Recommendation method and system based on artificial intelligence

Abstract

The invention discloses an artificial intelligence-based recommendation method and system, and relates to the technical field of artificial intelligence, and the artificial intelligence-based recommendation method comprises the following steps: S1, receiving a first recommendation request, and determining a recommendation strategy according to the first recommendation request; s2, acquiring first recommendation data according to the recommendation strategy, and sending the first recommendation data; and S3, receiving a second recommendation request sent according to the first recommendation data, obtaining second recommendation data according to the second recommendation request, and sending the second recommendation data. According to the method and the device, the plant information can be quickly pushed to the recommendation requester according to different human health conditions, different plant characteristics, different planting climate environments and / or different planting soil environments, the pushing adaptability is high, and the pushing accuracy is high.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based recommendation method and system thereof. Background Art

[0002] Artificial Intelligence (AI) is a technical science that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence. It is a theory, method, technology and application system that can perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0003] Among them, the recommendation method and system based on artificial intelligence is one of the important applications of artificial intelligence. With the increasing demand for health and comfort of living environment and the expansion of people's interests and hobbies, how to help people choose suitable plants for planting has become a complex and practical problem. However, the existing artificial intelligence recommendation method and system have the following problems:

[0004] (1) Existing artificial intelligence recommendation methods and systems are unable to quickly push plant information to recommendation requesters based on different human health conditions, different plant characteristics, different planting climate environments, and / or different planting soil environments, and the push adaptability and accuracy are poor.

[0005] (2) When using existing artificial intelligence recommendation methods and systems, the recommendation requester often needs to browse the large amount of mixed plant information searched, and the recommendation requester needs to have professional knowledge to select the required plant information from the large amount of mixed plant information. This method is time-consuming and labor-intensive, and is prone to information omissions and / or misjudgment. Summary of the invention

[0006] The purpose of this application is to provide an artificial intelligence-based recommendation method and system, which can quickly push plant information to the recommendation requester according to different human health conditions, different plant characteristics, different planting climate environments and / or different planting soil environments, and the push adaptability and accuracy are high.

[0007] To achieve the above-mentioned objectives, the present application provides an artificial intelligence-based recommendation method, comprising the following steps: S1: receiving a first recommendation request, and determining a recommendation strategy based on the first recommendation request; wherein the recommendation strategy includes: an indoor recommendation strategy and an outdoor recommendation strategy; the first recommendation request includes: a request time, and indoor planting or outdoor planting; when the first recommendation request is for indoor planting, the generated recommendation strategy is an indoor recommendation strategy; when the first recommendation request is for outdoor planting, the generated recommendation strategy is an outdoor recommendation strategy; S2: obtaining first recommendation data according to the recommendation strategy, and sending it; S3: receiving a second recommendation request sent according to the first recommendation data, obtaining second recommendation data according to the second recommendation request, and sending it; wherein the second recommendation request at least includes: required plant information; the second recommendation data includes: at least one push supplier.

[0008] As above, when the recommendation strategy is an indoor recommendation strategy, the first recommendation data is obtained according to the recommendation strategy, and the sub-steps of sending are as follows: S21: Access the indoor plant database and use all indoor plant data packets as initial indoor data; S22: Obtain human health data, and extract features of human health data through a feature extraction module based on artificial intelligence to obtain human pathological characteristics; S23: Analyze the indoor plant disease impact characteristics of each initial indoor data according to the human pathological characteristics, obtain indoor plant disease impact related values, and use the initial indoor data whose indoor plant disease impact related values ​​are greater than or equal to the preset indoor plant disease impact related threshold as the first indoor object to be eliminated, eliminate the first indoor object to be eliminated, and use the remaining initial indoor data as the first Indoor screening data; S24: acquiring functional requirement data, and extracting features from the functional requirement data through a feature extraction module based on artificial intelligence to obtain functional requirement features; S25: analyzing the indoor plant functional features of each first indoor screening data according to the functional requirement features to obtain indoor plant function-related values, and using the first indoor screening data whose indoor plant function-related values ​​are less than or equal to a preset indoor plant function-related threshold as the second indoor object to be eliminated, eliminating the second indoor object to be eliminated, and using the remaining first indoor screening data as the second indoor screening data; S26: acquiring indoor planting description data for each second indoor screening data, using each second indoor screening data and the indoor planting description data of each second indoor screening data as the first recommended data, and sending them.

[0009] As above, when the recommendation strategy is an outdoor recommendation strategy, the first recommendation data is obtained according to the recommendation strategy, and the sub-steps of sending are as follows: S21': Access the outdoor plant database and use all outdoor plant data packets as initial outdoor data; S22': Obtain human health data, and extract features of human health data through a feature extraction module based on artificial intelligence to obtain human pathological characteristics; S23': Analyze the outdoor plant disease impact characteristics of each initial outdoor data according to the human pathological characteristics, obtain the outdoor plant disease impact related value, and use the initial outdoor data whose outdoor plant disease impact related value is greater than or equal to the preset outdoor plant disease impact related threshold as the first outdoor object to be eliminated, eliminate the first outdoor object to be eliminated, and The remaining initial outdoor data is used as the first outdoor screening data; S24': acquiring functional requirement data, and extracting features from the functional requirement data through a feature extraction module based on artificial intelligence to obtain functional requirement features; S25': analyzing the outdoor plant functional features of each first outdoor screening data according to the functional requirement features to obtain outdoor plant function-related values, and using the first outdoor screening data whose outdoor plant function-related values ​​are less than or equal to a preset outdoor plant function-related threshold as the second outdoor object to be eliminated, eliminating the second outdoor object to be eliminated, and using the remaining first outdoor screening data as the second outdoor screening data; S26': acquiring the position to be planted, screening the second outdoor screening data according to the position to be planted, obtaining the first recommended data, and sending it.

