CN110807082B - Quality selective examination item determining method, system, electronic equipment and readable storage medium - Google Patents
Quality selective examination item determining method, system, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN110807082B CN110807082B CN201810866301.0A CN201810866301A CN110807082B CN 110807082 B CN110807082 B CN 110807082B CN 201810866301 A CN201810866301 A CN 201810866301A CN 110807082 B CN110807082 B CN 110807082B
- Authority
- CN
- China
- Prior art keywords
- evaluation
- evaluation data
- word
- quality
- article
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000011156 evaluation Methods 0.000 claims abstract description 252
- 230000011218 segmentation Effects 0.000 claims abstract description 88
- 238000012545 processing Methods 0.000 claims abstract description 19
- 239000013598 vector Substances 0.000 claims description 136
- 238000012549 training Methods 0.000 claims description 30
- 238000001914 filtration Methods 0.000 claims description 25
- 238000007689 inspection Methods 0.000 claims description 22
- 238000011157 data evaluation Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000010801 machine learning Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 description 17
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 230000014509 gene expression Effects 0.000 description 4
- 238000003908 quality control method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a quality selective examination item determining method, a system, electronic equipment and a readable storage medium, wherein the quality selective examination item determining method comprises the following steps: acquiring first evaluation data of an article; performing word segmentation processing on the first evaluation data to obtain a plurality of evaluation word segments; presetting a spot check project library; the selective examination item library stores a plurality of selective examination items to be detected and subject words corresponding to each selective examination item; calculating the similarity of each subject term and the evaluation segmentation term, and counting the frequency that the similarity is larger than a similarity threshold value; and selecting N items to be spot checked corresponding to the N subject words with the highest frequency as quality spot checked items, wherein N is a positive integer. According to the invention, the quality selective examination item is automatically selected according to the evaluation data of the user, so that dependence on professionals is eliminated, automatic and scientific selective examination is realized, the selective examination item depends on the evaluation data of the user, and the selective examination item is more representative and more reliable.
Description
Technical Field
The invention belongs to the field of big data processing, and particularly relates to a quality selective examination item determining method, a system, electronic equipment and a readable storage medium.
Background
The quality spot check of the articles is taken as an effective quality supervision and management method, is widely accepted and adopted by the national government quality supervision departments, industries and enterprises, and the Internet is taken as a platform and channel for article circulation, so that the spot check of the quality of the articles is also required. In general, an important link of the quality sampling inspection of the article is the selection of the sampling inspection item, in the existing sampling inspection item selection process, the selection mainly depends on the experience of quality control personnel, and according to the description of the article and the quality standard formulated by the national quality control department, the category of the article is judged, and the sampling inspection item of the sampling inspection or all the items are simply selected.
The sampling inspection method is seriously dependent on the related experience of quality control personnel, has higher professional requirements on people, and has poor popularization performance due to large number and frequent alternation of article types of an Internet platform, and meanwhile, the response time and the cost are difficult to control.
Disclosure of Invention
The invention aims to overcome the defects that the quality selective examination of internet articles mainly depends on the examination of quality control personnel to cause the selective examination efficiency to be reduced and the popularization is not available in the prior art, and provides a quality selective examination project determining method, a quality selective examination system, electronic equipment and a readable storage medium.
The invention solves the technical problems by the following technical scheme:
an item quality spot check item determination method, the item quality spot check item determination method comprising:
Acquiring first evaluation data of an article;
Performing word segmentation processing on the first evaluation data to obtain a plurality of evaluation word segments;
presetting a spot check project library; the selective examination item library stores a plurality of selective examination items to be detected and subject words corresponding to each selective examination item;
Calculating the similarity of each subject term and the evaluation segmentation term, and counting the frequency that the similarity is larger than a similarity threshold value;
and selecting N items to be spot checked corresponding to the N subject words with the highest frequency as quality spot checked items, wherein N is a positive integer.
Preferably, after the step of obtaining the first evaluation data of an article, the method for determining a quality sampling inspection item of the article further includes:
Judging whether the first evaluation data contains negative evaluation on the quality of the article, if so, filtering the first evaluation data which does not contain the negative evaluation on the quality of the article;
and in the step of performing word segmentation processing on the first evaluation data, performing word segmentation processing on the filtered first evaluation data.
Preferably, the step of determining whether the first evaluation data includes a negative evaluation of the quality of the article specifically includes:
acquiring second evaluation data of the target object within a preset time;
Assigning a target feature tag to the second evaluation data, the target feature tag being used to characterize whether the second evaluation data embodies that the target article has a quality problem;
creating a text information base for judging the quality of the article according to the second evaluation data;
Training according to the text information library and the target feature tag to obtain an article evaluation data judgment model;
And judging whether the first evaluation data comprises negative evaluation of the quality of the article or not by using the article evaluation data judging model.
Preferably, the step of creating a text information base for evaluating the quality of the article according to the evaluation data specifically includes:
presetting a word vector library; the word vector library stores a plurality of standard word segments and word vectors corresponding to each standard word segment;
performing word segmentation processing on the second evaluation data to obtain a plurality of segmented words;
Obtaining word segmentation vectors corresponding to the plurality of word segmentation words from the word vector library; the text information base comprises the word segmentation vector;
The step of obtaining the item evaluation data judgment model through training according to the text information library specifically comprises the following steps:
And inputting the word segmentation vector and the target feature label into a machine learning model as training samples, and training to obtain the article evaluation data judgment model.
