Commodity recommendation method and device, equipment and medium

Abstract

The invention provides a commodity recommendation method and device, equipment and a medium, and relates to the technical field of computers, in particular to the technical field of artificial intelligence and recommendation. According to the implementation scheme, user features of a target user are obtained, and the user features comprise first features capable of indicating the sensitivity of the target user to commodity production information; determining first score information of the plurality of candidate commodities based on target production information associated with the plurality of candidate commodities and user characteristics of the target user; and based on the first score information of the plurality of candidate commodities, determining a recommended commodity used for being recommended to the target user from the plurality of candidate commodities.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, in particular to the field of artificial intelligence technology and recommendation technology, and specifically to a product recommendation method, device, electronic device, and computer-readable storage medium. Background Art

[0002] With the development of computer technology and big data technology, people's consumption, entertainment, learning, travel and other behaviors in life are closely related to big data. In the operation process of business platforms, it is usually necessary to use user data to recommend products in order to improve corresponding sales performance.

[0003] The methods described in this section are not necessarily methods that have been previously conceived or employed. Unless otherwise indicated, it should not be assumed that any method described in this section is considered to be prior art simply because it is included in this section. Similarly, unless otherwise indicated, the issues mentioned in this section should not be considered to have been recognized in any prior art. Summary of the invention

[0004] The present disclosure provides a product recommendation method, device, electronic device, and computer-readable storage medium.

[0005] According to one aspect of the present disclosure, a product recommendation method is provided, comprising: obtaining user characteristics of a target user, wherein the user characteristics include a first feature that can indicate the target user's sensitivity to product production information; determining first rating information of a plurality of candidate products based on target production information associated with the plurality of candidate products and the user characteristics of the target user; and determining a recommended product for recommendation to the target user from the plurality of candidate products based on the first rating information of the plurality of candidate products.

[0006] According to another aspect of the present disclosure, a product recommendation device is provided, including: an acquisition unit configured to acquire user characteristics of a target user, wherein the user characteristics include a first feature that can indicate the target user's sensitivity to product production information; a first determination unit configured to determine first rating information of a plurality of candidate products based on target production information associated with the plurality of candidate products and the user characteristics of the target user; and a second determination unit configured to determine a recommended product for recommendation to the target user from among the plurality of candidate products based on the first rating information of the plurality of candidate products.

[0007] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned product recommendation method.

[0008] 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 enable a computer to execute the above-mentioned product recommendation method.

[0009] According to one or more embodiments of the present disclosure, it is possible to determine recommended commodities by taking into account the sensitivity of target users to commodity production information, so that the recommended commodities can better meet the needs of target users and improve the accuracy of commodity recommendations.

[0010] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The accompanying drawings exemplarily illustrate the embodiments and constitute a part of the specification, and together with the text description of the specification, are used to explain the exemplary implementation of the embodiments. The embodiments shown are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0012] Figure 1 A flowchart of a commodity recommendation method according to an exemplary embodiment of the present disclosure is shown;

[0013] Figure 2 A flowchart of a commodity recommendation method according to an exemplary embodiment of the present disclosure is shown;

[0014] Figure 3 A schematic diagram of the structure of a commodity recommendation model according to an exemplary embodiment of the present disclosure is shown;

[0015] Figure 4 A structural block diagram of a commodity recommendation device according to an exemplary embodiment of the present disclosure is shown;

[0016] Figure 5 A block diagram of an electronic device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0017] The following is a description of exemplary embodiments of the present disclosure 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 recognized by those of ordinary skill 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, the description of well-known functions and structures is omitted in the following description.

[0018] In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, timing relationship, or importance relationship of these elements, and such terms are only used to distinguish one element from another element. In some examples, the first element and the second element may refer to the same instance of the element, and in some cases, based on the description of the context, they may also refer to different instances.

[0019] The terms used in the description of various examples in this disclosure are only for the purpose of describing specific examples and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element can be one or more. In addition, the term "and / or" used in this disclosure covers any one of the listed items and all possible combinations.

[0020] In the related art, recommendations are usually made based on user preferences for different categories of goods, but the impact of the production process of the goods on the user experience is not taken into account.

[0021] Based on this, the present disclosure provides a product recommendation method, which determines recommended products in combination with the target user's sensitivity to product production information, so that the recommended products can better meet the user's needs.

