Agricultural product recommendation method, electronic device, storage medium, and program product
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA UNITED NETWORK COMM GRP CO LTD
- Filing Date
- 2026-02-13
- Publication Date
- 2026-06-05
Smart Images
Figure CN122153156A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data processing technology, and in particular to a method for recommending agricultural products, electronic devices, storage media, and program products. Background Technology
[0002] With the accelerated digital transformation of agriculture and the booming development of the agricultural e-commerce market, precise marketing and efficient distribution of agricultural products have become one of the core needs of the industry.
[0003] In existing technologies, agricultural products are mainly recommended to users by analyzing their historical shopping records.
[0004] However, the accuracy of the recommendations from the existing technologies mentioned above is insufficient, and they cannot fully adapt to the personalized needs of target users and the multi-dimensional attributes of agricultural products. Summary of the Invention
[0005] The agricultural product recommendation method, electronic device, storage medium, and program products provided in this application are used to improve the accuracy of agricultural product recommendations.
[0006] In a first aspect, embodiments of this application provide a method for recommending agricultural products, including:
[0007] Acquire first user data and first product data. The first user data is user data that has been authorized for use by the target user, and the first product data is agricultural product data produced in the target area.
[0008] The first user data is preprocessed to obtain the second user data;
[0009] Based on the second user data, determine the first user profile data of the target user;
[0010] Based on the data of the first product, a knowledge graph of the first product was determined;
[0011] Based on the first user profile data and the first product knowledge graph, determine the first recommended product list for the target user;
[0012] Obtain the second product data, which is the historical sales data of agricultural products produced in the target area;
[0013] Based on the second product data, the first recommended product list is modified to obtain the second recommended product list;
[0014] Output the second recommended product list.
[0015] Secondly, embodiments of this application provide an agricultural product recommendation device, comprising:
[0016] The acquisition module is used to acquire first user data and first product data. The first user data is user data that has been authorized for use by the target user, and the first product data is agricultural product data produced in the target area.
[0017] The processing module is used to preprocess the first user data to obtain the second user data;
[0018] The processing module is also used to determine the first user profile data of the target user based on the second user data;
[0019] The processing module is also used to determine the knowledge graph of the first product based on the first product data;
[0020] The processing module is also used to determine the first recommended product list for the target user based on the first user profile data and the first product knowledge graph;
[0021] The acquisition module is also used to acquire second product data, which is the historical sales data of agricultural products produced in the target area;
[0022] The processing module is also used to modify the first recommended product list based on the second product data to obtain the second recommended product list;
[0023] The output module is used to output the second list of recommended products.
[0024] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;
[0025] The memory stores the instructions that the computer executes;
[0026] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.
[0027] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.
[0028] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.
[0029] The agricultural product recommendation method, electronic device, storage medium, and program product provided in this application obtain target user authorization data and target area agricultural product data, construct accurate user profiles and build product knowledge graphs through preprocessing, combine the two to form an initial recommendation list, and then optimize and adjust it using historical sales data of agricultural products. This effectively solves the problem of insufficient recommendation accuracy in existing technologies, can fully adapt to the personalized needs of target users and the multi-dimensional attributes of agricultural products, and helps to improve the precision marketing effect and circulation efficiency of agricultural products. Attached Figure Description
[0030] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.
[0031] Figure 1 A flowchart illustrating the agricultural product recommendation method provided in this application embodiment. Figure 1 ;
[0032] Figure 2 A flowchart illustrating the agricultural product recommendation method provided in this application embodiment. Figure 2 ;
[0033] Figure 3 This is a schematic diagram of the structure of the agricultural product recommendation device provided in the embodiments of this application;
[0034] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.
[0035] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation
[0036] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0037] To address the shortcomings of existing technologies in providing accurate recommendations that fail to fully meet the personalized needs of target users and the multi-dimensional attributes of agricultural products, the following technical concept is proposed: First, acquire two types of core related data directly relevant to the recommendation scenario. Systematic preprocessing of one type of data extracts key information to construct a feature profile tailored to practical application needs. Simultaneously, deep analysis and organization of the other type of data clarifies the inherent relationships and attribute characteristics between data points, building a complete and comprehensive knowledge graph. Based on this, multi-dimensional matching is performed using the constructed feature profile and knowledge graph to generate an initial recommendation set. Then, historical related data is introduced as an optimization basis to dynamically adjust and correct deviations in this initial recommendation set, further improving the adaptability and accuracy of the recommendation results. Finally, recommendations that accurately meet core needs are output, thereby improving the accuracy of agricultural product recommendations.
[0038] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.
[0039] Figure 1 A flowchart illustrating the agricultural product recommendation method provided in this application embodiment. Figure 1 The methods described above in this application's embodiments can be applied to any electronic device. For example... Figure 1 As shown, the method includes:
[0040] S101. Obtain first user data and first product data. The first user data is user data that has been authorized for use by the target user, and the first product data is agricultural product data produced in the target area.
[0041] The first type of user data consists of user-related data that has been explicitly authorized by the target user and can be used for recommendation services, such as the user's dietary preferences, purchase history, age, gender, health status, and delivery address.
[0042] First product data: Data related to agricultural products produced within the target area (such as a specific county or production area), such as the name, category, origin, nutritional components, market launch time, specifications, and price of the agricultural products.
[0043] The target users are individuals or groups who receive agricultural product recommendation services (such as ordinary consumers and corporate buyers). The target area is a pre-defined agricultural product production area.
[0044] Specifically, data is collected through the interactive functions of electronic devices, such as reading user registration information, historical order data, and actively filled preference questionnaires after pop-up authorization; and through API calls to third-party authorized data (such as purchase records authorized by e-commerce platforms), ensuring that all data has been approved by the user and complies with privacy protection requirements. Data is also collected through reports from agricultural producers (farmers, cooperatives, and wholesale markets in the target area); through IoT devices (such as origin traceability devices); and through public channels (such as directories of agricultural products published by agricultural and rural affairs departments), ensuring that the data covers major agricultural products in the target area and that the information is authentic and valid.
[0045] Optional data may also include implicit user behavior data (page dwell time, clickstream events, search query logs), certification information (organic certification, geographical indication products, etc.), and cultural attributes (such as the history of Xiangshui rice as a tribute rice).
[0046] S102. Preprocess the first user data to obtain the second user data.
[0047] Specifically, the process involves cleaning the first user data (removing duplicate data, such as duplicate purchase records; deleting invalid data, such as blank preference options and incorrect contact information), standardizing the data (unifying the data format, such as unifying "age 25 years old" and "25" into numerical formats, and unifying "likes sweets" and "prefers sweets" into standardized expressions), completing the data (appropriately filling in missing key data, such as inferring unfilled dietary preferences through users' historical purchase records), and denoising the data (filtering out abnormal data, such as users mistakenly filling in "age 1000 years old"), ultimately outputting clean and standardized second user data.
[0048] S103. Based on the second user data, determine the first user profile data of the target user.
[0049] Specifically, feature extraction is performed on the second user data (such as extracting features like "purchase frequency" and "preferred product categories" from purchase records, and "age" and "gender" from personal information); then the extracted features are labeled ("purchases ≥ 2 times per month" are labeled as "high-frequency buyers", and "frequent purchases of spinach and lettuce" are labeled as "preferring leafy vegetables"); finally, all labels are integrated to form a complete first user profile data, ensuring that the profile can accurately match the user's real needs.
[0050] S104. Based on the data of the first product, determine the knowledge graph of the first product.
[0051] Specifically, core entities are extracted from the primary product data (e.g., agricultural product entities: apples, wheat; nutritional component entities: vitamin C, dietary fiber; geographical type entities: mountains, plains); then the relationships between entities are defined (e.g., the "richness" relationship between agricultural products and nutritional components, and the "origin type" relationship between agricultural products and geographical types); finally, through knowledge graph construction tools, the entities and relationships are entered to form a visualized or computable primary product knowledge graph, ensuring that the multi-dimensional attributes and related information of agricultural products can be quickly queried.
[0052] S105. Based on the first user profile data and the first product knowledge graph, determine the first recommended product list for the target user.
[0053] Specifically, the tags in the first user profile data (such as "prefer organic, weight loss, leafy greens") are matched with the agricultural product attribute tags in the first product knowledge graph. Algorithms are used to calculate the matching degree (such as cosine similarity and collaborative filtering). Agricultural products with high matching degrees are selected and sorted from highest to lowest to form the first recommended product list. For example, if the user profile tags are "organic, weight loss," the knowledge graph shows "organic spinach (low-calorie, high-fiber)" with the highest matching degree, so it is prioritized for inclusion in the list.
[0054] S106. Obtain the second product data, which is the historical sales data of agricultural products produced in the target area.
[0055] The second product data consists of historical sales data of agricultural products produced in the target region, such as sales volume, sales revenue, repurchase rate, peak sales periods, and user reviews of agricultural products in the target region over the past three months.
[0056] Specifically, historical sales data are extracted from the back-end systems of sales channels (e-commerce platforms, offline wholesale markets, supermarkets) in the target area; sales records are exported from sales terminals (such as POS systems and order management systems); and the extracted data is organized (e.g., summarized by agricultural product category and time dimension) to ensure that the data covers the main sales scenarios of agricultural products in the target area and that the time span is of reference value (usually 3-6 months).
[0057] S107. Based on the second product data, modify the first recommended product list to obtain the second recommended product list.
[0058] Specifically, the first recommended product list is adjusted by adding, deleting, and modifying items (removing agricultural products with extremely low historical sales and a repurchase rate of 0; adding agricultural products that are historically best-selling, have a high repurchase rate, and match user profiles), adjusting the ranking (promoting agricultural products from historical peak sales seasons and those with good user reviews), and optimizing suitability (e.g., based on historical data, if "weight-loss users prefer to buy small-packaged vegetables," then small-packaged products are prioritized in the list); and recalculating the recommendation priority using algorithms (such as logistic regression and decision trees) combined with historical sales data and the initial matching degree, ultimately generating the second recommended product list.
[0059] S108. Output the second recommended product list.
[0060] Specifically, the output should be displayed through the terminal device's interface, such as the application homepage recommendation bar, SMS push, articles, mini-program pop-ups, etc.; the output content should include the core information of the recommended product (name, place of origin, price, and reasons for recommendation, such as "suitable for weight loss, currently a hot seller"), to ensure that users can clearly obtain the recommendation information and place an order conveniently.
[0061] The agricultural product recommendation method, electronic device, storage medium, and program product provided in this application obtain target user authorization data and target area agricultural product data, construct accurate user profiles and build product knowledge graphs through preprocessing, combine the two to form an initial recommendation list, and then optimize and adjust it using historical sales data of agricultural products. This effectively solves the problem of insufficient recommendation accuracy in existing technologies, can fully adapt to the personalized needs of target users and the multi-dimensional attributes of agricultural products, and helps to improve the precision marketing effect and circulation efficiency of agricultural products.
[0062] Figure 2 A flowchart illustrating the agricultural product recommendation method provided in this application embodiment. Figure 2 .like Figure 2 As shown, the method includes:
[0063] S201. Obtain first user data and first product data. The first user data is user data that has been authorized for use by the target user, and the first product data is agricultural product data produced in the target area.
[0064] S202. Perform data cleaning on the first user data to obtain the third user data.
[0065] Specifically, the first user data is processed by removing duplicate data (such as questionnaires with the same preferences submitted multiple times by the user), deleting invalid data (such as blank fields, garbled characters, and illogical data, such as "purchase time in the future"), correcting erroneous data (such as correcting "age - 20 years" to invalid data and removing it, and correcting "misspelled delivery address" to the correct address), and handling missing data (such as marking "unknown" if the user did not fill in "dietary preferences" and not filling it in arbitrarily), and finally outputting the third user data without impurities.
[0066] S203. Standardize the third user data to obtain the fourth user data.
[0067] Specifically, the third user data undergoes numerical data standardization (e.g., unifying "age 25 years old, 30 years old" into integer format, unifying "height 175cm, 1.75m" into cm units, and mapping the values to the 0-1 range through normalization); textual data standardization (e.g., unifying "likes sweets, prefers sweets, likes sweets" into "prefers sweet flavors", and unifying "organic, organic food" into "organic"); and categorical data standardization (e.g., unifying "gender male, female" into "0, 1"), ultimately outputting fourth user data with a unified format.
[0068] S204. Perform feature vectorization on the fourth user data to obtain the second user data.
[0069] Specifically, core features (such as age, gender, purchase frequency, preferred product categories, and health status) are extracted from the fourth user data. Then, each feature is converted into a numerical vector using a feature vectorization algorithm. Finally, all feature vectors are concatenated and integrated to form a complete user feature vector (i.e., the second user data), ensuring that the computer can match user needs with product attributes through vector calculations. For example, "preferring leafy vegetables" is converted to [1,0,0] (1 represents "yes", 0 represents "no") using a preset encoding, and combined with other feature vectors to form a complete vector.
[0070] S205. Based on the second user data, determine the first user profile data of the target user.
[0071] In one possible implementation, the explicit tag data of the target user is determined by matching the second user data with a preset rule base; the second user data is input into a preset implicit tag generation model to obtain the implicit tag data of the target user; and the explicit tag data and the implicit tag data are integrated to obtain the first user profile data.
[0072] The preset rule base is a pre-defined set of rules used to match explicit user characteristics (the core being "explicit matching rules"), such as "age ≥ 60 years old → preference for soft agricultural products" and "frequently purchases infant food → attention to complementary food agricultural products." Explicit tag data consists of feature tags directly reflected by the user, obtained through matching the rule base (the core being "explicit needs," such as "over 60 years old, preference for soft agricultural products"). Implicit tag data consists of feature tags that are not directly reflected by the user but are potential, obtained through model mining (the core being "implicit needs," such as "frequently purchases weight-loss ingredients → potential need: low-calorie, high-fiber agricultural products").
[0073] Specifically, the second user data (feature vector) is input into a preset rule base. Through rule matching (e.g., "age feature 65 in vector → match 'elderly user' rule → generate 'preference for soft agricultural products' label"), explicit label data is output. The second user data is input into a preset implicit label generation model (e.g., a decision tree-based model). Through model calculation (e.g., analyzing the correlation of user purchase records, "purchase of weight loss tea + spinach → uncover the implicit demand of 'focus on healthy weight loss'"), implicit label data is output. The explicit and implicit label data are deduplicated and merged (e.g., removing duplicate "focus on health" labels and merging "explicit: preference for soft" and "implicit: focus on easily digestible" labels) to finally obtain the first user profile data.
[0074] S206. Based on the data of the first product, determine the knowledge graph of the first product.
[0075] In one possible implementation, the first product knowledge graph includes at least one of the following:
[0076] Data linking agricultural product entities and nutritional component entities;
[0077] Association data between agricultural product entities and geographical type entities;
[0078] Data linking agricultural products to suitable population groups;
[0079] Data linking agricultural products and cultural attributes.
[0080] Specifically, the association between agricultural products and nutritional components can be achieved by extracting the agricultural product entity and its corresponding nutritional components from the primary product data (e.g., "apple → vitamin C, dietary fiber") to construct a "rich" association; the association between agricultural products and geographical types can be achieved by extracting the agricultural product entity and its geographical type of origin (e.g., "mountain apple → mountain") to construct a "origin type" association; the association between agricultural products and suitable populations can be achieved by extracting the agricultural product entity and its suitable population (e.g., "soft pear → elderly") to construct a "suitable for consumption" association; and the association between agricultural products and cultural attributes can be achieved by extracting the agricultural product entity and its cultural attributes (e.g., "red dates → Spring Festival") to construct a "cultural heritage" association.
[0081] S207. Based on the first user profile data and the first product knowledge graph, determine the first recommended product list for the target user.
[0082] In one possible implementation, the first user profile data is input into a first processing channel to obtain a first score; the first user profile data and a first product knowledge graph are input into a second processing channel to obtain a second score; the first product knowledge graph is input into a third processing channel to obtain a third score; and based on the first score, the second score, and the third score, a first recommended product list for the target user is determined.
[0083] Specifically, the first user profile data is input into the first processing channel, where an algorithm (such as a user preference scoring model) is used to calculate the user's preference scores for various agricultural products (e.g., "user prefers leafy vegetables → first score for leafy vegetables is 9 points, and for fruits it is 6 points"), and the first score is output. The first user profile data and the first product knowledge graph are simultaneously input into the second processing channel, where a matching algorithm (such as cosine similarity) is used to calculate the matching degree score between the user and each agricultural product (e.g., "matching degree between user profile 'weight loss' and 'spinach' knowledge graph attribute 'low-calorie' → second score is 8.5 points"), and the second score is output. The first product knowledge graph is input into the third processing channel, where an algorithm (such as a product attribute scoring model) is used to calculate the quality score of the agricultural product's own attributes (e.g., "organic agricultural products have better attributes than ordinary agricultural products → third score is 9 points"), and the third score is output. Combining the three scores, agricultural products with higher comprehensive scores are selected to form the first recommended product list.
[0084] In one possible implementation, a first weight value for the first rating is determined based on the first rating, the second rating, and the third rating; a second weight value for the second rating is determined based on the first rating, the second rating, and the third rating; a third weight value for the third rating is determined based on the first rating, the second rating, and the third rating; and the first rating, the second rating, and the third rating are weighted and fused based on the first weight value, the second weight value, and the third weight value to obtain the first recommended product list for the target user.
[0085] Specifically, based on the importance of the three ratings, a weight value is determined for each rating (e.g., the second rating (matching degree) has a weight of 0.5, the first rating (user preference) has a weight of 0.3, and the third rating (product attributes) has a weight of 0.2), with the total weight value being 1; the comprehensive score is calculated, with the formula example being "Comprehensive Score = First Rating × 0.3 + Second Rating × 0.5 + Third Rating × 0.2"; all agricultural products are sorted from high to low according to their comprehensive scores, and the top N agricultural products (e.g., the top 10) are selected to form the first recommended product list.
[0086] S208. Obtain the second product data, which is the historical sales data of agricultural products produced in the target area.
[0087] S209. Based on the second product data, modify the first recommended product list to obtain the second recommended product list.
[0088] S210, Output the second recommended product list.
[0089] The preprocessing of the first user data is broken down into three ordered steps: data cleaning, standardization, and feature vectorization. This systematically removes impurities, redundancy, and errors from the original user data, standardizes the data format, and achieves numerical conversion of the data. It effectively avoids the deviation problems in subsequent processing caused by the disordered nature of the original data, and provides high-quality, reusable basic data support for core links such as user profile construction and recommendation model calculation, ensuring the accuracy and reliability of the processing results in subsequent links.
[0090] By combining explicit tag matching with implicit tag mining to construct the first user profile data of the target users, we can not only accurately capture the explicit needs clearly expressed by the target users, but also deeply explore the potential preferences that users have not directly expressed. Compared with the single-dimensional profile construction method, the final user profile is more in line with the real needs of the target users, laying the core foundation for the subsequent realization of personalized and precise agricultural product recommendations.
[0091] The first product knowledge graph clarifies the multi-dimensional related data types it covers, comprehensively covering the association information between agricultural products and nutritional components, geographical types, suitable populations, and cultural attributes. It breaks through the limitations of traditional product information presentation being fragmented and single-dimensional, and realizes the structured and systematic sorting of agricultural product attributes, which facilitates the rapid and accurate matching of user profile needs and improves the matching efficiency between products and user needs.
[0092] The first recommended product list is determined by a three-channel independent scoring method, which quantitatively evaluates the product from the dimensions of user preference, user-product matching, and product attributes. This overcomes the problem of one-sided and unreasonable recommendation evaluation caused by single-dimensional scoring, and can comprehensively consider the suitability of recommendations from multiple perspectives, making the generation of the initial recommendation list more comprehensive, objective and reasonable.
[0093] Based on the three-channel scoring, a weight allocation and weighted fusion mechanism is further introduced. The weight ratio of each scoring dimension can be dynamically determined according to the actual application scenario and recommendation needs. Through weighted calculation, a comprehensive score that is more in line with the actual recommendation needs is obtained, realizing accurate sorting and scientific screening of the first recommended product list. This effectively improves the adaptability and usability of the initial recommendation list and lays a solid foundation for the optimization of the final recommendation results.
[0094] Figure 3 This is a schematic diagram of the structure of the agricultural product recommendation device provided in the embodiments of this application, such as... Figure 3 As shown, the agricultural product recommendation device 30 provided in this embodiment includes an acquisition module 301, a processing module 302, and an output module 302.
[0095] The acquisition module 301 is used to acquire first user data and first product data. The first user data is user data that has been authorized for use by the target user, and the first product data is agricultural product data produced in the target area.
[0096] Processing module 302 is used to preprocess the first user data to obtain the second user data;
[0097] Processing module 302 is also used to determine the first user profile data of the target user based on the second user data;
[0098] Processing module 302 is also used to determine the knowledge graph of the first product based on the first product data;
[0099] Processing module 302 is also used to determine the first recommended product list for the target user based on the first user profile data and the first product knowledge graph;
[0100] The acquisition module 301 is also used to acquire second product data, which is the historical sales data of agricultural products produced in the target area;
[0101] The processing module 302 is also used to modify the first recommended product list based on the second product data to obtain the second recommended product list;
[0102] Output module 303 is used to output the second recommended product list.
[0103] In one possible implementation, the processing module 302 is specifically used for:
[0104] Data cleaning is performed on the first user's data to obtain the third user's data;
[0105] The third user data is standardized to obtain the fourth user data;
[0106] The fourth user data is vectorized to obtain the second user data.
[0107] In one possible implementation, the processing module 302 is specifically used for:
[0108] The explicit tag data of the target user is determined by matching the second user data with the preset rule base.
[0109] The second user data is input into the preset implicit label generation model to obtain the implicit label data of the target user;
[0110] By integrating explicit and implicit tag data, the first user profile data is obtained.
[0111] In one possible implementation, the first product knowledge graph includes at least one of the following:
[0112] Data linking agricultural product entities and nutritional component entities;
[0113] Association data between agricultural product entities and geographical type entities;
[0114] Data linking agricultural products to suitable population groups;
[0115] Data linking agricultural products and cultural attributes.
[0116] In one possible implementation, the processing module 302 is specifically used for:
[0117] The first user profile data is input into the first processing channel to obtain the first score;
[0118] The first user profile data and the first product knowledge graph are input into the second processing channel to obtain the second score;
[0119] The knowledge graph of the first product is input into the third processing channel to obtain the third score;
[0120] Based on the first, second, and third ratings, determine the first list of recommended products for the target user.
[0121] In one possible implementation, the processing module 302 is specifically used for:
[0122] Based on the first score, the second score, and the third score, determine the first weight value of the first score;
[0123] Based on the first, second, and third scores, determine the second weight value of the second score;
[0124] Based on the first, second, and third scores, determine the third weight value for the third score;
[0125] Based on the first weight value, the second weight value, and the third weight value, the first score, the second score, and the third score are weighted and fused to obtain the first recommended product list for the target user.
[0126] The agricultural product recommendation device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.
[0127] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the device 40 further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus.
[0128] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.
[0129] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.
[0130] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.
[0131] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.
[0132] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.
[0133] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.
[0134] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.
[0135] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.
[0136] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.
[0137] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.
[0138] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0139] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.
[0140] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0141] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.
[0142] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.
Claims
1. A method for recommending agricultural products, characterized in that, include: Acquire first user data and first product data, wherein the first user data is user data that has been authorized for use by the target user, and the first product data is agricultural product data produced in the target area; The first user data is preprocessed to obtain the second user data; Based on the second user data, the first user profile data of the target user is determined; Based on the first product data, a first product knowledge graph is determined; Based on the first user profile data and the first product knowledge graph, a first recommended product list for the target user is determined; Obtain the second product data, which is the historical sales data of agricultural products produced in the target area; Based on the second product data, the first recommended product list is modified to obtain the second recommended product list; Output the second recommended product list.
2. The method according to claim 1, characterized in that, The preprocessing of the first user data to obtain the second user data includes: The first user data is cleaned to obtain the third user data; The third user data is standardized to obtain the fourth user data; The fourth user data is vectorized to obtain the second user data.
3. The method according to claim 1, characterized in that, The step of determining the first user profile data of the target user based on the second user data includes: Based on the second user data, the explicit tag data of the target user is determined by matching it with a preset rule base; The second user data is input into a preset implicit tag generation model to obtain the implicit tag data of the target user; By integrating the explicit label data and the implicit label data, the first user profile data is obtained.
4. The method according to claim 1, characterized in that, The first product knowledge graph includes at least one of the following: Data linking agricultural product entities and nutritional component entities; Association data between agricultural product entities and geographical type entities; Data linking agricultural products to suitable population groups; Data linking agricultural products and cultural attributes.
5. The method according to claim 1, characterized in that, The step of determining the first recommended product list for the target user based on the first user profile data and the first product knowledge graph includes: The first user profile data is input into the first processing channel to obtain the first score; The first user profile data and the first product knowledge graph are input into the second processing channel to obtain the second score; The first product knowledge graph is input into the third processing channel to obtain the third score; Based on the first rating, the second rating, and the third rating, a first recommended product list for the target user is determined.
6. The method according to claim 5, characterized in that, The step of determining the first recommended product list for the target user based on the first rating, the second rating, and the third rating includes: Based on the first score, the second score, and the third score, a first weight value for the first score is determined; Based on the first score, the second score, and the third score, a second weight value for the second score is determined; Based on the first score, the second score, and the third score, a third weight value for the third score is determined; Based on the first weight value, the second weight value, and the third weight value, the first score, the second score, and the third score are weighted and fused to obtain the first recommended product list for the target user.
7. An agricultural product recommendation device, characterized in that, include: The acquisition module is used to acquire first user data and first product data. The first user data is user data that has been authorized for use by the target user, and the first product data is agricultural product data produced in the target area. The processing module is used to preprocess the first user data to obtain the second user data; The processing module is further configured to determine the first user profile data of the target user based on the second user data; The processing module is further configured to determine a first product knowledge graph based on the first product data; The processing module is further configured to determine the first recommended product list for the target user based on the first user profile data and the first product knowledge graph; The acquisition module is also used to acquire second product data, which is the historical sales data of agricultural products produced in the target area; The processing module is further configured to modify the first recommended product list based on the second product data to obtain a second recommended product list; The output module is used to output the second recommended product list.
8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.
10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-6.