Commodity verification method and device, storage medium and processor

Abstract

The present invention discloses a method and device for verifying a commodity, a storage medium and a processor. The method comprises: obtaining a target commodity that has completed an operation on a platform within a preset time period; obtaining search data associated with the target commodity, wherein the search data includes at least one search behavior data in the process of operating the target commodity on the platform; inputting each search behavior data into a search behavior model to obtain a behavior expression vector corresponding to each search behavior data; and determining whether the target commodity is an abnormal commodity based on the behavior expression vector corresponding to each search behavior data. The present invention solves the technical problem of low efficiency in verifying abnormal commodities due to the fact that commodity sellers often modify the information of abnormal commodities.

Description

Technical Field

[0001] The present invention relates to the field of commodity information processing, and in particular to a commodity verification method and device, a storage medium and a processor. Background Art

[0002] In the field of e-commerce, due to the platform attributes of new retail platforms, some merchants will release some illegal or abnormal products. Such illegal products include those that violate platform rules and sell fake products under the guise of selling mutton. Their characteristics are that they often use some synonyms or even words that are completely unrelated to the product as the text introduction content of the product to circumvent the platform's prevention and control mechanism. However, these products will still be found by some experienced buyers from a large number of similar products through their own search technology, causing such products to spread to a certain extent. Since these spreads will have a great impact on the platform and the shopping experience of users, the platform needs to use technical means to discover these illegal products.

[0003] In the related art, there are two main methods for detecting illegal products: the first is a method based on keyword interception combined with manual judgment: business experts summarize, summarize, and collect product sensitive dictionaries based on the characteristics of illegal products and confirmed illegal product information. On the basis of sensitive keyword interception, manual verification is performed, and keywords in the sensitive dictionary are adjusted and added according to business changes. Although this method can quickly intercept most illegal products, it is not flexible enough and relatively lagging. It is difficult to cope with new words or new descriptions created by sellers, and it is easy to cause misjudgment. The second category is a machine learning or deep learning method based on feature extraction, which uses the text information of the product as a feature to train a machine learning or deep learning classification model to understand the semantics contained in the product text. However, since the text content of illegal products is relatively obscure or similar to normal products, it is difficult for existing machine learning and deep learning models to capture effective semantics. Similarly, since these models require a large amount of training corpus, and these data from the product itself, such as product titles or descriptions, are difficult to quickly capture changes in them, the trained model is always an old model relative to the latest scenario, so online use cannot meet business needs very well.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a commodity verification method and device, a storage medium and a processor, so as to at least solve the technical problem that the verification efficiency of abnormal commodities is low due to the frequent modification of the information of abnormal commodities by commodity sellers.

[0006] According to one aspect of an embodiment of the present invention, a method for verifying a product is provided, comprising: obtaining a target product that has completed an operation on a platform within a preset time period; obtaining search data associated with the target product, wherein the search data includes at least one search behavior data during the operation of the target product on the platform; inputting each search behavior data into a search behavior model to obtain a behavior expression vector corresponding to each search behavior data; and determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data.

[0007] Furthermore, the search behavior data at least includes: product search terms and information on visited products, wherein the product search terms are search terms used by buyers to search for the target products when purchasing the target products on the platform, and the information on visited products refers to information on the products visited by the buyer in the search results after the platform returns the search results based on the product search terms. Inputting each search behavior data into the search behavior model, and obtaining a behavior expression vector corresponding to each search behavior data includes: obtaining a word vector matrix corresponding to the product search terms and a word vector matrix corresponding to the information on visited products; performing pooling processing on the word vector matrix corresponding to the product search terms to obtain a word vector matrix one, and performing pooling processing on the word vector matrix corresponding to the information on visited products to obtain a word vector matrix two; merging the word vector matrix one and the word vector matrix two to obtain a target matrix; and obtaining a behavior expression vector corresponding to each search behavior data based on the target matrix.

[0008] Furthermore, obtaining the word vector matrix corresponding to the product search term and the word vector matrix corresponding to the information of visiting the product includes: performing word segmentation processing on the product search term to obtain word segmentation set one; based on each word in the word segmentation set one, determining the word vector matrix corresponding to the product search term; performing word segmentation processing on the information of visiting the product to obtain word segmentation set two; based on each word in the word segmentation set two, determining the word vector matrix corresponding to the information of visiting the product.

[0009] Furthermore, determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data includes: passing the behavior expression vector through a fully connected neural network to obtain a label corresponding to the target product; and determining whether the target product is an abnormal product based on the label corresponding to the target product.

[0010] Furthermore, before inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data, the method also includes: obtaining information on illegal products detected on the platform within a preset period, and randomly selecting information on a preset number of normal products; obtaining purchase records of users who purchased the illegal products and the normal products; obtaining the user's search behavior characteristics based on the purchase records; and modeling based on the user's search behavior characteristics to obtain the search behavior model, wherein the search behavior model is a model generated by search training for abnormal products and / or safe products detected on the platform.

