CN116010707A - Commodity price anomaly identification method, device, equipment and storage medium - Google Patents
Commodity price anomaly identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN116010707A CN116010707A CN202310079950.7A CN202310079950A CN116010707A CN 116010707 A CN116010707 A CN 116010707A CN 202310079950 A CN202310079950 A CN 202310079950A CN 116010707 A CN116010707 A CN 116010707A
- Authority
- CN
- China
- Prior art keywords
- information
- commodity
- price
- pricing
- public opinion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The application relates to a commodity price anomaly identification method, a commodity price anomaly identification device, commodity price anomaly identification equipment and a storage medium. The method comprises the following steps: responding to pricing instructions aiming at target commodities, and extracting emotion information of commodity brands to which the target commodities belong from public opinion texts; the pricing instruction comprises current pricing information of the target commodity; acquiring reference price information of the target commodity; wherein the reference price information includes an official sale price, a historical deal price, and a historical successful pricing price for the target commodity; and identifying whether the current pricing information is abnormal according to the reference price information and the emotion information. In the pricing process of the target commodity, the emotion information of the brand of the target commodity and the reference price information of the target commodity, which are included in the public opinion text, are fully utilized, and whether the pricing of the target commodity is suspicious or not can be effectively identified, so that the accuracy of the commodity price anomaly identification result is improved.
Description
Technical Field
The present disclosure relates to the field of computer processing technologies, and in particular, to a method, an apparatus, a device, and a storage medium for identifying a price anomaly of a commodity.
Background
In the network platform, sellers may price the goods to thereby effect the transaction of the goods. However, there are cases where some sellers conduct a lot of frying on the goods, resulting in the pricing of the goods being far higher than the normal fluctuation interval of the goods, disturbing the normal transaction order of the platform.
In the related art, whether the commodity price is abnormal or not may be identified in combination with the periodic characteristics of the commodity price. However, in practical applications, the periodic variation of commodity prices is complex, and it is often difficult to find the periodicity law. Therefore, how to identify the abnormal price of the commodity in the network platform is a technical problem to be solved for those skilled in the art.
Disclosure of Invention
Based on the foregoing, embodiments of the present application provide a method, an apparatus, a device, and a storage medium for identifying a price anomaly of a commodity.
In a first aspect, an embodiment of the present application provides a method for identifying a price anomaly of a commodity, including:
responding to pricing instructions aiming at target commodities, and extracting emotion information of commodity brands to which the target commodities belong from public opinion texts; the pricing instruction comprises current pricing information of the target commodity;
acquiring reference price information of the target commodity; wherein the reference price information includes an official sale price, a historical deal price, and a historical successful pricing price for the target commodity;
and identifying whether the current pricing information is abnormal according to the reference price information and the emotion information.
In a second aspect, an embodiment of the present application provides a commodity price anomaly identification device, including:
the extraction module is used for responding to the pricing instruction for the target commodity and extracting emotion information of the commodity brand of the target commodity from the public opinion text; the pricing instruction comprises current pricing information of the target commodity;
the acquisition module is used for acquiring the reference price information of the target commodity; wherein the reference price information includes an official sale price, a historical deal price, and a historical successful pricing price for the target commodity;
and the identification module is used for identifying whether the current pricing information is abnormal or not according to the reference price information and the emotion information.
In a third aspect, an embodiment of the present application provides an electronic device, including: the method comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the commodity price anomaly identification method provided by the first aspect of the embodiment of the application when executing the computer program.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the commodity price anomaly identification method provided in the first aspect of the embodiments of the present application.
According to the technical scheme provided by the embodiment of the application, the pricing instruction for the target commodity is responded, the emotion information of the commodity brand to which the target commodity belongs is extracted from the public opinion text, the pricing instruction comprises the current pricing information of the target commodity, the reference price information of the target commodity is obtained, and whether the current pricing information is abnormal is identified according to the reference price information of the target commodity and the emotion information of the commodity brand to which the target commodity belongs. In the pricing process of the target commodity, the emotion information of the brand of the target commodity and the reference price information of the target commodity, which are included in the public opinion text, are fully utilized, and whether the pricing of the target commodity is suspicious or not can be effectively identified, so that the accuracy of the commodity price anomaly identification result is improved.