Preferably, the step of judging whether the first evaluation data includes a negative evaluation of the quality of the article by using the article evaluation data evaluation model specifically includes:
acquiring evaluation word segmentation vectors corresponding to the plurality of evaluation word segmentation words from the word vector library;
Inputting the evaluation word vector into the article evaluation data evaluation model, and outputting a characteristic label of the first evaluation data; the feature tag is used to characterize whether the first evaluation data contains a negative evaluation of the quality of the item;
and judging whether the first evaluation data comprises negative evaluation of the quality of the article according to the characteristic tag judgment model.
Preferably, before the step of querying the word vector library for word vectors corresponding to the plurality of word segments, the quality sampling test item determining method further includes:
Filtering stop words in the plurality of word segments;
and in the step of inquiring word segmentation vectors corresponding to the plurality of word segments from the word vector library, obtaining the corresponding word segmentation vectors for the plurality of filtered word segments.
Preferably, the step of calculating the similarity between each subject term and the evaluation term specifically includes:
acquiring a subject word vector corresponding to the subject word from the word vector library;
and calculating cosine similarity of the subject term vector and the evaluation word vector based on a cosine similarity algorithm to serve as the similarity.
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the quality spot check item determination method described above when executing the computer program.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the quality spot check item determination method described above.
The system comprises a data acquisition module, a word segmentation module, a similarity calculation module, a frequency statistics module, a quality selective examination item selection module and a selective examination item library; the selective examination item library stores a plurality of selective examination items to be detected and subject words corresponding to each selective examination item;
The data acquisition module is used for acquiring first evaluation data of an article;
the word segmentation module is used for carrying out word segmentation processing on the first evaluation data to obtain a plurality of evaluation word segments;
The similarity calculation module is used for calculating the similarity of each subject term and the evaluation segmentation term and calling the frequency statistics module;
The frequency statistics module is used for counting the frequency that the similarity of each subject term and the evaluation term is greater than a similarity threshold;
The quality selective examination item selection module is used for selecting N to-be-selective examination items corresponding to the N subject words with the highest frequency as quality selective examination items, wherein N is a positive integer.
Preferably, the system for determining the item quality sampling inspection item further comprises a judging module and a filtering module;
The judging module is used for judging whether the first evaluation data comprise negative evaluation on the quality of the article, and if so, the evaluation data filtering module is called;
the evaluation data filtering module is used for filtering first evaluation data which does not contain negative evaluation on the quality of the article;
the word segmentation module is used for carrying out word segmentation on the filtered first evaluation data.
Preferably, the judging module comprises an evaluation data acquisition unit, a label giving unit, a text information base creating unit and an article evaluation data judging model training unit;
the evaluation data acquisition unit is used for acquiring second evaluation data of the target object within a preset time;
The label giving unit is used for giving a target characteristic label to the second evaluation data, wherein the target characteristic label is used for representing whether the second evaluation data represents that the target object has quality problems;
The text information base creation unit is used for creating a text information base for judging the quality of the article according to the second evaluation data;
the article evaluation data judgment model training unit is used for training according to the text information base and the target feature tag to obtain an article evaluation data judgment model;
The judging module is used for judging whether the first evaluation data contains negative evaluation on the quality of the article or not by using the article evaluation data judging model.
Preferably, the judging module further comprises a word segmentation unit and a word vector library; the word vector library stores a plurality of standard word segments and word vectors corresponding to each standard word segment;
the word segmentation unit is used for carrying out word segmentation processing on the second evaluation data to obtain a plurality of segmented words;
The text information base creation unit is used for obtaining word segmentation vectors corresponding to the plurality of word segmentation from the word vector base; the text information base comprises the word segmentation vector;
the article evaluation data evaluation model training unit is used for inputting the word segmentation vector and the target feature label into a machine learning model as training samples, and training to obtain the article evaluation data evaluation model.
Preferably, the judging module further comprises a word vector obtaining unit and a label output unit;
The word vector acquisition unit is used for acquiring evaluation word segmentation vectors corresponding to the plurality of evaluation word segmentation words from the word vector library;
The label output unit is used for inputting the evaluation word vector into the article evaluation data evaluation model and outputting the characteristic label of the first evaluation data; the feature tag is used to characterize whether the first evaluation data contains a negative evaluation of the quality of the item;
the judging module is used for judging whether the first evaluation data contains negative evaluation on the quality of the article according to the characteristic tag judgment model.
Preferably, the text information base creation module further comprises an stop word filtering unit;
the stop word filtering unit is used for filtering stop words in the plurality of word segments;
the word vector obtaining unit is used for obtaining corresponding word segmentation vectors for the filtered multiple word segmentation.
Preferably, the word vector obtaining unit is further configured to obtain a subject word vector corresponding to the subject word from the word vector library;
the similarity calculation module is used for calculating cosine similarity of the subject term vector and the evaluation word vector based on a cosine similarity algorithm to serve as the similarity.