[0011] Furthermore, modeling is performed based on the search behavior characteristics of the user to obtain the search behavior model, including: modeling is performed based on the search behavior characteristics of the user using a bidirectional long short-term memory network or a recursive neural network to obtain the search behavior model.

[0012] Furthermore, the word vector matrix one and the word vector matrix two are merged, wherein the merging process is one of the following: splicing process and combining using a gate mechanism.

[0013] Furthermore, the pooling process is one of the following: average pooling and pooling based on an attention mechanism.

[0014] According to another aspect of an embodiment of the present invention, a commodity verification device is also provided, including: a first acquisition unit, used to acquire a target commodity that has completed an operation on a platform within a preset time period; an extraction unit, used to acquire search data associated with the target commodity, wherein the search data includes at least one search behavior data in the process of operating the target commodity on the platform; a second acquisition unit, used to input each search behavior data into a search behavior model to obtain a behavior expression vector corresponding to each search behavior data; and a determination unit, used to determine whether the target commodity is an abnormal commodity based on the behavior expression vector corresponding to each search behavior data.

[0015] Furthermore, the search behavior data at least includes: product search terms and information on visited products, wherein the product search terms are search terms used by buyers to search for the target products when purchasing the target products on the platform, and the information on visited products refers to information on the products visited by the buyer in the search results after the platform returns the search results based on the product search terms. The second acquisition unit includes: a first acquisition module, used to obtain a word vector matrix corresponding to the product search terms and a word vector matrix corresponding to the information on visited products; a first processing module, used to perform pooling processing on the word vector matrix corresponding to the product search terms to obtain a word vector matrix one, and to perform pooling processing on the word vector matrix corresponding to the information on visited products to obtain a word vector matrix two; a second processing module, used to merge the word vector matrix one and the word vector matrix two to obtain a target matrix; a second acquisition module, used to obtain a behavior expression vector corresponding to each search behavior data according to the target matrix.

[0016] Furthermore, the first acquisition module includes: a first processing submodule, used to perform word segmentation processing on the product search terms to obtain a word segmentation set one; a first determination submodule, used to determine the word vector matrix corresponding to the product search terms based on each word in the word segmentation set one; a second processing submodule, used to perform word segmentation processing on the information of accessing the product to obtain a word segmentation set two; a second determination submodule, used to determine the word vector matrix corresponding to the information of accessing the product based on each word in the word segmentation set two.

[0017] According to another aspect of an embodiment of the present invention, a storage medium is further provided, characterized in that the storage medium includes a stored program, wherein when the program is run, the device where the storage medium is located is controlled to execute any one of the above-mentioned commodity verification methods.

[0018] According to another aspect of an embodiment of the present invention, a processor is further provided, characterized in that the processor is used to run a program, wherein the program executes any one of the above-mentioned commodity verification methods when running.

[0019] In an embodiment of the present invention, a search behavior model is introduced, by obtaining a target product that has completed an operation on a platform within a preset time period; obtaining search data associated with the target product, wherein the search data includes at least one search behavior data in the process of operating the target product on the platform; inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data; determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data, thereby achieving the purpose of improving the verification efficiency of abnormal products. By inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data, and determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data, the technical effect of improving the accuracy of abnormal product detection is achieved, thereby solving the technical problem of low verification efficiency of abnormal products due to frequent modification of abnormal product information by product sellers. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0021] Figure 1 It is a hardware structure block diagram of a computer terminal (or mobile device) for implementing a commodity verification method according to an embodiment of the present application;

[0022] Figure 2 is a flow chart of a commodity verification method according to an embodiment of the present invention;

[0023] Figure 3 is a schematic diagram of a commodity authentication device according to an embodiment of the present invention; and

[0024] Figure 4 is a hardware structure block diagram of a computer terminal according to an embodiment of the present invention; DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following explanations:

[0028] Gradient Descent: Gradient descent is an iterative method that can be used to solve least squares problems (both linear and nonlinear). When solving model parameters of machine learning algorithms, that is, unconstrained optimization problems, gradient descent is one of the most commonly used methods. Another commonly used method is the least squares method.

[0029] Recurrent Neural Network (RNN): A recurrent neural network is a type of neural network that takes sequence data as input, performs recursion in the direction of sequence evolution, and all nodes (recurrent units) are connected in a chain.

[0030] User search behavior: the actions that users take before purchasing a product, such as submitting a query, clicking to browse products, etc.

[0031] Cross Entropy: Cross Entropy is an important concept in Shannon information theory. It is mainly used to measure the difference information between two probability distributions. The performance of language models is usually measured by cross entropy and perplexity.

[0032] Berry picking: An information search behavior model that describes how users iterate their queries based on the results returned by the search engine in order to find results that match their intent.

