Sorting model modeling method and device, commodity object searching method and device and medium

Abstract

The embodiment of the invention provides a sequencing model modeling method, a commodity object searching method, equipment and a medium. The method comprises the following steps: deleting search associated data of a target language site from search associated data to obtain target search associated data; screening feature data from the target search associated data according to a set rule, and constructing training data; constructing a sorting model; and training the sorting model based on the training data to obtain a sorting model meeting conditions. According to the method, the influence of the target language on the data can be removed, so that the target search associated data of each language is more balanced, non-uniform distribution of the data of each site to a small-flow language is eliminated, a sorting model supporting various languages is obtained, and the search of various languages can be accurately fed back.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method for constructing a sorting model, a method for searching for commodity objects, an electronic device, and a storage medium. Background Art

[0002] Cross-border e-commerce websites provide cross-border e-commerce services. Users can purchase goods from different countries on cross-border e-commerce websites. Since cross-border e-commerce websites support the sale of goods between different countries, they also support different languages.

[0003] When users search on cross-border e-commerce websites, they usually filter product objects based on the keywords they input. However, since e-commerce websites support multiple languages, the amount of data on users, products, etc. corresponding to different languages ​​on e-commerce websites varies greatly, which affects the sorting model. Product objects corresponding to languages ​​with more data are usually ranked higher in search sorting and are easier to provide feedback to users during search. Product objects in minority languages ​​tend to be ranked lower in search sorting due to the smaller amount of data, and have less feedback during search. This in turn affects the query and sales of product objects by some users who sell in minority languages. Summary of the invention

[0004] The embodiment of the present application provides a method for constructing a ranking model to improve the accuracy of the ranking model.

[0005] Correspondingly, the embodiments of the present application also provide a method for commodity objects, an electronic device and a storage medium to ensure the implementation and application of the above system.

[0006] In order to solve the above problems, the present application discloses a method for constructing a ranking model, which includes:

[0007] Deleting the search related data of the target language site from the search related data to obtain the target search related data;

[0008] From the target search related data, feature data is filtered according to set rules to construct training data;

[0009] Build a sorting model;

[0010] The sorting model is trained based on the training data to obtain a sorting model that meets the conditions.

[0011] Optionally, before deleting the search related data of the target language site from the search related data and obtaining the target search related data, the method further includes:

[0012] Get search related data of each language site;

[0013] The data volume information corresponding to each language site is determined based on the search-related data, and the target language site is determined based on the data volume information.

[0014] Optionally, filtering feature data from the target search related data according to set rules to construct training data includes:

[0015] Sampling information of target search related data based on each language site;

[0016] According to the adopted information, data of target features are collected from target search related data of each language site to construct training data.

[0017] Optionally, the sampling information of the target search related data based on each language site includes:

[0018] Based on the target search related data of each language site, determine the data volume information and behavior information of each language site;

[0019] Based on the data volume information and the behavior information, sampling information of the target search related data of each language site is determined.

[0020] Optionally, also include:

[0021] Extracting feature data of each feature from the associated data, and analyzing the correlation information between each feature data and the behavior information;

[0022] Target features are filtered based on the correlation information.

[0023] Optionally, the method further includes: constructing a site difference feature as a target feature for each commodity object.

[0024] Optionally, constructing site difference features for each commodity object includes:

[0025] Generate a difference feature based on at least one of a language feature, a commodity object feature, and a conversion feature;

[0026] Determine the site characteristics of the site corresponding to the commodity object;

[0027] The difference feature and the site feature are concatenated to obtain the corresponding site difference feature.

[0028] Optionally, generating a difference feature based on at least one of a language feature, a commodity object feature, and a conversion feature includes:

[0029] Obtaining correlation features of the product object, the correlation features include the interaction depth feature of the language, the conversion rate feature of the product object, and the service quality feature of the seller user;

[0030] Determining a conversion feature of the commodity object based on the association feature;

[0031] Conversion features, language features, and product object features are combined to determine the corresponding difference features.

[0032] Optionally, constructing a ranking model includes:

[0033] An activation function is determined, and the activation function is added to the neural network model to construct a sorting model.

[0034] Optionally, determining the target language site based on the data volume information includes:

[0035] When the data volume information corresponding to the site exceeds the target threshold, the site is determined to be a target language site, and the language of the site is the target language.

[0036] The present application also discloses a method for searching for commodity objects, the method comprising:

[0037] receiving a search request, wherein the search request carries a keyword;

[0038] Searching based on the keyword to determine multiple commodity objects;

[0039] Analyze multiple commodity objects based on a ranking model to filter commodity objects, wherein the ranking model is obtained based on training data, the training data is constructed based on feature data filtered according to set rules in target search related data, and the target search related data is determined by deleting a target language site from the search related data;

[0040] Generate search results based on the filtered product objects and feedback.

[0041] Optionally, analyzing the multiple commodity objects based on the ranking model to screen the commodity objects includes:

[0042] Collect target feature data from the feature data of the commodity object and input it into the sorting model;

[0043] The commodity objects are screened based on the output result of the ranking model.

[0044] An embodiment of the present application also discloses an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method described in the embodiment of the present application.

[0045] The embodiment of the present application further discloses a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method described in the embodiment of the present application.