The invention has the positive progress effects that: according to the invention, the quality selective examination item is automatically selected according to the evaluation data of the user, so that dependence on professionals is eliminated, automatic and scientific selective examination is realized, the selective examination item depends on the evaluation data of the user, and the selective examination item is more representative and more reliable.
Drawings
Fig. 1 is a flowchart of a method for determining a quality of an article item for spot inspection according to embodiment 1 of the present invention.
Fig. 2 is a flowchart of a method for determining a quality of an article item according to embodiment 2 of the present invention.
Fig. 3 is a specific flowchart of step 11 in the method for determining a quality of an article according to embodiment 2 of the present invention.
Fig. 4 is a specific flowchart of step 113 in the method for determining the item quality spot check item according to embodiment 2 of the present invention.
Fig. 5 is a specific flowchart of step 115 in the method for determining the quality of an article according to embodiment 2 of the present invention.
Fig. 6 is a specific flowchart of step 113 in the method for determining the item quality spot check item according to embodiment 3 of the present invention.
Fig. 7 is a specific flowchart of step 40 in the method for determining the item quality spot check item according to embodiment 3 of the present invention.
Fig. 8 is a schematic structural diagram of an electronic device according to embodiment 4 of the present invention.
Fig. 9 is a schematic block diagram of an item quality spot check item determination system according to embodiment 6 of the present invention.
Fig. 10 is a block diagram of an item quality spot check item determination system according to embodiment 7 of the present invention.
Fig. 11 is a schematic diagram of a judging module in the system for determining a quality of an article according to embodiment 7 of the present invention.
Fig. 12 is a schematic block diagram of a judging module in the system for determining a quality of an article according to embodiment 8 of the present invention.
Detailed Description
The invention is further illustrated by means of the following examples, which are not intended to limit the scope of the invention.
Example 1
As shown in fig. 1, the method for determining the item quality sampling inspection item includes:
Step 10, acquiring first evaluation data of an article; in the embodiment, the evaluation and refund repair content in the time range of the past 3 months corresponding to the article to be inspected is selected as the evaluation data, so that the content can better reflect the recent quality problem of the article;
step 20, performing word segmentation processing on the first evaluation data to obtain a plurality of evaluation word segments;
step 30, presetting a spot check project library; the selective examination item library stores a plurality of selective examination items to be detected and subject words corresponding to each selective examination item;
step 40, calculating the similarity of each subject term and the evaluation segmentation term, and counting the frequency that the similarity is larger than a similarity threshold value;
step 50, selecting N items to be spot checked corresponding to N subject words with highest frequency as quality spot checked items; n is a positive integer.
In the recommendation process, this embodiment may employ: arranging all the to-be-sampled items in descending order according to the frequency value, taking 5 before ranking, and setting limitation on the frequency, for example, the to-be-sampled items with the frequency more than 10 are required, and the threshold can be flexibly adjusted according to the actual application condition to be used as the final quality sampled items;
according to the embodiment, the quality selective examination item is automatically selected according to the evaluation data of the user, so that dependence on professionals is eliminated, automatic and scientific selective examination is realized, the selective examination item depends on the evaluation data of the user, the representativeness is better, and the selective examination item is more reliable.
Example 2
The method for determining the quality of the article according to the present embodiment is further improved on the basis of embodiment 1, as shown in fig. 2, after step 10, the method for determining the quality of the article further includes:
Step 11, judging whether the first evaluation data contains negative evaluation on the quality of the article, if so, executing step 12; if not, the data does not need to be filtered;
step 12, filtering out first evaluation data which does not contain negative evaluation of the quality of the article;
Further, replacing step 20 with step 20-1 specifically includes:
Step 20-1, performing word segmentation processing on the filtered first evaluation data to obtain a plurality of evaluation word segments;
It should be noted that, in general, the purpose of the quality sampling inspection is to be able to effectively monitor and manage the articles, especially the problematic articles, so in this embodiment, the comment data is screened and filtered in advance, and the sampling inspection is focused on the articles with quality problems, where, as shown in fig. 3, step 11 specifically includes:
Step 111, obtaining second evaluation data of the target object within a preset time;
Step 112, the second evaluation data is endowed with a target feature label; the target feature tag is used for representing whether the second evaluation data represents that the target object has a quality problem or not; here, the calibration of the target feature tag may be performed by picking part of the data from the existing evaluation data to perform manual marking, and marking whether the target feature tag belongs to a quality problem, for example, if the target feature tag belongs to an object quality problem, the target feature tag is marked as 1, otherwise, the target feature tag is marked as 0;
Step 113, creating a text information base for judging the quality of the article according to the second evaluation data;
step 114, training according to the text information library and the target feature label to obtain an article evaluation data judgment model;
Step 115, judging whether the first evaluation data includes a negative evaluation of the quality of the article by using the article evaluation data judging model.
