CN117273798A

CN117273798A - Crop supply and demand prediction system, method and device and storage medium

Info

Publication number: CN117273798A
Application number: CN202311026900.9A
Authority: CN
Inventors: 孙彤; 黄桂恒
Original assignee: Brick Suzhou Agricultural Internet Co ltd
Current assignee: Brick Suzhou Agricultural Internet Co ltd
Priority date: 2023-08-15
Filing date: 2023-08-15
Publication date: 2023-12-22

Abstract

The embodiment of the specification provides a crop supply and demand prediction system, a method, a device and a storage medium, wherein the method comprises the following steps: the data processing center acquires and stores the processed data; acquiring a query instruction of a user through a user terminal, and determining query data corresponding to the query instruction based on the query instruction, wherein the query data comprises supply data of crops; and feeding back the query data to the user terminal. The invention provides personalized inquiry scheme recommendation for the user facing the demand end, fully considers individual characteristics of crops at the supply end, and enables the user to control the supply and demand of the crops more accurately.

Description

Crop supply and demand prediction system, method and device and storage medium

Technical Field

The present disclosure relates to the field of agricultural management technologies, and in particular, to a system, a method, an apparatus, and a storage medium for crop supply and demand prediction.

Background

Because the planting type and the planting quantity of the agricultural products are difficult to control, a large amount of agricultural products are always marketed at the same time, and the supply and demand relationship of the agricultural products is unbalanced, so that the market price of the agricultural products is unstable, and the market order is affected under severe conditions. When crops are on the market at the same time, the supply and demand of crops in the market are larger than those of crops, and the selling price is reduced.

Aiming at the problem of optimizing information exchange between a supply end and a demand end, CN113822738A provides a multi-dimensional agricultural product supply and demand bidirectional personalized recommendation method, and the prior art carries out agricultural product supply and demand bidirectional personalized recommendation by constructing a bidirectional personalized recommendation model. The method can be used for recommending the social group with high customer similarity to the agricultural product demand party, so that the agricultural product demand party can conveniently conduct community communication and mutual recommendation; the method can also be used for accurately recommending similar customer groups to the agricultural product suppliers, providing demand preference information for the agricultural product suppliers, guiding the agricultural product suppliers to adjust and make production and sales decisions. This prior art, when oriented to the agricultural product suppliers, merely considers determining the supply type or supply amount of agricultural products according to consumer's consumption needs or preference characteristics, and the like, without considering individual characteristics of the agricultural products at the supply end, such as maturity, and the like. Therefore, the supply and demand conditions of crops can not be accurately estimated, and wrong prediction information is easily given, so that unnecessary loss is caused.

In order to solve the above problems, the present invention provides a crop supply and demand prediction system, method, device and storage medium, which can control the supply condition of crops more accurately and reduce unnecessary loss.

Disclosure of Invention

One of the embodiments of the invention provides a crop supply and demand prediction system, which comprises a data processing center, a data collection platform and a user terminal; the data collection platform is used for collecting processing data and sending the processing data to the data processing center for storage, wherein the processing data at least comprises planting data and culture data; the user terminal is used for receiving a query instruction of a user; receiving query data fed back by the data processing center and displaying the query data to the user; the data processing center is used for: acquiring a query instruction of the user through the user terminal, and determining query data corresponding to the query instruction based on the query instruction, wherein the query data comprises crop supply data; and feeding back the query data to the user terminal.

One of the embodiments of the present invention provides a crop supply and demand prediction method, which is executed by the data processing center, and includes: acquiring and storing processing data; acquiring a query instruction of a user through a user terminal, and determining query data corresponding to the query instruction based on the query instruction, wherein the query data comprises crop supply data; and feeding back the query data to the user terminal.

One embodiment of the invention provides a crop supply and demand prediction device, which comprises a processor and at least one memory, wherein the at least one memory is used for storing computer instructions; the at least one processor is configured to execute at least some of the computer instructions to implement the crop supply-demand prediction method described above.

One of the embodiments of the present specification provides a computer-readable storage medium storing computer instructions that, when read by a computer, the computer performs the crop supply and demand prediction method described above.

Some embodiments of the present description include at least the following benefits: (1) The query data is determined through the query instruction and displayed to the user, so that the query data required by the user can be fed back to the user, for example, when the user queries the supply data of crops, the supply data can be displayed in a targeted manner through the user terminal, the intellectualization of crop supply and demand prediction is realized, and the user can meet various query demands; (2) The recommended query items are determined through the related information of the user, personalized query function use experience can be provided for the user, the query operation of the user is optimized, and the query efficiency of the user is improved; (3) The market data is used for determining the regional demand data, and further, the supply data of crops is determined, so that a user can determine proper time and quantity of crops on the market based on the supply data, balance of supply and demand relations of the crops is maintained, and the supply quantity of the crops is controlled within the regional demand data as far as possible, so that unnecessary loss is avoided; (4) The regional demand data is determined through the regional demand prediction model, a rule can be found from a large amount of related data by utilizing the self-learning capability of the machine learning model, the association relationship between the related data and the regional demand data is obtained, and the accuracy and the efficiency of predicting the regional demand data are improved.

Drawings

The present specification will be further elucidated by way of example embodiments, which will be described in detail by means of the accompanying drawings. The embodiments are not limiting, in which like numerals represent like structures, wherein:

FIG. 1 is an exemplary flow chart of a crop supply and demand prediction method according to some embodiments of the present disclosure;

FIG. 2 is an exemplary diagram illustrating a determination of recommended query terms according to some embodiments of the present description;

FIG. 3 is an exemplary diagram illustrating determining offer data according to some embodiments of the present description.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings that are required to be used in the description of the embodiments will be briefly described below. It is apparent that the drawings in the following description are only some examples or embodiments of the present specification, and it is possible for those of ordinary skill in the art to apply the present specification to other similar situations according to the drawings without inventive effort. Unless otherwise apparent from the context of the language or otherwise specified, like reference numerals in the figures refer to like structures or operations.

It will be appreciated that "system," "apparatus," "unit" and/or "module" as used herein is one method for distinguishing between different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

As used in this specification and the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

A flowchart is used in this specification to describe the operations performed by the system according to embodiments of the present specification. It should be appreciated that the preceding or following operations are not necessarily performed in order precisely. Rather, the steps may be processed in reverse order or simultaneously. Also, other operations may be added to or removed from these processes.

