Food recommendation method and system based on spatio-temporal evolution knowledge association network

Abstract

The invention discloses a food recommendation method and system based on a spatio-temporal evolution knowledge association network, and relates to the technical field of recommendation. The method comprises the steps of collecting exclusive data of different crowds, generating a dynamic semantic coding cluster through cross-domain dynamic feature fusion, obtaining food data, generating a food demand association deduction rule set in combination with the dynamic semantic coding cluster, and constructing a spatio-temporal evolution knowledge association network according to the food demand association deduction rule set and preset time sequence situation parameters. And obtaining user data to be recommended, generating an adaptive recommendation strategy based on the spatio-temporal evolution knowledge association network, and performing dynamic reconstruction optimization on the spatio-temporal evolution knowledge association network through a bidirectional data iteration mechanism. According to the method, accurate matching of physiology-nutrition-scene can be realized, a recommendation strategy has dynamic adaptation capabilities of seasons, regions and the like, a self-adaptive recommendation strategy is generated in combination with real-time user data, recommendation efficiency and accuracy are improved, and autonomous evolution of the system is realized through a bidirectional data iteration mechanism.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of recommendation, in particular to a food recommendation method and system based on a spatio-temporal evolution knowledge association network. BACKGROUND

[0002] In the current intelligent recommendation field, there are technical bottlenecks such as insufficient depth of multi-dimensional data semantic fusion, weak real-time response ability of user dynamic needs, lack of spatio-temporal context association deduction mechanism, and low efficiency of system autonomous evolution. Traditional solutions cannot achieve semantic unified representation of cross-domain data, lack a cooperative processing mechanism for user state changes and dynamic characteristics of spatio-temporal environment, and do not build a knowledge association network with time evolution characteristics, resulting in the inability of the recommendation strategy to dynamically adjust to user needs and scene conditions, and problems such as limited cross-scene generalization ability and difficulty of recommendation logic to adapt to complex application environments. SUMMARY

[0003] The present application provides a food recommendation method based on a spatio-temporal evolution knowledge association network, comprising: Step S1, collecting exclusive data of different groups of people, and generating dynamic semantic coding clusters through cross-domain dynamic feature fusion; Step S2, obtaining food data, and generating a food demand association deduction rule set in combination with the dynamic semantic coding clusters; Step S3, constructing a spatio-temporal evolution knowledge association network according to the food demand association deduction rule set and preset time sequence context parameters; Step S4, obtaining real-time user data to be recommended, and generating an adaptive recommendation strategy based on the spatio-temporal evolution knowledge association network; Step S5, dynamically reconstructing and optimizing the spatio-temporal evolution knowledge association network through a bidirectional data iteration mechanism.

[0004] The food recommendation method based on a spatio-temporal evolution knowledge association network as described above, wherein collecting exclusive data of different groups of people and generating dynamic semantic coding clusters through cross-domain dynamic feature fusion comprises the following sub-steps: Step S11, collecting exclusive data of different groups of people; Step S12, performing dynamic clustering through cross-domain dynamic feature fusion to generate dynamic semantic coding clusters.

[0005] The food recommendation method based on a spatio-temporal evolution knowledge association network as described above, wherein obtaining food data and generating a food demand association deduction rule set in combination with the dynamic semantic coding clusters comprises the following sub-steps: Step S21, obtaining food data, extracting food multi-dimensional attributes, and constructing a food attribute knowledge base; Step S22, establishing food demand multi-dimensional association rules based on the food attribute knowledge base and the dynamic semantic coding clusters; Step S23, based on the food nutrition activity index and the historical dietary feedback of different groups of people, the priority of the food demand multi-dimensional association rule is adjusted in real time to form a food demand association deduction rule set.

[0006] The food recommendation method based on the spatio-temporal evolution knowledge association network as described above, wherein the spatio-temporal evolution knowledge association network is constructed according to the food demand association deduction rule set and the preset time sequence situation parameter, and includes the following sub-steps: Step S31, constructing a time sequence situation parameter according to a time sequence situation feature through a dynamic time warping algorithm; Step S32, constructing a spatio-temporal evolution knowledge association network based on the food demand association deduction rule set and the time sequence situation parameter.