[0022] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0023] Figure 1 FIG. 1 shows a flowchart of a commodity recommendation method 100 according to an exemplary embodiment of the present disclosure. Figure 1 As shown, the method 100 includes:

[0024] Step S101: obtaining user characteristics of a target user, wherein the user characteristics include a first characteristic that can indicate the sensitivity of the target user to commodity production information;

[0025] Step S102, determining first rating information of the plurality of candidate commodities based on target production information associated with the plurality of candidate commodities and user characteristics of target users; and

[0026] Step S103: Determine a recommended commodity for recommendation to a target user from among the plurality of candidate commodities based on the first rating information of the plurality of candidate commodities.

[0027] By applying the above method 100, a first score of a candidate commodity can be determined based on a first feature indicating the target user's sensitivity to commodity production information and current production data, so that the first score can indicate the target user's satisfaction with the production situation of the candidate commodity, and based on this, a recommended commodity can be determined so that the recommended commodity can better meet the needs of the user.

[0028] In some examples, the user characteristics of the target user may include a user portrait indicating information in multiple dimensions such as the user's personal characteristics and behavioral characteristics. For example, the user portrait of the target user's gender, age group, occupation, and resident area may also include the target user's historical consumption frequency, historical consumption unit price, and user behavioral characteristics of historical purchased goods information, so as to more accurately describe the characteristics of the target user and improve the accuracy of recommendations.

[0029] In some examples, the user features of the target user may be user feature vectors obtained using a specific encoding method, so that the user features may be learned using a machine learning method to achieve more efficient and accurate product recommendations.

[0030] According to some embodiments, the first feature is determined based on the target user's historical product evaluation information, which includes the target user's evaluation information on the production information of historically purchased products. Thus, mining can be performed based on the historical product evaluation information to more accurately determine the user's sensitivity to the product production time.

[0031] In some examples, historical product evaluation information may include evaluation text. In this case, for example, the first feature of the target user may be obtained by extracting keywords related to product production information preset in the evaluation text. For another example, keywords may be extracted from historical product evaluation information based on natural language semantic understanding technology. In one example, a label prediction model for the evaluation text may be used to determine the first label of the target user, and the first label may be used as the first feature of the target user. The first label may include, for example, a label indicating that the target user is very sensitive to product production information or a label indicating that the target user is not sensitive to product production information. In this case, the label prediction model may be obtained by supervised training using sample evaluation texts annotated with reference labels, so as to obtain the first feature of the target user more accurately and efficiently.

[0032] In some examples, product production information includes product production time, which may be the time from the current moment to the estimated production completion time of the product, so that recommended products that can be produced faster can be recommended to users based on this to fully meet user needs.

[0033] In some examples, product production information may also be other information, such as the required production time of the product, specifically the required production time from the start time of production to the completion time of production of the product. When the user expects to start production of the product after arriving at the store, the product production can be completed faster to meet the user's needs.

[0034] According to some embodiments, the target production information includes at least one of the following: information on the required production time of each candidate commodity, equipment status information indicating whether the equipment used to produce each candidate commodity is idle, raw material inventory information of the candidate commodity, personnel status information indicating whether the production personnel who can produce the candidate commodity are idle, and information on commodities that need to be produced within a preset time period in the future. Based on this, the production information of the commodity can be predicted more accurately.

[0035] It is understandable that the target production information may further include more information, such as information indicating whether the production processes of multiple candidate commodities can be combined, based on which the production information of the commodities can be predicted more accurately.

[0036] According to some embodiments, determining the first rating information of multiple candidate commodities based on the target production information associated with the multiple candidate commodities and the user characteristics of the target user in step S103 includes: determining the predicted production information of the multiple candidate commodities based on the target production information associated with the multiple candidate commodities; and determining the first rating information of the multiple candidate commodities based on the predicted production information of the multiple candidate commodities and the first characteristics of the target user. Thus, the predicted production information of the commodity can be determined first and then the corresponding first rating that can indicate the user experience can be determined, so as to improve the interpretability of the output result and facilitate more accurate adjustment of the commodity recommendation algorithm.

[0037] In some examples, the predicted production information includes predicted production time, so that the current product production schedule can be calculated through the target production information to determine the predicted production time of the candidate products, based on which more accurate first scoring information is obtained to improve the recommendation accuracy.