Some embodiments of the present disclosure provide a crop supply and demand prediction system. In some embodiments, a crop supply and demand prediction system may include a data processing center, a data collection platform, and a user terminal.

The data collection platform is a platform for collecting data. In some embodiments, the data collection platform may be used to collect process data and send it to a data processing center for storage. For more description of processing data, see FIG. 1 and its associated description.

A user terminal refers to one or more terminal devices or software used by a user. In some embodiments, the user terminal may be one or more users, including users who directly use the service, and may also include other related users. In some embodiments, a user may query data through a user terminal and the crop supply and demand prediction system may send a reminder to the user through the user terminal. Exemplary user types may be a commodity provider, commodity facilitator, commodity seller, etc.

In some embodiments, the user terminal may be one or any combination of a mobile device, tablet computer, laptop computer, desktop computer, and the like, other devices having input and/or output capabilities.

In some embodiments, the user terminal may be configured to receive a query instruction from a user. In some embodiments, the user terminal may be configured to receive the query data fed back by the data processing center and present it to the user.

In some embodiments, the manner in which the user terminal presents the query data may include at least one of text presentation, image presentation. For more description of query data, see FIG. 1 and its associated description.

The data processing center may be used to process data and/or information obtained from other components of the system. In some embodiments, the data processing center generates data interactions with the data collection platform, the user terminal. For example, the data processing center may receive and store incoming process data for the data collection platform. For another example, the data processing center obtains a query instruction of a user through the user terminal, determines query data corresponding to the query instruction based on the query instruction, and feeds back the query data to the user terminal. Wherein the query data may comprise crop feed data,

in some embodiments, the crop supply and demand prediction system may further include an information acquisition device.

The information acquisition device is a device for acquiring user information. Such as fingerprint acquisition devices, facial data acquisition devices, and the like.

In some embodiments, the user information acquired by the information acquisition device may be sent to a data processing center for storage and/or processing. More description about user information is given in fig. 2 and related description thereof.

In some embodiments, the crop supply and demand prediction system may further comprise a monitoring device.

The monitoring device is a device for capturing image data. In some embodiments, the monitoring device may be used to acquire image data of the crop. For example, the monitoring device may take an image of the crop during its growth phase to obtain image data.

In some embodiments, the image data acquired by the monitoring device may be sent to a data processing center for storage and/or processing. More description of image data is presented in fig. 1 and related description thereof.

In some embodiments of the present disclosure, a method for predicting crop supply and demand is provided, and a method for predicting crop supply and demand is described below with reference to fig. 1.

Fig. 1 is an exemplary flow chart of a crop supply and demand prediction method according to some embodiments of the present disclosure.

In some embodiments, the process 100 may be performed by a data processing center. As shown in fig. 1, the process 100 includes the following steps.

Step 110, the processing data is acquired and stored.

The processing data refers to data processed by the data processing center. In some embodiments, the processing data may include planting data and cultivation data.

The planting data refers to data related to the planting condition of crops. In some embodiments, the planting data includes at least a planting area, a planting time, and a planting number of the crop. In some embodiments, the planting data may also include crop type, planting density, and the like. The above description of the planting data is for illustrative purposes only and is not intended to limit the scope of the present description.

The cultivation data refers to data related to cultivation of crops. In some embodiments, the cultivation data includes at least fertilization data, watering data for the crop. The fertilization data refers to data related to fertilization, such as fertilization frequency, time per fertilization, and fertilizer amount. Watering data refers to data related to watering crops, such as time and water quantity of each watering, etc. The above description of the culture data is for illustrative purposes only and is not intended to limit the scope of the present disclosure.

In some embodiments, the data processing center may obtain process data from the data collection platform. The data collection platform may obtain the process data in a variety of ways. Such as user input, collection of related devices (e.g., monitoring apparatus, etc.), etc.

In some embodiments, the data processing center may read the process data from the storage device. The storage device may be a storage device of the data collection platform, or may be an external storage device that does not belong to the crop supply and demand prediction system, for example, a hard disk, an optical disk, or the like. In some embodiments, the process data may be read through interfaces including, but not limited to, program interfaces, data interfaces, transport interfaces, and the like. In some embodiments, the crop supply and demand prediction system may be operated to automatically extract process data from the interface. In some embodiments, the crop supply and demand prediction system may be invoked by an external other device or system, upon which the process data is passed to the crop supply and demand prediction system. In some embodiments, the process data may also be obtained in any manner known to those skilled in the art, and this description is not limiting.

In some embodiments, an internal database is provided in the data processing center, which can store the processing data.

Step 120, acquiring a query instruction of a user through the user terminal, and determining query data corresponding to the query instruction based on the query instruction.

A query instruction refers to an instruction request for querying relevant data.

The query data refers to data corresponding to a query instruction. One query instruction may correspond to one set of (individual) query data, or may correspond to multiple sets of query data. For example, the query instruction input by the user is "query for the planting time of the plant a", and the corresponding query data is "the planting time of the plant a". For another example, the query input by the user is "query plant a", and the corresponding query data may include "plant a has a planting area XXX", "plant a has a predicted maturity of XXX", "plant a has a predicted yield of XXX", and the like.

In some embodiments, the query data may include crop feed data.

The supply data refers to data related to the supply of the crop. In some embodiments, the feed data may include a feed period and a feed amount of the crop. The supply period refers to the period of the crop on the market, and the supply amount refers to the supply amount of the crop in the supply period.

For more explanation of determining the offer data, see fig. 3 and its associated description.

In some embodiments, the query instruction may include a grow data query instruction. Accordingly, the query data includes the growth data.

The growth data query instruction refers to an instruction related to querying growth data of crops, for example, the growth data query instruction may be "query growth data of plant a".

Growth data refers to data related to crop growth. In some embodiments, the growth data corresponding to the growth data query instruction includes at least one of planting data, estimated maturity, and estimated yield.

The estimated maturity refers to the estimated crop maturity. In some embodiments, the estimated maturity may include one or more.

The estimated yield refers to the yield of the crop predicted at the estimated maturity. In some embodiments, the predicted yield may be a predicted yield sequence of predicted crop yields at different predicted maturity.

In some embodiments, the data processing center may obtain planting data based on the image data. In some embodiments, the data processing center may determine the estimated maturity and the estimated yield based on the vector database.

Image data refers to images that contain crops. The image data may include images of various growth phases of the crop. The image data may be captured by the monitoring device and uploaded to the data processing center.