[0046] Compared with the prior art, the embodiments of the present application have the following advantages:

[0047] In an embodiment of the present application, the search associated data of the target language site is deleted from the search associated data to obtain the target search associated data, thereby removing the influence of the target language on the data and making the target search associated data of each language more balanced. Then, feature data is filtered from the target search associated data according to set rules to construct training data. The required feature data can be filtered as training data, and then a sorting model is constructed. The sorting model is trained based on the training data to obtain a sorting model that meets the conditions. This can eliminate the uneven distribution of data at each site for small traffic languages ​​and obtain a sorting model that supports various languages, so that searches in various languages ​​can be accurately fed back. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flowchart of the steps of an embodiment of a method for constructing a sorting model of the present application;

[0049] Figure 2 is a flowchart of another method for constructing a sorting model according to an embodiment of the present application;

[0050] Figure 3 It is a flowchart of the steps of an embodiment of a commodity object search method of the present application;

[0051] Figure 4 It is a schematic diagram of the structure of an exemplary device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0053] The embodiments of the present application can be applied to the search field. Cross-border e-commerce websites provide services for sellers and buyers in hundreds of countries and multiple languages, so websites can be set up in different countries for easy access. Different countries use different languages. For example, English-speaking countries such as the United Kingdom and the United States use English, and these websites can be called English sites. German websites use German, and the corresponding websites can be called German sites. Therefore, cross-border e-commerce websites include sites in various languages.

[0054] When training a sorting model, the embodiments of the present application can reduce the impact of big data language on sorting and improve the accuracy of the sorting model.

[0055] Step 102: Delete the search related data of the target language site from the search related data to obtain the target search related data.

[0056] The search-related data is obtained from the database of the cross-border e-commerce website. The search-related data includes various data related to the searched commodity object, such as commodity object data, keywords corresponding to the commodity object, search click data, store data, etc. Therefore, in the database of the cross-border e-commerce website, the search-related data of each language site can be obtained. Among them, the language of the site is mainly based on the default language of the site, for example, the corresponding sites of the United Kingdom and the United States are English sites, the Japanese site is a Japanese site, and the Korean site is a Korean site.

[0057] The embodiment of the present application builds a sorting model that supports searches in various languages. Among them, there are sites in some languages ​​with more data in cross-border e-commerce websites. If the sorting model is to be trained for all languages ​​together, these more data will affect the accuracy of the sorting model, resulting in inaccurate sorting results. Therefore, the language with more data can be determined as the target language, the search-related data of the target language site can be determined, and the search-related data of the target language site can be deleted from the search-related data. In the embodiment of the present application, the search-related data of the target language site is deleted, not all the data of the target language mentioned above, and there will also be search data corresponding to some target languages ​​in sites in other languages, so as not to affect the search sorting of the target language. For example, in the Korean site in South Korea, there will also be users who use the application English search, that is, the language of the site is the default language of the site, and there will be no restrictions on the language used by the user in the site. The embodiment of the present application can set the corresponding language as the target language, and the site with the target language as the reference language is the target language site, such as an English site. In some other embodiments, one or more target language sites can also be determined based on the corresponding data volume of the site.

[0058] In an optional embodiment, the search-related data of the target language site is deleted from the search-related data, and before the target search-related data is obtained, the method further includes: obtaining the search-related data of each language site; determining the data volume information corresponding to each language site based on the search-related data, and determining the target language site based on the data volume information. The search-related data of each language site is obtained based on log data, etc., and then the data volume information corresponding to each language site, such as the size of the data volume, the proportion of the data volume, etc., can be determined based on the search-related data. The target language site is determined based on the data volume information. Among them, the target language is usually one language, or more than one language, and is determined based on the data of each language site in the cross-border e-commerce website. In the embodiment of the present application, the determination of the target language site based on the data volume information includes: judging whether the data volume information corresponding to the language site exceeds the target threshold, and if the data volume information corresponding to a certain language site exceeds the target threshold, the language site is determined as the target language site. Among them, the target threshold is the threshold for judging the target site, which can be a threshold for the data volume or a threshold for the data proportion. For example, if the data volume of a certain site accounts for more than 60% of the total site data volume, or exceeds 50%, 40%, etc., it can be determined as a target language site.

[0059] In an example cross-border e-commerce website, the historical logs of English sites account for nearly 65% ​​of the entire website; while low-traffic sites such as Korean, Japanese, and Vietnamese account for only less than 2%. If the data of each language site is used for unified training, the characteristics of language sites in other small languages ​​may be completely dominated by the English site. Therefore, in the embodiment of the present application, the search-related data of the English site is deleted, and the search-related data of other language sites are used for training. It should be noted that there is still nearly 30% of search data corresponding to English keywords in the data traffic of each language site other than English, which can make up for the missing proportion of the search-related data of the English site, so that the sorting model supporting each language still has the ability to learn English keywords.

[0060] Step 104: filter feature data from the target search related data according to set rules to construct training data.

[0061] The target search associated data includes various data related to the search, but not all data has an impact on the ranking, and different types of data have different impacts on the ranking, so it is necessary to screen the required feature data and construct training samples.

