the entities and the skill categories, for example, the "AI technology" category corresponds to the entities such as "NLP", "CV", "machine learning", and the "business technology" category corresponds to the entities such as "maintenance", "welding", and the like, and when the "CV" entity is obtained, the skill category to which the "CV" entity belongs can be determined to be the "AI technology" according to the corresponding relations.

According to the embodiment of the application, the skill category to which the entity belongs can be quickly obtained according to the extracted entity and the corresponding relation between the entity and the skill category.

As an alternative embodiment, the encoding processing on the preset keyword by using the first model to obtain the encoded data includes:

processing preset keywords by using a word vector model to obtain word vectors;

and inputting the word vector into a first target network for coding to obtain coded data, wherein the first model comprises a word vector model and the first target network.

Optionally, in the embodiment of the present application, an encoding process is performed based on a first target network bilst (Bi-directional Long Short-Term Memory neural network), and more specifically, an original text is set to be' i meeting nlp technology, do you meet? ', where the complete text corresponds to w₁，w₂，w₃...w₁₂Totaling 12 characters in length, for convenience hereinafter simply denoted w₁To w₄To describe the method, in practice, 12 characters are complete, as shown in fig. 3.

As in FIG. 3, the original text w₁，w₂，w₃，w₄And (4) coding the word vector in a word vector word2voc model to obtain a word vector, and then sending the obtained word vector into a BilSTM for coding to obtain coded data. Where word2voc is a model that maps text to dense vectors.

As an alternative embodiment, decoding the encoded data using the second model to obtain the entity includes:

extracting the coded data by using a first target network to obtain hidden representation data;

and decoding the hidden characterization data by using a conditional random field model to obtain an entity, wherein the second model comprises a first target network and the conditional random field model.

Optionally, the coded data is extracted through a first target network BilSTM to obtain hidden characterization data, then the hidden characterization data is decoded by using a conditional random field model (CRF) to obtain an entity sequence, at this time, a NER task training calculation process is formed according to a model constructed by the BilSTM + CRF, and a fitting target formula is calculated by using a standard CRF-based likelihood logarithm:

Loss_ner(θ_ner)＝-crf_log_likelihood(Dense(Hiddenrepresentation),ae_seq,length)

wherein theta is_nerThe parameters of the NER model corresponding to training are shown, wherein Dense is a full-connection network, ae _ seq is a correct marking result of an input sequence, and length is a length of a single sentence sequence.

As an alternative embodiment, classifying the word group in the chat record according to the second scheme to obtain the text label of the target account includes:

training the hidden representation data by using a second target network, and outputting a probability numerical value of the phrase belonging to the text category;

and taking the text category corresponding to the maximum probability numerical value as a text label of the phrase.

As shown in fig. 3, after the hidden representation data is obtained, a second target network is used to train the hidden representation data, where the second target network may be a text convolutional neural network, and the second target network may be used to output probability values of the current conversational phrase belonging to each text category, and then the text category corresponding to the maximum probability value is set as the text label of the current conversational phrase, that is, the skill category of the target object.

The embodiment of the application can carry out convolution training on the hidden representation data of the word group based on the second target network, can directly output the probability numerical value of the word group belonging to each text category, takes the text category with the maximum probability numerical value as the text label of the current word group, and takes the text label as the judgment basis for subsequently assisting in judging that the target object has the working skill, thereby improving the judgment accuracy.

As an alternative embodiment, the hidden representation data is trained by using a second target network, and outputting a probability value that a phrase belongs to a text category includes:

carrying out feature extraction on the hidden characterization data by using a second target network to obtain first feature data;

inputting the first feature data into a pooling layer to obtain a first average pooling feature;

inputting the first average pooling feature into a second target network for feature extraction to obtain second feature data;

inputting the second feature data into the pooling layer to obtain a second average pooling feature;

and decoding the second average pooling characteristic by using a third target network to obtain a probability numerical value of the phrase belonging to the text category.