In some embodiments, the data processing center may analyze the image data to determine crop planting data via an image recognition algorithm. For example, the data processing center may process the image data through a plurality of image recognition algorithms such as an edge detection algorithm, a KNN algorithm, etc., to determine a planting area, a planting number, etc. of the crop. In some embodiments, the data processing center may obtain the planting time from the user terminal and/or the storage device.

In some embodiments, the planting data may be stored in an internal database of the data processing center before each planting, from which the data processing center may read the planting data.

In some embodiments, the data processing center builds a vector database based on historical data. The vector database comprises a reference growth characteristic vector constructed based on historical future weather data, historical planting data and historical culture data, and a reference maturity period and a reference seed yield ratio corresponding to the reference growth characteristic vector.

The historical planting data and the historical cultivating data are related data before a historical time point, and the historical future weather data are weather data after the historical time point.

The reference growth feature vector is a growth feature vector constructed based on historical planting data, historical cultivation data, and historical future weather data at a historical point in time. Weather data includes temperature, precipitation, ph, etc.

The reference maturity and reference seed yield ratio to which the reference growth feature vector corresponds may be determined based on historical data. For example, the time difference between the historical maturation time and the historical time point in the historical data may be determined as the reference maturation period, and the ratio of the number of historical plants in the historical data to the historical yield may be determined as the reference seed yield ratio.

In some embodiments, the data processing center may construct a growth feature vector based on future weather data, planting data, and cultivation data of the planting area, determine an associated growth feature vector that meets a preset matching condition from a vector database, and determine a reference maturity period corresponding to the associated growth feature vector as a predicted maturity period corresponding to the current growth feature vector; and determining the product of the planting quantity in the current planting data and the reference seed yield ratio corresponding to the reference growth characteristic vector as the estimated yield. The data processing center can access related websites of the weather forecast through the network to acquire future weather data. The preset matching condition may refer to a judgment condition for determining the reference growth characteristic vector. In some embodiments, the preset matching condition may include a vector distance less than a distance threshold, a vector distance minimum, and the like.

In some embodiments, the data processing center may determine a health score for the crop based on the image data; a first confidence in the estimated yield is determined based on the health score.

The health score is an indicator for assessing the health of a crop. The higher the health score, the higher the health of the crop.

In some embodiments, the data processing center may determine appearance characteristics of the crop by an image recognition algorithm based on the image data; based on the crop type and the appearance characteristics, determining the health score of the crop by querying a preset scoring table. For more details on image recognition algorithms, see the previous relevant description.

The appearance characteristics may include crop height ranges, leaf characteristics (e.g., leaf number ranges, leaf color, etc.), flower and fruit characteristics (e.g., fig, flower and fruit number ranges, etc.), and the like.

In some embodiments, the preset scoring table may include correspondence of different health scores to different appearance characteristics, different crop types.

In some embodiments, the preset scoring table may also include correspondence of different health scores to different crop types, different appearance characteristic sub-items. Wherein the appearance characteristic sub-item refers to one of the appearance characteristics. The preset scoring table may be determined based on a priori knowledge or historical data.

In some embodiments, the data processing center may query a preset scoring table to determine health sub-scores corresponding to different appearance feature sub-items based on crop type and appearance features, and determine health scores for the crop by statistically analyzing (e.g., adding, weighted summing, etc.) the health sub-scores corresponding to the plurality of appearance feature sub-items. For example, the health score of plant a correlates with leaf number and leaf color, with the leaf number of plant a corresponding to a health sub-score of x; if the health sub-score corresponding to the leaf color of the plant a is y, the health score of the plant a may be a weighted result (x×m+y×n)/2 of the health sub-score corresponding to the leaf number and m, n is a weighted weight, which may be preset based on a priori knowledge or historical data.

The first confidence level refers to a parameter for evaluating the accuracy of the estimated yield.

In some embodiments, the data processing center may determine a first confidence in the estimated yield by a preset rule based on the health score of the crop. For example, the preset rule may be that the higher the health score, the higher the first confidence.

According to the embodiment of the specification, the first confidence coefficient of the estimated yield is determined through the health score of the crops, the influence of the health degree of the crops on the estimated yield can be fully considered, the accuracy of determining the estimated yield is improved, the basis is provided for the subsequent determination of the supply data, and the accuracy of determining the supply data is improved.

According to the embodiment of the specification, the planting data, the estimated maturity and the estimated yield are determined through the image data and the vector database, so that the determination of the growth data is more accurate and quicker, the efficiency of determining the growth data is improved, and the growth data meeting the requirements is provided for users.

And 130, feeding back the query data to the user terminal.

In some embodiments, the data processing center may feed back the query data to the user terminal through various means, such as a network, a wired connection, and the like.

In some embodiments, the data processing center may display the query data in the user terminal through a specific display manner. The specific display means may include various kinds of display means, such as presentation by listing, sliding page change, etc., and is not limited herein.

In some embodiments, when the fed back query data exceeds the preset number, the data processing center may determine a display order of the query data in the user terminal, and display the query data according to the display order. Wherein the preset number may be preset based on historical experience.

In some embodiments, the data processing center may display the query data in a fixed order. The fixed order may be preset.

In some embodiments, the data processing center may determine the personalized display order based on the user preference data.

User preference data refers to data related to the habit of a user when viewing query data. For example, the user preference data may include a total time period and a total number of times the user views various types of query data, and the like. The user preference data may be derived based on historical viewing record statistics of the user at the user terminal.

When a user tends to preferentially view certain types of query data, it may be determined that the personalized display order is preferentially displayed for the types of query data.

In some embodiments, the data processing center may determine the display order of the query data based on a ranking model, which may be a machine learning model, such as a neural network model, a recurrent neural network model, or the like.

In some embodiments, the input of the ranking model may include user identity, user preference data, current time, query data; the output may include a display order of the query data. For more explanation about the identity of the user, see fig. 2 and its associated description.

In some embodiments, the ranking model may be trained based on a first training sample with a first tag. For example, a plurality of first training samples with first labels may be input into the initial ranking model, a loss function is constructed through the first labels and a prediction result of the initial ranking model, the initial ranking model is updated based on iteration of the loss function, and training is completed when the loss function of the initial ranking model meets a preset condition, where the preset condition may be that the loss function converges, the number of iterations reaches a threshold value, and so on.