[0062] Since data is obtained from each language site for training, and different language sites have different corresponding traffic, user behavior, etc., it is also necessary to determine sampling information such as the sampling ratio of the data corresponding to each language site, and sample the data according to the sampling information to train the data. The feature data is filtered from the target search related data according to the set rules to construct the training data, including: sampling information based on the target search related data of each language site; collecting target feature data from the search data of each language site according to the adopted information to construct the training data. The sampling information based on the target search related data of each language site includes: determining the data volume information and behavior information of each language site based on the target search related data of each language site; determining the sampling information of the target search related data of each language site based on the data volume information and behavior information.

[0063] Among them, the traffic corresponding to the data of different language sites is different. Some language sites have large traffic, that is, many visiting users, while some language sites have small traffic, that is, small access data. However, the amount of traffic is not proportional to the actual user interaction behavior of the website, that is, some websites have small traffic, but the conversion rate of commodity objects is high. Among them, the conversion rate refers to the ratio of the number of users after using a single interactive behavior and overall function in a commodity object to the number of users before using it. Therefore, based on the target search associated data of each language site, the data volume information of each language site, such as the proportion of data volume, etc., can be determined, and the conversion rate information corresponding to each language site can be determined based on the user's behavior information, and then based on the data volume information and the conversion rate information, the sampling information such as the sampling ratio of each language site can be determined. Therefore, the embodiment of the present application can adjust different sampling ratios of deep inquiry samples for different sites, wherein an inquiry refers to the process of verbally or in writing sending a transaction to the other party to explore transaction conditions in order to buy or sell a commodity object. In an e-commerce website, the buyer user's visit to the commodity object, the purchase and communication with the customer can all be called an inquiry.

[0064] Taking a cross-border e-commerce website as an example, based on the analysis of website log data, it can be found that the data distribution of each site is also different. Among the sites in other languages ​​other than English sites, the top five sites in terms of traffic are Spanish ( ES), French (French, FR), Russian (Russian, RU), Arabic (Arabic, AR), and Turkish (Turkish, TR) sites contribute nearly 60% of the traffic, especially the Spanish ES site, which accounts for 30% of the total traffic. However, from the perspective of user conversion, the conversion rate of sites such as Korean, Japanese, and Vietnamese, which account for less than 5%, is higher than the conversion rate of the top five sites in terms of traffic. In order to make the contribution of sites in different languages ​​more even, different sampling ratios of deep inquiry samples can be adjusted for different sites. For example, sites with low conversion rates of head traffic can sample behavioral data at a lower ratio, while sites with high conversion rates of tail traffic can sample behavioral data at a higher ratio, which can help alleviate the problem of sparse and long-tail behavioral data.

[0065] After determining the sampling information such as the sampling ratio, the data of the target feature can be collected from the search data of each language site according to the sampling information to construct the training data. Among them, since different data have different effects on the ranking, the target feature can also be analyzed and determined, and the data of the target feature can be collected.

[0066] In an optional embodiment, feature data of each feature is extracted from the associated data, the correlation information of each feature data and the behavior information is analyzed, and the target feature is screened based on the correlation information. Among them, the behavior information can be the behavior information of interaction with the commodity object, such as click behavior, visit behavior, add-to-cart behavior, purchase behavior, etc. The correlation information of each feature data and the behavior information can be analyzed. For example, the payment time and brand age of the order of the commodity object are weakly correlated features. Deleting such weakly correlated features can reduce the impact of irrelevant features. For example, the importance of various features of the interactive behavior can be scored by drop-rank (reduced-rank), the scoring interval is between 0-1, and the importance is arranged according to the branches from 0 to 1, and then the features with relatively small impact are eliminated according to the scoring and processing logic. The drop-rank (reduced-rank) method can make the probability of low-importance features appearing in the neural network low by adding perturbation variables. This method can take into account the combination between features, and it is more effective to screen the top N features, and the calculation amount is also relatively small (training the model once). Therefore, nearly 50% of weakly correlated features such as payment time and brand age can be deleted through the above method.

[0067] In the embodiment of the present application, keywords are also determined as target features based on historical data. It can obtain historical data within a specified time period, such as search keywords, such as 1 month, 3 months, half a year and other specified time periods, and then use language models (LM) such as uni, bi-gram and other models to train and analyze the historical data to calculate the difference, conversion rate, and popularity between keywords and score them, and obtain the keywords corresponding to the core attributes of each industry and each site as the target features. By adding keywords corresponding to these core attributes as target features, the proportion of important features in the training data can be increased, and the dominance of core features can be increased. By mining the core attributes of different language sites and industries based on user perspective demands, the click-through rate of commodity objects in actual applications is +13%, which verifies the correctness of this idea.

[0068] In the embodiment of the present application, a site difference feature is constructed as a target feature for each commodity object. Among them, the site difference feature is composed of a site feature combined with a difference feature, for example, the site feature and the difference feature are spliced ​​to obtain a difference feature. Among them, the site feature is the feature of the site of the commodity object. The difference feature can be generated by constructing the language feature, commodity object feature and conversion feature corresponding to each commodity object. Among them, the language feature is the feature corresponding to the language, for example, a corresponding feature value or vector value is set for each language as the feature of the language. The commodity object feature is the feature of the commodity object, which can be determined based on commodity object information, etc. The conversion feature is a feature obtained by converting multiple features of the commodity object, which is a feature obtained by integrating multiple features, for example, based on language, conversion rate, service quality and other information calculations, to determine the corresponding conversion feature. In an optional embodiment, the associated feature of the commodity object can be obtained, and the associated feature includes the interactive depth feature of the language, the conversion rate feature of the commodity object and the service quality feature of the seller user, and the conversion feature of the commodity object is determined based on the associated feature; the conversion feature, the language feature, and the commodity object feature are spliced ​​to determine the corresponding difference feature. Among them, the interaction depth characteristics of a language can be determined based on the interaction data corresponding to the language of the language, such as user consultation data on product objects, etc., for example, based on the depth characteristics of communication in each field under each multilingual site, basic information on product communication under each multilingual site, etc.