Drawings
Fig. 1 is a schematic flow chart of a commodity price anomaly identification method according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a commercial brand-based emotion information extraction process according to an embodiment of the present application;
FIG. 3 is a schematic flow chart of another method for identifying abnormal price of a commodity according to the embodiment of the present application;
fig. 4 is a schematic structural diagram of a commodity price anomaly identification device according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The present application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present application are shown in the drawings.
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions in the embodiments of the present application are further described in detail by the following embodiments with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The execution subject of the method embodiment described below may be a commodity price anomaly identification device, which may be implemented as part or all of an electronic device by software, hardware, or a combination of software and hardware. Alternatively, the electronic device may be a smart phone, a tablet computer, an electronic book reader, a vehicle-mounted terminal, or the like. Of course, the electronic device may also be an independent server or a server cluster, and the specific type of the electronic device is not limited in the embodiments of the present application. The following method embodiments are described taking an electronic device as an example of an execution subject.
Fig. 1 is a schematic flow chart of a commodity price anomaly identification method according to an embodiment of the present application. This embodiment relates to a specific process of how the electronic device recognizes whether the current pricing of the target commodity is abnormal.
As shown in fig. 1, the method may include:
s101, responding to pricing instructions for target commodities, and extracting emotion information of brands of the target commodities from the public opinion texts.
Specifically, the public opinion text may be original content of a user in the platform, commodity evaluation information and social platform public opinion text. By way of example, the public opinion text may be forums, microblogs, shopping websites, comments in an e-commerce platform, posts, etc. For example, the public opinion text may be rating data of goods by a user in a shopping network, or the like.
In practical applications, a user (e.g., a seller in a shopping platform) may trigger a pricing instruction through a corresponding control presented in the platform to price a target commodity, where the pricing instruction may include current pricing information of the target commodity by the user. For example, a user may price a certain shoe in a shopping platform.
After receiving the pricing instruction for the target commodity, the electronic device can acquire the public opinion text related to the target commodity and extract emotion information of the commodity brand to which the target commodity belongs from the acquired public opinion text. The public opinion text has entity information and emotion information, the entity information can be commodity names or brands, and the emotion information can represent the characteristics of commodity brands on certain attributes, such as quality, price and the like. In an alternative embodiment, entity words for representing entity information and attribute words for representing emotion information may be determined in public opinion text, so as to obtain emotion information of a commodity brand to which a target commodity belongs. The nearest heat condition of the target commodity can be effectively found through the emotion information, and whether the target commodity has a stir-frying trend or not is further predicted. For example, if the recent heat of the target commodity is found to be high by the emotion information, the suspicion of the target commodity is high, whereas if the heat of the target commodity is found to be not different from the usual heat by the emotion information, the suspicion of the target commodity is low.
S102, acquiring reference price information of the target commodity.
Specifically, a historical transaction log is obtained, and reference price information of the target commodity is extracted from the historical transaction log, wherein the reference price information comprises official suggested price, historical bargain price and historical successful pricing price of the target commodity. By improving the reference basis by taking the reference price information as the current pricing information of the target commodity, whether the current pricing information is abnormal or not can be predicted based on the gap between the current pricing information and the reference price information. For example, if the current pricing information deviates from the reference pricing information more, the higher the likelihood of abnormality of the current pricing information is predicted; conversely, if the current pricing information is offset by the reference pricing information less, the likelihood of abnormality of the current pricing information is predicted to be lower.
S103, identifying whether the current pricing information is abnormal or not according to the reference price information and the emotion information.
After the emotion information of the brand of the target commodity and the reference price information of the target commodity are obtained, comprehensively utilizing the reference price information and the emotion information of the brand of the target commodity to identify whether the current pricing information of the target commodity is abnormal or not. It can be understood that the above-mentioned reference price information is taken as a reference standard, a relatively reasonable price interval of the target commodity is provided, the above-mentioned emotion information can reflect the stir-frying trend of the brand of the target commodity, if the current pricing information deviates from the reference price information of the target commodity and the stir-frying suspicion of the target commodity is predicted to be greater through the emotion information, the current pricing information is more likely to be an abnormal price, otherwise, if the current pricing information deviates from the reference price information of the target commodity and the stir-frying suspicion of the target commodity is predicted to be less through the emotion information, the current pricing information is less likely to be an abnormal price.