In some embodiments, the first training sample may include a sample user identity of the sample user, sample user preference data, sample time, and sample query data; the first label is the order in which the sample user clicks and/or views the sample query data; the first training sample and the first tag may be obtained based on historical data. When the display mode of the query data is sliding page changing, the user can determine that the user views the page when the time of the user staying on the page exceeds a time threshold.

According to the embodiments of the present disclosure, by determining the display order of the query data in the user terminal, a personalized display order of the query data can be realized based on user preference or actual situation, which is helpful for improving the user experience. For example, when the current time is near the maturity of the crop, the maturity may be preferentially displayed in the query data fed back. The display sequence of the query data is determined through the ordering model, the self-learning capability of machine learning is utilized, rules can be found in a large amount of historical data, and the display sequence which meets the requirements of users better is predicted.

According to some embodiments of the present disclosure, query data is determined through a query instruction, and the query data is displayed to a user, so that query data required by the user can be fed back to the user, for example, when the user queries supply data of crops, the supply data can be displayed in a targeted manner through a user terminal, thereby realizing the intellectualization of crop supply and demand prediction, and being helpful for meeting various query requirements of the user.

It should be noted that the above description of the process 100 is for illustration and description only, and is not intended to limit the scope of applicability of the present disclosure. Various modifications and changes to the process 100 will be apparent to those skilled in the art in light of the present description. However, such modifications and variations are still within the scope of the present description.

FIG. 2 is an exemplary diagram illustrating a determination of recommended query terms according to some embodiments of the present description.

In some embodiments, a data processing center may be used to determine recommended query terms.

Recommended query terms are query terms recommended to the user. The query term is related to the crop. In some embodiments, the query term may include, but is not limited to, a maintenance guideline for the crop, a planting type, a planting area, a number of plants, a predicted yield, a predicted maturity, a storage period, a unit storage cost, a recommended sales time, a recommended sales price, and the like.

In some embodiments, the data processing center may determine some or all of the query terms as recommended query terms. In some embodiments, the data processing center may determine query terms that meet the current growth stage as recommended query terms based on the growth stages of different types of crops. For example, when a growth stage of a crop is about to mature, the data processing center may determine a planting area, a planting number, an estimated yield, an estimated mature time, a recommended sales time, and a recommended sales price of the crop as recommended query items. In some embodiments, the data processing center may also determine recommended query terms based on the current query terms. For example, the data processing center may determine query terms related to the current query term as recommended query terms. The correlation between query terms may be preset based on a priori knowledge or historical data. For example, the estimated maturity time, the recommended sales time, and the recommended sales price of the same agricultural product may be determined to have a correlation. For another example, the recommended sales time and the recommended sales price of a certain agricultural product and its bid may be determined to have a correlation.

The recommended query term may also be determined in other possible manners, which is not limited in this specification.

Referring to fig. 2, in some embodiments, a data processing center may determine a user identity 220 based on user information 210; determining a historical query record 230 and a recommended query pattern 240 for the user based on the user identity 220; determining a support 260 for the combination of items based on a preset algorithm 250 and the historical query record 230; responsive to the degree of support 260 for the combination of items being greater than the degree of support threshold, determining possible query items in the combination of items as candidate recommended query items 270; the preset recommendation condition 280 is determined based on the recommended query pattern 240, and candidate recommended query items 270 that meet the preset recommendation condition 280 are determined as recommended query items 290.

The user information may include fingerprint information, face information, age, school, etc. of the user. In some embodiments, the user information may be determined based on user input. In some embodiments, the user information may be obtained by an information acquisition device. For example, fingerprint information of a user may be acquired by a fingerprint acquisition device. For another example, the face information of the user may be acquired by a face data acquisition device.

In some embodiments, the user identity 220 may be determined based on the user information 210. In some embodiments, the data processing center may query the user identity library for determining the user identity corresponding to the currently collected user information based on the currently collected user information. The user identity library can be constructed based on data input by the user during initial login or registration. For example, after user information such as user face information is first input, a user identity library is constructed according to user identities input by the user, and when the user logs in by using the face information again, the user identities are determined by comparing the user identity library. The user identity may also be determined in other possible ways, without limitation.

The history query record refers to a query record when a user performs history query. The historical query record may include the category of the historical query term and the number of queries. The query times can be in a statistical range of years, quarters and months.

In some embodiments, the historical query record 230 may be determined by the user identity 220. When a user makes a query, the data processing center can make a record and bind with the identity of the user. In some embodiments, the data processing center may read the historical query record of the corresponding user from the storage device based on the user identity. For more description of storage devices, see FIG. 1 and its associated description.

The recommended query mode refers to a mode in which recommended query items are displayed to the user. For example, the recommended query pattern may include displaying all recommended query items, displaying a portion of recommended query items, not displaying recommended query items, and so forth.

In some embodiments, recommended query patterns may be ranked as recommended. For example, "displaying a part of recommended query items" is classified into a higher recommendation level, and "displaying all recommended query items" and "not displaying recommended query items" are classified into a lower recommendation level. Different recommendation levels indicate different degrees of recommendation for the recommended query pattern. The higher the recommendation level, the higher the recommendation level.

In some embodiments, the recommended query pattern 240 may be determined by the user identity 220. For example, a user identity may indicate that the user is older, has a lower academic, and may determine a recommended query pattern for a lower recommendation level. For another example, the user identity may reflect the proficiency of the user in using the crop supply and demand prediction system, such as when the user identity is an old user, the proficiency is high. For users with high proficiency, the recommended query mode can be adjusted according to the use habit of the users; for users with low proficiency, a recommended query mode with lower recommendation level can be set so as to reduce the operation difficulty of the users.

In some embodiments, the data processing center may calculate the support 260 for the item combinations through a preset algorithm 250 and a historical query record 230.

In some embodiments, the term combination may be a combination that includes the current query term and the possible query terms.

The current query term is the query term currently queried by the user.

The possible query items may be items that the user queries after querying the current query item, or may be items that the user queries while querying the current query item.

The support of the item combination is probability data reflecting that the current query item in the item combination is queried with the possible query item sequentially or simultaneously. For example, a combination of items includes A and B, where A represents the current query item and B represents the possible query item, and the support for this combination of items may be the probability that B becomes the next queried item after A is queried. For another example, the support of the term combination may be the probability that B is the term being queried while A is being queried.