[0069] By building language and site difference features, we can balance the learning of differentiated representations of commodity objects on each site, and also support the unified learning of differentiated features of different sites by the same model, thus optimizing resource utilization and conversion efficiency.

[0070] Then, according to the sampling information, data of target features can be collected from the target search-related data of each language site to construct training data.

[0071] Step 106, constructing a sorting model.

[0072] To sort the search results of commodity objects, it is necessary to first construct a sorting model, wherein the sorting model can be constructed based on various sorting algorithms, neural network models, linear sorting models, tree models, deep learning models, etc. The sorting model can be determined based on click-through rate, visit rate, etc.

[0073] In the embodiment of the present application, the sorting model can be constructed based on the neural network model, so the activation function can be determined, and the activation function is added to the neural network model to construct the sorting model. In the embodiment of the present application, a variety of activation functions can be set, for example, the activation function includes a linear activation function and a nonlinear activation function: sigmod, ReLU, ELU, Leaky ReLU, CReLU, relu6, SELU, softplus, etc. Among them, the activation function is a function added to the neural network to help the network learn complex patterns in the data. In neurons, the input acts on the activation function after a series of weighted summations, which is similar to the neuron-based model in the human brain. The activation function ultimately determines whether to pass the signal and the content to be emitted to the next neuron. Thus, the efficiency of the neural network can be improved by adjustment. Among them, sigmod is a smoothed step function for binary classification. ReLU (Rectified Linear Unit) is used to activate some neurons and increase sparsity. Leaky ReLU (Leaky ReLU), the difference between Leaky ReLU and ReLU is that the negative axis retains the constant leak, so that when the input information is less than 0, the information is not completely lost. ELU is an exponential linear unit. CReLU is a concatenated rectified linear unit, a rectified linear unit with ReLU6 taking 6, which is mainly used to have good numerical resolution when the mobile terminal has low precision. SELU is a balanced exponential linear unit. The Softplus function can be regarded as a smoothed version of the Relu function, and its derivative is just the Logistic function. Although the Softplus function also has the characteristics of unilateral inhibition and wide excitation boundaries, it does not have sparse activation and will not sparse the model.

[0074] The sorting model includes an embedding layer, which can convert input features into feature vectors. The embodiment of the present application can determine the embedding layer based on loading personalized vectors, correlation model learning, etc., so as to initialize the parameters of the model. For example, a personalized vector recall pre-trained identifier embedding layer can be loaded to perform vector conversion on the identifier. Load the text embedding layer trained by the language representation model (Bidirectional Encoder Representation from Transformers, BERT) to perform vector conversion on text fragments such as keywords. Load the embedding layer trained by the Graph Neural Networks (GNN) to perform vector conversion on the corresponding features.

[0075] In one example, for the target features of the input embedding layer, the text can be analyzed to determine the symbols and segment the text, determine each text segment, and then determine the tokens of the text segment and the identifier of each token. The semantic information and position information of each token are analyzed and converted into a corresponding feature vector.

[0076] In the sorting model, each feature vector can be input into the deep neural network (DNN) layer for dimensionality reduction, and the obtained dimensionality reduction vector can be input into the attention network. Based on the experience of correlation optimization, the embodiment of the present application can realize the self-attention optimization of keywords (query) and titles (title) in the sorting model, so as to make the representation of keywords-titles (query-title) stronger, and at the same time enhance the semantic matching ability in the sorting model to a certain extent.

[0077] In an embodiment of the present application, an optimizer may also be used during the training of the sorting model to improve the efficiency of the training, thereby making the training more stable and the model more robust.

[0078] Step 108: training the sorting model based on the training data to obtain a sorting model that meets the conditions.

[0079] Each training data may be input into the sorting model to determine the sorting result, and then a loss function is determined based on the sorting result and the training data, and the sorting model parameters are adjusted based on the loss function and training is continued until a sorting model that meets the conditions is obtained.

[0080] In summary, the search associated data of the target language site is deleted from the search associated data to obtain the target search associated data, thereby removing the influence of the target language on the data and making the target search associated data of each language more balanced. Then, the feature data is filtered from the target search associated data according to the set rules to construct the training data. The required feature data can be filtered as the training data, and then the sorting model is constructed. The sorting model is trained based on the training data to obtain a sorting model that meets the conditions, thereby eliminating the uneven distribution of data at each site for small traffic languages, and obtaining a sorting model that supports various languages, so that searches in various languages ​​can be accurately fed back.

[0081] Based on the above embodiments, the embodiments of the present application also provide a method for constructing a ranking model, which can delete the target language site data, delete weakly related features, add core key attributes, and eliminate the impact of uneven distribution of data from each site on low-traffic languages.

[0082] Reference Figure 2 , showing a step flow chart of another embodiment of a method for constructing a sorting model of the present application.