Of course, as an alternative embodiment, the reference price information and the emotion information may be set with corresponding weights, and whether the current pricing information of the target commodity is abnormal may be determined by using the reference price information, the emotion information, the weights corresponding to the reference price information, and the weights corresponding to the emotion information.
In the pricing process of the target commodity, the public opinion text issued by the user is fully utilized to mine the emotion information of the commodity brand of the target commodity, whether the target commodity has a suspicion of stir-frying or not is predicted through the emotion information, and the emotion information is taken as one of the characteristics to be added into the abnormality recognition process of the current pricing of the target commodity, so that the recognition result of the abnormality judgment of the current pricing information can be improved.
After the recognition result of the current pricing information is obtained, a corresponding control operation may also be performed according to the recognition result of the current pricing information.
Specifically, when the current pricing information is determined to be an abnormal price, the user may be denied the pricing operation for the target commodity, i.e., the target commodity is not allowed to be priced according to the current pricing information. Further, a prompt may be output to the user that the pricing failed to instruct the user to re-price the target merchandise.
When the current pricing information is determined to be the normal price, the user can agree with the pricing operation of the target commodity, namely, the target commodity is priced according to the current pricing information, and the current pricing information of the target commodity can be displayed, so that the transaction of the subsequent commodity is facilitated. Further, a prompt for successful pricing may also be output to the user.
Corresponding control operation is executed based on the identification result of the current pricing information, so that the accurate prevention and control of abnormal commodity pricing are realized, and the normal transaction order of the platform and the risk of commodity stir-frying prevention are maintained.
According to the commodity price anomaly identification method provided by the embodiment of the application, in response to a pricing instruction aiming at a target commodity, emotion information of a commodity brand to which the target commodity belongs is extracted from a public opinion text, the pricing instruction comprises current pricing information of the target commodity, reference price information of the target commodity is obtained, and whether the current pricing information is abnormal is identified according to the reference price information of the target commodity and the emotion information of the commodity brand to which the target commodity belongs. In the pricing process of the target commodity, the emotion information of the brand of the target commodity and the reference price information of the target commodity, which are included in the public opinion text, are fully utilized, and whether the pricing of the target commodity is suspicious or not can be effectively identified, so that the accuracy of the commodity price anomaly identification result is improved.
In one embodiment, as shown in fig. 2, the extracting, in S101, emotion information of a brand of the target commodity from the public opinion text may include:
s201, obtaining public opinion texts of all social platforms by utilizing a crawler technology.
With the rapid development of internet technology, the internet has become a carrier of a large amount of information, and various information including news, electronic commerce, commodity reviews, etc. can be viewed through the internet, so that crawler technology has been developed. Crawler technology refers to a program or script that automatically crawls information on the internet according to certain rules.
Because various information of the commodity can be distributed on various social platforms, such as forums, microblogs, shopping websites, various electronic commerce platforms and the like, the content of the related commodity information in each social platform can be crawled by adopting a crawler technology, so that a public opinion text is obtained. The social platform comprises an internal platform and an external platform.
S202, preprocessing the public opinion text to obtain the processed public opinion text.
After the public opinion text is obtained, the public opinion text can be cleaned, and some content irrelevant to commercial brands, such as advertisement stickers, non-commercial brand content posts and the like, can be filtered; furthermore, abnormal characters in the public opinion texts can be processed, a data basis is provided for accurate analysis of the follow-up public opinion texts, and influence of irrelevant contents on public opinion analysis results is avoided.
And S203, inputting the processed public opinion text into a pre-trained brand emotion analysis model to obtain emotion information of a commodity brand to which the target commodity included in the public opinion text belongs.
The framework of the brand emotion analysis model can be Bert+softmax and Bert+CRF, bert is a model obtained by pre-training two tasks of 'Fill in the blank task' and 'Next sentence prediction', the output of Bert can be mapped into a label set for labeling problems only by adding a full-connection layer on the basis of Bert and determining the output dimension of the full-connection layer, the output vector of a single token is processed by softmax or CRF, and the numerical value of each dimension represents the probability that the token is a certain label. Wherein CRF is a classical probability map model. That is, the addition of softmax or CRF based on the pre-training model can realize emotion judgment based on the commercial brand in the public opinion text and information labeling of the commercial brand entity. Of course, the brand emotion analysis model may be implemented by other architectures, which are not limited in this embodiment.