[0033] Example 1

[0034] According to an embodiment of the present invention, a method embodiment of commodity verification is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0035] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 The hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for verifying a commodity is shown. Figure 1 As shown, the computer terminal 10 (or mobile device 10) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0036] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0037] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the method for verifying the commodity in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the method for verifying the commodity of the above-mentioned application program is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0038] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0039] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0040] Under the above operating environment, this application provides Figure 2 The method of verification of the goods shown. Figure 2 The figure is a flow chart of a method for verifying a commodity according to the first embodiment of the present invention.

[0041] Step S101, obtaining target products that have completed operations on the platform within a preset time period.

[0042] The platform mentioned in this application refers to an e-commerce platform for selling goods. The above-mentioned preset time period can be 1 month, and the above-mentioned operations can include transaction operations, browsing, editing and forwarding operations. For example, obtain the target goods that have completed transaction operations on the e-commerce platform within 1 month.

[0043] Step S102, obtaining search data associated with the target product, wherein the search data includes at least one piece of search behavior data during operation of the target product on the platform.

[0044] For example, the above search data at least includes: the search terms entered by the user on the platform to search for target products, the search results returned by the platform after searching on the platform based on the search terms, and the user's access status based on the search results, etc.

[0045] Step S103: input each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data.

[0046] Optionally, in the method for verifying goods according to Example 1 of the present invention, before inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data, the method also includes: obtaining information on illegal goods detected on the platform within a preset period, and randomly selecting information on a preset number of normal goods; obtaining purchase records of users who purchased the illegal goods and the normal goods; obtaining the user's search behavior characteristics based on the purchase records; and modeling based on the user's search behavior characteristics to obtain the search behavior model, wherein the search behavior model is a model generated by search training for abnormal goods and / or safe goods detected on the platform.

[0047] Before training the search behavior model, we first collect data to find hidden products that have been successfully detected in the platform history (corresponding to the abnormal products mentioned above), and randomly select several normal products (corresponding to the safe products mentioned above) as counterexamples. For these products, we extract the behavioral information generated by the user before the purchase and model it in the form of a tree. Specifically, for each purchase record on the platform, we extract the search behavior of the corresponding user, take the user as the root, take all query sequences submitted by the user as branches, and the product sequence clicked corresponding to each query as the corresponding leaf. The entire process from input to output is an end-to-end neural network framework, with cross entropy as the objective function, and then train with stochastic gradient descent to train the search behavior model.

[0048] Optionally, in the method for verifying a product according to Example 1 of the present invention, modeling is performed based on the search behavior characteristics of the user to obtain the search behavior model, including: modeling is performed using a bidirectional long short-term memory network or a recursive neural network based on the search behavior characteristics of the user to obtain the search behavior model.

[0049] In the above scheme, a long short-term memory network (Bi-LSTM) or other recursive neural networks can be used for modeling, such as a gated recursive neuron GRU and a simple recursive neuron SRU; in addition to directly using average pooling for the output of the final recursive neural network RNN, pooling based on an attention mechanism can also be used.

[0050] Optionally, in the method for verifying goods according to the first embodiment of the present invention, the search behavior data at least includes: a product search term and information on visiting goods, wherein the product search term is a search term used by a buyer to search for a target product when the buyer purchases the target product on the platform, and the information on visiting goods refers to information on goods visited by the buyer in the search results after the platform returns the search results based on the product search term. Inputting each search behavior data into a search behavior model to obtain a behavior expression vector corresponding to each search behavior data includes: obtaining a word vector matrix corresponding to the product search term and a word vector matrix corresponding to the information on visiting goods; performing pooling processing on the word vector matrix corresponding to the product search term to obtain a word vector matrix one, and performing pooling processing on the word vector matrix corresponding to the information on visiting goods to obtain a word vector matrix two; merging the word vector matrix one and the word vector matrix two to obtain a target matrix; and obtaining a behavior expression vector corresponding to each search behavior data based on the target matrix.

[0051] The above-mentioned merging process can be to directly concatenate the word vector matrix 1 and the word vector matrix 2, or to add full connection for mapping after concatenation, or to use a gate mechanism to combine the word vector matrix 1 and the word vector matrix 2. It should be noted that there can be multiple product search terms on the platform, and there can also be multiple information for accessing products. Therefore, there can be multiple word vector matrices corresponding to product search terms, and there can be multiple word vector matrices corresponding to information for accessing products. When performing the above-mentioned merging process, multiple word vector matrices 1 and multiple word vector matrices 2 are merged to obtain the final behavior expression vector (a vector) of the entire tree (user).

[0052] Among them, obtaining the word vector matrix corresponding to the product search term and the word vector matrix corresponding to the information of visiting the product may include the following steps: performing word segmentation processing on the product search term to obtain word segmentation set one; based on each word in the word segmentation set one, determining the word vector matrix corresponding to the product search term; performing word segmentation processing on the information of visiting the product to obtain word segmentation set two; based on each word in the word segmentation set two, determining the word vector matrix corresponding to the information of visiting the product.