[0007] The food recommendation method based on the spatio-temporal evolution knowledge association network as described above, wherein the user data to be recommended is acquired in real time, and an adaptive recommendation strategy is generated based on the spatio-temporal evolution knowledge association network, and includes the following sub-steps: Step S41, acquiring and analyzing the user data to be recommended in real time to generate an interest index and a nutrition required index of the user; Step S42, acquiring an optimal food recommendation strategy according to the interest index, the nutrition required index of the user and the spatio-temporal evolution knowledge association network.

[0008] The application also provides a food recommendation system based on a spatio-temporal evolution knowledge association network, comprising: A dynamic semantic coding cluster generation module acquires exclusive data of different groups of people, and generates dynamic semantic coding clusters through cross-domain dynamic feature fusion; A food demand association deduction rule set generation module acquires food data, and generates a food demand association deduction rule set in combination with the dynamic semantic coding clusters; A spatio-temporal evolution knowledge association network module constructs a spatio-temporal evolution knowledge association network according to the food demand association deduction rule set and a preset time sequence situation parameter; An adaptive recommendation strategy generation module acquires user data to be recommended in real time, and generates an adaptive recommendation strategy based on the spatio-temporal evolution knowledge association network; A knowledge association network optimization module dynamically reconstructs and optimizes the spatio-temporal evolution knowledge association network through a bidirectional data iteration mechanism.

[0009] The food recommendation system based on the spatio-temporal evolution knowledge association network as described above, wherein the dynamic semantic coding cluster generation module specifically includes: An exclusive data acquisition sub-module acquires exclusive data of different groups of people; A dynamic clustering sub-module performs dynamic clustering through cross-domain dynamic feature fusion to generate dynamic semantic coding clusters.

[0010] The food recommendation system based on the spatio-temporal evolution knowledge association network as described above, wherein the food demand association deduction rule set generation module specifically comprises: The food attribute knowledge base construction submodule acquires food data, extracts food multi-dimensional attributes, and constructs a food attribute knowledge base; The association rule establishment submodule establishes food demand multi-dimensional association rules based on the food attribute knowledge base and dynamic semantic coding clusters; The priority dynamic adjustment submodule adjusts the priority of the food demand multi-dimensional association rules based on the food nutrition activity index and historical dietary feedback of different populations in real time, and forms the food demand association deduction rule set.

[0011] The food recommendation system based on the spatio-temporal evolution knowledge association network as described above, wherein the spatio-temporal evolution knowledge association network module specifically comprises: The time sequence context parameter construction submodule constructs time sequence context parameters through a dynamic time warping algorithm according to time sequence context characteristics; The knowledge association network construction submodule constructs a spatio-temporal evolution knowledge association network based on the food demand association deduction rule set and the time sequence context parameters.

[0012] The food recommendation system based on the spatio-temporal evolution knowledge association network as described above, wherein the adaptive recommendation strategy generation module specifically comprises: The user data acquisition and analysis submodule acquires and analyzes user data to be recommended in real time, generates interest indicators and nutrition required indicators of the user; The optimal food recommendation strategy acquisition submodule acquires an optimal food recommendation strategy in the spatio-temporal evolution knowledge association network according to the interest indicators, the nutrition required indicators of the user, and the spatio-temporal evolution knowledge association network.

[0013] The present application has the following beneficial effects: The present application can realize the precise matching of "physiology-nutrition-scene", so that the recommendation strategy has dynamic adaptation capabilities such as season and region; the adaptive recommendation strategy is generated in combination with real-time user data, which significantly improves the recommendation efficiency and accuracy; and the system self-evolution is realized through a bidirectional data iteration mechanism. The present application provides an intelligent recommendation solution with dynamic evolution capability for the fields of intelligent medical treatment and precise nutrition. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art according to these drawings.