[0038] According to some embodiments, based on target production information associated with multiple candidate commodities, determining predicted production information of multiple candidate commodities includes: inputting the target production information associated with the multiple candidate commodities into a production information prediction network to obtain predicted production information of the multiple candidate commodities, wherein the production information prediction network is trained using first sample data labeled with real production information. Thus, the trained production information prediction network can be used to predict production time to improve efficiency and accuracy.

[0039] In some examples, the first sample data includes sample production information associated with multiple sample commodities, and the training process of the conversion prediction network may include the following steps: inputting the first sample data into the initial production information prediction network to obtain predicted production information output by the initial production information prediction network, calculating the loss value of the initial production information prediction network based on the predicted production information and the actual production information, and then adjusting the parameters of the initial production information prediction network based on the loss value. By repeating the above training and parameter adjustment steps, a production information prediction network with optimized performance can be obtained.

[0040] When the predicted production information of the candidate products includes the predicted production time, the shorter the predicted time, the greater the possibility of recommendation. This is contrary to the monotonous trend of the predicted conversion information and is not conducive to simply and efficiently determining the recommendation results.

[0041] Based on this, according to some embodiments, the predicted production information of each candidate commodity includes a cumulative distribution function of the production time of the candidate commodity, and the cumulative distribution function indicates the probability distribution of the candidate commodity, and wherein, based on the predicted production information of multiple candidate commodities and the first feature of the target user, determining the first score information of multiple candidate commodities includes: determining the first score information of multiple candidate commodities based on the cumulative distribution function of the production time of multiple candidate commodities and the first feature of the target user. By determining the first score based on the cumulative distribution function, the normalization of the prediction results can be achieved. For example, when the first score is determined based on the inverse of the function value of the cumulative distribution function corresponding to the preset time, the calculation trend that the larger the first score, the greater the possibility of recommendation can be achieved, which is convenient for directly adding the predicted conversion information to determine the final score, and based on this, the recommendation result is determined simply and efficiently.

[0042] In some examples, the production information prediction network may be used to predict the distribution information of the production time of the candidate goods. The distribution information of the production time may include, for example, the variance of the production time and the mean of the production time, and then the cumulative distribution function of the production time is determined based on this. In some examples, the function value P(T) of the cumulative distribution function may be determined based on the preset time T, where P(T) indicates the probability that the production time of the candidate goods is not greater than the preset time T. In this case, the value of P is greater than 0 and less than 1. On this basis, the first scoring information may be obtained by calculating the difference between 1 and the function value P(T).

[0043] In some examples, determining the recommended product for recommendation to the target user from the multiple candidate products based on the first rating information of the multiple candidate products in step S103 includes: sorting the multiple candidate products based on the first rating information; and determining the recommended product for recommendation to the target user based on the sorting result. In this way, the recommended product can be determined more simply and accurately.

[0044] According to some embodiments, method 100 also includes: obtaining product features of multiple candidate products; determining predicted conversion information of multiple candidate products for the target user based on the product features of the multiple candidate products and the user features of the target user, and in step S103, based on the first rating information of the multiple candidate products and the predicted conversion information of the multiple candidate products, determining a recommended product for recommendation to the target user from the multiple candidate products.

[0045] In some examples, product features may include product category, price, historical sales information, discount information, etc.

[0046] According to some embodiments, the predicted conversion information includes at least one of the following: the probability of the target user placing an order to purchase each candidate product and the sales generated by the target user placing an order to purchase each candidate product. Thus, it is possible to determine recommended products based on the impact of the candidate products on actual sales performance, and then improve product sales performance through product recommendations to meet the needs of actual application scenarios.

[0047] According to some embodiments, at least one of the first rating information and the predicted conversion information is obtained using a product recommendation model, the product recommendation model includes at least one of a rating network and a conversion prediction network, wherein the rating network is used to output first rating information of multiple candidate products based on target production information associated with the multiple candidate products and first user features of target users, and the rating network is trained using second sample data annotated with reference rating information,

[0048] In this example, determining the first scoring information of the plurality of candidate commodities based on the target production information associated with the plurality of candidate commodities and the first feature of the target user in step S103 includes: inputting the target production information associated with the plurality of candidate commodities and the first feature of the target user into the scoring network to obtain the first scoring information of the plurality of candidate commodities. Thus, the first scoring information associated with the production information can be directly obtained using the end-to-end network, thereby improving efficiency and convenience.