[0053] In order to verify whether the target product is an abnormal product in the embodiment of the present application, the user search behavior data of each purchase record is organized into a tree shape; all search terms and text sequences of products on the platform are segmented, all the words that appear are counted to make a dictionary, and then each word is mapped to a corresponding id; the corresponding word embedding vector is found through the id of each word (it can be pre-trained or directly randomly initialized), so that each query word and product information can be represented as a matrix composed of word vectors. For example, the product search term is: small movie, and the word segmentation process of "small movie" is performed to obtain the word segmentation set of "small, electric, and movie"; based on each word in the word segmentation set of "small, electric, and movie", the word vector matrix corresponding to "small, electric, and movie" is determined. After searching on the platform based on the product search term "small movie", the platform returns the search results, wherein the search results include "small country movie", the user clicks on the search results of the "small country movie", and the word segmentation process of "small country movie" is performed to obtain the word segmentation set of "small country movie"; based on each word in the word segmentation set of "small country movie", the word vector matrix corresponding to "small country movie" is determined. The word vector matrix corresponding to "small, movie" is pooled to obtain word vector matrix one, and the word vector matrix corresponding to "small country movie" is pooled to obtain word vector matrix two, and then the word vector matrix one and the word vector matrix two are merged to obtain a target matrix; based on the target matrix, a behavior expression vector corresponding to each search behavior data is obtained.

[0054] Step S104: determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data.

[0055] Optionally, in the method for verifying a product according to Example 1 of the present invention, determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data includes: passing the behavior expression vector through a fully connected neural network to obtain a label corresponding to the target product; and determining whether the target product is an abnormal product based on the label corresponding to the target product.

[0056] The label corresponding to the above-mentioned target product may be a value between [0, 1]. For example, a target product with a value greater than or equal to 0.5 is determined as a safe product, and a target product with a value less than 0.5 is determined as an abnormal product.

[0057] It should be noted that the abnormal commodities mentioned in the embodiments of the present application are some hidden commodities, which have the characteristics of violating platform rules and selling fake products under false pretenses. Through the above scheme, the efficiency and accuracy of abnormal commodity detection can be improved. Although sellers can circumvent the existing review mechanism of the e-commerce system by changing the information of the commodity, sellers cannot control the buyer's search behavior. Usually, the buyer's first query is a direct description of the hidden commodity. When the search results are not ideal, they will constantly modify their queries to find their target commodities. Therefore, there is a lot of information to be mined in the buyer's query sequence and the clicked commodity sequence. The output of Bi-LSTM is connected to two fully connected neural networks to obtain the label of the target commodity, and it can be determined whether the target commodity is an abnormal commodity based on the label of the target commodity. In the case of identifying the target commodity as an abnormal commodity, the target commodity can be removed from the shelf to avoid the spread of the commodity on the platform.

[0058] In the method for verifying goods according to the first embodiment of the present invention, a search behavior model is introduced, by obtaining a target product that has completed an operation on a platform within a preset time period; obtaining search data associated with the target product, wherein the search data includes at least one search behavior data in the process of operating the target product on the platform; inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data; and determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data, thereby achieving the purpose of improving the verification efficiency of abnormal products. By inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data, and determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data, the technical effect of improving the accuracy of abnormal product detection is achieved, thereby solving the technical problem of low verification efficiency of abnormal products due to the frequent modification of abnormal product information by product sellers.

[0059] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0060] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0061] Example 2

[0062] According to an embodiment of the present invention, a device for implementing the above commodity verification method is also provided. Figure 3 As shown, the device includes: a first acquiring unit 301, an extracting unit 302, a second acquiring unit 303 and a determining unit 304.

[0063] Specifically, the first acquisition unit 301 is used to acquire the target product that has completed the operation on the platform within a preset time period.

[0064] The extraction unit 302 is used to obtain search data associated with the target product, wherein the search data includes at least one search behavior data in the process of operating the target product on the platform.

[0065] The second acquisition unit 303 is used to input each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data.

[0066] The determination unit 304 is used to determine whether the target product is an abnormal product according to the behavior expression vector corresponding to each search behavior data.

[0067] In summary, in the device for verifying goods according to the second embodiment of the present invention, the first acquisition unit 301 acquires the target goods that have completed the operation on the platform within a preset time period; the extraction unit 302 acquires the search data associated with the target goods, wherein the search data includes at least one search behavior data in the process of operating the target goods on the platform; the second acquisition unit 303 inputs each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data; the determination unit 304 determines whether the target goods are abnormal goods according to the behavior expression vector corresponding to each search behavior data, thereby achieving the purpose of improving the verification efficiency of abnormal goods, by inputting each search behavior data into the search behavior model to obtain the behavior expression vector corresponding to each search behavior data, and determining whether the target goods are abnormal goods based on the behavior expression vector corresponding to each search behavior data, thereby achieving the technical effect of improving the accuracy of abnormal goods detection, thereby solving the technical problem of low verification efficiency of abnormal goods due to the sellers of the goods often modifying the information of abnormal goods.