[0010] As above, wherein the second outdoor screening data is screened according to the location to be planted, and the sub-steps for obtaining the first recommended data are as follows: R1: Obtain the forecast data of the year in which the request time is located according to the location to be planted, wherein the forecast data includes: forecast climate data and forecast soil data; R2: Analyze the standard climate data of outdoor plants and the standard soil data of outdoor plants of each second outdoor screening data according to the forecast data, and obtain the outdoor planting adaptation value; R3: Use the second outdoor screening data whose outdoor planting adaptation value is greater than the preset outdoor planting adaptation threshold as the first recommended data.

[0011] As above, the sub-steps of obtaining the prediction data of the year in which the request time is located according to the location to be planted are as follows: R11: Traverse the historical climate database according to the location to be planted, and use the actual climate factor data corresponding to the climate area range containing the location to be planted as the historical climate data; R12: Analyze the historical climate data to obtain the prediction range values ​​of multiple climate factors, and use the prediction range values ​​of multiple climate factors as the predicted climate data, wherein the prediction range values ​​of multiple climate factors include at least: temperature prediction range value, rainfall prediction range value and light prediction range value; R13: Traverse the historical soil quality database according to the location to be planted, and use the actual soil quality factor data corresponding to the soil area range containing the location to be planted as the historical soil quality data; R14: Analyze the historical soil quality data to obtain the prediction range values ​​of multiple soil quality factors, and use the prediction range values ​​of multiple soil quality factors as the predicted soil quality data, wherein the prediction range values ​​of multiple soil quality factors include at least: pH prediction range value, multiple element content prediction range values ​​and multiple microbial content prediction range values.

[0012] As above, the expression of the predicted range value of the a-th climate factor is: in, The minimum value in the predicted range of the ath climate factor in the year of the request time; It is the maximum value in the prediction range of the ath climate factor in the year of the request time.

[0013] As above, the expression of the predicted range value of the b-th soil quality factor is: in, The minimum value in the predicted range of the bth soil quality factor in the year of the request time; It is the maximum value in the predicted range of the bth soil quality factor in the year of the request time.

[0014] As above, the expression of outdoor planting adaptation value is:

[0015] Among them, Spz h is the outdoor planting adaptation value between the predicted data and the hth second outdoor screening data; λ1 is the outdoor climate weight; η u is the standard climate impact weight corresponding to the u-th standard climate impact factor; qgz uis the judgment value between the standard climate impact range corresponding to the u-th standard climate impact factor and the predicted range value of the corresponding climate factor in the prediction data. If the predicted range value of the climate factor is within the standard climate impact range, the judgment value is X1; if the predicted range value of the climate factor is outside the standard climate impact range, the judgment value is X2, and X2 < X1; u ∈ [1, U], where U is the total number of standard climate impact factors; λ2 is the outdoor soil quality weight; μ f is the standard soil quality impact weight corresponding to the f-th standard soil quality impact factor; tgz f is the judgment value between the standard soil quality impact range corresponding to the f-th standard soil quality impact factor and the predicted range value of the corresponding soil factor in the prediction data. If the predicted range value of the soil factor is within the standard soil quality impact range, the judgment value is X3; if the predicted range value of the soil factor is outside the standard soil quality impact range, the judgment value is X4, and X4 < X3;

[0016] f ∈ [1, F], where F is the total number of standard soil quality impact factors.

[0017] This application also provides a recommendation system based on artificial intelligence, including: a recommendation service platform and an artificial intelligence recommendation center; wherein, the recommendation service platform: after completing login, sends a first recommendation request; receives the first recommendation data, and sends a second recommendation request according to the first recommendation data; receives the second recommendation data; the artificial intelligence recommendation center: is used to execute the above-mentioned artificial intelligence-based recommendation method.

[0018] As above, wherein, the artificial intelligence recommendation center includes: a transceiver unit, a first recommendation unit, a second recommendation unit, and a storage unit; wherein, the transceiver unit: receives the first recommendation request and sends the first recommendation request to the first recommendation unit; sends the first recommendation data to the recommendation service platform; receives the second recommendation request sent according to the first recommendation data and sends the second recommendation request to the second recommendation unit; sends the second recommendation data to the recommendation service platform; the first recommendation unit: determines a recommendation strategy according to the first recommendation request; obtains the first recommendation data according to the recommendation strategy and sends the first recommendation data to the transceiver unit; the second recommendation unit: obtains the second recommendation data according to the second recommendation request and sends the second recommendation data to the transceiver unit; the storage unit: is used to store an indoor plant database, an outdoor plant database, a historical climate database, and a historical soil database.

[0019] The beneficial effects achieved by this application are as follows:

[0020] (1) The artificial intelligence-based recommendation method and system of this application can quickly push plant information to the recommendation requester according to the differences in human health conditions, plant自身 characteristics, planting climate environments, and / or planting soil environments, and the push has high adaptability and high accuracy.

[0021] (2) When using the artificial intelligence-based recommendation method and system of the present application, the recommendation requester does not need to browse a large amount of mixed plant information, nor does he need to have professional knowledge to select the required plant information from a large amount of mixed plant information. This method saves time and effort, and is not prone to information omissions and / or misjudgments. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0023] Figure 1 A schematic diagram of the structure of an embodiment of an artificial intelligence-based recommendation system;

[0024] Figure 2 The present invention is a flowchart of an embodiment of an artificial intelligence-based recommendation method. DETAILED DESCRIPTION

[0025] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0026] like Figure 1 As shown, the present application provides an artificial intelligence-based recommendation system, comprising: a recommendation service platform 1 and an artificial intelligence recommendation center 2.

