CN115470352A

CN115470352A - Article quality determination method, device, equipment and medium

Info

Publication number: CN115470352A
Application number: CN202211130251.2A
Authority: CN
Inventors: 杨飞; 李国建; 洪进栋
Original assignee: Douyin Vision Co Ltd
Current assignee: Douyin Vision Co Ltd
Priority date: 2022-09-16
Filing date: 2022-09-16
Publication date: 2022-12-13

Abstract

The embodiment of the disclosure relates to a method, a device, equipment and a medium for determining article quality, wherein the method comprises the following steps: acquiring an article to be detected; inputting an article to be detected into a pre-constructed quality identification model to obtain a quality score; and performing quality grade conversion on the quality fraction to obtain the quality grade of the article to be detected. By adopting the technical scheme, the quality of the article to be tested is identified by adopting the quality identification model to obtain the quality score, and then the quality score is converted into the quality grade, so that the problem that the quality grade difference is inconsistent with the actual quality change of the article when the quality grade is directly evaluated is avoided, the accuracy of article quality determination is improved, and the output quality grade is beneficial to the rapid application of the subsequent other article service scenes.

Description

Article quality determination method, device, equipment and medium

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a medium for determining article quality.

Background

With the rapid development of computer technology and internet technology, reading various articles on the internet becomes an important way for users to acquire information or entertainment, and the quality of the articles needs to be identified.

In the related art, the quality recognition of the article can be determined by analyzing the writing quality of the author or analyzing the characteristics of the article content, but the above methods all have the problem that the accuracy of the quality of the recognized article is low.

Disclosure of Invention

In order to solve the technical problem, the present disclosure provides an article quality determination method, apparatus, device, and medium.

The embodiment of the disclosure provides an article quality determination method, which comprises the following steps:

acquiring an article to be detected;

inputting the article to be detected into a pre-constructed quality identification model to obtain a quality score;

and performing quality grade conversion on the quality scores to obtain the quality grade of the article to be detected.

An embodiment of the present disclosure further provides an article quality determining apparatus, where the apparatus includes:

the acquisition module is used for acquiring an article to be detected;

the score module is used for inputting the article to be detected into a pre-constructed quality identification model to obtain a quality score;

and the grade module is used for performing quality grade conversion on the quality scores to obtain the quality grade of the article to be tested.

An embodiment of the present disclosure further provides an electronic device, including: a processor; a memory for storing the processor-executable instructions; the processor is used for reading the executable instructions from the memory and executing the instructions to realize the article quality determination method provided by the embodiment of the disclosure.

The embodiment of the disclosure also provides a computer-readable storage medium, which stores a computer program for executing the article quality determination method provided by the embodiment of the disclosure.

Compared with the prior art, the technical scheme provided by the embodiment of the disclosure has the following advantages: the article quality determination scheme provided by the embodiment of the disclosure acquires an article to be detected; inputting an article to be detected into a pre-constructed quality identification model to obtain a quality score; and performing quality grade conversion on the quality fraction to obtain the quality grade of the article to be detected. By adopting the technical scheme, the quality of the article to be detected is identified by adopting the quality identification model to obtain the quality score, and then the quality score is converted into the quality grade, so that the problem that the quality grade difference is inconsistent with the actual quality change of the article when the quality grade is directly evaluated is avoided, the accuracy of determining the quality of the article is improved, and the output quality grade is beneficial to the rapid application of the subsequent service scenes of other articles.

Drawings

The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and features are not necessarily drawn to scale.

Fig. 1 is a schematic flow chart of an article quality determination method provided in some embodiments of the present disclosure;

fig. 2 is a schematic flow chart of another article quality determination method provided by some embodiments of the present disclosure;

FIG. 3 is a schematic illustration of article quality determination provided by some embodiments of the present disclosure;

fig. 4 is a schematic structural diagram of an article quality determination apparatus according to some embodiments of the present disclosure;

fig. 5 is a schematic structural diagram of an electronic device according to some embodiments of the present disclosure.

Detailed Description

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.

It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in a different order, and/or performed in parallel. Moreover, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect.

The term "include" and variations thereof as used herein are open-ended, i.e., "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions for other terms will be given in the following description.

It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence of the functions performed by the devices, modules or units.

It is noted that references to "a", "an", and "the" modifications in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that "one or more" may be used unless the context clearly dictates otherwise.