[0068] Optionally, in a device for verifying goods according to the second embodiment of the present invention, the search behavior data includes at least: product search terms and information on visiting goods, wherein the product search terms are search terms used by buyers to search for target goods when purchasing target goods on the platform, and the information on visiting goods refers to information on goods visited by buyers in the search results after the platform returns the search results based on the product search terms. The second acquisition unit includes: a first acquisition module, used to obtain a word vector matrix corresponding to the product search terms and a word vector matrix corresponding to the information on visiting goods; a first processing module, used to perform pooling processing on the word vector matrix corresponding to the product search terms to obtain word vector matrix one, and to perform pooling processing on the word vector matrix corresponding to the information on visiting goods to obtain word vector matrix two; a second processing module, used to merge word vector matrix one and word vector matrix two to obtain a target matrix; and a second acquisition module, used to obtain a behavior expression vector corresponding to each search behavior data according to the target matrix.

[0069] Optionally, in the device for verifying goods according to the second embodiment of the present invention, the first acquisition module includes: a first processing sub-module, used to perform word segmentation processing on the product search terms to obtain a word segmentation set one; a first determination sub-module, used to determine the word vector matrix corresponding to the product search terms according to each word in the word segmentation set one; a second processing sub-module, used to perform word segmentation processing on the information for accessing the product to obtain a word segmentation set two; and a second determination sub-module, used to determine the word vector matrix corresponding to the information for accessing the product according to each word in the word segmentation set two.

[0070] Optionally, in the device for verifying goods according to embodiment 2 of the present invention, the determination unit includes: a third acquisition module, used to pass the behavior expression vector through a fully connected neural network to obtain a label corresponding to the target product; and a determination module, used to determine whether the target product is an abnormal product based on the label corresponding to the target product.

[0071] Optionally, in the device for verifying goods according to the second embodiment of the present invention, the device also includes: a third acquisition unit, used to obtain information on illegal goods detected on the platform within a preset period, and randomly select information on a preset number of normal goods before inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data; a fourth acquisition unit, used to obtain purchase records of users who purchase illegal goods and normal goods; a fifth acquisition unit, used to obtain the user's search behavior characteristics based on the purchase records; a modeling unit, used to perform modeling based on the user's search behavior characteristics to obtain a search behavior model, wherein the search behavior model is a model generated by search training for abnormal goods and / or safe goods detected on the platform.

[0072] Optionally, in the device for verifying goods according to the second embodiment of the present invention, the modeling unit includes: a modeling module, which is used to model according to the user's search behavior characteristics using a bidirectional long short-term memory network or a recursive neural network to obtain a search behavior model.

[0073] Optionally, in the device for verifying goods according to the second embodiment of the present invention, in the second processing module, the merging process is one of the following: splicing process, and combining using a gate mechanism.

[0074] Optionally, in the device for verifying goods according to the second embodiment of the present invention, the pooling process is one of the following: average pooling and pooling based on an attention mechanism.

[0075] It should be noted that the first acquisition unit 301, the extraction unit 302, the second acquisition unit 303 and the determination unit 304 described above correspond to steps S101 to S104 in Example 1, and the examples and application scenarios implemented by the two modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules as part of the device can be run in the computer terminal 10 provided in Example 1.

[0076] Example 3

[0077] The embodiment of the present invention can provide a computer terminal, which can be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal can also be replaced by a terminal device such as a mobile terminal.

[0078] Optionally, in this embodiment, the computer terminal may be located in at least one network device among a plurality of network devices of the computer network.

[0079] In this embodiment, the above-mentioned computer terminal can execute the program code of the following steps in the method for verifying the goods of the application: obtaining a target product that has completed operations on the platform within a preset time period; obtaining search data associated with the target product, wherein the search data includes at least one search behavior data in the process of operating the target product on the platform; inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data; and determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data.

[0080] The above-mentioned computer terminal can also execute the program code of the following steps in the method for verifying the goods of the application program: the search behavior data at least includes: product search terms and information on visiting goods, wherein the product search terms are search terms used by buyers to search for the target goods when purchasing the target goods on the platform, and the information on visiting goods refers to the information on the goods visited by the buyer in the search results after the platform returns the search results based on the product search terms. Inputting each search behavior data into the search behavior model, and obtaining the behavior expression vector corresponding to each search behavior data includes: obtaining the word vector matrix corresponding to the product search terms and the word vector matrix corresponding to the information on visiting goods; performing pooling processing on the word vector matrix corresponding to the product search terms to obtain word vector matrix one, and performing pooling processing on the word vector matrix corresponding to the information on visiting goods to obtain word vector matrix two; merging the word vector matrix one and the word vector matrix two to obtain a target matrix; obtaining the behavior expression vector corresponding to each search behavior data based on the target matrix.