[0083] Step 202: Obtain search related data of each language site.

[0084] Step 204: determine the data volume information corresponding to each language site based on the search-related data.

[0085] Step 206: determine whether the data volume information corresponding to the site exceeds the target threshold.

[0086] If yes, then step 208 is executed, if no, then the language of other sites is continuously detected.

[0087] Step 208: determine that the site is a target language site and the language of the site is the target language.

[0088] Step 210: Delete the search related data of the target language site from the search related data to obtain the target search related data.

[0089] Step 212: searching for sample information of the associated data based on the target of each language site.

[0090] Wherein, based on the target search associated data of each language site, the data volume information and behavior information of each language site are determined; based on the data volume information and behavior information, the sampling information of the target search associated data of each language site is determined.

[0091] Step 214: extracting feature data of each feature from the associated data, and analyzing the correlation information between each feature data and the behavior information.

[0092] Step 216: Filter target features based on the correlation information.

[0093] Step 218: For each commodity object, construct a site difference feature as a target feature.

[0094] Among them, constructing a site difference feature for each commodity object includes: generating a difference feature based on at least one of a language feature, a commodity object feature and a conversion feature; determining a site feature of a site corresponding to the commodity object; and concatenating the difference feature with the site feature to obtain a corresponding site difference feature.

[0095] The generating of difference features based on at least one of the language feature, the commodity object feature and the conversion feature includes: obtaining associated features of the commodity object, the associated features including the interaction depth feature of the language, the conversion rate feature of the commodity object and the service quality feature of the seller user; determining the conversion feature of the commodity object based on the associated features; and concatenating the conversion feature, the language feature and the commodity object feature to determine the corresponding difference features.

[0096] Step 220 , collecting target feature data from target search related data of each language site according to the adopted information, and constructing training data.

[0097] Step 222, determine the activation function, and add the activation function to the neural network model to construct a sorting model.

[0098] Step 224: Train the sorting model based on the training data to obtain a sorting model that meets the conditions.

[0099] In each embodiment of the present application, the order of constructing the sorting model and determining the training data is not limited, and they can also be determined in parallel and can be determined according to needs.

[0100] Compared with the problem of uneven training of small languages ​​in a single model in unified modeling of the entire site, in the embodiment of the present application, the data of the target language site with a relatively large data proportion, such as the English site data, is removed, and the search data of the target language in the multilingual site is used to ensure that the model takes into account the target reasons. The municipal water conservancy in the present application also adjusts the sampling ratio of different deep inquiry samples for different language sites, so that sites with a large amount of access data are set to sample data at a lower ratio, while sites with a low amount of access data are sampled at a higher ratio, which can help alleviate the problem of sparse and long-tail behavior data.

[0101] Compared with the problem of too many models due to independent sorting modeling of sub-sites, which causes loss of resources and online performance. In addition, the amount of data in some small-traffic scenarios is small, and the model modeling is cross-cutting, which causes serious loss of retention and conversion. The embodiment of the present application can delete weakly related features when determining training data, and add core key attributes. It can also construct differentiated features for language sites, eliminating the impact of uneven distribution of data at each site on the effect of small-traffic languages.

[0102] Compared with regional modeling, there is a problem of inconsistent division of different regions, and using translated English difference information for matching will cause loss of semantic matching of different languages. The regions are close but the cultural customs are very different, resulting in the problem of lack of differentiation in site searches. Can the embodiment of the present application construct differentiated features of sites, so that the model can adapt to the language differences of each site when training. The activation function of the neural network in the model is also adjusted, and the efficiency and learning ability of model training are improved through the optimizer, thereby improving the robustness of multilingual expression difference modeling.

[0103] After the above ranking model is constructed, it can be used when searching on the website to improve the accuracy of ranking, so that searches in various languages ​​can obtain accurate and rich results.

[0104] Reference Figure 3 , shows a step flow chart of an embodiment of a commodity object search method embodiment of the present application.

[0105] Step 302: Receive a search request, where the search request carries a keyword.

[0106] When a user browses a certain language site on a cross-border e-commerce website, he or she can enter a corresponding keyword to search for a product object. The keyword that the user searches for on the website can be the basic language of the website or other languages.

[0107] Step 304: Search based on the keyword to determine multiple commodity objects.

[0108] A search can be performed based on the search keyword, wherein the searched commodity objects can be commodity objects published in the language or commodity objects published in other languages. Multiple commodity objects can be searched, for example, more than 50 or more than 100.

[0109] Step 306, analyzing the multiple commodity objects based on the ranking model and filtering the commodity objects, wherein the ranking model is obtained based on training data, the training data is constructed based on feature data filtered according to set rules in the target search related data, and the target search related data is determined by deleting the target language site from the search related data.

[0110] The characteristic data of each commodity object can be determined, and the target characteristic data can be collected from the characteristic data and input into the sorting model; the commodity objects can be screened based on the output result of the sorting model. For example, the sorting model can predict the click rate, visit rate, etc. of the commodity objects based on the target characteristic, and sort them. The top K commodity objects are screened.

[0111] Among them, feature data of each feature can be extracted from the associated data, and correlation information between each feature data and behavior information can be analyzed; and target features can be screened based on the correlation information.