[0015] Figure 1is a food recommendation method based on a spatio-temporal evolution knowledge association network provided by Embodiment One of the present application; Figure 2 is a food recommendation system schematic diagram based on a spatio-temporal evolution knowledge association network provided by Embodiment Two of the present application. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0017] Embodiment One As shown in the figure, Embodiment One of the present application provides a food recommendation method based on a spatio-temporal evolution knowledge association network, which comprises the following steps: Figure 1 Step S1, collecting exclusive data of different crowds, and generating dynamic semantic coding clusters through cross-domain dynamic feature fusion; Further, collecting exclusive data of different crowds and generating dynamic semantic coding clusters through cross-domain dynamic feature fusion comprises the following sub-steps: Step S11, collecting exclusive data of different crowds; Specifically, when collecting exclusive data of different crowds, first, the characteristics of the target crowd are determined, such as age, region, occupation, health status, etc. The samples are positioned through stratified sampling or snowball sampling, and the data are collected through multiple channels such as online questionnaire platforms, wearable devices, and offline research tools, covering dimensions such as basic attributes, physiological indicators, and psychological feedback of different crowds. Before data collection, the purpose, scope, and privacy protection measures are fully explained to the participating crowds, and explicit consent is obtained through electronic or paper authorization, ensuring that the data usage complies with regulations. After data collection, the data are cleaned and desensitized (such as anonymization processing and encryption of key information), stored according to the crowd characteristic labels, and finally structured experimental exclusive data of different crowds are generated. At the same time, a data usage log and permission management mechanism are established to ensure safe and compliant application of data in scientific research and product optimization scenarios. Step S12, dynamic clustering through cross-domain dynamic feature fusion to generate dynamic semantic coding clusters;

[0018] Step S12, dynamic clustering through cross-domain dynamic feature fusion to generate dynamic semantic coding clusters; Specifically, the attention mechanism is used to perform cross-domain weighted mapping on the exclusive data of different groups to obtain the fusion features of the exclusive data, a preset auto-encoding generation model is used to convert the fusion features into semantic feature vectors of a unified dimension, a time sequence graph convolution network is used to obtain the time correlation features in the exclusive data, the semantic feature vectors are dynamically clustered based on the time correlation features by using a density peak clustering algorithm, an encoding cluster of the "physiological state-nutritional demand" semantic label is formed, and the parameters of the encoding cluster are updated in real time by using an incremental learning algorithm.

[0019] Step S2, obtaining food data, combining dynamic semantic encoding cluster to generate food demand association deduction rule set; Further, obtaining food data, combining dynamic semantic encoding cluster to generate food demand association deduction rule set includes the following sub-steps: Step S21, obtaining food data, extracting food multi-dimensional attributes, and constructing a food attribute knowledge base; Specifically, food data is obtained through supply chain Internet of Things, food industry database and other channels, food freshness index, nutritional ingredients, processing process parameters and other food multi-dimensional attribute data are extracted from the food data, and a food attribute knowledge base is constructed.

[0020] Step S22, establishing food demand multi-dimensional association rules based on the food attribute knowledge base and the dynamic semantic encoding cluster; Specifically, the cross-space mapping relationship between "food attributes-group demand" is established by the attention reasoning engine according to the food attribute knowledge base and the dynamic semantic encoding cluster, and the food demand multi-dimensional association rules including the association rules of nutritional matching degree and metabolic adaptability are generated.