[0081] The above-mentioned computer terminal can also execute the program code of the following steps in the method for verifying the product of the application: obtaining the word vector matrix corresponding to the product search term and the word vector matrix corresponding to the information of accessing the product, including: performing word segmentation processing on the product search term to obtain word segmentation set one; based on each word in the word segmentation set one, determining the word vector matrix corresponding to the product search term; performing word segmentation processing on the information of accessing the product to obtain word segmentation set two; based on each word in the word segmentation set two, determining the word vector matrix corresponding to the information of accessing the product.

[0082] The above-mentioned computer terminal can also execute the program code of the following steps in the method of verifying the product of the application: determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data includes: passing the behavior expression vector through a fully connected neural network to obtain a label corresponding to the target product; determining whether the target product is an abnormal product based on the label corresponding to the target product.

[0083] The above-mentioned computer terminal can also execute the program code of the following steps in the method for verifying the goods of the application: before inputting each search behavior data into the search behavior model to obtain the behavior expression vector corresponding to each search behavior data, the method also includes: obtaining information on illegal goods detected on the platform within a preset period, and randomly selecting information on a preset number of normal goods; obtaining purchase records of users who purchased the illegal goods and the normal goods; obtaining the user's search behavior characteristics based on the purchase records; modeling based on the user's search behavior characteristics to obtain the search behavior model, wherein the search behavior model is a model generated by search training for abnormal goods and / or safe goods detected on the platform.

[0084] The above-mentioned computer terminal can also execute the program code of the following steps in the method of verifying the goods of the application: modeling based on the search behavior characteristics of the user, and obtaining the search behavior model includes: modeling based on the search behavior characteristics of the user using a bidirectional long short-term memory network or a recursive neural network to obtain the search behavior model.

[0085] The above-mentioned computer terminal can also execute the program code of the following steps in the method of verifying the product of the application: merging the word vector matrix one and the word vector matrix two, wherein the merging process is one of the following: splicing process, combining using a gate mechanism.

[0086] The computer terminal may also execute the program code of the following steps in the method for verifying the commodity of the application: the pooling process is one of the following: average pooling, pooling based on the attention mechanism.

[0087] Optionally, Figure 4 is a structural block diagram of a computer terminal according to an embodiment of the present invention. Figure 4 As shown, the computer terminal A may include: one or more (only one is shown in the figure) processors and a memory.

[0088] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the method and device for verifying the goods in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, the method for verifying the goods mentioned above is realized. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include a memory remotely arranged relative to the processor, and these remote memories can be connected to the terminal A via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0089] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: obtain the target product that has completed the operation on the platform within a preset time period; obtain the search data associated with the target product, wherein the search data includes at least one search behavior data in the process of operating the target product on the platform; input each search behavior data into the search behavior model to obtain the behavior expression vector corresponding to each search behavior data; and determine whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data.

[0090] Optionally, the processor may also execute the program code of the following steps: the search behavior data includes at least: product search terms and information on visited products, wherein the product search terms are search terms used by buyers to search for the target products when purchasing the target products on the platform, and the information on visited products refers to information on the products visited by the buyer in the search results after the platform returns the search results based on the product search terms. Inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data includes: obtaining a word vector matrix corresponding to the product search terms and a word vector matrix corresponding to the information on visited products; performing pooling processing on the word vector matrix corresponding to the product search terms to obtain a word vector matrix one, and performing pooling processing on the word vector matrix corresponding to the information on visited products to obtain a word vector matrix two; merging the word vector matrix one and the word vector matrix two to obtain a target matrix; and obtaining a behavior expression vector corresponding to each search behavior data based on the target matrix.

[0091] Optionally, the processor may also execute the program code of the following steps: obtaining the word vector matrix corresponding to the product search term and the word vector matrix corresponding to the information of accessing the product, including: performing word segmentation processing on the product search term to obtain word segmentation set one; determining the word vector matrix corresponding to the product search term based on each word in the word segmentation set one; performing word segmentation processing on the information of accessing the product to obtain word segmentation set two; determining the word vector matrix corresponding to the information of accessing the product based on each word in the word segmentation set two.

[0092] Optionally, the processor may also execute program code for the following steps: determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data, including: passing the behavior expression vector through a fully connected neural network to obtain a label corresponding to the target product; and determining whether the target product is an abnormal product based on the label corresponding to the target product.

[0093] Optionally, the processor may also execute program code for the following steps: before inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data, the method further includes: obtaining information on illegal products detected on the platform within a preset period, and randomly selecting information on a preset number of normal products; obtaining purchase records of users who purchased the illegal products and the normal products; obtaining the user's search behavior characteristics based on the purchase records; and modeling based on the user's search behavior characteristics to obtain the search behavior model, wherein the search behavior model is a model generated by search training for abnormal products and / or safe products detected on the platform.

[0094] Optionally, the processor may also execute the program code of the following steps: modeling based on the search behavior characteristics of the user to obtain the search behavior model, including: modeling based on the search behavior characteristics of the user using a bidirectional long short-term memory network or a recursive neural network to obtain the search behavior model.