In some embodiments, the preset algorithm 250 may determine, as the support 270 of the item combination, a ratio of the number of historical queries that the first historical query item and the second historical query item are queried sequentially or simultaneously in the historical query record 230 to the total number of historical queries contained in the historical query record. Wherein, the same historical query item as the current query item in the historical query record 230 is a first historical query item, and the same historical query item as the possible query item is a second historical query item.

The support threshold is a support-related threshold. When the support of the item combination is greater than the support threshold, the queried probability indicating the possible queried item is higher.

The support threshold may be preset by human or system. In some embodiments, the data processing center may determine the support threshold based on relevant data for the possible query terms.

In some embodiments, the data processing center may determine the support threshold based on the time interval of the crop from the maturity corresponding to the potential query term.

In some embodiments, the time interval between crop and maturity can be determined from image data captured by a monitoring device. For example, the current growth stage of the crop can be determined from the image data, and the time interval between the crop and the maturity can be determined from the current growth stage query.

In some embodiments, the shorter the interval of time between crops and maturity, the smaller the support threshold, i.e., the closer the crop is to maturity, the greater the probability that the corresponding possible query term for that crop will be queried.

In some embodiments of the present disclosure, by setting the support threshold negatively related to the interval time between the crop corresponding to the possible query item and the maturity period, individual features of the crop can be fully considered, the probability that the crop near maturity is recommended for query is improved, and the recommended query item is provided for the user more accurately.

Candidate recommended query terms are possible query terms that may be determined to be recommended query terms.

In some embodiments, when the support 260 of the combination of items is greater than a support threshold, the data processing center may determine possible query terms in the combination of items as candidate recommended query terms 270.

The preset recommendation condition is a condition for determining a recommended query term. In some embodiments, the preset recommendation condition may be determined by a system or an artificial preset.

In some embodiments, the data processing center may determine the preset recommendation condition 280 based on the recommendation query pattern 240. In some embodiments, the data processing center may determine the preset recommendation condition 280 by querying a preset condition lookup table based on the recommendation level of the recommended query pattern 240. The preset condition comparison table comprises corresponding relations between different recommended query modes and different preset recommended conditions, and the preset condition comparison table can be determined based on priori knowledge or historical data.

In some embodiments, the data processing center may determine the recommended query term 290 based on the candidate recommended query term 270 and the preset recommendation condition 280. In some embodiments, the data processing center may determine the candidate recommended query term 270 as the recommended query term 290 in response to the candidate recommended query term 270 meeting the preset recommendation condition 280.

In some embodiments, the data processing center may send a reminder to the user when the candidate recommended query term 270 meets the preset recommendation condition 280.

The reminder information may be used to provide the user with relevant important information about the crop. The reminder information may include regular reminders and advanced reminders.

Conventional reminders may be used to remind the user to perform normal process operations on the crop. For example, a user may be alerted to harvest a crop by a conventional reminder when the crop is approaching maturity. Conventional reminders may be determined by the system or by human presets.

In some embodiments, advanced alerts may include query missing alerts, query periodic alerts, and the like.

The query omission reminder can remind the user to confirm whether the query items of the daily query are omitted.

The query period refers to a period in which a reminder query is made at a plurality of important growth nodes of the crop. The user can be periodically reminded to inquire about the inquired items according to the inquired period.

In some embodiments, the data processing center may determine the query period based on a growth period of the crop. For example, the interrogation period may be set longer for crops with longer growth periods and shorter for crops with shorter growth periods.

The query cycle alert may alert a user to view the crop at a plurality of important growth nodes of the crop.

In some embodiments of the present description, by sending alert information to a user, the user's query experience may be optimized and important information related to crops may be provided to the user on a regular basis. By setting the query period, the user can be prevented from frequently carrying out query operation, and the query time of the user is reduced while the user is ensured to accurately grasp the growth condition of crops.

In some embodiments of the present disclosure, a recommended query item is determined according to related information of a user, so that personalized query function use experience can be provided for the user, a query operation of the user is optimized, and a query efficiency of the user is improved.

Referring to FIG. 3, in some embodiments, a data processing center may obtain market data 310 from a third party platform; determining regional demand data 330 by the regional demand prediction model 320 based at least on the market data 310; determining at least one supply period 340 for the crop based on the regional demand data 330; based on the at least one supply period 340 and the estimated production 350, a supply amount 360 corresponding to the at least one supply period 340 is determined.

The third party platform refers to other platforms that store market data.

Market data refers to data related to the crop market.

Referring to FIG. 3, in some embodiments, market data 310 includes at least historical sales data 311, historical storage cost data 312.

The historical sales data is historical data relating to sales of crops. The historical sales data may include historical sales prices, historical sales times, and historical sales volumes for the crop.

The historic storage cost data is data related to historic storage costs of crops. The historical storage cost data includes unit storage costs for different crops. The historical storage cost data may be a statistical period of years/seasons/months. The storage conditions for different kinds of crops may be different and the corresponding storage costs may be different.

In some embodiments, the data processing center may obtain historical sales data and historical storage cost data from a third party platform based on the network.

The above description of acquiring historical sales data and historical storage cost data is for illustrative purposes only and is not intended to limit the scope of the present description.

The regional demand data refers to demand data of the sales region for crops within a preset period of time. For example, the regional demand data may include a demand type, a demand amount, a demand time, etc. of crops by the sales region for a preset period of time.

The preset time period is a future time period. The preset time period may be one day or other time periods such as one month. In some embodiments, the region demand data may be composed of sub-demand data of a plurality of time intervals within a preset time period. The time interval may be pre-divided by the system or by human beings.

In some embodiments, the data processing center may determine the zone demand data 330 via the zone demand prediction model 320 based on historical sales prices, historical zone demand data, zone demographic data, seasonal data, temperature data.

In some embodiments, the region demand prediction model 320 may be a machine learning model, such as a deep neural network model, or the like.

In some embodiments, the inputs to the zone demand prediction model 320 may include historical sales prices for historical time periods, historical zone demand data, and zone demographic data, as well as season data, temperature data for preset time periods; the output may include zone demand data for a preset period of time. The season data and the temperature data are season data and temperature data in a preset time period corresponding to the regional demand data.

In some embodiments, the region demand prediction model 320 may be trained based on a second training sample with a second label. The region demand prediction model is trained in a similar manner to the ranking model, and for further description reference is made to FIG. 1 and its associated description.