Taking a brand emotion analysis model as a Bert+softmax and a Bert+CRF architecture as an example, after a processed public opinion text is obtained, inputting the processed public opinion text into a pre-trained brand emotion analysis model, coding the public opinion text through a Bert module in the brand emotion analysis model to obtain coding vectors corresponding to each token in the public opinion text, and then classifying and mapping the coding vectors through the softmax and the CRF, so that emotion information of a commodity brand to which a target commodity belongs in the public opinion text is extracted.
Optionally, before the crawler technology is used to obtain the public opinion text of each social platform, the brand emotion analysis model may be trained, and specifically, the brand emotion analysis model may be obtained through the following training process:
acquiring a first training data set; the first training data set comprises a plurality of sample public opinion texts and emotion information of commodity brands contained in each sample public opinion text; and training the brand emotion analysis model according to the first training data set.
Specifically, a plurality of sample public opinion texts are obtained, and then commodity brands and corresponding emotion polarities in the sample public opinion texts are marked to obtain emotion information of the commodity brands included in the sample public opinion texts. And then, taking a plurality of sample public opinion texts as input, taking emotion information of commodity brands included in the plurality of sample public opinion texts as expected output, determining actual output corresponding to the plurality of sample public opinion texts through a brand emotion analysis model, determining a loss value of a preset loss function based on the actual output and the expected output, and optimizing parameters of the brand emotion analysis model by using the loss value until the loss value reaches a preset convergence condition and is kept stable, so that a trained brand emotion analysis model is obtained.
Of course, after obtaining a plurality of sample public opinion texts, the sample public opinion texts can be cleaned, and some content irrelevant to commercial brands, such as advertising stickers, non-commercial brand content posts and the like, can be filtered; furthermore, abnormal characters in the sample public opinion text can be processed, and a data basis is provided for training of a subsequent brand emotion analysis model, so that accuracy of the brand emotion analysis model is ensured.
In this embodiment, since the brand emotion analysis model is obtained through training of a large amount of training data, the brand emotion analysis model can better learn the representation of the brand and emotion information of the commodity, and thus, the accuracy of emotion analysis based on the brand of the commodity can be improved by processing the public opinion text of the target commodity through the brand emotion analysis model.
In one embodiment, as shown in fig. 3, the process of S103 may be:
s301, inputting current pricing information, reference price information and emotion information into a pre-trained anomaly detection model to obtain anomaly probability of the current pricing information.
Specifically, the anomaly detection model may be a binary tree model, such as a LightGBM algorithm, an XGBoost algorithm, and the like. The abnormality detection model is trained based on a second training data set. The second training data set includes historical pricing information of a plurality of sample commodities, reference price information, emotion information of commodity brands to which the plurality of sample commodities belong, and tag information of whether the historical pricing information is abnormal.
When the anomaly detection model is trained, historical pricing information, reference price information and emotion information of brands of goods to which the plurality of sample goods belong can be used as inputs, label information of whether the historical pricing information of the plurality of sample goods is abnormal is used as expected output, actual output of whether the historical pricing information of the plurality of sample goods is abnormal is determined through the anomaly detection model, a loss value of a preset loss function is determined based on the actual output and the expected output, parameters of the anomaly detection model are optimized through the loss value until the loss value reaches a preset convergence condition and is kept stable, and therefore the anomaly detection model is obtained.
After the trained anomaly detection model is obtained, the current pricing information, the reference price information and the emotion information of the target commodity can be input into the anomaly detection model, the difference between the current pricing information and the reference price information is analyzed through the anomaly detection model, the emotion information of the commodity brand to which the target commodity belongs is analyzed, whether the target commodity is suspected of being stir-fried or not is predicted, and therefore the anomaly probability of the current pricing information is obtained.
S302, determining whether the abnormal probability of the current pricing information is larger than a preset threshold value.
The preset threshold value can be set based on actual requirements. If the abnormality probability of the current pricing information is greater than the preset threshold, the following S303 is executed, and if the abnormality probability of the current pricing information is less than or equal to the preset threshold, the following S304 is executed.
S303, determining the current pricing information as an abnormal price.
S304, determining the current pricing information as a normal price.