[0095] Optionally, the processor may also execute the program code of the following steps: merging the word vector matrix 1 and the word vector matrix 2, wherein the merging process is one of the following: splicing process, combining using a gate mechanism.

[0096] Optionally, the processor may also execute program code of the following steps: the pooling process is one of the following: average pooling, pooling based on an attention mechanism.

[0097] By adopting the embodiment of the present invention, a method for verifying a commodity is provided. By obtaining a target commodity that has completed an operation on a platform within a preset time period; obtaining search data associated with the target commodity, wherein the search data includes at least one search behavior data in the process of operating the target commodity on the platform; inputting each search behavior data into a search behavior model to obtain a behavior expression vector corresponding to each search behavior data; determining whether the target commodity is an abnormal commodity based on the behavior expression vector corresponding to each search behavior data, the purpose of improving the verification efficiency of abnormal commodities is achieved, by inputting each search behavior data into a search behavior model to obtain a behavior expression vector corresponding to each search behavior data, and determining whether the target commodity is an abnormal commodity based on the behavior expression vector corresponding to each search behavior data, thereby achieving the technical effect of improving the accuracy of abnormal commodity detection, thereby solving the technical problem of low verification efficiency of abnormal commodities due to the frequent modification of abnormal commodity information by commodity sellers.

[0098] It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only, and the computer terminal may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a handheld computer, a mobile Internet device (Mobile Internet Devices, MID), a PAD, or other terminal devices. Figure 4 The structure of the electronic device is not limited. For example, the computer terminal 10 may also include Figure 4 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 4 Different configurations shown.

[0099] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0100] Example 4

[0101] The embodiment of the present invention further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the commodity verification method provided in the first embodiment.

[0102] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0103] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: obtaining a target product that has completed an operation on the platform within a preset time period; obtaining search data associated with the target product, wherein the search data includes at least one search behavior data during the operation of the target product on the platform; inputting each search behavior data into a search behavior model to obtain a behavior expression vector corresponding to each search behavior data; and determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data.

[0104] Optionally, in this embodiment, the storage medium is also configured to store program codes for executing the following steps: the search behavior data includes at least: product search terms and information on visited products, wherein the product search terms are search terms used by buyers to search for the target products when purchasing the target products on the platform, and the information on visited products refers to information on the products visited by the buyer in the search results after the platform returns the search results based on the product search terms. Inputting each search behavior data into the search behavior model, and obtaining a behavior expression vector corresponding to each search behavior data includes: obtaining a word vector matrix corresponding to the product search terms and a word vector matrix corresponding to the information on visited products; performing pooling processing on the word vector matrix corresponding to the product search terms to obtain a word vector matrix one, and performing pooling processing on the word vector matrix corresponding to the information on visited products to obtain a word vector matrix two; merging the word vector matrix one and the word vector matrix two to obtain a target matrix; and obtaining a behavior expression vector corresponding to each search behavior data based on the target matrix.

[0105] Optionally, in this embodiment, the storage medium is also configured to store program codes for executing the following steps: obtaining the word vector matrix corresponding to the product search term and the word vector matrix corresponding to the information of accessing the product includes: performing word segmentation processing on the product search term to obtain word segmentation set one; based on each word in the word segmentation set one, determining the word vector matrix corresponding to the product search term; performing word segmentation processing on the information of accessing the product to obtain word segmentation set two; based on each word in the word segmentation set two, determining the word vector matrix corresponding to the information of accessing the product.

[0106] Optionally, in this embodiment, the storage medium is also configured to store program codes for executing the following steps: determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data includes: passing the behavior expression vector through a fully connected neural network to obtain a label corresponding to the target product; determining whether the target product is an abnormal product based on the label corresponding to the target product.

[0107] Optionally, in this embodiment, the storage medium is also configured to store program codes for executing the following steps: before inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data, the method also includes: obtaining information on illegal products detected on the platform within a preset period, and randomly selecting information on a preset number of normal products; obtaining purchase records of users who purchased the illegal products and the normal products; obtaining the user's search behavior characteristics based on the purchase records; and modeling based on the user's search behavior characteristics to obtain the search behavior model, wherein the search behavior model is a model generated by search training for abnormal products and / or safe products detected on the platform.

[0108] Optionally, in this embodiment, the storage medium is also configured to store program code for executing the following steps: modeling based on the search behavior characteristics of the user to obtain the search behavior model, including: modeling based on the search behavior characteristics of the user using a bidirectional long short-term memory network or a recursive neural network to obtain the search behavior model.

[0109] Optionally, in this embodiment, the storage medium is also configured to store program code for executing the following steps: merging the word vector matrix one and the word vector matrix two, wherein the merging process is one of the following: splicing process, combining using a gate mechanism.

[0110] Optionally, in this embodiment, the storage medium is also configured to store program codes for executing the following steps: the pooling process is one of the following: average pooling, pooling based on an attention mechanism.

