Model training method, device, equipment and storage medium

Abstract

The present disclosure provides a model training method, apparatus, device, and storage medium, relating to the fields of computer technology, particularly data processing, large language models, smart e-commerce, and other technical fields. A specific implementation scheme comprises: extracting multiple product conversion samples from multiple information samples; inputting the multiple product conversion samples into a conversion amount prediction model to be trained to obtain multiple first prediction values, each of which is related to the predicted product conversion amount and the minimum product amount; determining a loss function based on the multiple first prediction values; and training the conversion amount prediction model based on the loss function. The present disclosure can improve the accuracy of predicting product conversion amounts.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to technical fields such as data processing, large language models, and smart e-commerce. Background Art

[0002] In recent years, the rapid development of the e-commerce industry has led to an increase in product variety and a widening of price ranges. This has led to a decrease in the accuracy of product conversion amount predictions and the prediction results sometimes deviate from reality, which in turn affects the ranking of product information. Therefore, how to accurately predict product conversion amounts and ensure that the prediction results meet actual requirements is a pressing issue. Summary of the Invention

[0003] The present disclosure provides a model training method, apparatus, device, and storage medium.

[0004] According to one aspect of the present disclosure, a model training method is provided, comprising:

[0005] Extract multiple product conversion samples from multiple information samples;

[0006] Inputting a plurality of product conversion samples into a conversion amount prediction model to be trained to obtain a plurality of first prediction values, wherein the first prediction values ​​are related to the predicted conversion amount of the product and the minimum amount of the product;

[0007] determining a loss function based on the plurality of first prediction values;

[0008] The conversion amount prediction model is trained based on this loss function.

[0009] According to another aspect of the present disclosure, there is provided an information sorting method, comprising:

[0010] obtaining a plurality of first information;

[0011] Determine a first predicted value for each piece of first information using a pre-trained conversion amount prediction model;

[0012] Determining a predicted conversion amount of a product corresponding to each piece of first information based on the first predicted value of each piece of first information;

[0013] sorting the plurality of first information based on the predicted commodity conversion amounts corresponding to the respective first information;

[0014] The conversion amount prediction model is trained using any training method provided in the present disclosure.

[0015] According to another aspect of the present disclosure, there is provided a model training device, comprising:

[0016] A sample extraction module is used to extract multiple commodity conversion samples from multiple information samples;

[0017] An input module, configured to input a plurality of commodity conversion samples into a conversion amount prediction model to be trained to obtain a plurality of first prediction values, wherein the first prediction values ​​are related to the commodity conversion prediction amount and the commodity minimum amount;

[0018] A loss function determination module, configured to determine a loss function based on a plurality of first prediction values;

[0019] The training module is used to train the conversion amount prediction model based on the loss function.

[0020] According to another aspect of the present disclosure, there is provided an information sorting device, comprising:

[0021] An acquisition module, configured to acquire a plurality of first information;

[0022] A first determination module, configured to determine a first predicted value of each piece of first information using a pre-trained conversion amount prediction model;

[0023] A second determining module is configured to determine a predicted conversion amount of a commodity corresponding to each piece of first information based on the first predicted value of each piece of first information;

[0024] A sorting module, configured to sort the plurality of first information based on the predicted commodity conversion amount corresponding to each piece of first information;

[0025] The conversion amount prediction model is obtained by training using any training device provided in the present disclosure.

[0026] According to another aspect of the present disclosure, there is provided an electronic device, comprising:

[0027] at least one processor; and

[0028] a memory communicatively connected to the at least one processor; wherein,

[0029] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any method in the embodiments of the present disclosure.

[0030] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to execute any method according to the embodiments of the present disclosure.

[0031] According to another aspect of the present disclosure, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the computer program implements any one of the methods according to the embodiments of the present disclosure.

[0032] The present disclosure extracts commodity conversion samples from multiple information samples, and uses the conversion amount prediction model to be trained to predict multiple first prediction values ​​corresponding to the commodity conversion samples. The first prediction value is related to the commodity conversion prediction amount and the commodity minimum amount. Since the first prediction value is related to the commodity minimum amount, by limiting the value range of the first prediction value, the lower limit of the commodity conversion prediction amount corresponding to each commodity conversion sample can be indirectly limited, so that the commodity conversion prediction amount is not less than the commodity minimum amount, so that the prediction result meets the reasonable lower limit requirement, thereby improving the reliability of the prediction result.