Optionally, the specific step of outputting the probability value that the phrase belongs to the text category after the hidden representation data is trained by using the second target network is as follows: FIG. 3, (1) characterizes the data for concealment (x in FIG. 3)₁，x₂，x₃) Extracting characteristics through a CNN network (namely a second target network) to obtain first characteristic data; (2) Carrying out average pooling operation on the first characteristic data through a pooling layer AVG POOL layer to obtain a first average pooling characteristic of the first characteristic data; (3) performing characteristic purification on the first average pooling characteristic through one CNN convolution again to obtain second characteristic data; (4) the second feature data passes through the AVG POOL layer once again to obtain a second average pooling feature (x in FIG. 3)₄，x₅，x₆) (ii) a (5) And finally, decoding and judging the second average pooling feature through a feed-forward neural network (namely, a third target network), and performing probability judgment on seven skill categories in the embodiments, wherein the second target network outputs the maximum probability value of 7 category ranges by using softmax, namely the text label to which the current phrase belongs. During the training process, the fitting objective of the model is to minimize the negative likelihood logarithm of the probability, and the calculation formula is as follows:

wherein theta is_clsParameter, x, representing a hierarchy of classes_iRepresenting hidden token data vectors, z_iAnd representing the probability output of the classification network corresponding to each class through model prediction.

As an alternative embodiment, determining the work skill of the target object according to the skill category and the text label comprises:

and under the condition that the work skills corresponding to the skill category are consistent with the work skills corresponding to the text labels, determining that the target object has the work skills.

Optionally, the Loss obtained in the above embodiment_ner(θ_ner) And Loss_cls(θ_cls) And performing fusion to form multi-task training, wherein a fitting target formula of the multi-task training is as follows:

Loss_multi(θ_ner,θ_cls)＝αLoss_cls(θ_cls)+βLoss_ner(θ_ner)

wherein, alpha and beta are super-parameter weights used for adjusting the weight of loss, and the value range is dynamic adjustment of 0-1, and initially set to 0.5, 0.5, and the two hyperreferences are added to 1, which is done to help determine the skill entity as the skill of the current conversation person by classifying tasks, and avoid the skill designation of third party people, such as "i will not nlp technology, but zhang sanhui ", the" nlp technique "in such an example would be labeled by NER as" AI technique ", the intention of this phrase, however, is that the speaker will not be able to do this, and therefore this cannot be accounted for by the skills of the target object (conversation employee), therefore, the text classification by the text intention classification task is more accurate, and therefore, the skill label of the target object generated by the sentence is determined only when the label result (namely, the text label) of the text intention classification task is consistent with the NER labeling result (namely, the skill category to which the entity belongs).

The final goal of the multitask model training is to reduce Loss_multiAnd when the oscillation is within a certain lower threshold interval, the task model training is finished, such as oscillation within 1e-3 and 5e-3 threshold intervals.

According to the method and the device, the target object is determined to have the work skill only under the condition that the work skill corresponding to the skill category is consistent with the work skill corresponding to the text label, and therefore the accuracy of determining the work skill can be improved.

It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present application is not limited by the order of acts described, as some steps may occur in other orders or concurrently depending on the application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.

Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium (e.g., a ROM (Read-Only Memory)/RAM (Random Access Memory), a magnetic disk, an optical disk) and includes several instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the methods of the embodiments of the present application.

According to another aspect of the embodiments of the present application, there is also provided a work skill determination device for implementing the work skill determination method. Fig. 4 is a block diagram of an alternative apparatus for determining an operating skill according to an embodiment of the present application, and as shown in fig. 4, the apparatus may include:

an obtaining unit 401, configured to obtain a chat record stored in communication software, where the chat record is used to record a skill problem list generated by a target account, where the target account is an account used in the communication software, and the skill problem list is used to indicate a work skill of a target object corresponding to the target account;

an extracting unit 402, connected to the obtaining unit 401, configured to extract an entity from a phrase in the chat record by using a first scheme, and determine a skill category to which the entity belongs, where the first scheme is used to process a vocabulary in the phrase according to an input timing sequence, and extract the entity from the processed vocabulary;

a classifying unit 403, connected to the extracting unit 402, configured to classify the word groups in the chat record according to a second scheme to obtain text labels of the target account, where the number of the text categories is at least one, the second scheme is configured to perform text label classification processing on the word groups according to preset text category labels, and the preset text category labels are used to indicate the text categories to which the word groups belong;

and the determining unit 404 is connected to the classifying unit 403, and is configured to determine the work skills of the target object according to the skill category and the text label.