The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.

In the related technology, the quality identification of the article can be determined by analyzing the writing quality of the author, specifically, the writing quality score of the author can be judged from the grade, the authentication condition and the like of the author, and the writing quality score of the author is taken as the quality score of the current article, but the method has fluctuation in the quality of the text sent by the author, high creation quality of some subject matters and poor creation quality of other subject matters, if the quality score of the author represents the quality score of the article, a large error often exists, and in some scenes, effective author information is difficult to obtain, so that the method is not usable; or, the article is determined by analyzing the characteristics of the article content, but the problem of low accuracy of the quality of the identified article still exists in the method.

In order to solve the above problem, embodiments of the present disclosure provide an article quality determination method, which is described below with reference to specific embodiments.

Fig. 1 is a flowchart of an article quality determination method according to some embodiments of the present disclosure, which may be performed by an article quality determination apparatus, where the apparatus may be implemented by software and/or hardware, and may be generally integrated in an electronic device. As shown in fig. 1, the method includes:

step 101, obtaining an article to be detected.

The article to be tested may be at least one article that needs to be subjected to quality evaluation, the number of words of the article, the type of the article, the field of the article, the author of the document, and the like are not limited, and the article to be tested in the embodiment of the present disclosure takes the article with a longer number of words as an example, the article to be tested may be an article with a number of words greater than or equal to a word number threshold, and the word number threshold may be set according to an actual situation, for example, the word number threshold may be 1200, which is only an example.

In this embodiment, the article quality determining apparatus may acquire the article to be tested, and the specific acquisition source is not limited, for example, the article may be downloaded from the internet or uploaded locally.

And 102, inputting the article to be detected into a pre-constructed quality identification model to obtain a quality score.

The quality recognition model can be understood as a pre-trained deep learning model for quality evaluation or recognition of the article to be tested, and the specific deep learning model is not limited. The quality score can be an objective parameter for characterizing the quality of an article, and a larger quality score indicates better quality of the article.

In some embodiments, inputting the article to be tested into a pre-constructed quality identification model to obtain the quality score may include: determining an article vector of an article to be detected; and inputting the article vector into a quality identification model to obtain a quality score, wherein the quality identification model is determined based on a multilayer neural network.

In the embodiment of the present disclosure, the quality recognition model is exemplified by a multilayer neural network (MLP), which may be composed of an input layer, a hidden layer, and an output layer, each layer is composed of a plurality of neurons, the input layer transmits sample feature vectors to the next layer according to weights between connection points, such that one layer is transmitted forward, and the neural network f is trained to make f (X) and Y as close as possible given a sample feature and a target value pair (X, Y).

In some embodiments, determining the article vector of the article to be tested may include: dividing an article to be detected into a plurality of text blocks, and extracting a feature vector of each text block; and determining the average vector of the feature vectors of the text blocks as the article vector of the article to be detected.

The text blocks may be text units obtained by dividing or switching the articles to be tested, and the number of the text blocks may be multiple. The method and the device for segmenting the article to be detected can segment the article to be detected according to the preset word number as a unit, namely the word number of each text block is the preset word number, and the preset word number can be specifically determined according to actual conditions, for example, the preset word number can be 200. The article vector may be a characterization vector characterizing the article to be tested.

The feature vector may be a Representation vector of a text block, and is extracted through a language Representation model, in the embodiment of the present disclosure, the language Representation model is exemplified by a Bidirectional coding Representation model (BERT) based on a converter, and BERT is a pre-trained language Representation model in a natural language modeling algorithm, which emphasizes that a unidirectional language model or a method of shallow-layer splicing two unidirectional language models is not used for pre-training, but a new mask language model is used to enable generation of a deep Bidirectional language Representation.

Specifically, the article quality determining device may directly divide the article to be detected according to a preset number of words as a unit to obtain a plurality of text blocks, or may extract a text with a fixed number of words in front of the article to be detected when the number of words of the article to be detected is long, and divide the text according to the preset number of words as a unit to obtain a plurality of text blocks, where the fixed number of words may be set according to an actual situation, for example, the fixed number of words may be 2000. And then, inputting the plurality of text blocks into a language representation model to obtain a feature vector corresponding to each text block, summing and averaging all the feature vectors to obtain an average vector, and determining the average vector as an article vector of the article to be detected. And inputting the article vector into a pre-constructed quality identification model to obtain a quality score.