[0111] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0112] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0113] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0114] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0115] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0116] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0117] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A commodity verification method, characterized in that: include: Obtain target products that have completed operations on the platform within a preset time period; Acquire search data associated with the target product, wherein the search data includes at least one search behavior data in the process of operating the target product on the platform; Input each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data; Determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data; The search behavior data at least includes: product search terms and information about visited products, wherein the product search terms are search terms used by buyers to search for target products when purchasing target products on the platform, and the information about visited products refers to information about products visited by the buyer in the search results after the platform returns the search results based on the product search terms; Wherein, each search behavior data is input into the search behavior model, and the behavior expression vector corresponding to each search behavior data is obtained, including: Obtaining a word vector matrix corresponding to the product search term and a word vector matrix corresponding to the information of the accessed product; Pooling the word vector matrix corresponding to the product search term to obtain a word vector matrix 1, and pooling the word vector matrix corresponding to the information of the accessed product to obtain a word vector matrix 2; Merging the word vector matrix 1 and the word vector matrix 2 to obtain a target matrix; Obtaining a behavior expression vector corresponding to each search behavior data based on the target matrix; Determining whether the target product is an abnormal product based on the behavior expression vector corresponding to each search behavior data includes: Passing the behavior expression vector through a fully connected neural network to obtain a label corresponding to the target product; Determine whether the target product is an abnormal product based on the label corresponding to the target product.

2. The method according to claim 1, characterized in that Obtaining the word vector matrix corresponding to the product search term and the word vector matrix corresponding to the information of the accessed product includes: Performing word segmentation processing on the product search term to obtain a word segmentation set 1; Based on each word in the word segmentation set 1, determine the word vector matrix corresponding to the product search word; Performing word segmentation processing on the information of the visited products to obtain a second word segmentation set; Based on each word in the second word segmentation set, a word vector matrix corresponding to the information of the accessed product is determined.

3. The method according to claim 1, characterized in that Before inputting each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data, the method further includes: Obtain information about illegal products detected on the platform within a preset period, and randomly select information about a preset number of normal products; Obtaining purchase records of users who purchased the illegal products and the normal products; Acquire the user's search behavior characteristics based on the purchase record; Modeling is performed based on the search behavior characteristics of the user to obtain the search behavior model, wherein the search behavior model is a model generated by performing search training on abnormal products and / or safe products detected on the platform.

4. The method according to claim 3, characterized in that Modeling is performed based on the search behavior characteristics of the user, and the search behavior model obtained includes: Based on the search behavior characteristics of the user, a bidirectional long short-term memory network or a recursive neural network is used to perform modeling to obtain the search behavior model.

5. The method according to claim 1, characterized in that The word vector matrix 1 and the word vector matrix 2 are merged, wherein the merging process is one of the following: splicing process and combining by using a gate mechanism.

6. The method according to claim 1, characterized in that The pooling process is one of the following: average pooling and pooling based on an attention mechanism.

7. A commodity authentication device, characterized in that: include: A first acquisition unit is used to acquire a target product that has completed an operation on the platform within a preset time period; An extraction unit, configured to obtain search data associated with the target product, wherein the search data includes at least one search behavior data in a process of operating the target product on the platform; A second acquisition unit is used to input each search behavior data into the search behavior model to obtain a behavior expression vector corresponding to each search behavior data; A determination unit, configured to determine whether the target product is an abnormal product according to a behavior expression vector corresponding to each search behavior data; The search behavior data at least includes: product search terms and information about visited products, wherein the product search terms are search terms used by buyers to search for target products when purchasing target products on the platform, and the information about visited products refers to information about products visited by the buyer in the search results after the platform returns the search results based on the product search terms; The second acquiring unit includes: A first acquisition module, used to acquire a word vector matrix corresponding to the product search term and a word vector matrix corresponding to the information of accessing the product; A first processing module is used to perform pooling processing on the word vector matrix corresponding to the product search term to obtain a word vector matrix 1, and to perform pooling processing on the word vector matrix corresponding to the information of the accessed product to obtain a word vector matrix 2; A second processing module is used to merge the word vector matrix 1 and the word vector matrix 2 to obtain a target matrix; A second acquisition module, used for obtaining a behavior expression vector corresponding to each search behavior data according to the target matrix; The determination unit includes: a third acquisition module, used to pass the behavior expression vector through a fully connected neural network to obtain a label corresponding to the target product; and a determination module, used to determine whether the target product is an abnormal product according to the label corresponding to the target product.

8. The device according to claim 7, characterized in that The first acquisition module includes: A first processing submodule is used to perform word segmentation processing on the product search term to obtain a word segmentation set 1; A first determination submodule, configured to determine a word vector matrix corresponding to the product search term according to each word in the word segmentation set 1; The second processing submodule is used to perform word segmentation processing on the information of the visited products to obtain a second word segmentation set; The second determination submodule is used to determine the word vector matrix corresponding to the information of the accessed product according to each word in the second word segmentation set.

9. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the commodity verification method according to any one of claims 1 to 6.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the commodity verification method described in any one of claims 1 to 6.

Images