[0112] For each commodity object, a site difference feature is constructed as a target feature. The construction of the site difference feature for each commodity object includes: generating a difference feature based on at least one of a language feature, a commodity object feature, and a conversion feature; determining a site feature of a site corresponding to the commodity object; and concatenating the difference feature with the site feature to obtain a corresponding site difference feature.

[0113] The generating of difference features based on at least one of the language feature, the commodity object feature and the conversion feature includes: obtaining associated features of the commodity object, the associated features including the interaction depth feature of the language, the conversion rate feature of the commodity object and the service quality feature of the seller user; determining the conversion feature of the commodity object based on the associated features; and concatenating the conversion feature, the language feature and the commodity object feature to determine the corresponding difference features.

[0114] Step 308: Generate search results based on the filtered commodity objects and provide feedback.

[0115] A search result is generated based on the screened K product objects, and then the search result is fed back.

[0116] In summary, a search request is received, the search request carries a keyword, and then a search is performed based on the keyword to determine multiple commodity objects, and then the multiple commodity objects are analyzed based on a sorting model to filter the commodity objects, wherein the sorting model is obtained based on training of training data, the training data is constructed based on feature data filtered according to set rules in target search related data, the target search related data is determined based on deleting a target language site from the search related data, thereby supporting searches in various language sites, more accurately sorting commodity objects, generating search results based on the filtered commodity objects and feeding back, providing users with accurate feedback results, and improving user experience.

[0117] It should be noted that the embodiments of the present application may involve the use of user data. In actual applications, user-specific personal data can be used in the scheme described herein within the scope permitted by applicable laws and regulations, subject to the requirements of applicable laws and regulations of the country where the user is located (for example, with the user's explicit consent, effective notification to the user, etc.).

[0118] It should be noted that, for the method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

[0119] On the basis of the above embodiment, this embodiment further provides a device for constructing a ranking model, which is applied to an electronic device such as a server device, and the device includes:

[0120] A data screening module, used for deleting the search related data of the target language site from the search related data to obtain the target search related data;

[0121] A training data construction module, used to filter feature data from the target search related data according to set rules to construct training data;

[0122] Model building module, used to build the sorting model;

[0123] The training module is used to train the sorting model based on the training data to obtain a sorting model that meets the conditions.

[0124] In summary, the search associated data of the target language site is deleted from the search associated data to obtain the target search associated data, thereby removing the influence of the target language on the data and making the target search associated data of each language more balanced. Then, the feature data is filtered from the target search associated data according to the set rules to construct the training data. The required feature data can be filtered as the training data, and then the sorting model is constructed. The sorting model is trained based on the training data to obtain a sorting model that meets the conditions, thereby eliminating the uneven distribution of data at each site for small traffic languages, and obtaining a sorting model that supports various languages, so that searches in various languages ​​can be accurately fed back.

[0125] The data screening module is further used to obtain search-related data of each language site; determine data volume information corresponding to each language site based on the search-related data, and determine the target language site based on the data volume information.

[0126] The training data construction module is used to collect target feature data from the target search associated data of each language site based on the sampling information of the target search associated data of each language site according to the sampling information to construct the training data.

[0127] The training data construction module is used to determine the data volume information and behavior information of each language site based on the target search related data of each language site; and determine the sampling information of the target search related data of each language site based on the data volume information and behavior information.

[0128] The training data construction module is further used to extract feature data of each feature from the associated data, analyze the correlation information between each feature data and the behavior information; and filter the target feature based on the correlation information.

[0129] The training data construction module is also used to construct site difference features as target features for each commodity object.

[0130] The training data construction module is also used to generate a difference feature based on at least one of a language feature, a commodity object feature and a conversion feature; determine a site feature of a site corresponding to a commodity object; and concatenate the difference feature with the site feature to obtain a corresponding site difference feature.

[0131] The training data construction module is also used to obtain related features of commodity objects, which include interaction depth features of languages, conversion rate features of commodity objects, and service quality features of seller users; determine conversion features of commodity objects based on the related features; and splice conversion features, language features, and commodity object features to determine corresponding difference features.

[0132] The model building module is used to determine the activation function and add the activation function to the neural network model to build a sorting model.

[0133] The data screening module is used to determine that the site is a target language site and the language of the site is the target language when the data volume information corresponding to the site exceeds a target threshold.

[0134] Compared with the problem of uneven training of small languages ​​in a single model in unified modeling of the entire site, in the embodiment of the present application, the data of the target language site with a relatively large data proportion, such as the English site data, is removed, and the search data of the target language in the multilingual site is used to ensure that the model takes into account the target reasons. The municipal water conservancy in the present application also adjusts the sampling ratio of different deep inquiry samples for different language sites, so that sites with a large amount of access data are set to sample data at a lower ratio, while sites with a low amount of access data are sampled at a higher ratio, which can help alleviate the problem of sparse and long-tail behavior data.

[0135] Compared with the problem of too many models due to independent sorting modeling of sub-sites, which causes loss of resources and online performance. In addition, the amount of data in some small-traffic scenarios is small, and the model modeling is cross-cutting, which causes serious loss of retention and conversion. The embodiment of the present application can delete weakly related features when determining training data, and add core key attributes. It can also construct differentiated features for language sites, eliminating the impact of uneven distribution of data at each site on the effect of small-traffic languages.