[0021] Step S23, based on the food nutritional activity index and the historical diet feedback of different groups, the priority of the food demand multi-dimensional association rules is adjusted in real time to form a food demand association deduction rule set; Specifically, the calculation parameters of the association matching value are obtained based on the nutritional activity index of different foods and the historical diet feedback of different groups, and each parameter is processed without dimension, and the time feedback strong dynamic association matching formula The association matching value of different foods to different groups at different times is calculated in real time, wherein, is the association matching value of food and group at time is the food index, the value range of , is the number of food indexes, and one value represents one food, is the group index, the value range of , The number of crowd indexes, one value represents one crowd. for Moment Food and the crowd The associated matching value of It is the food population nutrition matching factor, For the crowd Sensitivity to nutritional activity, for Moment Food The nutritional activity value of For food The initial maximum nutritional activity value, For food For the crowd The nutritional value weight, is the nutrition matching factor adjustment index, is the historical diet feedback factor, for Moment Crowd Food Historical consumption feedback ratings, is the score mapping function, which maps the score to The weight of is the historical feedback factor adjustment index, is the reinforcement learning factor, For the crowd The reward learning rate parameter for the historical association matching value, To calculate the reward time length of the historical correlation matching value, The value range is , The reward time is The reward time decay coefficient of the historical associated matching value at time , For the historical moment Time crowd For edible food The priority of the multi-dimensional association rules of food demand is adjusted in real time based on the association matching values ​​of different foods for different groups of people at different times, forming a self-optimized food demand association deduction rule set.

[0022] Step S3: constructing a spatiotemporal evolution knowledge association network based on the food demand association deduction rule set and preset temporal context parameters; Furthermore, based on the food demand association deduction rule set and the preset temporal context parameters, the construction of the spatiotemporal evolution knowledge association network includes the following sub-steps: Step S31: constructing temporal context parameters using a dynamic time warping algorithm according to temporal context characteristics; Specifically, according to temporal context characteristics such as seasonal solar term cycles, user life cycles, food production cycles, and geographical locations, the discrete temporal context characteristics are converted into continuous temporal context parameters through a dynamic time warping algorithm.

[0023] Step S32: constructing a spatiotemporal evolution knowledge association network based on the food demand association deduction rule set and the temporal context parameters; Specifically, based on the food demand association deduction rule set and time series scenario parameters, the calculation parameters for food supply and demand adaptation are obtained and dimensionless processing is performed. Calculate the recommended weights of different foods for different groups of people at different times and places, where: For the moment Place Food For the crowd The recommended weight, for Moment Food and the crowd The associated matching value of is the time series attenuation coefficient, is the seasonal bias weight, For the moment The current month, For food The best season month, is the meal time bias weight, For the moment The current time point, For food The best time to eat For the moment Place Food The supply, For the location Food The weight of historical demand data, For the location Food Historical demand value, For the location Food The demand forecast value, is the minimum value, is the supply and demand elasticity index, is the number of cultural characteristics, The value range is , For the crowd On Location The first A self-optimizing spatiotemporal evolution knowledge association network is constructed based on the food demand association deduction rule set, temporal context parameters, and each recommended weight.

[0024] Step S4: acquiring the user data to be recommended in real time, and generating an adaptive recommendation strategy based on the spatiotemporal evolution knowledge association network; Furthermore, obtaining the user data to be recommended in real time and generating an adaptive recommendation strategy based on the spatiotemporal evolution knowledge association network includes the following sub-steps: Step S41: acquiring and analyzing the user data to be recommended in real time to generate the user's interest index and nutritional requirement index; Specifically, the historical interest data, current physiological data, situational data and other data of the user to be recommended are obtained in real time, analyzed, and the interest index of the user's favorite food at this moment and place and the nutritional requirement index suitable for the current physiological state are obtained.

[0025] Step S42: Obtaining the optimal food recommendation strategy based on the user's interest index, nutritional requirement index, and spatiotemporal evolution knowledge association network; Specifically, according to the nutritional indicators required, the association matching search is performed in the spatiotemporal evolution knowledge association network to obtain multiple matching targets that meet the nutritional indicators required by the user. Based on the user's interest indicators, the calculation parameters of the applicable recommendation degree are obtained and dimensionless processing is performed. Through the applicable recommendation formula Calculate the applicable recommendation degree of each matching target, where For users No. The applicable recommendation degree of matching targets, The value range is , For users The number of matching targets, For users The number of features of interest, For the The weight of the feature, is the similarity between the feature of interest and the matching target feature, For users No. Interest characteristics, To match the target No. Features, For applicable recommendation factors, matching targets greater than a preset applicable recommendation threshold are selected to generate an adaptive recommendation strategy and recommend it to the user.