[0033] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0035] Figure 1 is a schematic diagram of an application scenario according to an embodiment of the present disclosure;

[0036] Figure 2 is a flow chart for implementing a model training method according to an embodiment of the present disclosure;

[0037] Figure 3 is a schematic diagram of an overall process according to an embodiment of the present disclosure;

[0038] Figure 4 is a comparison diagram of a first Poisson distribution and a second Poisson distribution according to an embodiment of the present disclosure;

[0039] Figure 5 is a flowchart of an implementation method of information sorting according to an embodiment of the present disclosure;

[0040] Figure 6 is a structural diagram of a model training device 600 according to an embodiment of the present disclosure;

[0041] Figure 7 is a structural diagram of an information sorting device 700 according to an embodiment of the present disclosure;

[0042] Figure 8 A schematic block diagram of an example electronic device 800 is shown, which may be used to implement embodiments of the present disclosure. DETAILED DESCRIPTION

[0043] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be appreciated by those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0044] The “and / or” in the embodiments of the present disclosure indicates that there may be three relationships. For example, A and / or B may indicate three situations: A exists alone, A and B exist at the same time, and B exists alone. The term “at least one” herein indicates any combination of at least two of any one or more of a plurality of. For example, at least one of A, B, and C may indicate any one or more elements selected from the set consisting of A, B, and C. The terms “first” and “second” herein refer to and distinguish between multiple similar technical terms, and do not mean to limit the order or to limit the meaning to only two. For example, the first feature and the second feature refer to two categories / two features. The first feature may be one or more, and the second feature may also be one or more.

[0045] In recent years, the rapid development of the e-commerce industry has led to an increase in product categories and a widening price range within these categories. The distribution of transaction amounts, from high-end, expensive jewelry and luxury goods to low-end, affordable everyday consumer goods, has become multi-layered. Furthermore, increasing consumer purchasing behavior and the frequent holding of promotions have led to a single product often being presented in a variety of packaged packages. For example, a particular shampoo may be offered in a variety of price packages, such as a two-bottle or three-bottle pack. These phenomena complicate the prediction of product conversion amounts.

[0046] Currently, existing product conversion amount prediction methods include:

[0047] A method for predicting product conversion amounts based on a regression model. In the application scenario of a regression model, a truncation method is used to adjust the model output to ensure that the prediction results meet the lower limit constraint required in actual business operations. The basic idea of ​​this method is to truncate the model's prediction results. That is, when the predicted value falls below the preset lower limit, it is directly corrected to the lower limit. On the surface, this method can ensure that all prediction results meet the lower limit constraint.

[0048] However, a significant practical problem with the truncation method is that it forces a large number of predicted values ​​to fall near the lower limit. This tendency not only weakens the model's ability to distinguish differences in the actual conversion prices of different products, but also affects the accuracy and practicality of the predictions. This is because in real business environments, the conversion prices of products often exhibit a variety of characteristics, and the truncation method cannot accurately capture and reflect this variability.

[0049] To overcome the limitations of the truncation method, another approach is to constrain the model's output by selecting an appropriate activation function. As a crucial component of neural networks, the choice of activation function has a crucial impact on model performance. By selecting and designing an activation function, the model's output can be kept above a specified lower limit, thus satisfying the lower limit constraint.

[0050] However, the actual conversion price lower limits of different commodities are different. If a prediction model is built separately for each commodity and a corresponding activation function is set, it will not only greatly increase the complexity of model development and maintenance, but also reduce the flexibility and applicability of the model.

[0051] In addition, models based on deep learning, such as deep neural networks (DNN), convolutional neural networks (CNN), graph neural networks (GNN), or models using reinforcement learning technology, can directly predict the conversion amount of goods, but cannot effectively meet the lower limit constraint.

[0052] Therefore, the current existing product conversion amount prediction method cannot meet the lower limit constraint while making accurate predictions, which in turn affects the application of the product conversion amount. For example, the display order of the corresponding information of each product is determined by the product conversion amount corresponding to each product.

[0053] In order to solve the above problems, the embodiments of the present disclosure propose a model training method and an information sorting method. Figure 1 is a schematic diagram of an application scenario according to an embodiment of the present disclosure, such as Figure 1As shown, the application scenario diagram of the embodiment of the present disclosure may include but is not limited to a model training device 110 and an information sorting device 120, and the model training device 110 and the information sorting device 120 can communicate with each other through any type of wired or wireless network. Specifically, the model training device 110 can be used to train a conversion amount prediction model and send the trained conversion amount prediction model to the information sorting device 120; the information sorting device 120 can load and install the conversion amount prediction model, and after receiving the information corresponding to each product input by the user, each piece of information is input into the conversion amount prediction model, and the conversion amount prediction model outputs the product conversion predicted amount corresponding to each piece of information, and sorts each piece of information based on the product conversion predicted amount corresponding to each piece of information. Among them, the information sorting device 120 proposed in the embodiment of the present disclosure includes but is not limited to electronic devices such as mobile phones, computers, intelligent voice interaction devices, smart home appliances, car terminals, game consoles, e-book readers, multimedia playback devices, wearable devices, etc.; the model training device 110 may include an electronic device or server for providing background management for the information sorting device 120. In addition, the embodiments of the present disclosure do not impose specific restrictions on the number of model training devices 110 and information sorting devices 120. For example, the application scenario diagram of the embodiments of the present disclosure may include one or more model training devices 110 and one or more information sorting devices 120.