[0027] Among them, the recommendation service platform 1: after completing the login, sends a first recommendation request; receives the first recommendation data, and sends a second recommendation request according to the first recommendation data; receives the second recommendation data.

[0028] Artificial Intelligence Recommendation Center 2: used to execute the following artificial intelligence-based recommendation method.

[0029] Furthermore, the artificial intelligence recommendation center 2 includes: a transceiver unit, a first recommendation unit, a second recommendation unit and a storage unit.

[0030] Among them, the transceiver unit: receives a first recommendation request, and sends the first recommendation request to the first recommendation unit; sends the first recommendation data to the recommendation service platform; receives a second recommendation request sent according to the first recommendation data, and sends the second recommendation request to the second recommendation unit; sends the second recommendation data to the recommendation service platform.

[0031] The first recommendation unit: determines a recommendation strategy according to the first recommendation request; obtains first recommendation data according to the recommendation strategy, and sends the first recommendation data to the transceiver unit.

[0032] The second recommendation unit: obtains second recommendation data according to the second recommendation request, and sends the second recommendation data to the transceiver unit.

[0033] Storage unit: used to store indoor plant database, outdoor plant database, historical climate database and historical soil database.

[0034] The indoor plant database includes at least: a plurality of indoor plant data packets, each indoor plant data packet includes at least: indoor plant disease impact characteristics, indoor plant functional characteristics and indoor planting description data.

[0035] Specifically, the indoor plant disease impact characteristics are characteristics that can clearly describe which diseases the indoor plant has adverse effects on.

[0036] The functional characteristics of indoor plants are characteristics that can clearly describe the functions of the indoor plants, for example: having the function of purifying the air, having ornamental value and / or having medicinal value.

[0037] The indoor planting instruction data is data that can clearly describe the planting method of the indoor plant, and may be a file, an image, and / or a video.

[0038] Furthermore, the indoor planting instruction data includes at least: planting steps for indoor plants, recommended ventilation levels for indoor plants, recommended light levels for indoor plants, recommended soil data for indoor plants, recommended humidity levels for indoor plants, and recommended temperatures for indoor plants.

[0039] Specifically, the steps for planting indoor plants are to guide planters to complete the specific operational steps for planting indoor plants.

[0040] Recommended ventilation level for indoor plants: ventilation conditions that can ensure normal growth of indoor plants.

[0041] Recommended light levels for indoor plants: Light conditions that can ensure normal growth of indoor plants.

[0042] Recommended soil data for indoor plants: soil conditions that can ensure the normal growth of indoor plants.

[0043] Recommended humidity level for indoor plants: It is the moisture condition that can ensure the normal growth of indoor plants.

[0044] Recommended temperature for indoor plants: It is the temperature condition that can ensure the normal growth of indoor plants.

[0045] The outdoor plant database includes at least: a plurality of outdoor plant data packets, each of which includes at least: outdoor plant disease impact characteristics, outdoor plant functional characteristics, outdoor plant standard climate data, outdoor plant standard soil data and outdoor planting instructions data.

[0046] During the planting process, the actual ventilation level of indoor plants, the actual light level of indoor plants, the actual soil data of indoor plants, the actual humidity level of indoor plants and the actual temperature of indoor plants can all be regulated according to the recommended ventilation level of indoor plants, the recommended light level of indoor plants, the recommended soil data of indoor plants, the recommended humidity level of indoor plants and the recommended temperature of indoor plants. The specific regulation method can be manual regulation or regulation using artificial intelligence-based regulation equipment.

[0047] Specifically, the outdoor plant disease impact characteristics are characteristics that can clearly describe which diseases the outdoor plant has adverse effects on.

[0048] The functional characteristics of outdoor plants are characteristics that can clearly describe the functions of the outdoor plants.

[0049] The standard climate data of outdoor plants include at least: a plurality of standard climate influencing factors, each standard climate influencing factor corresponds to a standard climate influencing range and a standard climate influencing weight; the standard climate influencing factors include at least: temperature, rainfall and light.

[0050] The standard soil quality data for outdoor plants include at least: multiple standard soil quality influencing factors, each standard soil quality influencing factor corresponds to a standard soil quality influencing range and a standard soil quality influencing weight; the standard soil quality influencing factors include at least: soil pH, content of multiple elements and content of multiple microorganisms.

[0051] Among them, the historical climate database includes at least: multiple groups of actual climate factor data, one group of actual climate factor data corresponds to a climate region range, each group of actual climate factor data includes at least: multiple climate sub-data, one climate sub-data corresponds to a climate collection year, and each climate sub-data includes at least: actual temperature range value, actual rainfall range value and actual light range value.

[0052] Specifically, in, is the lowest actual temperature in the i-th group of actual climate factors; is the highest actual temperature in the i-th group of actual climate factors.

[0053] in, is the minimum actual rainfall intensity in the i-th group of actual climate factors; is the maximum actual rainfall intensity in the i-th group of actual climate factors.

[0054] in, is the minimum actual light intensity in the i-th group of actual climate factors; is the maximum actual light intensity in the i-th group of actual climate factors.

[0055] Among them, the historical soil database includes at least: multiple groups of actual soil factor data, one group of actual soil factor data corresponds to a soil area range, each group of actual soil factor data includes at least: multiple sub-soil data, each sub-soil data corresponds to a soil collection year, and each soil sub-data includes at least: actual soil pH range, multiple element content ranges and multiple microbial content ranges.