And 103, performing quality grade conversion on the quality scores to obtain the quality grade of the article to be tested.

The quality grade can be understood as a subjective parameter representing the quality of an article, and can be a parameter obtained by manually marking the quality of the article, and the quality grade can include multiple grades, in the embodiment of the disclosure, 5 grades are taken as an example, the higher the quality grade is, the better the quality of the article is, for example, the logic is more reasonable, the expression is clearer, the information amount is richer, and the like, the quality grades of 4 and 5 indicate that the quality of the article is high, the quality grade of 3 indicates that the quality of the article is general, and the quality grades of 1 and 2 indicate that the quality of the article is low. The quality grade and the quality score are two different article quality representation modes, the difference between the quality grade and the article quality is uneven, and the difference between the quality score and the article quality is even, so the quality score needs to be converted into the quality grade in the embodiment of the disclosure.

In some embodiments, the quality grade conversion of the quality score to obtain the quality grade of the article to be tested may include: determining a quality grade interval where the quality score is located; and inputting the quality scores and the quality grade interval into a preset score grade conversion formula, determining a quality grade continuous value, and determining the quality grade of the article to be tested based on the quality grade continuous value.

The quality class interval may be one interval composed of two quality classes, and the quality classes may include a plurality of quality classes, and the quality class interval may also include a plurality of quality class intervals, for example, when the quality classes are 5, the quality class intervals are 4. The score level conversion formula may be a preset formula for converting the quality score into a quality level continuous value, and may be a linear interpolation conversion formula in the embodiment of the present disclosure. The quality level continuation value may be a continuation value within a range of quality levels by which the quality level may be determined.

Specifically, after the article quality determination device inputs the article to be tested into a pre-constructed quality identification model to obtain a quality score, a quality grade interval in which the current quality score is located may be determined according to a corresponding relationship between the pre-determined quality score and a quality grade, where the quality grade interval may include a first quality grade and a second quality grade, the second quality grade is greater than the first quality grade, the first quality grade corresponds to the first quality score, the second quality grade corresponds to the second quality score, and the second quality score is greater than the first quality score; then, inputting the current quality fraction, and a first quality grade, a second quality grade, a first quality fraction and a second quality fraction corresponding to the quality grade interval into a preset grade conversion formula, obtaining a quality grade continuous value of the article to be tested by calculation, and obtaining the quality grade of the article to be tested by rounding the quality grade continuous value, for example, when the quality grade continuous value is 3.1, the quality grade is 3; the quality class continuation value is 3.8, and the quality class is 4.

Illustratively, when the fractional order conversion formula is a linear interpolation conversion formula, the linear interpolation conversion formula is expressed as

Z represents a quality grade continuous value of the article to be detected, the quality grade of the article to be detected is obtained by rounding Z, Y represents the quality fraction of the article to be detected, A represents a first quality grade of a quality grade interval in which the quality fraction of the article to be detected is located, B represents a second quality grade of the quality grade interval in which the quality fraction of the article to be detected is located, and the second quality grade isThe grade is larger than the first quality grade, t (A) represents a first quality score corresponding to the first quality grade, t (B) represents a second quality score corresponding to the second quality grade, and the second quality score is larger than the first quality score. The quality score of the article to be detected can be converted into a continuous value within the range of the quality grade by adopting a linear interpolation conversion formula, so that the corresponding quality grade is determined, and the method is convenient to be subsequently applied to other scenes.

Optionally, the score-grade conversion formula may also be an inverse formula of the grade-grade conversion formula, and the grade-grade conversion formula may be a formula for converting the quality grade into the quality score, and is required to be used when the quality recognition model is trained. When the score grade conversion formula is an inverse formula of the grade score conversion formula, the quality score of the article to be tested can be input into the inverse formula to obtain a corresponding quality grade continuous value, and the quality grade of the article to be tested is obtained by rounding the quality grade continuous value.

The quality grades are converted according to the quality scores to obtain the quality grades and/or the quality grade continuous values of the articles to be tested, and the quality grade continuous values are high in continuity, so that the quality grade continuous values can be used as the quality evaluation results of the articles to be tested in the scene with high continuity requirements according to actual service requirements, and the quality grades can be used as the quality evaluation results of the articles to be tested in the scene with low continuity requirements, so that the quality evaluation results of the articles in the embodiment of the disclosure can be suitable for various scenes, and the flexibility is high.