When the current pricing information is determined to be an abnormal price, the pricing operation of the target commodity can be refused, namely the target commodity cannot be priced according to the current pricing information indicated by the user; re-pricing the target commodity may also be indicated. When the current pricing information is determined to be the normal price, the pricing operation of the target commodity can be allowed, namely the target commodity can be priced according to the current pricing information indicated by the user, so that the accurate control of the abnormal commodity pricing operation is realized, and the normal transaction order of the platform is maintained.
In this embodiment, since the anomaly detection model is obtained through training of a large amount of training data, the anomaly detection model can better learn the mapping relationship between each feature and the recognition result, and thus, the accuracy of identifying the anomaly in the price of the commodity can be improved by processing the current pricing information of the target commodity through the anomaly detection model. And the emotion information of the commodity brand of the target commodity is input into the anomaly detection model as a characteristic, and the public opinion text of the target commodity is fully utilized, so that the anomaly detection model can effectively identify whether the pricing of the target commodity is suspicious, and the accuracy of the commodity price anomaly identification result is further improved.
Fig. 4 is a schematic structural diagram of a commodity price anomaly identification device according to an embodiment of the present application. As shown in fig. 4, the apparatus may include: an extraction module 401, an acquisition module 402 and an identification module 403.
Specifically, the extracting module 401 is configured to extract emotion information of a brand of the target commodity from the public opinion text in response to a pricing instruction for the target commodity; the pricing instruction comprises current pricing information of the target commodity;
the acquisition module 402 is configured to acquire reference price information of the target commodity; wherein the reference price information includes an official suggested price, a historical transaction price, and a historical successful pricing price for the target commodity;
the identifying module 403 is configured to identify whether the current pricing information is abnormal according to the reference price information and the emotion information.
According to the commodity price anomaly identification device provided by the embodiment of the application, the pricing instruction for the target commodity is responded, the emotion information of the commodity brand to which the target commodity belongs is extracted from the public opinion text, the pricing instruction comprises the current pricing information of the target commodity, the reference price information of the target commodity is obtained, and whether the current pricing information is abnormal or not is identified according to the reference price information of the target commodity and the emotion information of the commodity brand to which the target commodity belongs. In the pricing process of the target commodity, the emotion information of the brand of the target commodity and the reference price information of the target commodity, which are included in the public opinion text, are fully utilized, and whether the pricing of the target commodity is suspicious or not can be effectively identified, so that the accuracy of the commodity price anomaly identification result is improved.
On the basis of the above embodiment, optionally, the extracting module 401 is specifically configured to obtain public opinion text of each social platform by using a crawler technology; preprocessing the public opinion text to obtain a processed public opinion text; and inputting the processed public opinion text into a pre-trained brand emotion analysis model to obtain emotion information of the commodity brand to which the target commodity belongs, wherein the emotion information is included in the public opinion text.
On the basis of the above embodiment, optionally, the apparatus further includes: and a training module.
Specifically, the training module is configured to obtain a first training data set before the public opinion text of each social platform is obtained by using a crawler technology; training the brand emotion analysis model according to the first training data set; the first training data set comprises a plurality of sample public opinion texts and emotion information of commodity brands contained in each sample public opinion text.
On the basis of the above embodiment, optionally, the identification module 403 is specifically configured to input the current pricing information, the reference price information, and the emotion information into a pre-trained anomaly detection model, so as to obtain anomaly probability of the current pricing information; when the anomaly probability is greater than a preset threshold, determining that the current pricing information is an anomaly price; and when the abnormal probability is smaller than or equal to the preset threshold value, determining that the current pricing information is a normal price.
On the basis of the above embodiment, optionally, the anomaly detection model is a classification tree model, and the anomaly detection model is obtained based on training of a second training data set; the second training data set comprises historical pricing information of a plurality of sample commodities, reference price information, emotion information of commodity brands to which the plurality of sample commodities belong and label information of whether the historical pricing information is abnormal or not.
On the basis of the above embodiment, optionally, the apparatus further includes: and a processing module.
Specifically, the processing module is used for executing corresponding control operation according to the identification result of the current pricing information.
On the basis of the above embodiment, optionally, the public opinion text includes at least one of the following:
user original content, commodity evaluation information and social platform public opinion text in the platform.
Fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application, as shown in fig. 5, where the device includes a processor 510, a memory 520, an input device 530, and an output device 540; the number of processors 510 in the device may be one or more, one processor 510 being taken as an example in fig. 5; the processor 510, memory 520, input means 530 and output means 540 in the device may be connected by a bus or other means, for example by a bus connection in fig. 5.
The memory 520 is a computer-readable storage medium, and may be used to store a software program, a computer-executable program, and modules, such as program instructions/modules (e.g., the extraction module 401, the acquisition module 402, and the identification module 403 in the commodity price anomaly identification device) corresponding to the commodity price anomaly identification method in the embodiment of the present application. The processor 510 executes various functional applications and data processing of the above-described devices by running software programs, instructions and modules stored in the memory 520, i.e., implements the above-described commodity price anomaly identification method.
The input means 530 may be used to receive input numeric or character information and to generate key signal inputs related to user settings and function control of the device described above. The output 540 may include a display device such as a display screen.
In one embodiment, there is also provided a computer-readable storage medium having stored thereon a computer program for executing a commodity price anomaly identification method when executed by a processor, the method comprising:
responding to pricing instructions aiming at target commodities, and extracting emotion information of commodity brands to which the target commodities belong from public opinion texts; the pricing instruction comprises current pricing information of the target commodity;
acquiring reference price information of the target commodity; wherein the reference price information includes an official suggested price, a historical transaction price, and a historical successful pricing price for the target commodity;
and identifying whether the current pricing information is abnormal according to the reference price information and the emotion information.
Of course, the computer-readable storage medium provided in the embodiments of the present application is not limited to the method operations described above when the computer program is executed, and may also perform the related operations in the commodity price anomaly identification method provided in any embodiment of the present application.
From the above description of embodiments, it will be clear to a person skilled in the art that the present application may be implemented by means of software and necessary general purpose hardware, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a FLASH Memory (FLASH), a hard disk, or an optical disk of a computer, where the instructions include a number of instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the method described in the embodiments of the present application.
It should be noted that, in the above-mentioned embodiments of the search apparatus, each unit and module included are only divided according to the functional logic, but not limited to the above-mentioned division, as long as the corresponding functions can be implemented; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present application.
Note that the above is only a preferred embodiment of the present application and the technical principle applied. Those skilled in the art will appreciate that the present application is not limited to the particular embodiments described herein, but is capable of numerous obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the present application. Therefore, while the present application has been described in connection with the above embodiments, the present application is not limited to the above embodiments, but may include many other equivalent embodiments without departing from the spirit of the present application, the scope of which is defined by the scope of the appended claims.
Claims (10)
1. A commodity price anomaly identification method, comprising:
responding to pricing instructions aiming at target commodities, and extracting emotion information of commodity brands to which the target commodities belong from public opinion texts; the pricing instruction comprises current pricing information of the target commodity;
acquiring reference price information of the target commodity; wherein the reference price information includes an official suggested price, a historical transaction price, and a historical successful pricing price for the target commodity;
and identifying whether the current pricing information is abnormal according to the reference price information and the emotion information.
2. The method of claim 1, wherein the extracting emotion information of the brand to which the target commodity belongs from the public opinion text comprises:
obtaining public opinion texts of all social platforms by utilizing a crawler technology;
preprocessing the public opinion text to obtain a processed public opinion text;
and inputting the processed public opinion text into a pre-trained brand emotion analysis model to obtain emotion information of the commodity brand to which the target commodity belongs, wherein the emotion information is included in the public opinion text.
3. The method of claim 2, further comprising, prior to the obtaining public opinion text for each social platform using crawler technology:
acquiring a first training data set; the first training data set comprises a plurality of sample public opinion texts and emotion information of commodity brands contained in each sample public opinion text;
and training the brand emotion analysis model according to the first training data set.
4. The method of claim 1, wherein the identifying whether the current pricing information is abnormal based on the reference price information and the emotion information comprises:
inputting the current pricing information, the reference price information and the emotion information into a pre-trained abnormality detection model to obtain abnormality probability of the current pricing information;
when the anomaly probability is greater than a preset threshold, determining that the current pricing information is an anomaly price;
and when the abnormal probability is smaller than or equal to the preset threshold value, determining that the current pricing information is a normal price.