[0056] Specifically, in, is the minimum pH value in the jth group of actual soil factor data; It is the maximum pH value in the jth group of actual soil factor data.

[0057] in, is the minimum value of the content of the nth element in the jth group of actual soil factor data; It is the maximum value of the content of the nth element in the jth group of actual soil factor data.

[0058] in, is the minimum value of the content of the mth microorganism in the jth group of actual soil quality factor data; It is the maximum value of the content of the mth microorganism in the jth group of actual soil factor data.

[0059] The outdoor planting instruction data is the specific operation steps for guiding the growers to complete the outdoor plant planting.

[0060] like Figure 2 As shown, the present application provides a recommendation method based on artificial intelligence, comprising the following steps:

[0061] S1: Receive a first recommendation request, and determine a recommendation strategy based on the first recommendation request; wherein the recommendation strategy includes: an indoor recommendation strategy and an outdoor recommendation strategy; the first recommendation request includes: a request time, and indoor planting or outdoor planting; when the first recommendation request is for indoor planting, the generated recommendation strategy is an indoor recommendation strategy; when the first recommendation request is for outdoor planting, the generated recommendation strategy is an outdoor recommendation strategy.

[0062] Specifically, the request time is the time when the first recommendation request is sent.

[0063] Indoor Recommendation Strategy: Method used to recommend plants suitable for indoor cultivation.

[0064] Outdoor Recommendation Strategy: A method used to recommend plants suitable for outdoor growing.

[0065] S2: Obtain the first recommendation data according to the recommendation strategy and send it.

[0066] Further, as an embodiment, when the recommendation strategy is an indoor recommendation strategy, the sub-steps of obtaining the first recommendation data according to the recommendation strategy and sending the first recommendation data are as follows:

[0067] S21: Access the indoor plant database and take all indoor plant data packets as initial indoor data.

[0068] S22: Acquire human health data, and extract features from the human health data through an artificial intelligence-based feature extraction module to obtain human pathological features.

[0069] Specifically, human pathological characteristics are characteristics that can clearly describe the relevant conditions of the diseases currently suffered by the grower. The relevant conditions of the diseases at least include: allergic diseases (e.g., pollen allergy, allergy to specific plants), respiratory diseases (e.g., asthma, chronic obstructive pulmonary disease) and immune system diseases (e.g., low immunity, autoimmune diseases).

[0070] S23: Analyze the indoor plant disease impact characteristics of each initial indoor data according to the human pathological characteristics, obtain the indoor plant disease impact related value, and use the initial indoor data whose indoor plant disease impact related value is greater than or equal to the preset indoor plant disease impact related threshold as the first indoor object to be eliminated, eliminate the first indoor object to be eliminated, and use the remaining initial indoor data as the first indoor screening data.

[0071] Specifically, the indoor plant disease impact correlation value between the indoor plant disease impact features of the initial indoor data can be obtained through a pre-trained artificial intelligence-based correlation calculation model or a neural network model.

[0072] The specific value of the indoor plant disease impact related threshold is set according to the actual situation. If the indoor plant disease impact related value is greater than or equal to the preset indoor plant disease impact related threshold, it means that the indoor plant corresponding to the initial indoor data has an adverse effect on the planter. Among them, the larger the indoor plant disease impact related value is, the greater the degree to which the indoor plant corresponding to the initial indoor data will have an adverse effect on the planter's body.

[0073] For example: human pathological characteristics indicate that the planter has a pollen allergy, and indoor plant disease impact characteristics indicate that the indoor plant is a flowering plant (for example: wind-pollinated plants and flowers with large amounts of pollen). If the indoor plant disease impact correlation value between the human pathological characteristics and the indoor plant disease impact characteristics is greater than or equal to a preset indoor plant disease impact correlation threshold, the planter should avoid planting such indoor plants. Therefore, the data of such indoor plants cannot be pushed to the planter.

[0074] The human pathological characteristics indicate that the planter is allergic to specific plants, the indoor plant disease impact characteristics indicate that the indoor plant is a plant to which the planter is allergic, and if the indoor plant disease impact correlation value between the human pathological characteristics and the indoor plant disease impact characteristics is greater than or equal to the preset indoor plant disease impact correlation threshold, the planter should avoid planting such indoor plants. Therefore, the data of such indoor plants cannot be pushed to the planter.

[0075] The human pathological characteristics indicate that the planter suffers from asthma, and the indoor plant disease impact characteristics indicate that the indoor plant is a plant with a strong irritating odor (for example, tuberose and geranium). If the indoor plant disease impact correlation value between the human pathological characteristics and the indoor plant disease impact characteristics is greater than or equal to the preset indoor plant disease impact correlation threshold, the planter should avoid planting such indoor plants. Therefore, the data of such indoor plants cannot be pushed to the planter.

[0076] The human pathological characteristics indicate that the planter suffers from chronic obstructive pulmonary disease, and the indoor plant disease impact characteristics indicate that the indoor plant is a plant that is prone to mold growth (for example: plants with moist soil due to long-term excessive watering and plants with a large amount of fluff or pollen). If the indoor plant disease impact correlation value between the human pathological characteristics and the indoor plant disease impact characteristics is greater than or equal to the preset indoor plant disease impact correlation threshold, the planter should avoid planting such indoor plants. Therefore, the data of such indoor plants cannot be pushed to the planter.

[0077] The human pathological characteristics indicate that the planters are people with low immunity, and the indoor plant disease impact characteristics indicate that the indoor plants are plants that easily carry pathogens or pests (for example, wild plants or plants that have not been sterilized). If the indoor plant disease impact correlation value between the human pathological characteristics and the indoor plant disease impact characteristics is greater than or equal to the preset indoor plant disease impact correlation threshold, the planters should avoid planting such indoor plants. Therefore, the data of such indoor plants cannot be pushed to the planters.