[0054] Figure 2 This is a flowchart of an implementation of a model training method according to an embodiment of the present disclosure, including:

[0055] S210, extracting multiple commodity conversion samples from multiple information samples;

[0056] S220: Input multiple product conversion samples into a conversion amount prediction model to be trained to obtain multiple first prediction values, where the first prediction values ​​are related to the predicted product conversion amount and the minimum product amount;

[0057] S230, determining a loss function according to the multiple first prediction values;

[0058] S240: Training the conversion amount prediction model based on the loss function.

[0059] It can be seen from the above training method that during the training process of the conversion amount prediction model, the model can output the first prediction value corresponding to each commodity conversion sample. The first prediction value is related to the commodity conversion prediction amount and the commodity minimum amount. Since the first prediction value is related to the commodity minimum amount, by limiting the value range of the first prediction value, the lower limit of the commodity conversion prediction amount corresponding to each commodity conversion sample can be indirectly limited, so that the commodity conversion prediction amount is not less than the commodity minimum amount, so that the prediction result meets the reasonable lower limit requirements, thereby improving the reliability of the prediction result.

[0060] Figure 3 It is a schematic diagram of the overall process according to an embodiment of the present disclosure. In the overall process of the embodiment of the present disclosure, the goal is to calculate the effective cost per mille (eCPM) corresponding to each product information. In this process, the conversion rate (Central Video Recorder, CVR) prediction module (also known as the CVR tower) is mainly used to predict the CVR of the product click sample, and the CVR tower is trained based on the prediction results; and the conversion amount prediction model is used to predict the gross merchandise volume (GMV) of the product conversion sample, and the conversion amount prediction model is trained based on the prediction results. In the embodiment of the present disclosure, the conversion amount prediction model can also be called the GMV tower.

[0061] In some implementations, extracting multiple commodity conversion samples from multiple information samples includes:

[0062] Obtain multiple information samples and determine the user behavior corresponding to each information sample;

[0063] When the user behavior corresponding to the information sample is a conversion behavior, the information sample is determined to be a product conversion sample;

[0064] Extract multiple product conversion samples.

[0065] In some embodiments, further comprising:

[0066] Record the actual amount of product conversion corresponding to each product conversion sample.

[0067] First, it is necessary to collect a wide range of information samples from various channels and sources. These information samples may come from users' online behavior records, transaction records, browsing records, search records, click records, etc. Each information sample contains user data, which is helpful for subsequent analysis of user behavior and then determining product conversion samples.

[0068] After acquiring multiple information samples, the sample screening module 311 within the sample processing module 310 analyzes each sample to determine the corresponding user behavior. User behaviors include browsing, clicking, adding to cart, and conversions. This process utilizes data analysis techniques and algorithms, such as machine learning and deep learning. These techniques can help extract valuable information from massive amounts of data and accurately determine the user behavior corresponding to each information sample.

[0069] After determining the user behavior corresponding to each information sample, the product samples need to be further tagged with user behavior using the sample tagging module 312. Click behavior refers to a user clicking on a piece of product information; conversion behavior refers to the process by which a user transitions from initial behavior such as browsing or clicking to final behavior such as actual purchase or payment.

[0070] To identify product conversion samples, clear conversion behavior definitions and identification criteria can be set. For example, a user submitting an order and successfully paying based on a product's corresponding information sample can be defined as a conversion behavior, and the corresponding information sample can be identified as a product conversion sample.

[0071] After determining the product conversion samples, these samples need to be extracted from the original information samples, and the product conversion amount corresponding to each product conversion sample needs to be recorded for subsequent analysis and processing.

[0072] By identifying user behavior, this method can distinguish click samples and further extract conversion samples from these click samples, providing a basis for screening product conversion samples. Furthermore, product click samples can be used to train the CVR tower, and product conversion samples can be used to train the GMV tower, thereby improving the prediction accuracy of each prediction model.

[0073] In some embodiments, the conversion amount prediction model includes an embedding module, a neural gate module, and a prediction module;

[0074] Input multiple product conversion samples into the conversion amount prediction model to be trained to obtain multiple first prediction values, including:

[0075] For each product conversion sample, perform the following operations:

[0076] The product conversion sample is input into the embedding module, and the embedding module is used to extract the feature vector of the product conversion sample; the neural gate module is used to set the weight corresponding to each element in the feature vector; and the prediction module is used to determine the first prediction value based on the feature vector and the weight corresponding to each element.

[0077] In the embodiment of the present disclosure, the product click samples and product conversion samples are the outputs of the previous steps, and contain user purchase behavior data. These data are the basis for subsequent analysis and prediction. In order to extract the features of each product click sample and product conversion sample from these data, the sample mapping unit 320 is required.

[0078] The sample mapping unit 320 includes an embedding module 321 and a neural gate module 322 .

[0079] The core function of the embedding module 321 is to convert high-dimensional, sparse product sample data into highly dense feature vectors. This process not only reduces the data's dimensionality but also preserves the key information and inherent structure of the original data. Through the processing of the embedding module 321, each product sample is mapped to a feature vector of fixed length.

[0080] After obtaining the feature vector, it is necessary to assign weights to the elements in the feature vector, which can be achieved through the neural gate module 322 in the model.