[0026] Step S5: Dynamically reconstruct and optimize the spatiotemporal evolution knowledge association network through a bidirectional data iteration mechanism; Specifically, the bidirectional data iteration mechanism includes inner-loop real-time fine-tuning and outer-loop global reconstruction. The inner-loop real-time fine-tuning is to construct a feedback data stream by collecting user interaction data in real time, and to update the recommendable weights of the food and its related foods corresponding to the spatiotemporal evolution knowledge association network in real time; the outer-loop global reconstruction is to reconstruct the spatiotemporal evolution knowledge association network by regularly collecting new foods and their data as well as change data of existing foods.

[0027] Example 2 like Figure 2 As shown, the second embodiment of the present application provides a food recommendation system based on a spatiotemporal evolution knowledge association network, including: The dynamic semantic coding cluster generation module 21 collects exclusive data of different groups of people and generates dynamic semantic coding clusters through cross-domain dynamic feature fusion; Furthermore, the dynamic semantic coding cluster generation module 21 includes the following submodules: Exclusive data collection submodule 211 collects exclusive data of different groups of people; Dynamic clustering submodule 212 performs dynamic clustering by cross-domain dynamic feature fusion to generate dynamic semantic coding clusters; A food demand association deduction rule set generation module 22 acquires food data and generates a food demand association deduction rule set in combination with a dynamic semantic coding cluster; Furthermore, the food demand association deduction rule set generation module 22 includes the following submodules: The food attribute knowledge base construction submodule 221 acquires food data, extracts multidimensional attributes of food, and constructs a food attribute knowledge base; The association rule establishment submodule 222 establishes multidimensional association rules for food demand based on the food attribute knowledge base and the dynamic semantic coding cluster; The priority dynamic adjustment submodule 223 adjusts the priority of the multi-dimensional association rules of food demand in real time based on the food nutritional activity index and historical dietary feedback of different groups of people to form a food demand association deduction rule set; The spatiotemporal evolution knowledge association network module 23 constructs a spatiotemporal evolution knowledge association network based on a food demand association deduction rule set and preset temporal context parameters; Furthermore, the spatiotemporal evolution knowledge association network module 23 includes the following submodules: The temporal context parameter construction submodule 231 constructs temporal context parameters according to temporal context features through a dynamic time warping algorithm; The knowledge association network construction submodule 232 constructs a spatiotemporal evolution knowledge association network based on the food demand association deduction rule set and the temporal context parameters; The adaptive recommendation strategy generation module 24 obtains the user data to be recommended in real time, and generates an adaptive recommendation strategy based on the spatio-temporal evolution knowledge association network; Further, the adaptive recommendation strategy generation module 24 comprises the following sub-modules: The user data acquisition and analysis sub-module 241 obtains the user data to be recommended in real time, and performs analysis to generate an interest index and a nutrition requirement index of the user; The optimal food recommendation strategy acquisition sub-module 242 acquires an optimal food recommendation strategy from the spatio-temporal evolution knowledge association network according to the interest index, the nutrition requirement index and the spatio-temporal evolution knowledge association network of the user; The knowledge association network optimization module 25 performs dynamic reconstruction and optimization on the spatio-temporal evolution knowledge association network through a bidirectional data iteration mechanism; Corresponding to the above-mentioned embodiments, the embodiments of the present application provide a computer storage medium, comprising at least one memory and at least one processor. The memory is used for storing one or more program instructions. The processor is used for running one or more program instructions to execute a food recommendation method based on a spatio-temporal evolution knowledge association network.

[0028] Corresponding to the above-mentioned embodiments, the embodiments of the present application provide a computer readable storage medium, and the computer storage medium comprises one or more program instructions, and the one or more program instructions are used for a food recommendation method based on a spatio-temporal evolution knowledge association network.