It should be noted that the obtaining unit 401 in this embodiment may be configured to execute the step S201, the extracting unit 402 in this embodiment may be configured to execute the step S202, the classifying unit 403 in this embodiment may be configured to execute the step S203, and the determining unit 404 in this embodiment may be configured to execute the step S204.

Through the module, chat records stored in communication software are obtained, wherein the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating work skills of a target object corresponding to the target account; extracting entities from phrases in the chat records by using a first scheme, and determining skill categories to which the entities belong, wherein the first scheme is used for processing vocabularies in the phrases according to an input time sequence and extracting the entities from the processed vocabularies; classifying the word groups in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the word groups according to preset text labels, and the preset text labels are used for indicating the text types to which the word groups belong; and determining the working skill of the target object according to the skill category and the text label. The method and the device for determining the working skills of the employees automatically generate the working skills of the employees by combining entity recognition and classification of phrase text categories, so that the working skills of each employee can be quickly and accurately obtained, the technical effect of promoting cross-department communication to be more efficient and smooth is achieved, and the problems that more manpower resources are wasted in manual arrangement and details are easily omitted or arrangement errors occur in the related technology are solved.

As an alternative embodiment, the extraction unit comprises: the acquisition module is used for acquiring preset keywords in the phrases, wherein the preset keywords are vocabularies with the association degree with the skill category larger than a preset threshold; the encoding module is used for encoding preset keywords by using a first model to obtain encoded data; the decoding module is used for decoding the coded data by using the second model to obtain an entity, wherein the entity is data used for representing that the target object has the working skill in the phrase; the first determining module is used for determining the skill category to which the entity belongs according to preset conditions, wherein the preset conditions are used for representing the corresponding relation between the entity and the skill category.

As an alternative embodiment, the encoding module comprises: the processing subunit is used for processing the preset keywords by using the word vector model to obtain word vectors; and the coding subunit is used for inputting the word vector into the first target network for coding to obtain coded data, wherein the first model comprises a word vector model and the first target network.

As an alternative embodiment, the decoding module comprises: the first extraction subunit is used for extracting the coded data by using a first target network to obtain hidden representation data; and the first decoding subunit is used for decoding the hidden characterization data by using the conditional random field model to obtain an entity, wherein the second model comprises a first target network and the conditional random field model.

As an alternative embodiment, the classifying unit includes: the training module is used for training the hidden representation data by utilizing a second target network and outputting a probability numerical value of the phrase belonging to the text category; and the setting module is used for taking the text category corresponding to the maximum probability numerical value as the text label of the phrase.

As an alternative embodiment, the training module comprises: the second extraction subunit is used for extracting the features of the hidden representation data by using a second target network to obtain first feature data; the first input subunit is used for inputting the first characteristic data into the pooling layer to obtain a first average pooling characteristic; the second input subunit is used for inputting the first average pooling feature into a second target network for feature extraction to obtain second feature data; the third input subunit is used for inputting the second characteristic data into the pooling layer to obtain a second average pooling characteristic; and the second decoding subunit is used for decoding the second average pooling feature by using a third target network to obtain a probability numerical value of the phrase belonging to the text category.

As an alternative embodiment, the determining unit includes: and the second determining module is used for determining that the target object has the work skill under the condition that the work skill corresponding to the skill category is consistent with the work skill corresponding to the text label.

It should be noted here that the modules described above are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the disclosure of the above embodiments. It should be noted that the modules described above as a part of the apparatus may be operated in a hardware environment as shown in fig. 1, and may be implemented by software, or may be implemented by hardware, where the hardware environment includes a network environment.

According to still another aspect of the embodiments of the present application, there is also provided an electronic device for implementing the method for determining work skills, where the electronic device may be a server, a terminal, or a combination thereof.

Fig. 5 is a block diagram of an alternative electronic device according to an embodiment of the present application, as shown in fig. 5, including a processor 501, a communication interface 502, a memory 503, and a communication bus 504, where the processor 501, the communication interface 502, and the memory 503 are communicated with each other through the communication bus 504, where,

a memory 503 for storing a computer program;

the processor 501, when executing the computer program stored in the memory 503, implements the following steps:

s1, obtaining chat records stored in the communication software, wherein the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating the work skill of a target object corresponding to the target account;

s2, extracting entities from the phrases in the chat records by using a first scheme, and determining the skill categories of the entities, wherein the first scheme is used for processing the words in the phrases according to the input time sequence and extracting the entities from the processed words;

s3, classifying phrases in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the phrases according to preset text labels, and the preset text labels are used for indicating the text types to which the phrases belong;

and S4, determining the work skill of the target object according to the skill category and the text label.