Further, in this embodiment, as shown in fig. 4, step 113 specifically includes:
Step 1131, presetting a word vector library; the word vector library stores a plurality of standard word segments and word vectors corresponding to each standard word segment;
step 1132, performing word segmentation processing on the second evaluation data to obtain a plurality of segmented words;
Step 1133, obtaining word segmentation vectors corresponding to the plurality of word segmentation from a word vector library; the text information base comprises the word segmentation vector;
in step 114, the word segmentation vector and the target feature label are input into a machine learning model as training samples, and the article evaluation data judgment model is obtained through training.
It should be noted that, the term vector refers to using a multidimensional array to represent a term, and by using the term vector, the distance between the adjacent terms can be closer when the cosine distance is calculated, and the generation of the term vector has a relatively mature open source implementation technology. The generation of the word vector library in this embodiment may divide words and learn word vector expression modes of each word for all feedback text contents under each category through collected item category information, item description information, item feedback information and the like, and finally generate a word vector library corresponding to each item category.
Further, after training to obtain the item evaluation data evaluation model, as shown in fig. 5, step 115 specifically includes:
Step 1151, obtaining an evaluation word segmentation vector corresponding to a plurality of evaluation word segments from a word vector library;
Step 1152, inputting the evaluation word segmentation vector into an article evaluation data evaluation model, and outputting a feature tag of the first evaluation data; the feature tag is used to characterize whether the first evaluation data contains a negative evaluation of the quality of the item;
Step 1153, determining whether the first evaluation data includes a negative evaluation of the quality of the article according to the feature tag evaluation model.
In this embodiment, after the second comment data is segmented, a corresponding segmented word vector is obtained based on a word vector library, then the segmented word vector and the target feature tag are used as training corpus to train to obtain an article evaluation data judgment model, and then the first evaluation data is judged based on the article evaluation data judgment model.
Example 3
The method for determining the quality of the article for the spot check item of the present embodiment is further improved on the basis of embodiment 2, as shown in fig. 6, since a part of the user evaluation data contains many irregular expressions, punctuation marks, invalid contents, etc., after the evaluation data is segmented, such as stop words and symbols are removed to improve the accuracy of the model, therefore, before step 1133, step 113 further includes:
step 1134, filtering stop words in the plurality of word segments;
further, step 1133 is replaced with step 1133-1, specifically comprising:
Step 1133-1, obtaining corresponding word segmentation vectors from the plurality of filtered word segmentation vectors in the word vector library;
In addition, the term vector library may be used to query the vector representation of the subject term and the evaluation word, so that the degree of adjacency between the two terms may be obtained by calculating the cosine similarity of the vector, specifically, as shown in fig. 7, step 40 specifically includes:
step 401, acquiring a subject word vector corresponding to a subject word from a word vector library;
And step 402, calculating cosine similarity of the subject term vector and the evaluation word vector based on a cosine similarity algorithm to serve as similarity.
In this embodiment, in a specific implementation process, the cosine distance between each word and each subject word in the evaluation data can be calculated according to the subject word involved in the specific quality project requirements defined in the national and industry quality standards, if the cosine distance between a certain word and the subject word is greater than a certain threshold (for example, the set threshold is 0.8), it is indicated that the problem of the record feedback is close to the subject word in a very large probability, so the subject count is increased by 1, and finally, the count number corresponding to each quality subject can be obtained.
Example 4
An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the method of item quality spot check item determination of any one of embodiments 1 to 3 when the computer program is executed.
Fig. 8 is a schematic structural diagram of an electronic device according to embodiment 4 of the present invention. Fig. 8 illustrates a block diagram of an exemplary electronic device 90 suitable for use in implementing embodiments of the present invention. The electronic device 90 shown in fig. 8 is merely an example, and should not be construed as limiting the functionality and scope of use of embodiments of the present invention.
As shown in fig. 8, the electronic device 90 may be embodied in the form of a general purpose computing device, which may be a server device, for example. Components of the electronic device 90 may include, but are not limited to: at least one processor 91, at least one memory 92, a bus 93 connecting the different system components, including the memory 92 and the processor 91.
The bus 93 includes a data bus, an address bus, and a control bus.
The memory 92 may include volatile memory such as Random Access Memory (RAM) 921 and/or cache memory 922, and may further include Read Only Memory (ROM) 923.
Memory 92 may also include a program tool 925 having a set (at least one) of program modules 924, such program modules 924 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment.
The processor 91 executes various functional applications and data processing by running a computer program stored in the memory 92.
The electronic device 90 may also communicate with one or more external devices 94 (e.g., keyboard, pointing device, etc.). Such communication may occur through an input/output (I/O) interface 95. Also, the electronic device 90 may communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet, through a network adapter 96. The network adapter 96 communicates with other modules of the electronic device 90 via the bus 93. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with electronic device 90, including, but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, data backup storage systems, and the like.
It should be noted that although several units/modules or sub-units/modules of an electronic device are mentioned in the above detailed description, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more units/modules described above may be embodied in one unit/module in accordance with embodiments of the present application. Conversely, the features and functions of one unit/module described above may be further divided into ones that are embodied by a plurality of units/modules.
Example 5
A computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method for determining a quality of an item for spot inspection of any one of embodiments 1-3.
More specifically, among others, readable storage media may be employed including, but not limited to: portable disk, hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device, or any suitable combination of the foregoing.