In some embodiments, the second training samples include sample historical sales prices, sample historical region demand data, and sample region population data for a sample historical time period, and sample season data and sample temperature data for the sample time period; the second label is the regional demand data of the sample time period corresponding to the second training sample. Wherein the sample period is located after the sample history period.

In some embodiments, the second training sample and the second label may be obtained based on historical data, and the second label may be manually labeled.

In some embodiments, the data processing center may also determine the region demand data and its confidence level through a region demand prediction model. The confidence of the region demand data is also referred to as a second confidence. The second confidence level refers to a parameter for evaluating the accuracy of the predicted area demand data. The second confidence may be a confidence sequence including a confidence corresponding to each sub-demand data in the region demand data.

In some embodiments, the regional demand prediction model may include a demand prediction layer and a second confidence determination layer.

In some embodiments, the inputs to the demand prediction layer may include historical sales prices for a historical time period, historical regional demand data and regional population data, as well as seasonal data, temperature data for a preset time period; the output may include zone demand data for a preset period of time. In some embodiments, the demand prediction layer may be a deep neural network model or the like.

In some embodiments, the input to the second confidence determination layer may include area demand data, an amount of data for a historical sales price, an amount of data for a historical area demand data, a sequence of time points of interval, a standard deviation for a historical sales price, a standard deviation for a historical area demand data; the output may include a second confidence sequence of the region demand data. In some embodiments, the second confidence determination layer may be a neural network model or the like.

The interval time point sequence is a sequence formed by a plurality of time intervals of a plurality of time intervals from the current time in a preset time period corresponding to the region demand data.

In some embodiments, the demand prediction layer may be trained based on a second training sample with a second label. For training of the demand prediction layer, reference may be made to the previous relevant description.

In some embodiments, the second confidence determination layer may be trained based on a third training sample with a third label. The training process of the second confidence level determination layer is similar to the training process of the ranking model, and more description can be found in relation to FIG. 1.

In some embodiments, the third training sample may include sample area demand data for a sample period, a data amount of sample historical sales price for a sample historical period, a data amount of sample historical area demand data for a sample historical period, a sequence of sample interval time points, a standard deviation of sample historical sales price, and a standard deviation of sample historical area demand data, and the third training sample may be obtained based on the historical data.

In some embodiments, the third label may be a second confidence level corresponding to the third training sample. The third tag may be used to reflect the accuracy of the predicted area demand data, e.g., the third tag may be a real number between 0 and 1. In some embodiments, the third tag may be obtained based on historical data. For example, the demand prediction layer may be utilized to predict regional demand data for the first historical period. The prediction mode comprises the following steps: and inputting the historical sales price, the historical regional demand data and the regional population data of the second historical time period, and the season data and the temperature data of the first historical time period into a demand prediction layer to obtain the predicted regional demand data of the first historical time period. Wherein the first historical period is located after the second historical period. And comparing the predicted area demand data output by the demand prediction layer with the actual area demand data of the first historical time period to determine a third label. When the predicted area demand data is closer to the actual area demand data, the third label is larger, and the corresponding second confidence is higher. An exemplary third tag is determined in the following manner: third label=1- (|prediction area demand data-actual area demand data|)/actual area demand data.

In some embodiments, the data processing center may treat at least one preset time period in which the region demand data is greater than the demand threshold as at least one supply period. In some embodiments, the data processing center may treat at least one time interval in which the sub-demand data is greater than the demand threshold as at least one supply period.

The demand threshold may be preset by the system or by human beings. In some embodiments, if the total demand for all supply periods greater than the demand threshold is less than the estimated yield, the demand threshold may be reduced until the total demand for all supply periods greater than the demand threshold is no less than the estimated yield to ensure that the crop can be fully sold. For more details on the estimated yield, see fig. 1 and its associated description.

In some embodiments, the data processing center may determine a supply amount corresponding to the at least one supply period based on the at least one supply period, the zone demand data, and the estimated production.

In some embodiments, the data processing center may determine a dispensing ratio for each supply period based on the zone demand data for at least one supply period, and determine the supply amount based on the dispensing ratio and the estimated production. For example, the data processing center may determine a ratio of the required amount corresponding to each supply period to the total required amount, and determine a product of the estimated yield and the ratio corresponding to each supply period as the supply amount corresponding to each supply period. The larger the area demand data of the supply period is, the larger the supply amount allocated to the supply period is.

For more description of the zone demand data, see the previous relevant description.

In some embodiments of the present disclosure, the market data is used to determine the regional demand data, and further determine the supply data of the crops, so that the user can determine the appropriate time and quantity of the crops on the basis of the supply data, which is helpful for maintaining the balance of the supply and demand relationship of the crops, and the supply quantity of the crops is controlled within the regional demand data as far as possible, so as to avoid unnecessary loss. The regional demand data is determined through the regional demand prediction model, a rule can be found from a large amount of related data by utilizing the self-learning capability of the machine learning model, the association relationship between the related data and the regional demand data is obtained, and the accuracy and the efficiency of predicting the regional demand data are improved.

It will be appreciated that the amount of data of the historical sales price, the amount of data of the historical regional demand data may affect the second confidence of the predicted regional demand data. The larger the data volume, the higher the second confidence. For example, the second confidence level predicted using only the past 5 days of historical sales prices and the past 30 days of historical sales prices is different. The time intervals of the preset time periods corresponding to the region demand data from the current time are different, and the second confidence degrees are also different. For example, the second confidence of the area demand data after 5 days of prediction is higher than the second confidence of the area demand data after 10 days of prediction. The standard deviation of the historical sales price and the standard deviation of the historical area demand data may reflect the fluctuation range of the historical sales price data and the historical area demand data, and the smaller the fluctuation range is, the higher the second confidence is. In some embodiments of the present disclosure, by taking the above data as input to the second confidence determining layer, multiple factors affecting the second confidence may be taken into account, and an accurate second confidence may be obtained.

In some embodiments, the data processing center may further determine predicted sales data for a preset period of time based on the sales price prediction model, and determine at least one supply period and its corresponding supply amount based on the predicted sales data.

The predicted sales data refers to data related to sales of crops by a sales area for a preset period of time. For example, the forecast sales data may include sales prices or the like for the sales area to sell the crop for a preset period of time.

In some embodiments, the predicted sales data may be composed of sub-sales data for a plurality of time intervals within a preset time period. For example, the forecast sales data may include sales prices for crops over a plurality of time intervals over a preset period of time.

Further description regarding the preset time period is referred to in the foregoing related description.