[0049] In some examples, the second sample data includes sample production information associated with multiple sample commodities and first sample features of sample users, and the training process of the conversion prediction network may include the following steps: inputting the second sample data into the initial scoring network to obtain the predicted scoring information output by the initial scoring network, calculating the loss value of the initial scoring network based on the predicted scoring information and the reference scoring information, and then adjusting the parameters of the initial scoring network based on this. By repeating the above training and parameter adjustment steps, a scoring network with optimized performance can be obtained.

[0050] According to some embodiments, the conversion prediction network is used to output predicted conversion information of multiple candidate products for target users based on product features of multiple candidate products and user features of target users, and the conversion prediction network is trained using third sample data labeled with real conversion information.

[0051] In this example, the product features of multiple candidate products and the user features of target users can be input into the conversion prediction network to obtain predicted conversion information of multiple candidate products. Thus, the trained conversion prediction network can be used to predict revenue to improve efficiency and accuracy.

[0052] In some examples, the third sample data includes sample product features of multiple sample products and sample user features of sample users, and the training process of the conversion prediction network may include the following steps: inputting the third sample data into the initial conversion prediction network to obtain predicted conversion information output by the initial conversion prediction network, calculating the loss value of the initial conversion prediction network based on the predicted conversion information and the actual conversion information, and then adjusting the parameters of the initial conversion prediction network based on this. By repeating the above training and parameter adjustment steps, a conversion prediction network with optimized performance can be obtained.

[0053] In some examples, the initial conversion prediction network may be constructed based on one or more of a cascade network layer, a multi-layer perceptron, and an activation layer, and the present disclosure is not limited to the specific network structure.

[0054] In some examples, the predicted conversion information of each candidate product may be obtained by calculation based on a preset formula including user characteristics and product characteristics to improve efficiency.

[0055] In some examples, based on the predicted conversion information of multiple candidate products and the first rating information of multiple candidate products, determining the recommended product for recommendation to the target user from the multiple candidate products includes: performing weighted summation on the predicted conversion information of the multiple candidate products and the first rating information of the multiple candidate products to obtain the recommended rating information of each candidate product; sorting the multiple candidate products based on the recommended rating information; and determining the recommended product for recommendation to the target user based on the sorting result. In this way, the recommended product can be determined more simply and accurately.

[0056] In some examples, the weights of the predicted conversion information of the candidate products and the first rating information of the candidate products can be determined according to actual needs to meet the needs of actual application scenarios.

[0057] Figure 2 FIG. 2 shows a flowchart of a commodity recommendation method 200 according to an exemplary embodiment of the present disclosure. Figure 2 As shown, the method 200 includes:

[0058] Step S201: obtaining user characteristics of a target user and commodity characteristics of a plurality of candidate commodities, wherein the user characteristics include a first characteristic capable of indicating the sensitivity of the target user to commodity production information;

[0059] Step S202: inputting product features of multiple candidate products and user features of target users into a conversion prediction network to obtain predicted conversion information of the multiple candidate products;

[0060] Step S203: inputting target production information associated with a plurality of candidate commodities and a first feature of a target user into a scoring network to obtain first scoring information of the plurality of candidate commodities;

[0061] Step S204: performing weighted summation on the predicted conversion information of the plurality of candidate commodities and the first score information of the plurality of candidate commodities to obtain the recommended score information of each candidate commodity;

[0062] Step S205: sorting the multiple candidate commodities based on the recommendation rating information; and

[0063] Step S206: Determine the recommended products to be recommended to the target user based on the ranking result.

[0064] By applying the above method 200, it is possible to more efficiently and accurately determine recommended products for target users, so as to specifically meet the needs of target users and improve the accuracy of recommendations.

[0065] Figure 3 A structural schematic diagram of a commodity recommendation model according to an exemplary embodiment of the present disclosure is shown.

[0066] In this example, the product recommendation model can be constructed based on a dual-tower structure, specifically including a first tower for outputting predicted conversion information and a second tower for outputting first rating information. In some examples, the first tower can be similar in structure and function to the conversion prediction network described above, and the second tower can be similar in structure and function to the rating network described above.

[0067] See also Figure 3 As shown, in this example, the input side of the first tower and the second tower includes a coding layer and a cascade network layer for embedding and encoding user features, product features, and target production information to obtain a feature vector that can be processed by a neural network model. When the encoding of the target production information is completed, the first feature of the target user and the feature vector corresponding to the target production information are jointly output to the next network layer of the second tower, so that the second tower can make predictions based on the first feature of the target user. By inputting the encoded feature vector into the hidden layer, the input features can be mapped to more complex feature dimensions to obtain output results based on this. By further connecting the multi-layer perceptron and the activation layer and the regression layer, the features output by the hidden layer can be mapped to the output data set to obtain the predicted conversion information output by the first tower and the first scoring information output by the second tower.