The article quality determining scheme provided by the embodiment of the disclosure acquires an article to be detected; inputting an article to be detected into a pre-constructed quality identification model to obtain a quality score; and performing quality grade conversion on the quality scores to obtain the quality grade of the article to be detected. By adopting the technical scheme, the quality of the article to be tested is identified by adopting the quality identification model to obtain the quality score, and then the quality score is converted into the quality grade, so that the problem that the quality grade difference is inconsistent with the actual quality change of the article when the quality grade is directly evaluated is avoided, the accuracy of article quality determination is improved, and the output quality grade is beneficial to the rapid application of the subsequent other article service scenes.

For example, fig. 2 is a schematic flow chart of another article quality determination method provided in some embodiments of the present disclosure, as shown in fig. 2, in a possible implementation manner, the quality identification model may be generated as follows:

step 201, sample data is obtained, wherein the sample data comprises a plurality of sample articles and sample quality grades corresponding to the sample articles.

The sample data may be a quantity with a large data size for model training, and the sample data in the embodiment of the present disclosure may include a plurality of sample articles and a sample quality level manually labeled in advance for each sample article.

Step 202, converting the sample quality grade of each sample article into a corresponding sample quality score.

Since the quality level difference is not uniform with the article quality, for example, a quality level from 3 to 4 means that the article quality is greatly improved, while a quality level from 2 to 3 means that the quality level is slightly improved, i.e., the article quality difference between 3 and 4 is significantly larger than the article quality difference between 2 and 3. Therefore, before training through sample data, the sample quality grades of the sample articles can be converted into corresponding sample quality scores, so that the difference of the sample quality scores and the difference of the corresponding article quality are ensured to be consistent as much as possible.

In some embodiments, converting the sample quality rating for each sample article to a corresponding sample quality score may include: and inputting the sample quality grade of each sample article into a preset grade score conversion formula to obtain a corresponding sample quality score, wherein the grade score conversion formula is an index function. The sample quality rating may be a quality rating corresponding to the sample article, and the sample quality score may be a quality score corresponding to the sample article.

Inputting the marked quality grade of each sample article into a preset grade score conversion formula to obtain a corresponding sample quality score; the grade score conversion formulaThe formula may be a formula for converting the quality grade into the quality score, a specific formula may be set according to actual conditions, and the embodiment of the present disclosure takes the grade score conversion formula as an example of an index function, which may be expressed as

Wherein t (x) represents the quality score corresponding to the quality grade x, x represents the quality grade, and when x specifically represents the sample quality grade, t (x) represents the sample quality score, and the corresponding relationship between the quality score and the quality grade obtained by the above formula is shown in table 1.

TABLE 1 corresponding relationship table of quality fraction and quality grade

Quality grade	Mass fraction of
		1	-20.09
2	-2.72
		3	0.00
4	20.09
		5	148.41

As shown in table 1, table 1 is a correspondence table of quality scores and quality ranks, in which the difference between the quality scores corresponding to quality ranks 1 and 3 is 17.73, and the difference between the quality scores corresponding to quality ranks 3 and 4 is 20.09, which are substantially equal. After the sample quality grades of the sample articles are converted into the sample quality scores, the difference of the sample quality scores and the difference of the corresponding article quality can be ensured to be consistent as much as possible.

And step 203, training the initial multilayer neural network by taking each sample article as input data and the sample mass fraction of each sample article as output data to obtain a quality identification model.

The initial multi-layer neural network may include four layers, the number of neurons in each layer may be 768, 256, 64, and 1, and the loss function of the initial multi-layer neural network may be determined according to actual situations, for example, a Mean Absolute Error (MAE) may be used as an example, and since the quality level of the artificially labeled sample includes noise, the training effect may be better by using the MAE.

The article quality determination device can train the initial multilayer neural network by taking each sample article in the sample data as input data and taking the sample quality score of each sample article as output data, and obtain the multilayer neural network with the loss function meeting the requirement, namely the quality recognition model.