[0081] The neural gate module 322 is responsible for dynamically adjusting the weights of each element in the feature vector. The weight assignment is not fixed but is determined based on the feature vector itself, the requirements of the prediction task, and the model parameters. Through the processing of the neural gate module 322, feature elements that have a significant impact on the prediction result are assigned higher weights, while feature elements with less influence are assigned lower weights. Furthermore, the weights corresponding to the feature elements will also change dynamically as the model is iterated.

[0082] After obtaining the weighted feature vector, the multi-task prediction unit 330 can be used to predict the CVR of the product click sample and the GMV of the product conversion sample respectively. In order to predict the CVR of each product click sample, the CVR tower prediction module 331 can be used. This module makes predictions based on the feature vector of the product click sample and the weights corresponding to its internal elements; in order to predict the GMV of each product conversion sample, the GMV tower prediction module 332 can be used. This module makes predictions based on the conversion amount, feature vector and the weights corresponding to its internal elements of each product conversion sample. Both the CVR tower prediction module 331 and the GMV tower prediction module 332 contain multiple neural network structures such as fully connected layers and activation functions. These structures work together to convert feature vectors and weights into final prediction values.

[0083] For the CVR tower, the individual CVR prediction values ​​output by the CVR tower prediction module 331 can be used to construct a cross-entropy loss. This can then be used to train the CVR tower, thereby improving the accuracy of the CVR tower's predictions for product CVRs. Similarly, for the GMV tower, the individual first prediction values ​​output by the GMV tower prediction module 332 can be used to construct a loss function. This can then be used to train the GMV tower, thereby improving the accuracy of product GMV predictions. Through continuous training and optimization, the CVR tower prediction module 331 and the GMV tower prediction module 332 can learn how to extract useful information from feature vectors and accurately predict target values.

[0084] After completing the training of the CVR tower and GMV tower, you can input the product information into the CVR tower and GMV tower respectively, output the CVR and GMV corresponding to the product information, and then calculate the eCPM of the product.

[0085] By converting product samples into feature vectors using this method, we not only reduce the data dimension but also preserve the key information and internal structure of the original data, thereby improving feature extraction efficiency. Furthermore, by assigning weights to each element in the feature vector, we can focus the model on features that influence the prediction results, thus increasing its flexibility.

[0086] In the disclosed embodiment, the loss function construction and training process of the GMV tower are introduced in detail.

[0087] In some implementations, the first predicted value is related to the predicted conversion amount of the product and the minimum amount of the product, including:

[0088] The first predicted value satisfies a first functional relationship with the predicted amount of product conversion and the minimum amount of the product.

[0089] In some implementations, the first predicted value obtained by the conversion amount prediction model belongs to a first value range, and the first value range is related to the first functional relationship.

[0090] In some embodiments, the first functional relationship includes: a first predicted value is equal to a first ratio minus 1, where the first ratio is a ratio of the predicted conversion amount of the product to the minimum amount of the product;

[0091] The first value range is greater than or equal to 0.

[0092] In the embodiment of the present disclosure, in order to ensure that the prediction results of the conversion amount prediction model reach a reasonable minimum standard, a first functional relationship is defined to specifically illustrate the correlation between the first predicted value, the predicted conversion amount of the product, and the minimum amount of the product. The first functional relationship is shown in formula (1):

[0093]

[0094] Where PV represents the first predicted value, GMV represents the total transaction volume (GMV) determined by the predicted conversion amount, and min_sku_price represents the minimum price for the product. If the product is sold individually, min_sku_price is the product price itself. If the product is sold as a package or combination, min_sku_price is the minimum price for the package or combination. If the first predicted value is greater than or equal to 0, the predicted conversion amount is greater than or equal to the minimum price.

[0095] Through the above method, the product conversion prediction amount predicted by the model can be reasonable, logical and practical. In addition, this method can narrow the model's prediction range for the first prediction value, thereby improving the accuracy of the product conversion prediction amount and providing a data basis for subsequent GMV tower training.

[0096] In some embodiments, determining a loss function based on a plurality of first prediction values ​​includes:

[0097] Fitting the distribution of the plurality of first prediction values ​​to a first Poisson distribution;

[0098] Determine the actual conversion amount of each product conversion sample, and determine the first real value of each product conversion sample based on the actual conversion amount, the minimum amount of the product, and the first functional relationship;

[0099] Fitting the distribution of the plurality of first true values ​​to a second Poisson distribution;

[0100] A loss function is determined based on the first Poisson distribution and the second Poisson distribution.

[0101] The conversion amount prediction model is used to predict the first predicted values ​​for multiple commodity conversion samples. In order to deeply understand the distribution characteristics of these first predicted values, the Poisson distribution in statistics is used for fitting.

[0102] The Poisson distribution is a discrete probability model that describes the number of random events occurring within a fixed time and space. In the disclosed embodiments, it can be applied to describe the distribution of multiple first predicted values. Through fitting, a first Poisson distribution corresponding to the multiple first predicted values ​​can be obtained, which reveals the distribution pattern of the first predicted values ​​in the population.