[0029] The embodiments disclosed in the present application provide a computer readable storage medium, and the computer readable storage medium stores computer program instructions, and when the computer program instructions run on a computer, the computer executes the above-mentioned food recommendation method based on a spatio-temporal evolution knowledge association network.

[0030] In the embodiments of the present application, the processor can be an integrated circuit chip with a signal processing capability. The processor can be a general purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0031] The disclosed methods, steps, and logic block diagrams in the embodiments of the present application can be implemented or performed with a general- purpose processor, a special purpose processor, or any other processor. The steps of the methods disclosed in the embodiments of the present application can be directly embodied to a hardware code, a processor code, or a software code for execution by a processor. The software code can reside in the storage media such as the random access memory (RAM), the flash memory, the read only memory (ROM), the programmable read only memory (PROM), the electrically programmable read only memory (EPROM), the electrically erasable programmable read only memory (EEPROM), the compact disk (CD), the digital versatile disk (DVD), the Blu-ray disk, the hard disk drive (HDD), or any other storage medium. The processor reads information in the storage medium, and performs the steps of the methods in combination with hardware of the processor.

[0032] The storage medium can be the memory, for example, the volatile memory or the non-volatile memory, or can include both the volatile and non-volatile memory.

[0033] The non-volatile memory can be the read only memory (ROM), the programmable read only memory (PROM), the erasable programmable read only memory (EPROM), the electrically EPROM (EEPROM), or the flash memory.

[0034] The volatile memory can be the random access memory (RAM) used as the external cache. By way of example, and not limitation, many forms of RAM are available, for example, the static random access memory (SRAM), the dynamic random access memory (DRAM), the synchronous dynamic random access memory (SDRAM), the double data rate SDRAM (DDR SDRAM), the enhanced SDRAM (ESDRAM), the Synchlink DRAM (SLDRAM), and the direct Rambus RAM (DRRAM).

[0035] The storage medium described in the embodiments of the present application is intended to include, but not limited to, these and any other suitable types of memory.

[0036] Those skilled in the art should be aware that, in one or more examples described above, functions described by the present application can be implemented in combination of hardware and software. When the software is applied, the corresponding functions can be stored in a computer readable medium or transmitted as one or more instructions or codes on a computer readable medium. The computer readable medium includes computer storage medium and communication medium, wherein the communication medium includes any medium that facilitates transfer of a computer program from one place to another. The storage medium can be any available medium accessible by a general or special purpose computer.

[0037] The above detailed description further describes the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made on the basis of the technical solutions of the present application should be included in the protection scope of the present application.

Claims

1. A food recommendation method based on spatiotemporal evolution knowledge association network, characterized in that: include: Step S1: Collect exclusive data of different groups of people and generate dynamic semantic coding clusters through cross-domain dynamic feature fusion; Step S2: Acquire food data and generate a food demand association deduction rule set based on dynamic semantic coding clusters; Step S3: constructing a spatiotemporal evolution knowledge association network based on the food demand association deduction rule set and preset temporal context parameters; Step S4: acquiring the user data to be recommended in real time, and generating an adaptive recommendation strategy based on the spatiotemporal evolution knowledge association network; Step S5: Dynamically reconstruct and optimize the spatiotemporal evolution knowledge association network through a bidirectional data iteration mechanism.

2. The food recommendation method based on spatiotemporal evolution knowledge association network according to claim 1, characterized in that: Collecting exclusive data from different groups of people and generating dynamic semantic coding clusters through cross-domain dynamic feature fusion includes the following sub-steps: Step S11: collecting exclusive data of different groups of people; Step S12: Perform dynamic clustering through cross-domain dynamic feature fusion to generate dynamic semantic coding clusters.