[0068] By applying the above dual-tower network, step S202 and step S203 in the above method 200 are implemented.

[0069] Based on the predicted conversion information and the first rating information of each candidate product for the target user output by the first tower and the second tower respectively, the two can be weighted calculated based on the preset weights in the output layer to obtain the recommended rating information of each candidate product, thereby realizing the above-mentioned step S204.

[0070] In some examples, multiple candidate products may be sorted in the output layer based on the weighted calculation results, and the sorting results may be used as a basis for selecting recommended products, thereby implementing the above-mentioned steps S205 to S206.

[0071] The product recommendation model can be obtained by supervised training using fourth sample data annotated with real conversion information and reference rating information. The fourth sample data may specifically include sample product features of multiple sample products, sample user features of sample users, and sample production information related to multiple sample products. It can be understood that during the training process, it is necessary to explicitly obtain the predicted conversion information and the first rating information output by the first tower and the second tower respectively, so as to calculate the loss value based on the predicted results and the annotated reference results, and then adjust the parameters of the initial product recommendation model. By repeating the training and parameter adjustment steps multiple times, a product recommendation model with optimized performance can be obtained.

[0072] After the product recommendation model training is completed, there is no need to obtain the output results of the first tower and the second tower during the model application process. It is only necessary to obtain the recommendation score information of the output layer or the candidate product ranking results.

[0073] According to one aspect of the present disclosure, a commodity recommendation device is also provided. Figure 4 FIG. 4 shows a structural block diagram of a commodity recommendation device 400 according to an exemplary embodiment of the present disclosure. Figure 4 As shown, the device 400 includes:

[0074] The acquisition unit 401 is configured to acquire user characteristics of a target user, wherein the user characteristics include a first characteristic that can indicate the sensitivity of the target user to the production time of the product;

[0075] A first determining unit 402 is configured to determine, for each candidate commodity among the plurality of candidate commodities, a first score of the candidate commodity based on current production information associated with the candidate commodity and user characteristics of a target user; and

[0076] The second determining unit 403 is configured to determine a recommended commodity for recommendation to a target user from among the plurality of candidate commodities based on the first rating information of the plurality of candidate commodities.

[0077] According to some embodiments, the first feature is determined based on historical product evaluation information of the target user, where the historical product evaluation information includes evaluation information of the target user on production information of historically purchased products.

[0078] According to some embodiments, the second determination unit 403 includes: a first determination subunit, configured to determine predicted production information of multiple candidate commodities based on target production information associated with the multiple candidate commodities; and a second determination subunit, configured to determine first scoring information of the multiple candidate commodities based on the predicted production information of the multiple candidate commodities and the first feature of the target user.

[0079] According to some embodiments, the target production information includes at least one of the following items: production time information of the candidate commodity, equipment status information indicating whether the equipment used to produce the candidate commodity is idle, raw material inventory information of the candidate commodity, personnel status information indicating whether the production personnel who can produce the candidate commodity are idle, and information on commodities that need to be produced within a preset time period in the future, and the predicted production information includes predicted production time, which indicates the time between the current moment and the predicted production completion moment of the candidate commodity.

[0080] According to some embodiments, the predicted production information of each candidate commodity includes a cumulative distribution function of the production time of the candidate commodity, the cumulative distribution function indicating the probability distribution of the production time of the candidate commodity, and wherein the second determination subunit is configured to determine first scoring information of multiple candidate commodities based on the cumulative distribution function of the production time of the multiple candidate commodities and the first feature of the target user.

[0081] According to some embodiments, the first determination subunit is configured to input target production information associated with multiple candidate commodities into a production information prediction network to obtain predicted production information of the multiple candidate commodities, wherein the production information prediction network is trained using first sample data labeled with real production information.

[0082] According to some embodiments, the acquisition unit 401 is also configured to acquire product features of multiple candidate products, and the device 400 also includes: a third determination unit 404, configured to determine the predicted conversion information of multiple candidate products for the target user based on the product features of the multiple candidate products and the user characteristics of the target user, and wherein the second determination unit 403 is configured to: determine the recommended product for recommendation to the target user from the multiple candidate products based on the first rating information of the multiple candidate products and the predicted conversion information of the multiple candidate products.