[0103] Next, we need to determine the actual conversion amounts for each product conversion sample. These actual amounts represent the actual economic value of the product conversions. We also need to determine the minimum conversion amount for each product. Based on the actual conversion amounts and the minimum conversion amount for each product, combined with the defined first functional relationship, we can calculate the first true value for each product conversion sample. With these first true values, we again use the Poisson distribution to fit the data, resulting in a second Poisson distribution that describes the distribution of the first true values ​​in the population.

[0104] Based on the first Poisson distribution and the second Poisson distribution, a loss function can be constructed. The loss function is a function used in fields such as machine learning to evaluate the difference between the model prediction result and the actual result. In the embodiment of the present disclosure, the first Poisson distribution can be used as the prediction result, and the second Poisson distribution can be used as the actual result. By comparing the difference between the first Poisson distribution and the second Poisson distribution, the Poisson distribution negative log-likelihood loss (PoissonNLLLoss) between the two Poisson distributions is used as the loss function. A loss function that can reflect the prediction accuracy and error can be constructed.

[0105] In some embodiments, determining a loss function based on a first Poisson distribution and a second Poisson distribution includes:

[0106] Compute the Poisson negative log-likelihood loss between the first and second Poisson distributions;

[0107] The Poisson distribution negative log-likelihood loss is used as the loss function.

[0108] Figure 4 : is a comparison diagram of the first Poisson distribution and the second Poisson distribution according to an embodiment of the present disclosure. Figure 4 As shown, the first Poisson distribution is X1~λ1, and the second Poisson distribution is X2~λ2. X1 and X2 are random variables in the first predicted value and the first true value, respectively. λ1 and λ2 represent the parameters of the first and second Poisson distributions, respectively. The horizontal axis represents the predicted target, and the vertical axis represents the probability of that target occurring. For example, since the predicted conversion amount and the minimum purchase amount are similar, and the actual conversion amount and the minimum purchase amount are also similar, the first predicted value and the first true value are more likely to be around 0 due to the constraints of the first functional relationship.

[0109] Correspondingly, the probability density function of the first Poisson distribution is:

[0110]

[0111] The probability density function of the second Poisson distribution is:

[0112]

[0113] The Poisson distribution negative log-likelihood loss is calculated based on the first Poisson distribution and the second Poisson distribution. If λ1 is the probability distribution parameter fitted by the model output, and no normalization or logarithmization is performed, the loss L corresponding to the kth sample is n for:

[0114] L n =―logP(X1=k)=λ1―klogλ1+log(k!) (4)

[0115] After normalization and logarithmization, the loss L corresponding to the kth sample is n for:

[0116] L n =―logP(X1=k)=exp(λ1)―kλ1+log(k!) (5)

[0117] Based on the above calculation results, the Poisson distribution negative log-likelihood loss is used as the loss function, and the conversion amount prediction model is iteratively trained based on this function.

[0118] This method can reflect the difference between the model's predicted probability distribution and the true distribution. Based on this difference, the Poisson distribution negative log-likelihood loss is determined to guide the model training.

[0119] The present disclosure also provides an information sorting method. Figure 5 This is a flowchart of an implementation method of information sorting according to an embodiment of the present disclosure, including:

[0120] S510, obtaining multiple first information;

[0121] S520: Determine a first predicted value for each piece of first information using a pre-trained conversion amount prediction model;

[0122] S530: Determine the predicted conversion amount of the product corresponding to each piece of first information based on the first predicted value of each piece of first information;

[0123] S540: Sort the plurality of first information based on the predicted commodity conversion amounts corresponding to the respective first information;

[0124] Among them, the conversion amount prediction model is trained using the above training method.

[0125] Using this method, a pre-trained conversion amount prediction model is used to predict each piece of first information and determine the corresponding first predicted value. Since this first predicted value is related to the minimum price of the product, by limiting the value range of this first predicted value, it is possible to indirectly control the product conversion predicted amount corresponding to each piece of first information to be no less than the minimum price of the product. Finally, multiple pieces of first information are sorted based on the product conversion predicted amount, improving the accuracy and reliability of the first information sorting. The first information in this embodiment of the disclosure includes advertisements, introductions, and other information about the product.

[0126] In some embodiments, the first predicted value satisfies a first functional relationship with the predicted amount of product conversion and the minimum amount of the product;

[0127] Determining the commodity conversion prediction amount corresponding to each first information according to the first prediction value of each first information includes: determining the commodity conversion prediction amount according to the first prediction value, the first functional relationship and the minimum amount of the commodity.

[0128] In some implementations, the first functional relationship includes: the first predicted value is equal to a first ratio minus 1, where the first ratio is the ratio of the predicted conversion amount of the product to the minimum amount of the product.