[0136] Compared with regional modeling, there is a problem of inconsistent division of different regions, and using translated English difference information for matching will cause loss of semantic matching of different languages. The regions are close but the cultural customs are very different, resulting in the problem of lack of differentiation in site searches. Can the embodiment of the present application construct differentiated features of sites, so that the model can adapt to the language differences of each site when training. The activation function of the neural network in the model is also adjusted, and the efficiency and learning ability of model training are improved through the optimizer, thereby improving the robustness of multilingual expression difference modeling.

[0137] On the basis of the above embodiment, this embodiment further provides a commodity object search device, which is applied to electronic devices such as a server device, and the device includes:

[0138] A receiving module, configured to receive a search request, wherein the search request carries a keyword;

[0139] A search module, used to search based on the keyword to determine multiple commodity objects;

[0140] A sorting module, used to analyze multiple commodity objects based on a sorting model and filter commodity objects, wherein the sorting model is obtained based on training data, the training data is constructed based on feature data filtered according to set rules in target search related data, and the target search related data is determined by deleting a target language site from the search related data;

[0141] The feedback module is used to generate search results and provide feedback based on the filtered product objects.

[0142] The sorting module is used to collect target feature data from the feature data of the commodity objects and input the data into the sorting model; and screen the commodity objects based on the output results of the sorting model.

[0143] In summary, a search request is received, the search request carries a keyword, and then a search is performed based on the keyword to determine multiple commodity objects, and then the multiple commodity objects are analyzed based on a sorting model to filter the commodity objects, wherein the sorting model is obtained based on training of training data, the training data is constructed based on feature data filtered according to set rules in target search related data, the target search related data is determined based on deleting a target language site from the search related data, thereby supporting searches in various language sites, more accurately sorting commodity objects, generating search results based on the filtered commodity objects and feeding back, providing users with accurate feedback results, and improving user experience.

[0144] The embodiment of the present application also provides a non-volatile readable storage medium, which stores one or more modules (programs). When the one or more modules are applied to a device, the device can execute instructions (instructions) of each method step in the embodiment of the present application.

[0145] The embodiment of the present application further provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method described in the embodiment of the present application.

[0146] The embodiment of the present application also provides an electronic device, comprising: a processor, and a memory connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method described in the embodiment of the present application. In the embodiment of the present application, the electronic device includes a server, a terminal device, and other devices.

[0147] The embodiments of the present disclosure may be implemented as a device configured as desired using any appropriate hardware, firmware, software, or any combination thereof, and the device may include electronic devices such as a server (cluster), a terminal, etc. Figure 4 An exemplary apparatus 400 that may be used to implement various embodiments described in this application is schematically illustrated.

[0148] For one embodiment, Figure 4 An exemplary apparatus 400 is shown having one or more processors 402, a control module (chip set) 404 coupled to at least one of the processor(s) 402, a memory 406 coupled to the control module 404, a non-volatile memory (NVM) / storage device 408 coupled to the control module 404, one or more input / output devices 410 coupled to the control module 404, and a network interface 412 coupled to the control module 404.

[0149] Processor 402 may include one or more single-core or multi-core processors, and processor 402 may include any combination of general-purpose processors or special-purpose processors (such as graphics processors, application processors, baseband processors, etc.). In some embodiments, device 400 can be used as a server, terminal, or other device described in the embodiments of the present application.

[0150] In some embodiments, the apparatus 400 may include one or more computer-readable media (e.g., memory 406 or NVM / storage device 408) having instructions 414 and one or more processors 402 combined with the one or more computer-readable media and configured to execute the instructions 414 to implement a module to perform the actions described in the present disclosure.

[0151] For one embodiment, control module 404 may include any suitable interface controller to provide any suitable interface to at least one of processor(s) 402 and / or any suitable device or component in communication with control module 404 .

[0152] The control module 404 may include a memory controller module to provide an interface to the memory 406. The memory controller module may be a hardware module, a software module, and / or a firmware module.

[0153] The memory 406 may be used, for example, to load and store data and / or instructions 414 for the device 400. For one embodiment, the memory 406 may include any suitable volatile memory, such as a suitable DRAM. In some embodiments, the memory 406 may include a double data rate type four synchronous dynamic random access memory (DDR4 SDRAM).

[0154] For one embodiment, control module 404 may include one or more input / output controllers to provide interfaces to NVM / storage device 408 and input / output device(s) 410 .

[0155] For example, NVM / storage 408 may be used to store data and / or instructions 414. NVM / storage 408 may include any suitable non-volatile memory (e.g., flash memory) and / or may include any suitable non-volatile storage device(s) (e.g., one or more hard disk drives (HDDs), one or more compact disk (CD) drives, and / or one or more digital versatile disk (DVD) drives).

[0156] NVM / storage device 408 may include storage resources that are part of the device on which apparatus 400 is installed, or it may be accessible to the device without being part of the device. For example, NVM / storage device 408 may be accessed via (one or more) input / output devices 410 over a network.

[0157] (One or more) input / output devices 410 may provide an interface for the apparatus 400 to communicate with any other appropriate device, and the input / output device 410 may include a communication component, an audio component, a sensor component, etc. The network interface 412 may provide an interface for the apparatus 400 to communicate through one or more networks, and the apparatus 400 may wirelessly communicate with one or more components of a wireless network according to any of one or more wireless network standards and / or protocols, for example, accessing a wireless network based on a communication standard, such as Bluetooth, WiFi, 2G, 3G, 4G, 5G, etc., or a combination thereof for wireless communication.