[0078] The human pathological characteristics indicate that the planter suffers from an autoimmune disease, and the indoor plant disease impact characteristics indicate that the indoor plant is a plant with specific plant components that make this type of person sensitive (for example, people with systemic lupus erythematosus may be sensitive to certain light-sensitive plants and should avoid raising these plants to avoid aggravating their condition). If the indoor plant disease impact correlation value between the human pathological characteristics and the indoor plant disease impact characteristics is greater than or equal to the preset indoor plant disease impact correlation threshold, the planter should avoid planting this type of indoor plant. Therefore, the data of this type of indoor plant cannot be pushed to the planter.

[0079] S24: Acquire functional requirement data, and perform feature extraction on the functional requirement data through an artificial intelligence-based feature extraction module to obtain functional requirement features.

[0080] Specifically, the artificial intelligence-based feature extraction module is a pre-built and trained module for feature extraction.

[0081] S25: Analyze the indoor plant function characteristics of each first indoor screening data according to the functional requirement characteristics to obtain the indoor plant function related value, and use the first indoor screening data whose indoor plant function related value is less than or equal to the preset indoor plant function related threshold as the second indoor object to be eliminated, eliminate the second indoor object to be eliminated, and use the remaining first indoor screening data as the second indoor screening data.

[0082] Specifically, the specific value of the indoor plant function-related threshold is set according to the actual situation. If the indoor plant function-related value is greater than the preset indoor plant function-related threshold, it means that the indoor plants corresponding to the first indoor screening data meet the requirements of the planters for the functions of the planted plants.

[0083] S26: Acquire indoor planting description data of each second indoor screening data, take each second indoor screening data and the indoor planting description data of each second indoor screening data as first recommendation data, and send them.

[0084] Specifically, after obtaining the first recommendation data, the artificial intelligence recommendation center sends the first recommendation data to the recommendation service platform, which is displayed by the recommendation service platform.

[0085] Further, as another embodiment, when the recommendation strategy is an outdoor recommendation strategy, the sub-steps of obtaining the first recommendation data according to the recommendation strategy and sending the first recommendation data are as follows:

[0086] S21': Access the outdoor plant database and take all outdoor plant data packets as initial outdoor data.

[0087] S22': Acquire human health data, and extract features from the human health data through a feature extraction module based on artificial intelligence to obtain human pathological features.

[0088] S23': Analyze the outdoor plant disease impact characteristics of each initial outdoor data according to the human pathological characteristics, obtain the outdoor plant disease impact related value, and use the initial outdoor data whose outdoor plant disease impact related value is greater than or equal to the preset outdoor plant disease impact related threshold as the first outdoor object to be eliminated, eliminate the first outdoor object to be eliminated, and use the remaining initial outdoor data as the first outdoor screening data.

[0089] Specifically, the outdoor plant disease impact correlation value between the human pathological characteristics and the outdoor plant disease impact characteristics of the initial outdoor data can be obtained through a pre-trained artificial intelligence-based correlation calculation model or a neural network model.

[0090] The specific value of the outdoor plant disease impact related threshold is set according to the actual situation. If the outdoor plant disease impact related value is greater than or equal to the preset outdoor plant disease impact related threshold, it means that the outdoor plants corresponding to the initial outdoor data have adverse effects on the planters. Among them, the larger the outdoor plant disease impact related value, the greater the degree to which the outdoor plants corresponding to the initial outdoor data will have adverse effects on the body of the planters.

[0091] S24': Acquire functional requirement data, and extract features from the functional requirement data through a feature extraction module based on artificial intelligence to obtain functional requirement features.

[0092] Specifically, the artificial intelligence-based feature extraction module is a pre-built and trained module for feature extraction.

[0093] S25': Analyze the outdoor plant function characteristics of each first outdoor screening data according to the functional requirement characteristics to obtain the outdoor plant function related value, and use the first outdoor screening data whose outdoor plant function related value is less than or equal to the preset outdoor plant function related threshold as the second outdoor object to be eliminated, eliminate the second outdoor object to be eliminated, and use the remaining first outdoor screening data as the second outdoor screening data.

[0094] Specifically, the specific value of the outdoor plant function-related threshold is set according to the actual situation. If the outdoor plant function-related value is greater than the preset outdoor plant function-related threshold, it means that the outdoor plants corresponding to the first outdoor screening data meet the requirements of the planters for the functions of the planted plants.

[0095] S26': Acquire the location to be planted, filter the second outdoor screening data according to the location to be planted, obtain the first recommendation data, and send it.

[0096] Specifically, the location to be planted is the location where the planter intends to plant outdoor plants. The artificial intelligence recommendation center sends a request to obtain the location to be planted to the recommendation service platform, and the recommendation service platform receives and executes the request to obtain the location to be planted, and sends the location to be planted to the artificial intelligence recommendation center.

[0097] Further, the second outdoor screening data is screened according to the location to be planted, and the sub-steps of obtaining the first recommended data are as follows:

[0098] R1: Obtain the forecast data for the year of the request time according to the location to be planted, where the forecast data includes: forecast climate data and forecast soil quality data.

[0099] Furthermore, the sub-steps of obtaining the forecast data for the year of the request time according to the location to be planted are as follows:

[0100] R11: Traverse the historical climate database according to the location to be planted, and use the actual climate factor data corresponding to the climate region range containing the location to be planted as the historical climate data.