[0083] According to some embodiments, at least one of the first rating information and the predicted conversion information is obtained using a product recommendation model, the product recommendation model includes at least one of a rating network and a conversion prediction network, wherein the rating network is used to output first rating information of multiple candidate products based on target production information associated with the multiple candidate products and first user characteristics of the target user, the rating network is trained using second sample data labeled with reference rating information, and wherein the conversion prediction network is used to output predicted conversion information of multiple candidate products for the target user based on product characteristics of the multiple candidate products and user characteristics of the target user, and the conversion prediction network is trained using third sample data labeled with real conversion information.

[0084] According to some embodiments, the predicted conversion information includes at least one of the following items: the probability that the target user places an order to purchase each candidate commodity and the sales amount generated by the target user placing an order to purchase each candidate commodity.

[0085] According to another aspect of the present disclosure, an electronic device is also provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned product recommendation method.

[0086] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is also provided, wherein the computer instructions are used to enable a computer to execute the above-mentioned product recommendation method.

[0087] See also Figure 5 , a block diagram of an electronic device 500 that can be used as the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device can be a computer device of different types, such as a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, 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 required herein.

[0088] Figure 5 1 shows a block diagram of an electronic device according to an embodiment of the present disclosure. Figure 5 As shown, the electronic device 500 may include at least one processor 501 , a working memory 502 , an I / O device 504 , a display device 505 , a storage device 506 , and a communication interface 507 that can communicate with each other via a system bus 503 .

[0089] The processor 501 may be a single processing unit or multiple processing units, all of which may include a single or multiple computing units or multiple cores. The processor 501 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuits, and / or any device that manipulates signals based on operating instructions. The processor 501 may be configured to obtain and execute computer-readable instructions stored in a working memory 502, a storage device 506, or other computer-readable media, such as program codes of an operating system 502a, program codes of an application 502b, and the like.

[0090] Working memory 502 and storage device 506 are examples of computer-readable storage media for storing instructions, which are executed by processor 501 to implement the various functions described above. Working memory 502 may include both volatile memory and non-volatile memory (e.g., RAM, ROM, etc.). In addition, storage device 506 may include a hard disk drive, a solid-state drive, a removable medium, including external and removable drives, memory cards, flash memory, a floppy disk, an optical disk (e.g., CD, DVD), a storage array, a network attached storage, a storage area network, etc. Working memory 502 and storage device 506 may all be collectively referred to herein as memory or computer-readable storage media, and may be a non-transitory medium capable of storing computer-readable, processor-executable program instructions as computer program code, which may be executed by processor 501 as a specific machine configured to implement the operations and functions described in the examples herein.

[0091] The I / O device 504 may include an input device and / or an output device. The input device may be any type of device capable of inputting information to the electronic device 500, and may include but is not limited to a mouse, a keyboard, a touch screen, a track pad, a track ball, a joystick, a microphone, and / or a remote controller. The output device may be any type of device capable of presenting information, and may include but is not limited to a video / audio output terminal, a vibrator, and / or a printer.

[0092] The communication interface 507 allows the electronic device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver and / or a chipset, such as a Bluetooth™ device, an 802.11 device, a WiFi device, a WiMax device, a cellular communication device and / or the like.

[0093] The application 502b in the working register 502 can be loaded to execute the various methods and processes described above, such as Figure 1In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via the storage device 506 and / or the communication interface 507. When the computer program is loaded and executed by the processor 501, one or more steps of the commodity recommendation method described above may be executed.

[0094] Various implementations of the systems and techniques described above 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), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including 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.

[0095] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0096] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may 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 may 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.

[0097] 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).

[0098] The systems and techniques described herein may 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 with 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 may 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.

[0099] A computing system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0100] It should be understood that the various forms of 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 different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0101] Although the embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it should be understood that the above-mentioned methods, systems and devices are merely exemplary embodiments or examples, and the scope of the present invention is not limited by these embodiments or examples, but only by the claims after authorization and their equivalent scope. Various elements in the embodiments or examples can be omitted or replaced by their equivalent elements. In addition, each step can be performed in an order different from that described in the present disclosure. Further, the various elements in the embodiments or examples can be combined in various ways. It is important that with the evolution of technology, many elements described herein can be replaced by equivalent elements that appear after the present disclosure.