[0158] For one embodiment, at least one of the processor(s) 402 may be packaged together with the logic of one or more controllers (e.g., a memory controller module) of the control module 404. For one embodiment, at least one of the processor(s) 402 may be packaged together with the logic of one or more controllers of the control module 404 to form a system-in-package (SiP). For one embodiment, at least one of the processor(s) 402 may be integrated on the same die with the logic of one or more controllers of the control module 404. For one embodiment, at least one of the processor(s) 402 may be integrated on the same die with the logic of one or more controllers of the control module 404 to form a system-on-chip (SoC).

[0159] In various embodiments, the device 400 may be, but is not limited to, a terminal device such as a server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.). In various embodiments, the device 400 may have more or fewer components and / or different architectures. For example, in some embodiments, the device 400 includes one or more cameras, a keyboard, a liquid crystal display (LCD) screen (including a touch screen display), a non-volatile memory port, multiple antennas, a graphics chip, an application-specific integrated circuit (ASIC), and a speaker.

[0160] Among them, the main control chip can be used as a processor or control module in the detection device, sensor data, location information, etc. are stored in a memory or NVM / storage device, the sensor group can be used as an input / output device, and the communication interface may include a network interface.

[0161] The embodiment of the present application also provides an electronic device, including: a processor; and a memory, on which executable code is stored, and when the executable code is executed, the processor executes one or more methods as described in the embodiment of the present application. In the embodiment of the present application, the memory can store various data, such as target files, file and application association data, and user behavior data, so as to provide a data basis for various processing.

[0162] The embodiments of the present application also provide one or more machine-readable media on which executable codes are stored. When the executable codes are executed, the processor executes one or more methods described in the embodiments of the present application.

[0163] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0164] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0165] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0166] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0167] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0168] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0169] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0170] The hand modeling method, a processing method based on the hand model, an electronic device and a storage medium provided by the present application are introduced in detail above. The principles and implementation methods of the present application are explained in this article using specific examples. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for constructing a ranking model, characterized in that: The method comprises: Deleting the search related data of the target language site from the search related data to obtain the target search related data; From the target search related data, feature data is filtered according to set rules to construct training data; Build a sorting model; The sorting model is trained based on the training data to obtain a sorting model that meets the conditions.

2. The method according to claim 1, characterized in that Before deleting the search related data of the target language site from the search related data to obtain the target search related data, the following further includes: Get search related data of each language site; The data volume information corresponding to each language site is determined based on the search-related data, and the target language site is determined based on the data volume information.

3. The method according to claim 2, characterized in that The target search related data is searched for feature data according to set rules to construct training data, including: Sampling information of target search related data based on each language site; According to the adopted information, data of target features are collected from target search related data of each language site to construct training data.

4. The method according to claim 3, characterized in that The sample information of the target search related data based on each language site includes: Based on the target search related data of each language site, determine the data volume information and behavior information of each language site; Based on the data volume information and the behavior information, sampling information of the target search related data of each language site is determined.

5. The method according to claim 3, characterized in that: Also includes: Extracting feature data of each feature from the associated data, and analyzing the correlation information between each feature data and the behavior information; Target features are filtered based on the correlation information.

6. The method according to claim 3, characterized in that Also includes: For each commodity object, a site difference feature is constructed as the target feature.

7. The method according to claim 6, characterized in that The site difference features are constructed for each commodity object, including: Generate a difference feature based on at least one of a language feature, a commodity object feature, and a conversion feature; Determine the site characteristics of the site corresponding to the commodity object; The difference feature and the site feature are concatenated to obtain the corresponding site difference feature.

8. The method according to claim 7, characterized in that The generating of difference features based on at least one of the language feature, the commodity object feature and the conversion feature includes: Obtaining correlation features of the product object, the correlation features include the interaction depth feature of the language, the conversion rate feature of the product object, and the service quality feature of the seller user; Determining a conversion feature of the commodity object based on the association feature; Conversion features, language features, and product object features are combined to determine the corresponding difference features.

9. The method according to claim 1, characterized in that: The constructing of the sorting model comprises: An activation function is determined, and the activation function is added to the neural network model to construct a sorting model.

10. The method according to claim 2, characterized in that Determining a target language site based on the data volume information includes: When the data volume information corresponding to the site exceeds the target threshold, the site is determined to be a target language site, and the language of the site is the target language.

11. A commodity object search method, characterized in that: The method comprises: receiving a search request, wherein the search request carries a keyword; Searching based on the keyword to determine multiple commodity objects; Analyze multiple commodity objects based on a ranking model to filter commodity objects, wherein the ranking model is obtained based on training data, the training data is constructed based on feature data filtered according to set rules in target search related data, and the target search related data is determined by deleting a target language site from the search related data; Generate search results based on the filtered product objects and feedback.

12. The method according to claim 11, characterized in that The analyzing the multiple commodity objects based on the sorting model and screening the commodity objects includes: Collect target feature data from the feature data of the commodity object and input it into the sorting model; The commodity objects are screened based on the output result of the ranking model.

13. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 12.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method according to any one of claims 1 to 12.

Images