[0129] In the embodiment of the present disclosure, in order to set a reasonable lower limit for the prediction results of the conversion amount prediction model, a first functional relationship is defined, which specifically describes the intrinsic relationship between the first prediction value, the predicted conversion amount of the product, and the minimum amount of the product. Among them, the predicted conversion amount of the product is:

[0130] GMV=(PV+1)min_sku_price (6)

[0131] Among them, GMV represents the total transaction amount of the product determined by the predicted conversion amount of the product, PV represents the first predicted value, min_sku_price is the minimum price of the product. If the product is sold as a single piece, min_sku_price is the product price itself. If the product is sold as a package or combination, min_sku_price is the minimum price of the package or combination.

[0132] By adopting the above method and establishing the relationship between GMV and the minimum amount of goods, GMV can be made greater than the minimum amount of goods, thereby improving the reliability of the model's GMV prediction and providing data support for subsequent information sorting.

[0133] The above describes the process of determining the predicted conversion amount of the product corresponding to each piece of first information. For each piece of first information, the corresponding predicted revenue needs to be determined. Finally, the first information is ranked based on the predicted revenue corresponding to each piece of first information.

[0134] The predicted return of the first information is determined by the following values:

[0135] (1) The predicted amount of commodity conversion, i.e., GMV, can be determined using the conversion amount prediction model proposed in this disclosure;

[0136] (2) Click-Through-Rate (CTR) of the first information, which can be obtained by an external CTR prediction model;

[0137] (3) Return on Investment (ROI) is the net benefit obtained by users through the first information, which can be determined based on the parameters set by the user;

[0138] (4) The CVR of a product is the proportion of users who have visited or browsed the first information and have purchased the product corresponding to the first information, which can be determined based on the above CVR tower.

[0139] In the embodiment of the present disclosure, the predicted revenue can be determined by eCPM. The revenue from one effective exposure is:

[0140] M=CTR×CVR×GMV / ROI (7)

[0141] Based on the revenue from one effective exposure, eCPM can be determined in two ways:

[0142] (1) Multiply the revenue from one effective exposure by one thousand to directly get the eCPM;

[0143] (2) The conversion amount prediction model is used to predict the conversion amount of the first information multiple times, and the effective exposure revenue is calculated once using each prediction result, and the eCPM is determined based on the multiple effective exposure revenues.

[0144] Finally, the first information can be ranked according to the eCPM corresponding to each first information. Generally speaking, the larger the eCPM value, the higher the corresponding first information is ranked.

[0145] The present disclosure also provides a model training device. Figure 6 : is a schematic diagram of the structure of a model training device 600 according to an embodiment of the present disclosure, comprising:

[0146] The sample extraction module 610 is used to extract multiple commodity conversion samples from multiple information samples;

[0147] An input module 620 is configured to input a plurality of commodity conversion samples into a conversion amount prediction model to be trained to obtain a plurality of first prediction values, wherein the first prediction values ​​are related to the commodity conversion prediction amount and the commodity minimum amount;

[0148] a loss function determination module 630, configured to determine a loss function based on the plurality of first prediction values;

[0149] The training module 640 is used to train the conversion amount prediction model based on the loss function.

[0150] In some embodiments, the sample extraction module 610 is configured to:

[0151] Obtain multiple information samples and determine the user behavior corresponding to each information sample;

[0152] When the user behavior corresponding to the information sample is a conversion behavior, the information sample is determined to be a product conversion sample;

[0153] Extract multiple product conversion samples.

[0154] In some embodiments, the sample extraction module 610 is further configured to:

[0155] Record the actual amount of product conversion corresponding to each product conversion sample.

[0156] In some embodiments, the conversion amount prediction model includes an embedding module, a neural gate module, and a prediction module;

[0157] The input module 620 is configured to:

[0158] For each product conversion sample, perform the following operations:

[0159] The product conversion sample is input into the embedding module, and the embedding module is used to extract the feature vector of the product conversion sample; the neural gate module is used to set the weight corresponding to each element in the feature vector; and the prediction module is used to determine the first prediction value based on the feature vector and the weight corresponding to each element.

[0160] In some implementations, the first predicted value is related to the predicted conversion amount of the product and the minimum amount of the product, including:

[0161] The first predicted value satisfies a first functional relationship with the predicted amount of product conversion and the minimum amount of the product.

[0162] In some implementations, the first predicted value obtained by the conversion amount prediction model belongs to a first value range, and the first value range is related to the first functional relationship.

[0163] In some embodiments, the first functional relationship includes: the first predicted value is equal to the first ratio minus 1, and the first ratio is the ratio of the predicted conversion amount of the product to the minimum amount of the product;

[0164] The first value range is greater than or equal to 0.

[0165] In some embodiments, the loss function determination module 630 is configured to:

[0166] Fitting the distribution of the plurality of first prediction values ​​to a first Poisson distribution;

[0167] Determine the actual conversion amount of each product conversion sample, and determine the first real value of each product conversion sample based on the actual conversion amount, the minimum amount of the product, and the first functional relationship;

[0168] Fitting the distribution of the plurality of first true values ​​to a second Poisson distribution;

[0169] A loss function is determined based on the first Poisson distribution and the second Poisson distribution.