[0101] R12: Analyze historical climate data to obtain prediction range values ​​of multiple climate factors, and use the prediction range values ​​of multiple climate factors as predicted climate data, wherein the prediction range values ​​of multiple climate factors at least include: temperature prediction range value, rainfall prediction range value and light prediction range value.

[0102] Furthermore, the expression of the predicted range value of the a-th climate factor is:

[0103] in, The minimum value in the predicted range of the ath climate factor in the year of the request time; It is the maximum value in the prediction range of the ath climate factor in the year of the request time.

[0104] Furthermore,

[0105] in, It is the minimum value in the actual range of the ath climate factor in the climate collection year before the year of the request time.

[0106] It is the minimum value in the actual range of the ath climate factor in the vth climate collection year.

[0107] It is the minimum value in the actual range of the ath climate factor in the v-1th climate collection year; where the v-1th climate collection year is the year before the vth climate collection year; v∈[1,V], V is the total number of historical climate data.

[0108] The actual range values ​​of climate factors include at least: actual temperature range values, actual rainfall range values ​​and actual light range values.

[0109]

[0110] in, It is the maximum value in the actual range of the ath climate factor in the climate collection year before the year of the request time.

[0111] It is the maximum value in the actual range of the ath climate factor in the vth climate collection year.

[0112] It is the maximum value in the actual range of the ath climate factor in the v-1th climate collection year.

[0113] R13: Traverse the historical soil quality database according to the location to be planted, and use the actual soil quality factor data corresponding to the soil quality area range containing the location to be planted as the historical soil quality data.

[0114] R14: Analyze historical soil quality data to obtain predicted range values ​​of multiple soil quality factors, and use the predicted range values ​​of multiple soil quality factors as predicted soil quality data, wherein the predicted range values ​​of multiple soil quality factors at least include: pH predicted range values, predicted range values ​​of multiple element contents and predicted range values ​​of multiple microbial contents.

[0115] Furthermore, the expression of the predicted range value of the b-th soil quality factor is:

[0116] in, The minimum value in the predicted range of the bth soil quality factor in the year of the request time; It is the maximum value in the predicted range of the bth soil quality factor in the year of the request time.

[0117] Furthermore,

[0118] in, It is the minimum value in the actual range of the bth soil quality factor in the soil quality collection year before the request time year.

[0119] It is the minimum value in the actual range of the bth soil factor in the eth soil collection year.

[0120] is the minimum value in the actual range of the bth soil factor in the e-1th soil collection year; where the e-1th soil collection year is the year before the eth soil collection year; e∈[1,E], E is the total number of historical soil data.

[0121] The actual range values ​​of soil factors include at least: actual soil pH range, multiple element content range and multiple microbial content range.

[0122]

[0123] in, It is the maximum value in the actual range of the bth soil quality factor in the soil quality collection year before the year of the request time.

[0124] It is the maximum value in the actual range of the bth soil factor in the eth soil collection year.

[0125] It is the maximum value in the actual range of the bth soil factor in the e-1th soil collection year.

[0126] R15: Use the predicted climate data and predicted soil quality data as the predicted data for the year of the request time.

[0127] R2: Analyze the standard climate data of outdoor plants and the standard soil data of outdoor plants of each second outdoor screening data according to the predicted data to obtain the outdoor planting adaptation value.

[0128] Furthermore, the expression of outdoor planting adaptation value is:

[0129]

[0130] Among them, Spz h is the outdoor planting adaptation value between the predicted data and the hth second outdoor screening data; λ1 is the outdoor climate weight; η u is the standard climate impact weight corresponding to the u-th standard climate impact factor; qgz uis the judgment value between the standard climate impact range corresponding to the u-th standard climate impact factor and the predicted range value of the corresponding climate factor in the predicted data. If the predicted range value of the climate factor is within the standard climate impact range, the judgment value is X1; if the predicted range value of the climate factor is outside the standard climate impact range, the judgment value is X2, and X2 < X1; u ∈ [1, U], where U is the total number of standard climate impact factors; λ2 is the outdoor soil weight; μ f is the standard soil impact weight corresponding to the f-th standard soil impact factor; tgz f is the judgment value between the standard soil impact range corresponding to the f-th standard soil impact factor and the predicted range value of the corresponding soil factor in the predicted data. If the predicted range value of the soil factor is within the standard soil impact range, the judgment value is X3; if the predicted range value of the soil factor is outside the standard soil impact range, the judgment value is X4, and X4 < X3; f ∈ [1, F], where F is the total number of standard soil impact factors.

[0131] Specifically, the outdoor climate weight λ1 is used to represent the influence degree of climate factors on the growth of outdoor plants corresponding to the second outdoor screening data. The specific value of the outdoor climate weight λ1 is set according to the actual situation, for example: according to the actual planting experience of this type of outdoor plants.

[0132] The standard climate impact weight η corresponding to the u-th standard climate impact factor u is used to represent the influence degree of this standard climate impact factor on the growth of outdoor plants corresponding to the second outdoor screening data among all climate factors. The standard climate impact weight η u The specific value is set according to the actual situation, for example: according to the actual planting experience of this type of outdoor plants.

[0133] The outdoor soil weight λ2 is used to represent the influence degree of soil factors on the growth of outdoor plants corresponding to the second outdoor screening data, for example: according to the actual planting experience of this type of outdoor plants.

[0134] The standard soil impact weight μ corresponding to the f-th standard soil impact factor f is used to represent the influence degree of this standard soil impact factor on the growth of outdoor plants corresponding to the second outdoor screening data among all soil factors. The standard soil impact weight μ f The specific value is set according to the actual situation, for example: according to the actual planting experience of this type of outdoor plants.

[0135] The judgment value of X1, the judgment value of X2, the judgment value of X3, and the judgment value of X4 are all preset fixed values, and the specific values are set according to the actual situation.