In some embodiments, the sales price prediction model may be a machine learning model, such as a deep neural network model, or the like.

In some embodiments, the inputs to the sales price prediction model may include historical sales prices and regional average income for a historical time period, regional demand data for a preset time period; the output may include predicted sales data for a preset period of time.

The sales price prediction model may be trained based on a fourth training sample with a fourth label, similar to the training process of the ranking model, for further description see fig. 1.

In some embodiments, the fourth training sample may include sample historical sales prices and sample area average revenue for a sample historical time period, sample area demand data for a sample time period; the fourth label is the actual sales data of the fourth training sample corresponding to the sample period. Wherein the sample period is located after the sample history period. The fourth training sample and the fourth tag may be obtained based on historical data.

In some embodiments, the data processing center may determine at least one supply period for the crop based on the storage period for the crop, the predicted sales data, the predicted unit storage cost; a candidate supply sequence corresponding to at least one supply period is determined, and a target supply sequence is determined therefrom.

The storage period of a crop refers to the maximum time that the crop can be stored after it has matured and been picked. In some embodiments, the data processing center may take as the storage period of the crop a mean value of historical storage times in the historical storage data of the crop.

The predicted unit storage cost refers to the unit storage cost of the predicted crop to be stored after maturity. In some embodiments, the data processing center may take the average of the historical unit storage costs in the historical storage data of the crop as the predicted unit storage cost of the crop.

In some embodiments, the data processing center may treat at least one preset time period and/or time interval within the storage period for which the profit value per unit is greater than the value threshold as at least one supply period. Wherein the unit profit value is the difference of the unit sales price minus the unit total storage cost, which is the product of the predicted unit storage cost and the storage period.

In some embodiments, the data processing center may determine a candidate supply corresponding to the at least one supply period based on the at least one supply period, the zone demand data, and the estimated production, resulting in at least one candidate supply sequence. Each candidate supply amount sequence comprises at least one candidate supply amount corresponding to the supply period. See above for further explanation of determining the supply amount corresponding to the supply period.

In some embodiments, the data processing center may determine a harvest value of at least one candidate supply sequence based on the unit profit value, and determine a candidate supply sequence having a maximum harvest value as the target supply sequence. Where harvest value = candidate supply volume sequence × unit profit value sequence, which is a sequence of unit profit values for different supply periods. For example, the unit profit value sequence is (y ₁ ,y ₂ ,y ₃ ),y ₁ 、y ₂ 、y ₃ When the unit profit value corresponding to the supply period 1, the supply period 2, and the supply period 3 respectively is determined, a certain candidate supply quantity sequence (x ₁ ,x ₂ ,x ₃ ) When the harvest value of (a) is maximum, the candidate supply sequence (x ₁ ,x ₂ ,x ₃ ) A sequence of supply amounts for the target.

In some embodiments, the data processing center may construct a target equation and its constraints based on the unit profit value and the at least one candidate supply volume sequence, converting the determined target supply volume sequence into a problem that optimizes the solution of the target equation. In some embodiments, the objective equation and its constraints are constructed as follows:

max z＝x ₁ *y ₁ +x ₂ *y ₂ +…+x _n *y _n

wherein y is ₁ 、y ₂ 、……、y _n For a known per-unit profit value, x, for each supply period ₁ 、x ₂ 、……、x _n Candidate supply amounts corresponding to the supply periods in the candidate supply amount sequence; constraint (1) indicates the actual supply amount x in the ith supply period _i Predicted area demand data m of the supply period or less _i The method comprises the steps of carrying out a first treatment on the surface of the Constraint (2) indicates that the total supply amount is equal to or less than the total estimated yield X.

Some embodiments of the present disclosure predict predicted sales data at a future point in time via a sales price prediction model, and determine a supply period and a supply amount based on the predicted sales data, while considering the influence of cost and income while determining the supply data, and facilitate maximizing the predicted income while selling all crops.

In some embodiments, the data processing center may feed back the previously determined provisioning data to the user terminal. In some embodiments, the user terminal may also display a third confidence level of the offer data at the same time as the offer data.

The third confidence level refers to a parameter for evaluating the accuracy of the supplied data.

In some embodiments, the data processing center may determine the third confidence in the offer data in a variety of ways based on the second confidence in the area demand data and the fourth confidence in the forecast sales data. For example, at least one of the mean value of the second confidence of the area demand data and the fourth confidence of the predicted sales data, and the weighted mean value is determined as the third confidence of the supply data.

For more on the second confidence level see the previous relevant description.

The fourth confidence level refers to a parameter for evaluating the accuracy of the predicted sales data. The fourth confidence level may be a confidence level sequence including a confidence level for predicting a respective sales price in the sales data.

In some embodiments, the data processing center may determine a fourth confidence level of the predicted sales data based on the sales price prediction model. In some embodiments, the sales price prediction model may include a price prediction layer and a fourth confidence determination layer.

In some embodiments, the input of the price prediction layer may include historical sales price and regional average income for a historical period of time, regional demand data for a preset period of time; the output may include predicted sales data for a preset period of time. In some embodiments, the price prediction layer may be a deep neural network model or the like.

In some embodiments, the input to the fourth confidence determination layer may include predicted sales data for a preset period of time, a data amount for a historical sales price for a historical period of time, a sequence of time points of interval, a standard deviation for the historical sales price; the output may include a fourth confidence level of the predicted sales data. In some embodiments, the fourth confidence determination layer may be a neural network model or the like. See above for further description of the sequence of spaced time points.

In some embodiments, the price prediction layer may be trained based on a fourth training sample with a fourth label. For training of the price prediction layer, reference may be made to the previous relevant description.

In some embodiments, the fourth confidence determination layer may be trained based on a fifth training sample with a fifth label. The training process of the fourth confidence level determination layer is similar to the training process of the ranking model, and more description can be found in relation to FIG. 1.

In some embodiments, the fifth training sample may include sample forecast sales data for a sample period, a data amount for a sample historical sales price for a sample historical period, a sample interval time point sequence, a standard deviation for a sample historical sales price. The fifth training sample may be obtained based on historical data.

In some embodiments, the fifth label may be a fourth confidence level corresponding to the fifth training sample. The fifth tag may be used to reflect the accuracy of the predicted sales data, e.g., the fifth tag may be a real number between 0 and 1. In some embodiments, the fifth tag may be obtained based on historical data. The fifth tag may be obtained in a similar manner to the third tag, and further description is given above, and will not be repeated here.