[0170] In some embodiments, the loss function determination module 630 is configured to:

[0171] Compute the Poisson negative log-likelihood loss between the first and second Poisson distributions;

[0172] The Poisson distribution negative log-likelihood loss is used as the loss function.

[0173] The present disclosure also provides an information sorting device. Figure 7 FIG. 7 is a schematic diagram of the structure of an information sorting device 700 according to an embodiment of the present disclosure, comprising:

[0174] An acquisition module 710 is configured to acquire a plurality of first information;

[0175] A first determination module 720 is configured to determine a first predicted value of each piece of first information using a pre-trained conversion amount prediction model;

[0176] The second determining module 730 is configured to determine the predicted conversion amount of the product corresponding to each piece of first information based on the first predicted value of each piece of first information;

[0177] A sorting module 740 is configured to sort the plurality of first information based on the predicted commodity conversion amount corresponding to each piece of first information;

[0178] The conversion amount prediction model is obtained by training using the above-mentioned training device.

[0179] In some embodiments, the first predicted value satisfies a first functional relationship with the predicted amount of product conversion and the minimum amount of the product;

[0180] The second determination module 730 is configured to determine a predicted conversion amount of the product based on the first predicted value, the first functional relationship, and the minimum amount of the product.

[0181] In some implementations, the first functional relationship includes: the first predicted value is equal to the first ratio minus 1, and the first ratio is the ratio of the predicted amount of product conversion to the minimum amount of the product.

[0182] For the description of specific functions and examples of each module and submodule of the device in the embodiment of the present disclosure, please refer to the relevant description of the corresponding steps in the above method embodiment, which will not be repeated here.

[0183] In the technical solution disclosed herein, the acquisition, storage and application of personal information of users involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0184] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0185] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0186] like Figure 8 As shown, the device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. Various programs and data required for the operation of the device 800 can also be stored in the RAM 803. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0187] Various components in device 800 are connected to I / O interface 805, including: an input unit 806, such as a keyboard, mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, optical disk, etc.; and a communication unit 809, such as a network card, modem, wireless communication transceiver, etc. The communication unit 809 allows device 800 to exchange data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0188] The computing unit 801 can be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as the detection method. For example, in some embodiments, the detection method can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the detection method described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform the detection method in any other appropriate manner (e.g., by means of firmware).

[0189] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system comprising at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0190] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0191] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0192] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0193] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0194] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0195] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0196] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A model training method, comprising: Acquire multiple information samples and determine user behaviors corresponding to each of the information samples; If the user behavior corresponding to the information sample is a conversion behavior, the information sample is determined to be a product conversion sample; and a plurality of the product conversion samples are extracted; wherein the conversion behavior includes a process in which the user changes from an initial behavior of at least one of browsing and clicking to a final behavior of at least one of actual purchase and payment; Inputting the plurality of commodity conversion samples into a conversion amount prediction model to be trained to obtain a plurality of first prediction values, wherein the first prediction values ​​are related to the commodity conversion prediction amount and the commodity minimum amount; determining a loss function based on the plurality of first prediction values; The conversion amount prediction model is trained based on the loss function.

2. The method according to claim 1, further comprising: Record the actual amount of commodity conversion corresponding to each of the commodity conversion samples.

3. The method according to claim 2, wherein: The conversion amount prediction model includes an embedding module, a neural gate module and a prediction module; Inputting the plurality of commodity conversion samples into the conversion amount prediction model to be trained to obtain a plurality of first prediction values ​​includes: For each product conversion sample, perform the following operations: The commodity conversion sample is input into the embedding module, and the embedding module is used to extract the feature vector of the commodity conversion sample; the neural gate module is used to set the weight corresponding to each element in the feature vector; and the prediction module is used to determine the first prediction value based on the feature vector and the weight corresponding to each element.

4. The method according to any one of claims 1 to 3, wherein: The first predicted value is related to the predicted conversion amount of the product and the minimum amount of the product, including: The first predicted value satisfies a first functional relationship with the predicted conversion amount of the product and the minimum amount of the product.

5. The method according to claim 4, wherein The first predicted value obtained by the conversion amount prediction model belongs to a first value range, and the first value range is related to the first functional relationship.

6. The method according to claim 5, wherein: The first functional relationship includes: the first predicted value is equal to a first ratio minus 1, and the first ratio is the ratio of the predicted conversion amount of the product to the minimum amount of the product; The first value range is greater than or equal to 0.

7. The method according to claim 6, wherein: The determining of the loss function according to the plurality of first prediction values ​​comprises: Fitting the distribution of the plurality of first prediction values ​​to a first Poisson distribution; Determining the actual commodity conversion amount of each commodity conversion sample, and determining a first real value of each commodity conversion sample based on the actual commodity conversion amount, the minimum commodity amount, and the first functional relationship; Fitting the distribution of the plurality of first true values ​​to a second Poisson distribution; The loss function is determined according to the first Poisson distribution and the second Poisson distribution.