In a possible implementation manner, the present invention may also be implemented in the form of a program product, which comprises program code for causing a terminal device to carry out the steps of implementing the method for determining a quality of an item of choice as described in any of the embodiments 1-3, when said program product is run on the terminal device.
Wherein the program code for carrying out the invention may be written in any combination of one or more programming languages, which program code may execute entirely on the user device, partly on the user device, as a stand-alone software package, partly on the user device and partly on the remote device or entirely on the remote device.
Example 6
As shown in fig. 9, the system for determining the quality of the article for sampling inspection comprises a data acquisition module 1, a word segmentation module 2, a similarity calculation module 3, a frequency statistics module 4, a quality for sampling inspection item selection module 5 and a sampling inspection item library 6; the spot check item library 6 stores a plurality of spot check items to be detected and subject words corresponding to the spot check items to be detected;
The data acquisition module 1 is used for acquiring first evaluation data of an article; in the embodiment, the evaluation and refund repair content in the time range of the past 3 months corresponding to the article to be inspected is selected as the evaluation data, so that the content can better reflect the recent quality problem of the article;
The word segmentation module 2 is used for carrying out word segmentation processing on the first evaluation data to obtain a plurality of evaluation word segments;
the similarity calculation module 3 is configured to calculate the similarity between each subject term and the evaluation segmentation term, and call the frequency statistics module 4;
The frequency statistics module 4 is used for counting the frequency that the similarity of each subject term and the evaluation segmentation term is greater than a similarity threshold;
The quality selective examination item selection module 5 is configured to select N to-be-selective examination items corresponding to the N subject words with the highest frequency as quality selective examination items, where N is a positive integer.
In the spot check item determination process, the present embodiment may employ: arranging all the to-be-sampled items in descending order according to the frequency value, taking 5 before ranking, and setting limitation on the frequency, for example, the to-be-sampled items with the frequency more than 10 are required, and the threshold can be flexibly adjusted according to the actual application condition to be used as the final quality sampled items;
according to the embodiment, the quality selective examination item is automatically selected according to the evaluation data of the user, so that dependence on professionals is eliminated, automatic and scientific selective examination is realized, the selective examination item depends on the evaluation data of the user, the representativeness is better, and the selective examination item is more reliable.
Example 7
The system for determining the quality of the article for the spot check in this embodiment is further improved on the basis of embodiment 6, as shown in fig. 10, and the system for determining the quality of the article for the spot check further includes a judging module 7 and a filtering module 8;
The judging module 7 is configured to judge whether the first evaluation data includes a negative evaluation on the quality of the article, and if yes, call the evaluation data filtering module 8;
the evaluation data filtering module 8 is configured to filter out first evaluation data that does not include a negative evaluation of the quality of the item;
the word segmentation module 2 is used for carrying out word segmentation processing on the filtered first evaluation data.
In general, the objective of the quality sampling inspection is to effectively monitor and manage the articles, especially the articles with problems need to be emphasized, so in this embodiment, the comment data is screened and filtered in advance, and the sampling inspection is focused on the articles with quality problems, specifically, as shown in fig. 11, the judging module 7 includes an evaluation data acquiring unit 71, a label giving unit 72, a text information base creating unit 73 and an article evaluation data evaluating model training unit 74;
The evaluation data acquisition unit 71 is configured to acquire second evaluation data of the target object within a preset time;
The tag assigning unit 72 is configured to assign a target feature tag to the second evaluation data, where the target feature tag is used to characterize whether the second evaluation data represents that the target article has a quality problem; here, the calibration of the target feature tag may be performed by picking part of the data from the existing evaluation data to perform manual marking, and marking whether the target feature tag belongs to a quality problem, for example, if the target feature tag belongs to an object quality problem, the target feature tag is marked as 1, otherwise, the target feature tag is marked as 0;
The text information base creation unit 73 is configured to create a text information base for evaluating the quality of the article based on the second evaluation data;
The article evaluation data evaluation model training unit 74 is configured to train to obtain an article evaluation data evaluation model according to the text information base and the target feature tag;
the judging module 7 is configured to judge whether the first evaluation data includes a negative evaluation of the quality of the article by using the article evaluation data evaluation model.
Referring to fig. 11, the judging module 7 further includes a word segmentation unit 75 and a word vector library 76; the word vector library 76 stores a plurality of standard word segments and word vectors corresponding to each standard word segment;
the word segmentation unit 75 is configured to perform word segmentation processing on the second evaluation data to obtain a plurality of word segments;
The text information base creation unit 73 is configured to acquire word segmentation vectors corresponding to the plurality of word segments from the word vector base 76; the text information base comprises the word segmentation vector;
The article evaluation data evaluation model training unit 74 is configured to input the word segmentation vector and the target feature label as training samples into a machine learning model, and train to obtain the article evaluation data evaluation model.
It should be noted that, the term vector refers to using a multidimensional array to represent a term, and by using the term vector, the distance between the adjacent terms can be closer when the cosine distance is calculated, and the generation of the term vector has a relatively mature open source implementation technology. The generation of the word vector library 76 in this embodiment may divide words and learn the word vector expression mode of each word for all the feedback text contents under each category through the collected item category information, item description information, item feedback information and the like, and finally generate the word vector library 76 corresponding to each item category.