Claims

1. A product recommendation method, comprising: Acquire user characteristics of a target user, wherein the user characteristics include a first characteristic that can indicate the sensitivity of the target user to commodity production information; determining first rating information of the plurality of candidate commodities based on target production information associated with the plurality of candidate commodities and user characteristics of the target user; and Based on the first rating information of the plurality of candidate commodities, a recommended commodity for recommendation to the target user is determined from the plurality of candidate commodities.

2. The method of claim 1, wherein: The first feature is determined based on historical commodity evaluation information of the target user, where the historical commodity evaluation information includes evaluation information of the target user on production information of historically purchased commodities.

3. The method of claim 1, wherein: The determining of first scoring information of the plurality of candidate commodities based on target production information associated with the plurality of candidate commodities and user characteristics of the target user comprises: determining predicted production information of the plurality of candidate commodities based on target production information associated with the plurality of candidate commodities; and Based on the predicted production information of the plurality of candidate commodities and the first characteristic of the target user, first scoring information of the plurality of candidate commodities is determined.

4. The method of claim 3, wherein: The target production information includes at least one of the following: production time information of the candidate product, equipment status information indicating whether the equipment used to produce the candidate product is idle, raw material inventory information of the candidate product, personnel status information indicating whether the production personnel who can produce the candidate product are idle, and information of the product that needs to be produced within a preset time period in the future, The predicted production information includes a predicted production duration, where the predicted production duration indicates a duration between a current time and a predicted production completion time of the candidate commodity.

5. The method of claim 3, wherein: The predicted production information of each candidate commodity includes a cumulative distribution function of the production time of the candidate commodity, wherein the cumulative distribution function indicates the probability distribution of the production time of the candidate commodity. And wherein, determining the first scoring information of the plurality of candidate commodities based on the predicted production information of the plurality of candidate commodities and the first feature of the target user comprises: Based on the cumulative distribution function of the production time of the plurality of candidate commodities and the first feature of the target user, first scoring information of the plurality of candidate commodities is determined.

6. The method of claim 3, wherein: The determining, based on the target production information associated with the plurality of candidate commodities, predicted production information of the plurality of candidate commodities comprises: The target production information associated with the multiple candidate commodities is input into a production information prediction network to obtain predicted production information of the multiple candidate commodities, wherein the production information prediction network is trained using first sample data labeled with real production information.

7. The method according to any one of claims 1 to 6, further comprising: Acquire commodity features of the plurality of candidate commodities; Determine predicted conversion information of the plurality of candidate commodities for the target user based on commodity features of the plurality of candidate commodities and user features of the target user, And wherein, determining the recommended product for recommendation to the target user from the multiple candidate products based on the first rating information of the multiple candidate products comprises: Based on the first rating information of the plurality of candidate commodities and the predicted conversion information of the plurality of candidate commodities, a recommended commodity for recommendation to the target user is determined from the plurality of candidate commodities.

8. The method of claim 7, wherein: At least one of the first rating information and the predicted conversion information is obtained by using a product recommendation model, wherein the product recommendation model includes at least one of a rating network and a conversion prediction network. The scoring network is used to output first scoring information of the plurality of candidate commodities based on target production information associated with the plurality of candidate commodities and the first user feature of the target user, and the scoring network is trained using second sample data annotated with reference scoring information. And wherein, the conversion prediction network is used to output predicted conversion information of the multiple candidate products for the target user based on the product features of the multiple candidate products and the user features of the target user, and the conversion prediction network is trained using third sample data labeled with real conversion information.

9. The method of claim 7, wherein: The predicted conversion information includes at least one of the following items: the probability of the target user placing an order to purchase each candidate commodity and the sales amount generated by the target user placing an order to purchase each candidate commodity.

10. A commodity recommendation device, comprising: an acquisition unit, configured to acquire user characteristics of a target user, wherein the user characteristics include a first characteristic capable of indicating the sensitivity of the target user to commodity production information; A first determining unit is configured to determine first scoring information of the plurality of candidate commodities based on target production information associated with the plurality of candidate commodities and user characteristics of the target user; and The second determining unit is configured to determine a recommended commodity for recommending to the target user from the plurality of candidate commodities based on the first scoring information of the plurality of candidate commodities.

11. An electronic device, comprising: at least one processor; as well as a memory communicatively coupled to the at least one processor; in 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 9.

12. 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 1-9.

Images