[0136] R3: The second outdoor screening data whose outdoor planting adaptation value is greater than a preset outdoor planting adaptation threshold is used as the first recommended data.

[0137] Specifically, the outdoor planting adaptation threshold is set according to the actual situation. The larger the outdoor planting adaptation value is, the greater the push accuracy is.

[0138] S3: receiving a second recommendation request sent according to the first recommendation data, obtaining second recommendation data according to the second recommendation request, and sending the second recommendation data; wherein the second recommendation request at least includes: required plant information; the second recommendation data includes: at least one push supplier.

[0139] Specifically, the required plant information is basic information and / or images of indoor plants and / or outdoor plants confirmed from the first recommendation data.

[0140] Further, a second recommendation request sent according to the first recommendation data is received, and the sub-steps of obtaining the second recommendation data according to the second recommendation request are as follows:

[0141] S31: Generate a search instruction according to the second recommendation request, where the search instruction at least includes: a search object.

[0142] Specifically, the search object is the basic information and / or image of the required indoor plants or outdoor plants. The basic information and / or image of the indoor plants or outdoor plants are data used to characterize the type of the indoor plants or outdoor plants.

[0143] S32: Acquire multiple initial suppliers from the Internet according to the search instruction.

[0144] Specifically, the initial supplier refers to the supplier who is selling the search object.

[0145] S33: Obtain the historical reputation value of each initial supplier, and calculate the average reputation value of the historical reputation value of each initial supplier.

[0146] S34: The initial suppliers whose average reputation value is greater than a preset reputation threshold are selected as push suppliers, and the push suppliers are selected as second recommendation data.

[0147] Specifically, the preset reputation threshold is set according to the actual situation. The larger the average reputation value is, the better the reputation of the supplier in the transaction is, and the greater the probability that the indoor plants and / or outdoor plants sold are the same as the ones sold.

[0148] The beneficial effects achieved by this application are as follows:

[0149] (1) The AI-based recommendation method and system of the present application can quickly push plant information to the person requesting the recommendation based on different human health conditions, different plant characteristics, different planting climate environments, and / or different planting soil environments, and the push information has high adaptability and accuracy.

[0150] (2) When using the artificial intelligence-based recommendation method and system of the present application, the recommendation requester does not need to browse a large amount of mixed plant information, nor does he need to have professional knowledge to select the required plant information from a large amount of mixed plant information. This method saves time and effort, and is not prone to information omissions and / or misjudgments.

[0151] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic creative concepts. Therefore, the scope of protection of the present application is intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application. Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the protection scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A recommendation method based on artificial intelligence, characterized in that: The steps include: S1: receiving a first recommendation request, and determining a recommendation strategy according to the first recommendation request; wherein the recommendation strategy includes: an indoor recommendation strategy and an outdoor recommendation strategy; the first recommendation request includes: a request time, and indoor planting or outdoor planting; when the first recommendation request is for indoor planting, the generated recommendation strategy is an indoor recommendation strategy; when the first recommendation request is for outdoor planting, the generated recommendation strategy is an outdoor recommendation strategy; S2: Obtain the first recommended data according to the recommendation strategy and send it; S3: receiving a second recommendation request sent according to the first recommendation data, obtaining second recommendation data according to the second recommendation request, and sending the second recommendation data; wherein the second recommendation request at least includes: required plant information; the second recommendation data includes: at least one push supplier.

2. The artificial intelligence-based recommendation method according to claim 1, characterized in that: When the recommendation strategy is an indoor recommendation strategy, the sub-steps of obtaining the first recommendation data according to the recommendation strategy and sending the data are as follows: S21: Accessing the indoor plant database, and taking all the indoor plant data packets as initial indoor data; S22: Acquire human health data, and extract features from the human health data through a feature extraction module based on artificial intelligence to obtain human pathological features; S23: analyzing the indoor plant disease impact characteristics of each initial indoor data according to the human pathological characteristics, obtaining an indoor plant disease impact related value, and taking the initial indoor data whose indoor plant disease impact related value is greater than or equal to a preset indoor plant disease impact related threshold as the first indoor object to be eliminated, eliminating the first indoor object to be eliminated, and taking the remaining initial indoor data as the first indoor screening data; S24: Acquire functional requirement data, and extract features from the functional requirement data using a feature extraction module based on artificial intelligence to obtain functional requirement features; S25: Analyze the indoor plant function characteristics of each first indoor screening data according to the functional requirement characteristics to obtain indoor plant function related values, and use the first indoor screening data whose indoor plant function related values ​​are less than or equal to the preset indoor plant function related threshold as the second indoor object to be eliminated, eliminate the second indoor object to be eliminated, and use the remaining first indoor screening data as the second indoor screening data; S26: Acquire indoor planting description data of each second indoor screening data, take each second indoor screening data and the indoor planting description data of each second indoor screening data as first recommendation data, and send them.