In this embodiment, referring to fig. 11, the judging module 7 further includes a word vector obtaining unit 77 and a tag output unit 78;
The word vector obtaining unit 77 is configured to obtain, from the word vector library 76, evaluation word vectors corresponding to the plurality of evaluation word segments;
The tag output unit 78 is configured to input the evaluation word vector into the item evaluation data evaluation model, and output a feature tag of the first evaluation data; the feature tag is used to characterize whether the first evaluation data contains a negative evaluation of the quality of the item;
the judging module 7 is configured to judge whether the first evaluation data includes a negative evaluation of the quality of the article according to the feature tag evaluation model.
In this embodiment, after the second comment data is segmented, a corresponding segmented word vector is obtained based on the word vector library 76, then the segmented word vector and the target feature tag are trained as training corpus to obtain an item evaluation data evaluation model, and then the first evaluation data is judged based on the item evaluation data evaluation model.
Example 8
The system method for determining the quality of the article and the spot inspection project in this embodiment is further improved on the basis of embodiment 6, and since a part of the user evaluation data contains many irregular expressions, punctuation marks, invalid contents and the like, after the evaluation data is segmented, stop words and symbols are removed to improve the accuracy of the model, as shown in fig. 12, the judging module further includes a stop word filtering unit 79;
the stop word filtering unit 79 is configured to filter stop words in the plurality of word segments;
the word vector obtaining unit 77 is configured to obtain corresponding word vectors for the plurality of filtered word segments.
In addition, based on the word vector library 76, the vector representations of the subject words and the evaluation segmentation words can be queried, so that the degree of adjacency between two words, specifically, can be obtained by cosine similarity calculation of the vectors:
the word vector obtaining unit 77 is further configured to obtain a subject word vector corresponding to the subject word from the word vector library 76;
the similarity calculation module is used for calculating cosine similarity of the subject term vector and the evaluation word vector based on a cosine similarity algorithm to serve as the similarity.
In this embodiment, in a specific implementation process, the cosine distance between each word and each subject word in the evaluation data can be calculated according to the subject word involved in the specific quality project requirements defined in the national and industry quality standards, if the cosine distance between a certain word and the subject word is greater than a certain threshold (for example, the set threshold is 0.8), it is indicated that the problem of the record feedback is close to the subject word in a very large probability, so the subject count is increased by 1, and finally, the count number corresponding to each quality subject can be obtained.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the principles and spirit of the invention, but such changes and modifications fall within the scope of the invention.
Claims (12)
1. The method for determining the item of the selective examination of the quality of the article is characterized by comprising the following steps:
Acquiring first evaluation data of an article;
Judging whether the first evaluation data contains negative evaluation on the quality of the article, if so, filtering the first evaluation data which does not contain the negative evaluation on the quality of the article;
Performing word segmentation processing on the filtered first evaluation data to obtain a plurality of evaluation word segments;
presetting a spot check project library; the selective examination item library stores a plurality of selective examination items to be detected and subject words corresponding to each selective examination item;
Calculating the similarity of each subject term and the evaluation segmentation term, and counting the frequency that the similarity is larger than a similarity threshold value;
Selecting N items to be spot checked corresponding to the N subject words with the highest frequency as quality spot checked items, wherein N is a positive integer;
The step of determining whether the first evaluation data includes a negative evaluation of the quality of the item specifically includes:
acquiring second evaluation data of the target object within a preset time;
Assigning a target feature tag to the second evaluation data, the target feature tag being used to characterize whether the second evaluation data embodies that the target article has a quality problem;
creating a text information base for judging the quality of the article according to the second evaluation data;
Training according to the text information library and the target feature tag to obtain an article evaluation data judgment model;
And judging whether the first evaluation data comprises negative evaluation of the quality of the article or not by using the article evaluation data judging model.
2. The method for determining a quality of an item according to claim 1, wherein the step of creating a text information base for evaluating the quality of the item according to the second evaluation data comprises:
presetting a word vector library; the word vector library stores a plurality of standard word segments and word vectors corresponding to each standard word segment;
performing word segmentation processing on the second evaluation data to obtain a plurality of segmented words;
Obtaining word segmentation vectors corresponding to the plurality of word segmentation words from the word vector library; the text information base comprises the word segmentation vector;
The step of obtaining the item evaluation data judgment model through training according to the text information library specifically comprises the following steps:
And inputting the word segmentation vector and the target feature label into a machine learning model as training samples, and training to obtain the article evaluation data judgment model.
3. The method of claim 2, wherein the step of determining whether the first evaluation data includes a negative evaluation of the quality of the article using the article evaluation data evaluation model specifically comprises:
acquiring evaluation word segmentation vectors corresponding to the plurality of evaluation word segmentation words from the word vector library;
Inputting the evaluation word vector into the article evaluation data evaluation model, and outputting a characteristic label of the first evaluation data; the feature tag is used to characterize whether the first evaluation data contains a negative evaluation of the quality of the item;
and judging whether the first evaluation data comprises negative evaluation of the quality of the article according to the characteristic tag judgment model.