In some embodiments, the third confidence of the offer data may also be related to the first confidence of the estimated yield. For example, when the first confidence in the estimated production is less than the confidence threshold, the data processing center may decrease the third confidence in the supplied data accordingly.

For more on the first confidence level see fig. 1 and its associated description.

In some embodiments of the present disclosure, since the supply data is determined based on the estimated yield, by correlating the third confidence of the supply data with the first confidence of the estimated yield, the third confidence may be appropriately reduced and the accuracy of the third confidence may be improved when the accuracy of the estimated yield is low.

According to some embodiments of the present disclosure, based on determining the regional demand data of the future time period, further determining the predicted sales data, and determining the supply data of the crops in combination with the predicted sales data, it is possible to effectively increase sales profits of the agricultural products, help to maintain balance of supply and demand relationships of the crops, and increase profits of users while determining suitable time and volume of the crops on the market. The predicted sales data is determined by the sales price prediction model, a rule can be found from a large amount of related data by utilizing the self-learning capability of the machine learning model, the association relationship between the related data and the predicted sales data is obtained, and the accuracy and the efficiency of determining the predicted sales data are improved.

It will be appreciated that the amount of data for the historical sales price may affect the fourth confidence in the predicted sales data. The greater the amount of data for the historical sales price, the higher the corresponding fourth confidence. For example, the fourth confidence level predicted using only the past 5 days of historical sales prices and the past 30 days of historical sales prices is different. The time interval between the preset time period corresponding to the region demand data and the current time is different, and the fourth confidence coefficient is also different. For example, the fourth confidence of the predicted sales data after 5 days is higher than the fourth confidence of the predicted sales data after 10 days. The standard deviation of the historical sales price may reflect a fluctuation range of the historical sales price, the smaller the fluctuation range, the higher the fourth confidence level. In some embodiments of the present disclosure, by taking the above data as input to the fourth confidence determining layer, multiple factors affecting the fourth confidence may be taken into consideration, and an accurate fourth confidence may be obtained.

According to the method and the device for controlling the crop feeding condition, the second confidence coefficient of the feeding data is displayed simultaneously when the feeding data is displayed to the user, the user can determine the actual feeding period and the actual feeding quantity based on the accuracy of the feeding data, and the user can control the crop feeding condition more accurately.

While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements, and adaptations to the present disclosure may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within this specification, and therefore, such modifications, improvements, and modifications are intended to be included within the spirit and scope of the exemplary embodiments of the present invention.

The embodiments in this specification are for illustration and description only and do not limit the scope of applicability of the specification. Various modifications and changes may be made by those skilled in the art in light of the present description while remaining within the scope of the present description.

Certain features, structures, or characteristics of one or more embodiments of the present description may be combined as suitable.

The embodiments described in this specification are merely illustrative of the principles of the embodiments of this specification. Other variations are possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present specification may be considered as consistent with the teachings of the present specification. Accordingly, the embodiments of the present specification are not limited to only the embodiments explicitly described and depicted in the present specification.

Claims

1. A crop supply and demand prediction system, which is characterized by comprising a data processing center, a data collection platform and a user terminal;

the data collection platform is used for collecting processing data and sending the processing data to the data processing center for storage, wherein the processing data at least comprises planting data and culture data;

the user terminal is used for receiving the query instruction of the user, receiving the query data fed back by the data processing center and displaying the query data to the user;

the data processing center is used for:

acquiring a query instruction of the user through the user terminal, and determining query data corresponding to the query instruction based on the query instruction, wherein the query data comprises crop supply data;

and feeding back the query data to the user terminal.

2. The system of claim 1, further comprising an information acquisition device for acquiring user information; the data processing center is further configured to:

determining a user identity based on the user information;

determining a historical query record and a recommended query mode of the user based on the user identity;

determining the support degree of an item combination based on a preset algorithm and the historical query record, wherein the item combination comprises a current query item and a possible query item;

determining the possible query terms in the term combination as candidate recommended query terms in response to the degree of support of the term combination being greater than a degree of support threshold;

and determining a preset recommendation condition based on the recommendation query mode, and determining the candidate recommendation query items meeting the preset recommendation condition as recommendation query items.

3. The system of claim 1, further comprising a monitoring device for capturing image data; the query instruction comprises a growth data query instruction, growth data corresponding to the growth data query instruction at least comprises at least one of planting data, estimated maturity and estimated yield, and the data processing center is further used for:

Determining the planting data based on the image data;

and determining the estimated maturity and the estimated yield based on a vector database, wherein the vector database is constructed based on historical data, and the vector database comprises reference growth feature vectors constructed based on historical future weather data, historical planting data and historical culture data, and reference maturity and reference seed yield ratios corresponding to the reference growth feature vectors.

4. The system of claim 1, wherein the data processing center is further configured to:

obtaining market data from a third party platform, wherein the market data at least comprises historical sales data and historical storage cost data;

determining regional demand data through a regional demand prediction model based at least on the market data, the regional demand prediction model being a machine learning model;

determining at least one supply period for the crop based on the regional demand data;

and determining a supply amount corresponding to the at least one supply period based on the at least one supply period, the regional demand data and the estimated yield.

5. A crop supply and demand prediction method, the method being performed by the data processing center and comprising:

Acquiring and storing processing data;

acquiring a query instruction of a user through a user terminal, and determining query data corresponding to the query instruction based on the query instruction, wherein the query data comprises crop supply data;

and feeding back the query data to the user terminal.

6. The method of claim 5, wherein the method further comprises:

determining a user identity based on the user information;

7. The method of claim 5, wherein the query instruction comprises a growth data query instruction, the growth data corresponding to the growth data query instruction comprising at least one of the planting data, a predicted maturity, and a predicted yield; the determining the query data corresponding to the query instruction further includes:

Determining the planting data based on the image data shot by the monitoring device;

8. The method of claim 5, wherein the determining the query data corresponding to the query instruction further comprises:

9. A crop supply and demand prediction apparatus, the apparatus comprising at least one processor and at least one memory;

The at least one memory is configured to store computer instructions;

the at least one processor is configured to execute at least some of the computer instructions to implement the crop supply-demand prediction method of any one of claims 5-8.

10. A computer-readable storage medium storing computer instructions that, when read by a computer in the storage medium, the computer performs the crop supply-demand prediction method according to any one of claims 5 to 8.