According to the scheme, when the quality recognition model is trained, the quality grade of the article in the sample data is converted into the quality score through the preset conversion function and then the training is carried out, the consistency of the quality score and the article quality is ensured, the problem that the model accuracy is low due to the fact that the labeled quality grade is directly adopted for model training is avoided, the accuracy of the quality recognition model is improved, and the accuracy of follow-up article quality assessment is improved.

The article quality determination method of the embodiment of the present disclosure is further described below by a specific example. Fig. 3 is a schematic diagram of article quality determination provided in some embodiments of the present disclosure, and as shown in fig. 3, the diagram shows a complete process of article quality determination, which may specifically include a training phase 301 and an inference phase 302 in the diagram, where the training phase 301 is used to train the quality recognition model, and the inference phase 302 is used to input an article to be tested to obtain a corresponding quality level.

As shown in fig. 3, the training phase 301 may include three steps: extracting an expression vector, converting quality scores and training MLP by Bert, segmenting each sample article into texts to obtain a plurality of text blocks in the step of extracting the expression vector by Bert, extracting the expression vector of each text block by adopting a Bert model and calculating an average vector to obtain an article vector corresponding to each sample article; in the quality score conversion step, the sample quality grades of the sample articles are converted into sample quality scores, and the specific conversion mode refers to the embodiment; and training the MLP by taking the article vector of each sample article as input data and the corresponding sample mass score as output data, calculating fitting loss according to the article to the fitting value passing through the MLP, returning and updating parameters of the MLP, and determining the trained MLP as the quality recognition model.

The inference phase 302 may also include three steps: the method comprises the steps of Bert extraction of an expression vector, MLP inference and quality score inverse transformation, wherein a Bert model is adopted for determining a corresponding article vector for an article to be tested in the Bert extraction of the expression vector step, the article vector of the article to be tested is input into a quality recognition model trained in a training stage in the MLP inference step for inference, a quality score is output, the quality score is inversely transformed into a quality grade continuous value in the quality score inverse transformation step, and the quality grade of the article to be tested is obtained based on the quality grade continuous value.

In the scheme, starting from the text of the article, the quality score of the text of the article is evaluated through a training model, and for a long text article, the quality of the article mainly depends on the text instead of matching pictures and titles, so that the method is more reasonable and has smaller error; in addition, because the quality grade is inconsistent with the change of the article quality, the effect of directly using the quality grade training model is poor, and the quality grade is converted into the quality score for training in the scheme, so that the consistency of the quality score and the change of the article quality is ensured, and the accuracy of the quality recognition model is improved; meanwhile, after the quality of the article is evaluated by the application model, the quality grade continuous value and/or the quality grade can be converted through the inverse conversion of the quality fraction, so that the rapid application of other subsequent article service scenes is facilitated, and the universality is improved.

Fig. 4 is a schematic structural diagram of an article quality determination apparatus provided in some embodiments of the present disclosure, which may be implemented by software and/or hardware, and may be generally integrated in an electronic device. As shown in fig. 4, the apparatus includes:

an obtaining module 401, configured to obtain an article to be detected;

a score module 402, configured to input the article to be tested into a pre-constructed quality identification model to obtain a quality score;

and a grade module 403, configured to perform quality grade conversion on the quality scores to obtain quality grades of the articles to be tested.

Optionally, the score module 402 includes:

the vector unit is used for determining the article vector of the article to be detected;

and the determining unit is used for inputting the article vector into the quality identification model to obtain a quality score, wherein the quality identification model is determined based on a multilayer neural network.

Optionally, the vector unit is configured to:

dividing the article to be detected into a plurality of text blocks, and extracting the feature vector of each text block;

and determining the average vector of the feature vectors of the text blocks as the article vector of the article to be detected.

Optionally, the level module 403 is specifically configured to:

determining a quality grade interval where the quality score is located;

and inputting the quality scores and the quality grade interval into a preset score grade conversion formula, determining a quality grade continuous value, and determining the quality grade of the article to be tested based on the quality grade continuous value.

Optionally, the fractional rank conversion formula is a linear interpolation conversion formula.

Optionally, the apparatus further comprises a training module, configured to:

acquiring sample data, wherein the sample data comprises a plurality of sample articles and sample quality grades corresponding to the sample articles;

converting the sample quality grade of each sample article into a corresponding sample quality score;

and training the initial multilayer neural network by taking the sample articles as input data and the sample mass fraction of the sample articles as output data to obtain the quality identification model.