8. The method according to claim 7, wherein: The determining the loss function according to the first Poisson distribution and the second Poisson distribution includes: Calculating a Poisson distribution negative log-likelihood loss between the first Poisson distribution and the second Poisson distribution; The Poisson distribution negative log-likelihood loss is used as the loss function.

9. A method for sorting information, comprising: obtaining a plurality of first information; Determine a first predicted value for each piece of the first information using a pre-trained conversion amount prediction model; Determining a predicted conversion amount of a product corresponding to each piece of the first information based on the first predicted value of each piece of the first information; sorting the plurality of first information based on the predicted commodity conversion amounts corresponding to the respective first information; Wherein, the conversion amount prediction model is trained using any training method described in claims 1-8.

10. The method according to claim 9, wherein: A first functional relationship is satisfied between the first predicted value, the predicted conversion amount of the product, and the minimum amount of the product; Determining the commodity conversion prediction amount corresponding to each piece of the first information based on the first prediction value of each piece of the first information includes: determining the commodity conversion prediction amount based on the first prediction value, the first functional relationship and the minimum amount of the commodity.

11. The method according to claim 9 or 10, wherein: The first functional relationship includes: the first predicted value is equal to a first ratio minus 1, and the first ratio is the ratio of the predicted conversion amount of the product to the minimum amount of the product.

12. A model training device comprising: A sample extraction module is used to obtain multiple information samples and determine the user behavior corresponding to each of the information samples; If the user behavior corresponding to the information sample is a conversion behavior, the information sample is determined to be a product conversion sample; and a plurality of the product conversion samples are extracted; wherein the conversion behavior includes a process in which the user changes from an initial behavior of at least one of browsing and clicking to a final behavior of at least one of actual purchase and payment; An input module, configured to input the plurality of commodity conversion samples into a conversion amount prediction model to be trained to obtain a plurality of first prediction values, wherein the first prediction values ​​are related to the commodity conversion prediction amount and the commodity minimum amount; a loss function determining module, configured to determine a loss function based on the plurality of first prediction values; A training module is used to train the conversion amount prediction model based on the loss function.

13. The apparatus according to claim 12, wherein the sample extraction module is further configured to: Record the actual amount of commodity conversion corresponding to each of the commodity conversion samples.

14. The device according to claim 13, wherein The conversion amount prediction model includes an embedding module, a neural gate module and a prediction module; The input module is used to: For each product conversion sample, perform the following operations: The commodity conversion sample is input into the embedding module, and the embedding module is used to extract the feature vector of the commodity conversion sample; the neural gate module is used to set the weight corresponding to each element in the feature vector; and the prediction module is used to determine the first prediction value based on the feature vector and the weight corresponding to each element.

15. The device according to any one of claims 12 to 14, wherein: The first predicted value is related to the predicted conversion amount of the product and the minimum amount of the product, including: The first predicted value satisfies a first functional relationship with the predicted conversion amount of the product and the minimum amount of the product.

16. The device according to claim 15, wherein The first predicted value obtained by the conversion amount prediction model belongs to a first value range, and the first value range is related to the first functional relationship.

17. The device according to claim 16, wherein The first functional relationship includes: the first predicted value is equal to a first ratio minus 1, and the first ratio is the ratio of the predicted conversion amount of the product to the minimum amount of the product; The first value range is greater than or equal to 0.

18. The device according to claim 17, wherein The loss function determination module is used to: Fitting the distribution of the plurality of first prediction values ​​to a first Poisson distribution; Determining the actual commodity conversion amount of each commodity conversion sample, and determining a first real value of each commodity conversion sample based on the actual commodity conversion amount, the minimum commodity amount, and the first functional relationship; Fitting the distribution of the plurality of first true values ​​to a second Poisson distribution; The loss function is determined according to the first Poisson distribution and the second Poisson distribution.

19. The device according to claim 18, wherein The loss function determination module is used to: Calculating a Poisson distribution negative log-likelihood loss between the first Poisson distribution and the second Poisson distribution; The Poisson distribution negative log-likelihood loss is used as the loss function.

20. An information sorting device, comprising: An acquisition module, configured to acquire a plurality of first information; a first determination module, configured to determine a first prediction value of each of the first information using a pre-trained conversion amount prediction model; a second determining module, configured to determine a predicted conversion amount of a commodity corresponding to each piece of the first information based on the first predicted value of each piece of the first information; A sorting module, configured to sort the plurality of first information based on the predicted commodity conversion amount corresponding to each piece of first information; Wherein, the conversion amount prediction model is obtained by training using any one of the training devices described in claims 12-19.

21. The device according to claim 20, wherein A first functional relationship is satisfied between the first predicted value, the predicted conversion amount of the product, and the minimum amount of the product; The second determination module is configured to determine the predicted conversion amount of the commodity based on the first predicted value, the first functional relationship, and the minimum amount of the commodity.

22. The device according to claim 20 or 21, wherein The first functional relationship includes: the first predicted value is equal to a first ratio minus 1, and the first ratio is the ratio of the predicted conversion amount of the product to the minimum amount of the product.

23. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 11.

24. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-11.

25. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 11.

Images