3. The artificial intelligence-based recommendation method according to claim 1, characterized in that: When the recommendation strategy is an outdoor recommendation strategy, the sub-steps of obtaining the first recommendation data according to the recommendation strategy and sending the data are as follows: S21': access the outdoor plant database and take all outdoor plant data packets as initial outdoor data; S22': Acquire human health data, and extract features from the human health data through a feature extraction module based on artificial intelligence to obtain human pathological features; S23': Analyze the outdoor plant disease impact characteristics of each initial outdoor data according to the human pathological characteristics, obtain the outdoor plant disease impact related value, and use the initial outdoor data whose outdoor plant disease impact related value is greater than or equal to the preset outdoor plant disease impact related threshold as the first outdoor object to be eliminated, eliminate the first outdoor object to be eliminated, and use the remaining initial outdoor data as the first outdoor screening data; S24': Acquire functional requirement data, and extract features from the functional requirement data through a feature extraction module based on artificial intelligence to obtain functional requirement features; S25': Analyze the outdoor plant function characteristics of each first outdoor screening data according to the functional requirement characteristics, obtain the outdoor plant function related value, and use the first outdoor screening data whose outdoor plant function related value is less than or equal to the preset outdoor plant function related threshold as the second outdoor object to be eliminated, eliminate the second outdoor object to be eliminated, and use the remaining first outdoor screening data as the second outdoor screening data; S26': Acquire the location to be planted, filter the second outdoor screening data according to the location to be planted, obtain the first recommendation data, and send it.

4. The artificial intelligence-based recommendation method according to claim 3, characterized in that: The sub-steps of filtering the second outdoor screening data according to the location to be planted and obtaining the first recommended data are as follows: R1: Obtain the forecast data of the requested year based on the location to be planted, where the forecast data includes: forecast climate data and forecast soil quality data; R2: Analyze the standard climate data and standard soil data of outdoor plants of each second outdoor screening data according to the predicted data to obtain the outdoor planting adaptation value; R3: The second outdoor screening data whose outdoor planting adaptation value is greater than a preset outdoor planting adaptation threshold is used as the first recommended data.

5. The artificial intelligence-based recommendation method according to claim 4, characterized in that: The sub-steps for obtaining the forecast data for the year of the request time according to the location to be planted are as follows: R11: Traverse the historical climate database according to the location to be planted, and use the actual climate factor data corresponding to the climate region range containing the location to be planted as the historical climate data; R12: Analyze historical climate data to obtain prediction range values ​​of multiple climate factors, and use the prediction range values ​​of the multiple climate factors as predicted climate data, wherein the prediction range values ​​of the multiple climate factors at least include: temperature prediction range value, rainfall prediction range value and light prediction range value; R13: Traverse the historical soil quality database according to the location to be planted, and use the actual soil quality factor data corresponding to the soil quality area range containing the location to be planted as the historical soil quality data; R14: Analyze historical soil quality data to obtain predicted range values ​​of multiple soil quality factors, and use the predicted range values ​​of multiple soil quality factors as predicted soil quality data, wherein the predicted range values ​​of multiple soil quality factors at least include: pH predicted range values, predicted range values ​​of multiple element contents and predicted range values ​​of multiple microbial contents.

6. The artificial intelligence-based recommendation method according to claim 5, characterized in that: The expression of the prediction range value of the a-th climate factor is: in, The minimum value in the predicted range of the ath climate factor in the year of the request time; It is the maximum value in the prediction range of the ath climate factor in the year of the request time.

7. The artificial intelligence-based recommendation method according to claim 5, characterized in that: The expression of the prediction range value of the bth soil quality factor is: in, The minimum value in the predicted range of the bth soil quality factor in the year of the request time; It is the maximum value in the predicted range of the bth soil quality factor in the year of the request time.

8. The artificial intelligence-based recommendation method according to claim 4, characterized in that: The expression of outdoor planting adaptation value is: Among them, Spz h is the outdoor planting adaptation value between the predicted data and the h-th second outdoor screening data; λ1 is the outdoor climate weight; η u is the standard climate impact weight corresponding to the u-th standard climate impact factor; qgz u is the judgment value between the standard climate impact range corresponding to the u-th standard climate impact factor and the predicted range value of the corresponding climate factor in the predicted data. If the predicted range value of the climate factor is within the standard climate impact range, the judgment value is X1. If the predicted range value of the climate factor is outside the standard climate impact range, the judgment value is X2, and X2 < X1; u ∈ [1, U], where U is the total number of standard climate impact factors; λ2 is the outdoor soil quality weight; μ f is the standard soil quality impact weight corresponding to the f-th standard soil quality impact factor; tgz f is the judgment value between the standard soil quality impact range corresponding to the f-th standard soil quality impact factor and the predicted range value of the corresponding soil quality factor in the predicted data. If the predicted range value of the soil quality factor is within the standard soil quality impact range, the judgment value is X3. If the predicted range value of the soil quality factor is outside the standard soil quality impact range, the judgment value is X4, and X4 < X3; f ∈ [1, F], where F is the total number of standard soil quality impact factors.

9. An artificial intelligence-based recommendation system, characterized in that: include: Recommendation service platform and artificial intelligence recommendation center; The recommendation service platform: after completing the login, sends a first recommendation request; receives the first recommendation data, and sends a second recommendation request according to the first recommendation data; receives the second recommendation data; Artificial intelligence recommendation center: used to execute the artificial intelligence-based recommendation method described in any one of claims 1-8.

10. The artificial intelligence-based recommendation system according to claim 9, characterized in that: The artificial intelligence recommendation center includes: a transceiver unit, a first recommendation unit, a second recommendation unit and a storage unit; The transceiver unit: receives a first recommendation request and sends the first recommendation request to a first recommendation unit; sends first recommendation data to a recommendation service platform; receives a second recommendation request sent according to the first recommendation data, and sends the second recommendation request to a second recommendation unit; sends the second recommendation data to the recommendation service platform; The first recommendation unit: determines a recommendation strategy according to the first recommendation request; obtains first recommendation data according to the recommendation strategy, and sends the first recommendation data to the transceiver unit; The second recommendation unit: obtains the second recommendation data according to the second recommendation request, and sends the second recommendation data to the transceiver unit; Storage unit: used to store indoor plant database, outdoor plant database, historical climate database and historical soil database.

Images