4. The article quality spot inspection item determining method according to claim 2, wherein prior to the step of obtaining word segmentation vectors corresponding to the plurality of word segments from the word vector library, the article quality spot inspection item determining method further comprises:
Filtering stop words in the plurality of word segments;
In the step of obtaining word segmentation vectors corresponding to the plurality of word segments from the word vector library, corresponding word segmentation vectors are obtained for the plurality of filtered word segments.
5. The method for determining a quality of an item according to claim 3, wherein the step of calculating a similarity between each subject term and the evaluation segmentation term comprises:
acquiring a subject word vector corresponding to the subject word from the word vector library;
and calculating cosine similarity of the subject term vector and the evaluation word vector based on a cosine similarity algorithm to serve as the similarity.
6. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of item quality spot check item determination of any one of claims 1 to 5 when the computer program is executed by the processor.
7. A computer readable storage medium having stored thereon a computer program, characterized in that the program when executed by a processor implements the steps of the item quality spot check item determination method of any of claims 1 to 5.
8. The system for determining the quality spot check items of the articles is characterized by comprising a data acquisition module, a word segmentation module, a similarity calculation module, a frequency statistics module, a quality spot check item selection module and a spot check item library; the selective examination item library stores a plurality of selective examination items to be detected and subject words corresponding to each selective examination item;
The data acquisition module is used for acquiring first evaluation data of an article;
the word segmentation module is used for carrying out word segmentation processing on the first evaluation data to obtain a plurality of evaluation word segments;
The similarity calculation module is used for calculating the similarity of each subject term and the evaluation segmentation term and calling the frequency statistics module;
The frequency statistics module is used for counting the frequency that the similarity of each subject term and the evaluation term is greater than a similarity threshold;
The quality selective examination item selection module is used for selecting N to-be-selective examination items corresponding to the N subject words with the highest frequency as quality selective examination items, wherein N is a positive integer;
The system for determining the quality spot check project of the article further comprises a judging module;
The judging module is used for judging whether the first evaluation data comprises negative evaluation on the quality of the article;
the judging module comprises an evaluation data acquisition unit, a label giving unit, a text information base creating unit and an article evaluation data judging model training unit;
the evaluation data acquisition unit is used for acquiring second evaluation data of the target object within a preset time;
The label giving unit is used for giving a target characteristic label to the second evaluation data, wherein the target characteristic label is used for representing whether the second evaluation data represents that the target object has quality problems;
The text information base creation unit is used for creating a text information base for judging the quality of the article according to the second evaluation data;
the article evaluation data judgment model training unit is used for training according to the text information base and the target feature tag to obtain an article evaluation data judgment model;
The judging module is used for judging whether the first evaluation data comprise negative evaluation on the quality of the article or not by utilizing the article evaluation data judging model;
the system for determining the quality spot check project of the article further comprises an evaluation data filtering module;
If the first evaluation data comprises negative evaluation of the quality of the article, invoking the evaluation data filtering module;
the evaluation data filtering module is used for filtering first evaluation data which does not contain negative evaluation on the quality of the article;
the word segmentation module is used for carrying out word segmentation on the filtered first evaluation data.
9. The item quality spot check item determination system of claim 8, wherein the judgment module further comprises a word segmentation unit and a word vector library; the word vector library stores a plurality of standard word segments and word vectors corresponding to each standard word segment;
the word segmentation unit is used for carrying out word segmentation processing on the second evaluation data to obtain a plurality of segmented words;
The text information base creation unit is used for obtaining word segmentation vectors corresponding to the plurality of word segmentation from the word vector base; the text information base comprises the word segmentation vector;
the article evaluation data evaluation model training unit is used for inputting the word segmentation vector and the target feature label into a machine learning model as training samples, and training to obtain the article evaluation data evaluation model.
10. The item quality spot check item determination system according to claim 9, wherein the judgment module further comprises a word vector acquisition unit and a tag output unit;
The word vector acquisition unit is used for acquiring evaluation word segmentation vectors corresponding to the plurality of evaluation word segmentation words from the word vector library;
The label output unit is used for inputting the evaluation word vector into the article evaluation data evaluation model and outputting the characteristic label of the first evaluation data; the feature tag is used to characterize whether the first evaluation data contains a negative evaluation of the quality of the item;
the judging module is used for judging whether the first evaluation data contains negative evaluation on the quality of the article according to the characteristic tag judgment model.
11. The item quality spot check item determination system of claim 9, wherein the determination module further comprises an stop word filtering unit;
the stop word filtering unit is used for filtering stop words in the plurality of word segments;
the word vector obtaining unit is used for obtaining corresponding word segmentation vectors for the filtered multiple word segmentation.