Optionally, the training module further includes:

and inputting the sample quality grade of each sample article into a preset grade score conversion formula to obtain a corresponding sample quality score, wherein the grade score conversion formula is an exponential function.

The article quality determining device provided by the embodiment of the disclosure can execute the article quality determining method provided by any embodiment of the disclosure, and has corresponding functional modules and beneficial effects of the executing method.

Embodiments of the present disclosure also provide a computer program product comprising a computer program/instructions, which when executed by a processor, implement the article quality determination method provided by any of the embodiments of the present disclosure.

Fig. 5 is a schematic structural diagram of an electronic device according to some embodiments of the present disclosure. Referring now specifically to fig. 5, a schematic diagram of an electronic device 500 suitable for use in implementing embodiments of the present disclosure is shown. The electronic device 500 in the disclosed embodiment may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), a vehicle terminal (e.g., a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. The electronic device shown in fig. 5 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.

As shown in fig. 5, electronic device 500 may include a processing means (e.g., central processing unit, graphics processor, etc.) 501 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 502 or a program loaded from a storage means 508 into a Random Access Memory (RAM) 503. In the RAM503, various programs and data necessary for the operation of the electronic apparatus 500 are also stored. The processing device 501, the ROM 502, and the RAM503 are connected to each other through a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.

Generally, the following devices may be connected to the I/O interface 505: input devices 506 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 507 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage devices 508 including, for example, magnetic tape, hard disk, etc.; and a communication device 509. The communication means 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. While fig. 5 illustrates an electronic device 500 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may be alternatively implemented or provided.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program containing program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 509, or installed from the storage means 508, or installed from the ROM 502. The computer program performs the above-described functions defined in the article quality determination method of the embodiment of the present disclosure when executed by the processing apparatus 501.

It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (HyperText Transfer Protocol), and may interconnect with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.

The computer readable medium may be embodied in the electronic device; or may be separate and not incorporated into the electronic device.

The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquiring an article to be detected; inputting the article to be detected into a pre-constructed quality identification model to obtain a quality score; and performing quality grade conversion on the quality scores to obtain the quality grade of the article to be detected.

Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including but not limited to an object oriented programming language such as Java, smalltalk, C + +, including conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in the embodiments of the present disclosure may be implemented by software or hardware. Wherein the name of an element does not in some cases constitute a limitation on the element itself.

The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems on a chip (SOCs), complex Programmable Logic Devices (CPLDs), and the like.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

It is understood that, before the technical solutions disclosed in the embodiments of the present disclosure are used, the type, the use range, the use scenario, etc. of the information related to the present disclosure should be informed to the user and obtain the authorization of the user in a proper manner according to the relevant laws and regulations.

The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other combinations of features described above or equivalents thereof without departing from the spirit of the disclosure. For example, the above features and the technical features disclosed in the present disclosure (but not limited to) having similar functions are replaced with each other to form the technical solution.

Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.

Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

1. A method for determining the quality of an article, comprising:

acquiring an article to be detected;

2. The method of claim 1, wherein inputting the article to be tested into a pre-constructed quality recognition model to obtain a quality score comprises:

determining an article vector of the article to be detected;

and inputting the article vector into the quality identification model to obtain a quality score, wherein the quality identification model is determined based on a multilayer neural network.

3. The method of claim 2, wherein determining the article vector for the article to be tested comprises:

4. The method of claim 1, wherein performing quality grade conversion on the quality scores to obtain quality grades of the articles to be tested comprises:

determining a quality grade interval where the quality score is located;

5. The method of claim 4, wherein the fractional order conversion formula is a linear interpolation conversion formula.

6. The method of claim 1, wherein the quality identification model is generated by:

and training an initial multilayer neural network by taking the sample articles as input data and the sample mass fraction of the sample articles as output data to obtain the quality recognition model.

7. The method of claim 6, wherein converting the sample quality rating for each of the sample articles to a corresponding sample quality score comprises:

8. An article quality determination apparatus, comprising:

the acquisition module is used for acquiring an article to be detected;

9. An electronic device, characterized in that the electronic device comprises:

a processor;

a memory for storing the processor-executable instructions;

the processor is configured to read the executable instructions from the memory and execute the instructions to implement the article quality determination method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that the storage medium stores a computer program for executing the article quality determination method of any one of claims 1-7.