3. The food recommendation method based on spatiotemporal evolution knowledge association network according to claim 1, characterized in that: Obtaining food data and generating a food demand association deduction rule set based on dynamic semantic coding clusters includes the following sub-steps: Step S21: Acquire food data, extract food multidimensional attributes, and build a food attribute knowledge base; Step S22: establishing multidimensional association rules for food demand based on the food attribute knowledge base and the dynamic semantic coding cluster; Step S23: Based on the food nutritional activity index and historical dietary feedback of different groups of people, the priority of the multidimensional association rules of food demand is adjusted in real time to form a food demand association deduction rule set.

4. The food recommendation method based on spatiotemporal evolution knowledge association network according to claim 1, characterized in that: Based on the food demand association deduction rule set and the preset temporal scenario parameters, the construction of the spatiotemporal evolution knowledge association network includes the following sub-steps: Step S31: constructing temporal context parameters using a dynamic time warping algorithm according to temporal context characteristics; Step S32: constructing a spatiotemporal evolution knowledge association network based on the food demand association deduction rule set and temporal context parameters.

5. The food recommendation method based on spatiotemporal evolution knowledge association network according to claim 1, characterized in that: Acquiring the user data to be recommended in real time and generating an adaptive recommendation strategy based on the spatiotemporal evolution knowledge association network includes the following sub-steps: Step S41: acquiring and analyzing the user data to be recommended in real time to generate the user's interest index and nutritional requirement index; Step S42: Obtain the optimal food recommendation strategy based on the user's interest index, nutritional requirement index and spatiotemporal evolution knowledge association network.

6. A food recommendation system based on spatiotemporal evolution knowledge association network, characterized by: include: The dynamic semantic coding cluster generation module collects exclusive data of different groups of people and generates dynamic semantic coding clusters through cross-domain dynamic feature fusion; The food demand association deduction rule set generation module obtains food data and generates a food demand association deduction rule set based on dynamic semantic coding clusters; The spatiotemporal evolution knowledge association network module constructs a spatiotemporal evolution knowledge association network based on the food demand association deduction rule set and preset temporal scenario parameters; The adaptive recommendation strategy generation module obtains the user data to be recommended in real time and generates an adaptive recommendation strategy based on the spatiotemporal evolution knowledge association network; The knowledge association network optimization module dynamically reconstructs and optimizes the spatiotemporal evolution knowledge association network through a bidirectional data iteration mechanism.

7. A food recommendation system based on a spatiotemporal evolution knowledge association network as claimed in claim 6, characterized in that: Dynamic semantic coding cluster generation module, specifically including: Exclusive data collection submodule, collecting exclusive data of different groups of people; The dynamic clustering submodule performs dynamic clustering through cross-domain dynamic feature fusion to generate dynamic semantic coding clusters.

8. The food recommendation system based on spatiotemporal evolution knowledge association network according to claim 6, characterized in that: The food demand association deduction rule set generation module specifically includes: The food attribute knowledge base construction submodule obtains food data, extracts food multidimensional attributes, and constructs a food attribute knowledge base; The association rule establishment submodule establishes multidimensional association rules for food demand based on the food attribute knowledge base and dynamic semantic coding clusters; The priority dynamic adjustment submodule adjusts the priority of the multidimensional association rules of food demand in real time based on the food nutritional activity index and historical dietary feedback of different populations to form a set of food demand association deduction rules.

9. The food recommendation system based on spatiotemporal evolution knowledge association network according to claim 6, characterized in that: The spatiotemporal evolution knowledge association network module specifically includes: The temporal context parameter construction submodule constructs temporal context parameters based on temporal context characteristics through a dynamic time warping algorithm; The knowledge association network construction submodule constructs a spatiotemporal evolution knowledge association network based on the food demand association deduction rule set and temporal scenario parameters.

10. The food recommendation system based on spatiotemporal evolution knowledge association network according to claim 6, characterized in that: Adaptive recommendation strategy generation module, specifically including: The user data collection and analysis submodule acquires and analyzes the user data to be recommended in real time, generating user interest indicators and nutritional requirement indicators; The optimal food recommendation strategy acquisition submodule obtains the optimal food recommendation strategy based on the user's interest indicators, nutritional requirements indicators and spatiotemporal evolution knowledge association network.

Images