12. The article quality spot inspection item determination system according to claim 10, wherein the word vector acquisition unit is further configured to acquire a subject word vector corresponding to the subject word from the word vector library;
the similarity calculation module is used for calculating cosine similarity of the subject term vector and the evaluation word vector based on a cosine similarity algorithm to serve as the similarity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810866301.0A CN110807082B (en) | 2018-08-01 | 2018-08-01 | Quality selective examination item determining method, system, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810866301.0A CN110807082B (en) | 2018-08-01 | 2018-08-01 | Quality selective examination item determining method, system, electronic equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110807082A CN110807082A (en) | 2020-02-18 |
| CN110807082B true CN110807082B (en) | 2024-06-18 |
Family
ID=69486760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810866301.0A Active CN110807082B (en) | 2018-08-01 | 2018-08-01 | Quality selective examination item determining method, system, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110807082B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113283768A (en) * | 2021-06-01 | 2021-08-20 | 平安国际智慧城市科技股份有限公司 | Food detection item extraction method, device, equipment and storage medium |
| CN114254951A (en) * | 2021-12-27 | 2022-03-29 | 南方电网物资有限公司 | Power grid equipment arrival sampling inspection method based on digitization technology |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101408966A (en) * | 2008-11-20 | 2009-04-15 | 汤溪蔚 | Method and system for evaluation or questionnaire inquisition of brands through network |
| CN103123633A (en) * | 2011-11-21 | 2013-05-29 | 阿里巴巴集团控股有限公司 | Generation method of evaluation parameters and information searching method based on evaluation parameters |
| CN105590176A (en) * | 2016-03-07 | 2016-05-18 | 杭州国家电子商务产品质量监测处置中心 | Product quality risk monitoring sampling method based on Internet |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101334787B (en) * | 2008-07-22 | 2011-11-23 | 刘宜云 | Objects evaluation information enquiry system and method |
| CN103020851B (en) * | 2013-01-10 | 2015-10-14 | 山大地纬软件股份有限公司 | A kind of metric calculation method supporting comment on commodity data multidimensional to analyze |
| WO2015198436A1 (en) * | 2014-06-26 | 2015-12-30 | 楽天株式会社 | Information processing device, information processing method, and program |
| US20170139918A1 (en) * | 2015-11-13 | 2017-05-18 | Salesforce.Com, Inc. | Managing importance ratings related to event records in a database system |
| CN107203841A (en) * | 2017-05-08 | 2017-09-26 | 中车青岛四方机车车辆股份有限公司 | The method of inspection and device of product quality |
| CN108269169A (en) * | 2017-12-29 | 2018-07-10 | 武汉璞华大数据技术有限公司 | A kind of shopping guide method and system |
-
2018
- 2018-08-01 CN CN201810866301.0A patent/CN110807082B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101408966A (en) * | 2008-11-20 | 2009-04-15 | 汤溪蔚 | Method and system for evaluation or questionnaire inquisition of brands through network |
| CN103123633A (en) * | 2011-11-21 | 2013-05-29 | 阿里巴巴集团控股有限公司 | Generation method of evaluation parameters and information searching method based on evaluation parameters |
| CN105590176A (en) * | 2016-03-07 | 2016-05-18 | 杭州国家电子商务产品质量监测处置中心 | Product quality risk monitoring sampling method based on Internet |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110807082A (en) | 2020-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112163424A (en) | Data labeling method, device, equipment and medium | |
| CN113238922B (en) | Log analysis method and device, electronic equipment and medium | |
| CN112181490B (en) | Method, device, equipment and medium for identifying function category in function point evaluation method | |
| CN110826494A (en) | Method and device for evaluating quality of labeled data, computer equipment and storage medium | |
| CN112613569A (en) | Image recognition method, and training method and device of image classification model | |
| CN111125116B (en) | Method and system for positioning code field in service table and corresponding code table | |
| CN112445775A (en) | Fault analysis method, device, equipment and storage medium of photoetching machine | |
| CN111427974A (en) | Data quality evaluation management method and device | |
| CN110807082B (en) | Quality selective examination item determining method, system, electronic equipment and readable storage medium | |
| CN106610932A (en) | Corpus processing method and device and corpus analyzing method and device | |
| CN114428677A (en) | Task processing method, processing device, electronic equipment and storage medium | |
| CN113569988B (en) | Algorithm model evaluation method and system | |
| CN113592017B (en) | Deep learning model standardized training method, management system and processing terminal | |
| CN113868419B (en) | Text classification method, device, equipment and medium based on artificial intelligence | |
| CN114385694A (en) | Data processing method and device, computer equipment and storage medium | |
| CN117093556A (en) | Log classification method, device, computer equipment and computer readable storage medium | |
| CN110956030A (en) | Method and system for comparing configuration information of remote machine of transformer substation | |
| CN115936748A (en) | Business big data analysis method and system | |
| CN112598226B (en) | Equipment checking method, device, equipment and storage medium | |
| CN113849618A (en) | Strategy determination method and device based on knowledge graph, electronic equipment and medium | |
| CN111353087A (en) | Hot word statistical method and device, storage medium and electronic terminal | |
| CN115808504B (en) | Online drift compensation method for gas sensor for concentration prediction | |
| CN117150439B (en) | Automobile manufacturing parameter detection method and system based on multi-source heterogeneous data fusion | |
| CN115904964A (en) | Error reporting test script classification method and device | |
| CN117972170A (en) | Slow query identification method, device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |