An agricultural product online and offline traceability marketing system and method

Abstract

The application relates to the technical field of product marketing, in particular to an online and offline traceability marketing system and method for agricultural products, which comprises a quality monitoring unit, an product traceability unit and a code scanning and recording unit.The quality monitoring unit is used for monitoring the environmental data of target agricultural products in real time, and determining the quality attenuation coefficient of the target agricultural products according to the environmental data.The product traceability unit is used for generating a traceability code according to the quality attenuation coefficient and the space-time information of the target agricultural products.The code scanning and recording unit is used for printing the traceability code on the packaging of the target agricultural products.Through real-time monitoring of the environmental data of target agricultural products and generation of a traceability code, consumers can obtain detailed information of the agricultural products by scanning the code, including the production environment, quality control, attenuation condition and the like.Transparent traceability information enhances the trust of consumers in product quality, especially in the background of increasing attention to food safety and the source of agricultural products.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of product marketing, and in particular to an online and offline traceability marketing system and method for agricultural products. BACKGROUND

[0002] Traditional systems often lack real-time and transparent traceability functions for agricultural products, and consumers lack sufficient information about the production environment, quality control and decay of agricultural products, and cannot obtain detailed information through scanning codes or other methods, which makes it difficult for consumers to make more rational purchasing decisions and may have lower trust in product quality. In addition, traditional systems usually promote products to all consumers in a single way, lack real-time monitoring and analysis of consumer behavior, and traditional systems often cannot dynamically adjust and accurately segment according to the purchasing intentions and historical behavior of consumers, resulting in less personalized marketing strategies and even missing potential high-intention consumers. In addition, traditional systems usually rely on experience or rough sales data for inventory management and sales forecasting, lacking accurate prediction of consumer purchasing behavior and purchase probability, which makes it difficult for businesses to accurately predict which products will quickly sell out and which products have low demand, easily leading to inventory accumulation and increasing the risk of unsalable or expired products. In traditional systems, marketing strategies are often rough and cannot be tailored to individual consumers' interests, needs and purchasing intentions, and lack of personalized marketing not only reduces consumer engagement but also may lead to a decrease in purchase conversion rates. SUMMARY

[0003] The technical problem to be solved by the present application is to overcome the shortcomings of the prior art and provide an online and offline traceability marketing system and method for agricultural products.

[0004] The technical solution adopted to solve the above technical problems is an online and offline traceability marketing system for agricultural products, comprising:

[0005] A quality monitoring unit is configured to monitor environmental data of target agricultural products in real time and determine a quality decay coefficient of the target agricultural products based on the environmental data.

[0006] A product traceability unit is configured to generate a traceability code based on the quality decay coefficient and the spatio-temporal information of the target agricultural products.

[0007] A code scanning recording unit is configured to print the traceability code on the packaging of target agricultural products, deploy the traceability code to an online page, and obtain the code scanning records of consumers based on the packaging of the target agricultural products and the online page.

[0008] a purchase probability unit configured to predict a purchase probability of the consumer according to the code scanning records of the consumer and a purchase conversion rate corresponding to each code scanning;

[0009] an intention classification unit configured to classify all consumers into a first consumer set and a second consumer set according to the purchase probabilities;

[0010] a product marketing unit configured to perform a price penetration strategy on the first consumer set and perform a value education strategy on the second consumer set.

[0011] Preferably, determining the quality decay coefficient of the target agricultural product according to the environmental data comprises:

[0012] calculating a dynamic spoilage rate of the target agricultural product according to the environmental data;

[0013] determining the quality decay coefficient of the target agricultural product according to the dynamic spoilage rate of the target agricultural product.

[0014] Preferably, the calculation formula of the dynamic spoilage rate is as follows:

[0015] k=k0·exp(α(T-T opt )+β(H-H opt ) 2 +γSigmoid(L-L th ));

[0016] wherein k represents the dynamic spoilage rate of the target agricultural product, k0 represents a basic spoilage rate, i.e. a spoilage rate under standard environmental conditions, α, β and γ represent environmental sensitivity coefficients, respectively reflecting the influence of temperature, humidity and light changes on the spoilage rate, T, H and L represent temperature data, humidity data and light data in the environmental data, T opt , H opt and L th represent optimal storage conditions of the target agricultural product;

[0017] The calculation formula of the quality decay coefficient is as follows:

[0018]

[0019] wherein Q(t) represents the quality decay coefficient at a current sampling time t, Q0 represents an initial quality coefficient, t now represents a current time, and t harvest represents a time when the target agricultural product is picked.

[0020] Preferably, generating a traceability code according to the quality decay coefficient and the spatio-temporal information of the target agricultural product comprises:

[0021] Obtain the CRC16 checksum of the quality attenuation coefficient, and generate a quality channel code based on the quality attenuation coefficient and the corresponding CRC16 checksum.

[0022] The spatiotemporal information of the target agricultural product is obtained to generate a spatiotemporal channel code;

[0023] The quality channel code and the spatiotemporal channel code are combined in a dual-channel manner to obtain the traceability code.

[0024] Preferably, the calculation formula for the quality channel code is as follows:

[0025]

[0026] Among them, Code Q This represents the quality channel code, and CRC16(Q(t)) represents the CRC16 checksum of the quality attenuation coefficient.

[0027] The calculation formula for the spatiotemporal channel encoding is as follows:

[0028]

[0029] Among them, Code ST This indicates the spatiotemporal channel encoding. Geohash(x,y,9) represents the geographic coordinates (x,y) encoded as Geohash with a precision of 9 bits. This represents the XOR operation in bitwise operations, applied to a combination of geographic location and timestamp. UnixTime(t now ,ms) indicates converting the current time in milliseconds to Unix time format;

[0030] The formula for calculating the traceability code is as follows:

[0031] Code dual =Code Q ||Code ST ;

[0032] Among them, Code dual This represents the source code.

[0033] Preferably, predicting the consumer's purchase probability based on the consumer's scanning records and the purchase conversion rate corresponding to each scan includes:

[0034] The decay rate is determined based on the current time and the scanning time in the consumer's scanning record;

[0035] The consumer's trust value is determined based on the purchase conversion rate corresponding to each scan and the decay rate.

[0036] determine a social influence of the consumer according to a PageRank algorithm;

[0037] predict a purchase probability of the consumer according to the quality decay coefficient, a trust value of the consumer and the social influence of the consumer.

[0038] Preferably, the calculation formula of the decay speed is as follows:

[0039] ω(t)=exp(-λ(t now -t scan(i) ));

[0040] wherein ω(t) represents the decay speed at the current sampling time, t scan(i) represents the scanning code time of the i-th scanning code in the scanning code record, and λ represents a decay factor.

[0041] The calculation formula of the trust value is as follows:

[0042]

[0043] wherein Trust(u) represents the trust value of the consumer, n represents the number of scanning codes in the scanning code record, and Cr(i) represents the purchase conversion rate of the i-th scanning code in the scanning code record.

[0044] The calculation formula of the social influence is as follows:

[0045]

[0046] wherein Social(u) represents the social influence of the consumer, d represents a damping coefficient, N represents the total number of consumers, B u represents a friend set of the consumer u, and L(v) represents the out-link number of the consumer u, i.e. the number of people followed by the consumer u.

[0047] The prediction formula of the purchase probability is as follows:

[0048]

[0049] wherein P buy (u) represents the purchase probability of the consumer, a1, a2 and a3 represent preset weight coefficients, and a1+a2+a3=1, and σ represents a Sigmoid function, which is used to compress the purchase probability to the interval [0, 1].

[0050] Preferably, all consumers are classified according to the purchase probability to obtain a first consumer set and a second consumer set, comprising:

[0051] comparing the purchase probability with a preset first intention threshold and a preset second intention threshold.

[0052] If the purchase probability is greater than the preset first intention threshold, the consumer corresponding to the purchase probability is added to an empty first consumer set;

[0053] If the purchase probability is less than the preset second intention threshold, the consumer corresponding to the purchase probability is added to an empty second consumer set;

[0054] The above operation is repeated until all consumers are processed to obtain the first consumer set and the second consumer set, wherein the first consumer set represents a high-purchase-intention consumer set, and the second consumer set represents a low-purchase-intention consumer set.

[0055] Preferably, a price penetration strategy is performed on the first consumer set, and a value education strategy of agricultural products is performed on the second consumer set, including:

[0056] A dynamically adjusted price is determined according to the average purchase probability of the consumers in the first consumer set, and a calculation formula of the dynamically adjusted price is as follows:

[0057]

[0058] wherein p represents the dynamically adjusted price, represents the average purchase probability of the consumers in the first consumer set, and p base represents a reference price of the target agricultural product;

[0059] The price penetration strategy is performed on the first consumer set according to the dynamically adjusted price;

[0060] The value education information of agricultural products is constructed according to the environmental data and the farming information of the target agricultural product, and the value education strategy of agricultural products is performed on the second consumer set according to the value education information of agricultural products.

[0061] The technical solution adopted to solve the above technical problem is: an online and offline traceability marketing method for agricultural products, which is applicable to the online and offline traceability marketing system for agricultural products and includes:

[0062] Real-time monitoring of environmental data of a target agricultural product, determination of a quality attenuation coefficient of the target agricultural product according to the environmental data;

[0063] Generation of a traceability code according to the quality attenuation coefficient and the spatiotemporal information of the target agricultural product;

[0064] Printing of the traceability code on the packaging of the target agricultural product, deployment of the traceability code to an online page, and acquisition of a scanning code record of a consumer according to the packaging of the target agricultural product and the online page;

[0065] predict the purchase probability of the consumer according to the code scanning record of the consumer and the purchase conversion rate corresponding to each code scanning;

[0066] intention classification is performed on all consumers according to the purchase probability to obtain a first consumer set and a second consumer set;

[0067] a price penetration strategy is performed on the first consumer set, and an agricultural product value education strategy is performed on the second consumer set.

[0068] The beneficial effects of the present application are as follows: (1) The present application generates a traceability code by real-time monitoring of the environmental data of the target agricultural product, and consumers can obtain detailed information of the agricultural product, including its production environment, quality control, and decay, through code scanning. The transparent traceability information enhances the trust of consumers in product quality, especially in the context of increasing attention to food safety and the source of agricultural products. The purchase probability of consumers can be accurately predicted by dynamically adjusting the real-time environmental data and the code scanning record of consumers. Through this data, merchants can quickly identify and segment consumer groups, so as to implement more accurate marketing strategies for different groups. The first consumer set can implement a price penetration strategy, and the second consumer set can increase their awareness and interest in products through an agricultural product value education strategy; (2) The present application predicts the purchase probability according to the code scanning record of consumers and the purchase conversion rate, providing accurate sales prediction for merchants. In this way, merchants can better predict which products will be quickly sold and which products have low demand, thereby optimizing inventory management and reducing the risk of unsalable and expired products. Consumers are divided into a first consumer set and a second consumer set according to the purchase probability, and different marketing strategies can be adopted. High-intention consumers can be attracted through a price penetration strategy to quickly increase market share. Consumers with low purchase intention can be shown the added value of agricultural products through educational content to increase their interest in products, thereby promoting sales; (3) The present application continuously tracks the code scanning record of consumers and the purchase conversion rate after each code scanning. The system can continuously optimize the marketing strategy, and the personalized marketing strategy of improving the conversion rate enables the needs of each consumer to be accurately matched, thereby improving the sales conversion. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 a system architecture schematic diagram of the overall system in an embodiment proposed by the present application;

[0070] Figure 2 a step flow schematic diagram of the overall method in an embodiment proposed by the present application.

[0071] Label: 1, quality monitoring unit; 2, product traceability unit; 3, code scanning recording unit; 4, purchase probability unit; 5, intention classification unit; 6, product marketing unit. DETAILED DESCRIPTION

[0072] Example one, as shown in the figure, the present application proposes an agricultural product online and offline traceability marketing system, comprising: Figure 1

[0073] Quality monitoring unit 1, quality monitoring unit 1 is used for real-time monitoring of the environmental data of target agricultural products, and determining the quality attenuation coefficient of target agricultural products according to the environmental data;

[0074] Product traceability unit 2, product traceability unit 2 is used for generating traceability code according to the quality attenuation coefficient and the space-time information of target agricultural products;

[0075] Code scanning recording unit 3, code scanning recording unit 3 is used for printing traceability code on the packaging of target agricultural products, deploying traceability code to online page, and obtaining the code scanning record of consumers according to the packaging of target agricultural products and online page;

[0076] Purchase probability unit 4, purchase probability unit 4 is used for predicting the purchase probability of consumers according to the code scanning record of consumers and the purchase conversion rate corresponding to each code scanning;

[0077] Intention classification unit 5, intention classification unit 5 is used for intention classification of all consumers according to the purchase probability, to obtain the first consumer set and the second consumer set;

[0078] Product marketing unit 6, product marketing unit 6 is used for executing price penetration strategy on the first consumer set and executing agricultural product value education strategy on the second consumer set.

[0079] ​In the present application, the quality decay coefficient refers to the rate or degree of quality decline of agricultural products during storage and transportation due to environmental changes (such as temperature, humidity, etc.). The quality decay coefficient helps monitor and assess changes in product quality, thereby providing early warning and ensuring that the product remains in optimal quality at the time of sale. The traceability code is a unique identifier generated through technical means (such as a two-dimensional code, RFID tag, etc.) to track and record detailed information about the product's origin, production, processing, transportation, etc. It allows consumers to understand the detailed background and history of agricultural products by scanning the traceability code. The space-time information includes time and space data of agricultural products, such as production time, picking time, transportation path, storage location, etc. These information are important basis for generating traceability code. The purchase conversion rate refers to the proportion of consumers who actually make purchases among those who scan the traceability code within a certain time period. For example, if 100 consumers scan the code and 10 of them eventually purchase the product, the purchase conversion rate is 10%. The purchase probability refers to the system's prediction of whether a consumer is likely to make a purchase based on their scanning records and behavior (such as scanning frequency, scanning time, etc.). The purchase probability is a predicted value derived from historical data and consumer behavior analysis. The intention classification refers to grouping consumers based on their purchase intentions. Typically, different consumer groups are formed, and through purchase probability, consumers can be divided into high purchase intention group (first consumer group) and low purchase intention group (second consumer group). This classification helps develop more precise marketing strategies. The price penetration strategy refers to attracting consumers to purchase products through lower prices. The agricultural product value education strategy aims to educate consumers to understand the value of agricultural products.

[0080] In embodiment two, the present application proposes an online and offline traceability marketing system for agricultural products. Compared with embodiment one, this embodiment further includes determining the quality decay coefficient of the target agricultural product based on environmental data, which includes:

[0081] Calculating the dynamic spoilage rate of the target agricultural product based on environmental data;

[0082] Determining the quality decay coefficient of the target agricultural product based on the dynamic spoilage rate of the target agricultural product.

[0083] In this embodiment, the dynamic spoilage rate refers to the speed at which the quality of agricultural products declines due to changes in environmental factors such as microbial activity, temperature, humidity, etc. during storage, transportation, or sales. The spoilage rate is not a fixed value, but changes with environmental conditions such as temperature, humidity, oxygen concentration, etc., hence the term dynamic spoilage rate.

[0084] In an optional embodiment, the formula for calculating the dynamic spoilage rate is as follows:

[0085] k = k0 · exp (α (T - Topt )+ β (H - H opt ) 2 + γ sigmoid (L - L th ) ) ;

[0086] wherein k represents a dynamic spoilage rate of the target agricultural product, k0 represents a basic spoilage rate, i.e., a spoilage rate under standard environmental conditions, a, β and γ represent environmental sensitivity coefficients, respectively reflecting the influence of temperature, humidity and light change on the spoilage rate, T, H and L represent temperature data, humidity data and light data in the environmental data, T opt , H opt and L th represent optimal storage conditions of the target agricultural product;

[0087] The calculation formula of the quality attenuation coefficient is as follows:

[0088]

[0089] wherein Q (t) represents the quality attenuation coefficient at the current sampling time t, Q0 represents an initial quality coefficient, t now represents the current time, t harvest represents the time of picking the target agricultural product.

[0090] In an optional embodiment, the traceability code is generated according to the quality attenuation coefficient and the spatio-temporal information of the target agricultural product, comprising:

[0091] obtaining a CRC16 check code of the quality attenuation coefficient, and generating a quality channel code according to the quality attenuation coefficient and the corresponding CRC16 check code;

[0092] obtaining the spatio-temporal information of the target agricultural product to generate a spatio-temporal channel code;

[0093] double-channel synthesizing the quality channel code and the spatio-temporal channel code to obtain the traceability code.

[0094] It should be noted that CRC (Cyclic Redundancy Check) is a common error detection encoding method, usually used to ensure the integrity of data transmission, CRC16 is a variant of CRC, which uses a 16-bit long check code; quality channel encoding is to encode the quality attenuation coefficient and its corresponding CRC16 check code to generate a unique identifier (or code), which can be used to track the quality information of agricultural products. By combining the quality attenuation coefficient with the CRC16 check code, a unique quality channel code can be created to identify the quality status of a specific agricultural product; space-time channel encoding is to encode the space-time information of the target agricultural product to generate a unique identifier, which contains the time and spatial location information of the agricultural product at different stages. Space-time channel encoding can provide detailed traceability information about agricultural products, helping users understand the product's flow path, production background, and historical information.

[0095] In an optional embodiment, the calculation formula of the quality channel code is as follows:

[0096]

[0097] Code Q = CRC16(Q(t))

[0098] The calculation formula of the space-time channel code is as follows:

[0099]

[0100] Code ST = Geohash(x, y, 9) XOR UnixTime(t , ms) now ;

[0101] The calculation formula of the traceability code is as follows:

[0102] Code dual = Code Q || Code ST ;

[0103] Code dual ;

[0104] In an optional embodiment, the purchase probability of a consumer is predicted based on the consumer's scan code record and the purchase conversion rate corresponding to each scan, including:

[0105] determine the decay rate according to the current time and the scan code time in the scan code record of the consumer;

[0106] determine the trust value of the consumer according to the purchase conversion rate corresponding to each scan code of the consumer and the decay rate;

[0107] determine the social influence of the consumer according to the PageRank algorithm;

[0108] predict the purchase probability of the consumer according to the quality decay coefficient, the trust value of the consumer and the social influence of the consumer.

[0109] It should be noted that the trust value of the consumer refers to the degree of trust of the consumer to a certain brand or product; in the social network, the PageRank algorithm can be used to measure the influence of individuals in the social network, by calculating the social connection and influence between consumers, the PageRank algorithm can assign a "social influence score" to each consumer, which reflects the potential influence of the consumer on the behavior of others in the social circle. The consumer with greater social influence is more likely to affect the purchase decision of other consumers.

[0110] In an optional embodiment, the calculation formula of the decay rate is as follows:

[0111] ω(t)=exp(-λ(t now -t scan (i)));

[0112] Wherein, ω(t) represents the decay rate at the current sampling time, t scan(i) represents the scan code time of the i-th scan code in the scan code record, and λ represents the decay factor;

[0113] The calculation formula of the trust value is as follows:

[0114]

[0115] Wherein, Trust(u) represents the trust value of the consumer, n represents the number of scan codes in the scan code record, and Cr(i) represents the purchase conversion rate of the i-th scan code in the scan code record;

[0116] The calculation formula of the social influence is as follows:

[0117]

[0118] Wherein, Social(u) represents the social influence of the consumer, d represents the damping coefficient, N represents the total number of consumers, B u represents the friend set of consumer u, and L(v) represents the outlink number of consumer u, i.e. the number of people he follows.

[0119] The prediction formula of the purchase probability is as follows:

[0120]

[0121] wherein P buy (u) represents the purchase probability of the consumer, a1, a2 and a3 represent preset weight coefficients, and a1+a2+a3=1, and σ represents a Sigmoid function for compressing the purchase probability to the interval [0, 1].

[0122] In an optional embodiment, all consumers are classified according to the purchase probability to obtain a first consumer set and a second consumer set, comprising:

[0123] comparing the purchase probability with a preset first intention threshold and a preset second intention threshold;

[0124] if the purchase probability is greater than the preset first intention threshold, the consumer corresponding to the purchase probability is added to the empty first consumer set;

[0125] if the purchase probability is less than the preset second intention threshold, the consumer corresponding to the purchase probability is added to the empty second consumer set;

[0126] repeating the above operation until all consumers are processed to obtain the first consumer set and the second consumer set, wherein the first consumer set represents a high purchase intention consumer set, and the second consumer set represents a low purchase intention consumer set.

[0127] It should be noted that the first intention threshold is a preset critical value for distinguishing the purchase intention of the consumer to the product, which represents a standard of strong purchase intention, and if the purchase probability of the consumer is greater than the threshold, it is considered that the consumer shows high purchase intention and belongs to one of the target consumer groups. The threshold is preset according to historical data, market analysis, marketing strategy and other factors to help the marketing team identify high-potential customer groups; the second intention threshold is another preset critical value for distinguishing consumers with low purchase intention, and consumers with a purchase probability lower than the threshold are considered to have low purchase intention and are less likely to purchase the product. It also reflects the purchase intention of the consumer, but is usually a low intention.

[0128] In an optional embodiment, a price penetration strategy is performed on the first consumer set, and an agricultural product value education strategy is performed on the second consumer set, comprising:

[0129] determining a dynamic adjustment price according to the average purchase probability of the consumers in the first consumer set, wherein the calculation formula of the dynamic adjustment price is as follows:

[0130]

[0131] Where p represents the dynamically adjusted price. p represents the average purchase probability of consumers in the first consumer set. base Indicates the benchmark price of the target agricultural product;

[0132] A price penetration strategy is implemented on the first consumer set based on dynamically adjusted prices.

[0133] Agricultural product value education information is constructed based on environmental data and agricultural information of target agricultural products, and agricultural product value education strategies are implemented for the second consumer set based on agricultural product value education information.

[0134] It should be noted that agricultural information typically includes various information related to the agricultural production process, such as planting time, weather conditions, fertilization, and harvest time. This information is crucial for understanding the production background, quality assurance, and supply cycle of agricultural products.

[0135] Example 3, as Figure 2 As shown, the present invention proposes an online and offline traceability marketing method for agricultural products, which is applicable to the aforementioned online and offline traceability marketing system for agricultural products, comprising:

[0136] S1. Monitor the environmental data of the target agricultural product in real time, and determine the quality decay coefficient of the target agricultural product based on the environmental data;

[0137] S2. Generate a traceability code based on the quality decay coefficient and the spatiotemporal information of the target agricultural product;

[0138] S3. Print the traceability code on the packaging of the target agricultural product, deploy the traceability code on the online page, and obtain the consumer's scanning record based on the packaging of the target agricultural product and the online page;

[0139] S4. Predict the probability of a consumer's purchase based on the consumer's scanning records and the purchase conversion rate corresponding to each scan.

[0140] S5. Classify all consumers according to their purchase probability to obtain the first consumer set and the second consumer set;

[0141] S6. Implement a price penetration strategy for the first consumer set and an agricultural product value education strategy for the second consumer set.

[0142] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. An online and offline traceability and marketing system for agricultural products, characterized in that, include: A quality monitoring unit (1) is used to monitor the environmental data of the target agricultural product in real time and determine the quality decay coefficient of the target agricultural product based on the environmental data. Product traceability unit (2), the product traceability unit (2) is used to generate traceability code based on the quality decay coefficient and the spatiotemporal information of the target agricultural product; The scanning and recording unit (3) is used to print the traceability code on the packaging of the target agricultural product, deploy the traceability code on the online page, and obtain the consumer's scanning record based on the packaging of the target agricultural product and the online page. Purchase probability unit (4), the purchase probability unit (4) is used to predict the purchase probability of the consumer based on the consumer's scanning record and the purchase conversion rate corresponding to each scanning; Intention classification unit (5), the intention classification unit (5) is used to classify the intentions of all consumers according to the purchase probability to obtain a first consumer set and a second consumer set; Product marketing unit (6), which is used to implement a price penetration strategy on the first consumer set and an agricultural product value education strategy on the second consumer set; Determining the quality degradation coefficient of the target agricultural product based on the environmental data includes: The dynamic spoilage rate of the target agricultural product is calculated based on the environmental data. The quality decay coefficient of the target agricultural product is determined based on the dynamic spoilage rate of the target agricultural product; The formula for calculating the dynamic putrefaction rate is as follows: k=k0·exp(α(T-T opt )+β(H-H opt ) 2 +γSigmoid(L-L th )); Where k represents the dynamic spoilage rate of the target agricultural product, k0 represents the baseline spoilage rate, i.e., the spoilage rate under standard environmental conditions, α, β, and γ represent environmental sensitivity coefficients, reflecting the effects of temperature, humidity, and light changes on the spoilage rate, respectively, and T, H, and L represent the temperature, humidity, and light data in the environmental data, respectively. opt H opt and L th Indicates the optimal storage conditions for the target agricultural product; The formula for calculating the mass attenuation coefficient is as follows: Where Q(t) represents the mass attenuation coefficient at the current sampling time t, Q0 represents the initial mass coefficient, and t now t represents the current time. harvest Indicates the time when the target agricultural product was harvested; Predicting the purchase probability of the consumer based on the consumer's scanning records and the purchase conversion rate corresponding to each scan includes: The decay rate is determined based on the current time and the scanning time in the consumer's scanning record; The consumer's trust value is determined based on the purchase conversion rate corresponding to each scan and the decay rate. The consumer's social influence is determined using the PageRank algorithm; The probability of a consumer's purchase is predicted based on the quality decay coefficient, the consumer's trust value, and the consumer's social influence. The formula for calculating the attenuation rate is as follows: ω(t)=exp(-λ(t now -t scan(i) )); Where ω(t) represents the decay rate at the current sampling time, t scan(i) λ represents the scanning time of the i-th scan in the scan record, and λ represents the attenuation factor. The formula for calculating the trust value is as follows: Where Trust(u) represents the consumer's trust value, n represents the number of times the code is scanned in the scanning record, and Cr(i) represents the purchase conversion rate of the i-th scan in the scanning record. The formula for calculating social influence is as follows: Where Social(u) represents the consumer's social influence, d represents the damping coefficient, N represents the total number of consumers, and B u Let L(v) represent the set of friends pointing to consumer u, and let L(v) represent the number of outgoing links of consumer u, i.e. the number of people it follows. The formula for predicting the purchase probability is as follows: Among them, P buy (u) represents the consumer's purchase probability, a1, a2 and a3 represent preset weighting coefficients, and a1+a2+a3=1. σ represents the Sigmoid function, which is used to compress the purchase probability to the interval [0,1].