Alternatively, in this embodiment, the communication bus may be a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 5, but this is not intended to represent only one bus or type of bus.

The communication interface is used for communication between the electronic equipment and other equipment.

The memory may include RAM, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory. Alternatively, the memory may be at least one memory device located remotely from the processor.

As an example, as shown in fig. 5, the memory 503 may include, but is not limited to, an obtaining unit 401, an extracting unit 402, a classifying unit 403, and a determining unit 404 in the work skill determining device. In addition, other module units in the above work skill determination device may also be included, but are not limited to these, and are not described in detail in this example.

The processor may be a general-purpose processor, and may include but is not limited to: a CPU (Central Processing Unit), an NP (Network Processor), and the like; but also a DSP (Digital Signal Processing), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component.

In addition, the electronic device further includes: and the display is used for displaying the determination result of the work skill.

Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment is not described herein again.

It can be understood by those skilled in the art that the structure shown in fig. 5 is only an illustration, and the device implementing the method for determining work skills may be a terminal device, and the terminal device may be a terminal device such as a smart phone (e.g., an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD, and the like. Fig. 5 is a diagram illustrating a structure of the electronic device. For example, the terminal device may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in FIG. 5, or have a different configuration than shown in FIG. 5.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by a program instructing hardware associated with the terminal device, where the program may be stored in a computer-readable storage medium, and the storage medium may include: flash disk, ROM, RAM, magnetic or optical disk, and the like.

According to still another aspect of an embodiment of the present application, there is also provided a storage medium. Alternatively, in the present embodiment, the storage medium may be a program code for executing the method for determining a work skill.

Optionally, in this embodiment, the storage medium may be located on at least one of a plurality of network devices in a network shown in the above embodiment.

Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps:

s1, obtaining chat records stored in the communication software, wherein the chat records are used for recording a skill problem list generated by a target account, the target account is an account used in the communication software, and the skill problem list is used for indicating the work skill of a target object corresponding to the target account;

s2, extracting entities from the phrases in the chat records by using a first scheme, and determining the skill categories of the entities, wherein the first scheme is used for processing the words in the phrases according to the input time sequence and extracting the entities from the processed words;

s3, classifying phrases in the chat records according to a second scheme to obtain text labels of the target account, wherein the number of the text labels is at least one, the second scheme is used for performing text label classification processing on the phrases according to preset text labels, and the preset text labels are used for indicating the text types to which the phrases belong;

and S4, determining the work skill of the target object according to the skill category and the text label.

Optionally, the specific example in this embodiment may refer to the example described in the above embodiment, which is not described again in this embodiment.

Optionally, in this embodiment, the storage medium may include, but is not limited to: various media capable of storing program codes, such as a U disk, a ROM, a RAM, a removable hard disk, a magnetic disk, or an optical disk.

According to yet another aspect of an embodiment of the present application, there is also provided a computer program product or a computer program comprising computer instructions stored in a computer readable storage medium; the computer instructions are read by a processor of the computer device from a computer-readable storage medium, and the computer instructions are executed by the processor to cause the computer device to perform the method steps of determining a work skill in any of the embodiments described above.

The above-mentioned serial numbers of the embodiments of the present application are merely for description and do not represent the merits of the embodiments.

The integrated unit in the above embodiments, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in the above computer-readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in the form of a software product, stored in a storage medium, including instructions for causing one or more computer devices (which may be personal computers, servers, network devices, or the like) to execute all or part of the steps of the method for determining the work skills according to the embodiments of the present application.

In the above embodiments of the present application, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed client may be implemented in other manners. The above-described embodiments of the apparatus are merely illustrative, and for example, a division of a unit is merely a division of a logic function, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, and may also be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The foregoing is only a preferred embodiment of the present application and it should be noted that those skilled in the art can make several improvements and modifications without departing from the principle of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.