5. The method of claim 4, wherein the anomaly detection model is a binary tree model and the anomaly detection model is trained based on a second training dataset;
the second training data set comprises historical pricing information of a plurality of sample commodities, reference price information, emotion information of commodity brands to which the plurality of sample commodities belong and label information of whether the historical pricing information is abnormal or not.
6. The method according to any one of claims 1 to 5, further comprising:
and executing corresponding control operation according to the identification result of the current pricing information.
7. The method of any one of claims 1 to 5, wherein the public opinion text comprises at least one of:
user original content, commodity evaluation information and social platform public opinion text in the platform.
8. A commodity price anomaly identification method, comprising:
the extraction module is used for responding to the pricing instruction for the target commodity and extracting emotion information of the commodity brand of the target commodity from the public opinion text; the pricing instruction comprises current pricing information of the target commodity;
the acquisition module is used for acquiring the reference price information of the target commodity; wherein the reference price information includes an official suggested price, a historical transaction price, and a historical successful pricing price for the target commodity;
and the identification module is used for identifying whether the current pricing information is abnormal or not according to the reference price information and the emotion information.
9. An electronic device, comprising: a memory storing a computer program, and a processor implementing the steps of the method according to any one of claims 1 to 7 when the computer program is executed by the processor.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310079950.7A CN116010707A (en) | 2023-01-17 | 2023-01-17 | Commodity price anomaly identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310079950.7A CN116010707A (en) | 2023-01-17 | 2023-01-17 | Commodity price anomaly identification method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116010707A true CN116010707A (en) | 2023-04-25 |
Family
ID=86030176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310079950.7A Pending CN116010707A (en) | 2023-01-17 | 2023-01-17 | Commodity price anomaly identification method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116010707A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117235677A (en) * | 2023-11-10 | 2023-12-15 | 邦邦汽车销售服务(北京)有限公司 | Automobile accessory price anomaly identification detection method |
-
2023
- 2023-01-17 CN CN202310079950.7A patent/CN116010707A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117235677A (en) * | 2023-11-10 | 2023-12-15 | 邦邦汽车销售服务(北京)有限公司 | Automobile accessory price anomaly identification detection method |
| CN117235677B (en) * | 2023-11-10 | 2024-02-20 | 邦邦汽车销售服务(北京)有限公司 | Automobile accessory price anomaly identification detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107945024B (en) | Method for identifying internet financial loan enterprise operation abnormity, terminal equipment and storage medium | |
| WO2018040944A1 (en) | System, method, and device for identifying malicious address/malicious purchase order | |
| US8682739B1 (en) | Identifying objects in video | |
| WO2020001106A1 (en) | Classification model training method and store classification method and device | |
| CA2917256C (en) | Screenshot-based e-commerce | |
| CN111107048A (en) | Phishing website detection method and device and storage medium | |
| CN110827112B (en) | Deep learning commodity recommendation method and device, computer equipment and storage medium | |
| US9104746B1 (en) | Identifying contrarian terms based on website content | |
| US20230089850A1 (en) | Real-time product environmental impact scoring | |
| CA3144405A1 (en) | Text information recognizing method, extracting method, devices and system | |
| US9460163B1 (en) | Configurable extractions in social media | |
| CN116010707A (en) | Commodity price anomaly identification method, device, equipment and storage medium | |
| CN116664306A (en) | Intelligent recommendation method and device for wind control rules, electronic equipment and medium | |
| US20230100172A1 (en) | Item matching and recognition system | |
| CN115795345A (en) | Information processing method, device, equipment and storage medium | |
| CN112001792B (en) | Configuration information consistency detection method and device | |
| US10534782B1 (en) | Systems and methods for name matching | |
| CN112686676A (en) | Industrial Internet identification chain processing method, device and equipment | |
| CN113139115A (en) | Information recommendation method, search method, device, client, medium and equipment | |
| US10102560B1 (en) | Identifying child node correlations | |
| CN112949752B (en) | Training method and device of business prediction system | |
| US20230245138A1 (en) | Systems and methods for authenticating online sales using machine learning | |
| CN111199439B (en) | Commodity information processing method and device | |
| US20230099904A1 (en) | Machine learning model prediction of interest in an object | |
| TWI640944B (en) | Automated intention extraction device and method thereof |
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 |