2. The online and offline traceability and marketing system for agricultural products according to claim 1, characterized in that, Generate a traceability code based on the quality decay coefficient and the spatiotemporal information of the target agricultural product, including: Obtain the CRC16 checksum of the quality attenuation coefficient, and generate a quality channel code based on the quality attenuation coefficient and the corresponding CRC16 checksum. The spatiotemporal information of the target agricultural product is obtained to generate a spatiotemporal channel code; The quality channel code and the spatiotemporal channel code are combined in a dual-channel manner to obtain the traceability code.

3. The online and offline traceability and marketing system for agricultural products according to claim 2, characterized in that, The formula for calculating the quality channel code is as follows: Among them, Code Q This represents the quality channel code, and CRC16(Q(t)) represents the CRC16 checksum of the quality attenuation coefficient. The calculation formula for the spatiotemporal channel encoding is as follows: Among them, Code ST This indicates the spatiotemporal channel encoding. Geohash(x,y,9) represents the geographic coordinates (x,y) encoded as Geohash with a precision of 9 bits. This represents the XOR operation in bitwise operations, applied to a combination of geographic location and timestamp. UnixTime(t now ,ms) indicates converting the current time in milliseconds to Unix time format; The formula for calculating the traceability code is as follows: Code dual =Code Q ||Code ST ; Among them, Code dual This represents the source code.

4. The online and offline traceability and marketing system for agricultural products according to claim 3, characterized in that, Based on the purchase probability, all consumers are categorized by intention to obtain a first consumer set and a second consumer set, including: The purchase probability is compared with a preset first intention threshold and a preset second intention threshold; If the purchase probability is greater than the preset first intention threshold, then the consumer corresponding to the purchase probability is added to an empty first consumer set. If the purchase probability is less than the preset second intention threshold, then the consumer corresponding to the purchase probability is added to an empty second consumer set. Repeat the above operation until all consumers have been processed to obtain a first consumer set and a second consumer set, where the first consumer set represents the set of consumers with high purchase intention and the second consumer set represents the set of consumers with low purchase intention.

5. The online and offline traceability and marketing system for agricultural products according to claim 4, characterized in that, A price penetration strategy is implemented for the first consumer set, and an agricultural product value education strategy is implemented for the second consumer set, including: The dynamically adjusted price is determined based on the average purchase probability of consumers in the first consumer set, wherein the calculation formula for the dynamically adjusted price is as follows: Where p represents the dynamically adjusted price. p represents the average purchase probability of consumers in the first consumer set. base Indicates the benchmark price of the target agricultural product; A price penetration strategy is applied to the first consumer set based on the dynamically adjusted price. Agricultural product value education information is constructed based on the environmental data and agricultural information of the target agricultural product, and an agricultural product value education strategy is implemented on the second consumer set based on the agricultural product value education information.

6. A method for online and offline traceability marketing of agricultural products, applicable to the online and offline traceability marketing system for agricultural products as described in claim 5, characterized in that, include: Real-time monitoring of environmental data of the target agricultural product, and determination of the quality decay coefficient of the target agricultural product based on the environmental data; A traceability code is generated based on the quality decay coefficient and the spatiotemporal information of the target agricultural product; The traceability code is printed on the packaging of the target agricultural product, and the traceability code is deployed on the online page. Consumers' scanning records are obtained based on the packaging of the target agricultural product and the online page. The probability of a consumer making a purchase is predicted based on the consumer's scanning records and the purchase conversion rate corresponding to each scan. Based on the purchase probability, all consumers are classified according to their intentions to obtain a first consumer set and a second consumer set; A price penetration strategy is applied to the first consumer set, and an agricultural product value education strategy is applied to the second consumer set.

Images