An intelligent analysis method, platform and medium for environmental behavior data

Abstract

The present invention relates to an intelligent analysis method and system for environmental behavior data, the method comprising: locating resonance nodes, dividing them into a first resonance node set involving a positive resonance relationship, and a second resonance node set involving a reverse resonance relationship; monitoring the i1th resonance node and the i2th resonance node to determine whether the environmental data for the first object has changed effectively; and pushing environmental data interference information based on the effective change. The present invention changes the traditional environmental monitoring means through the application of information technology, uses communication network technology means to implement long-term, continuous and effective monitoring of environmental protection behaviors, scientifically and accurately, comprehensively and efficiently monitors and manages the environmental conditions of the area under its jurisdiction, so that the environmental management work of the environmental protection department can achieve scientific monitoring and efficient management.

Description

[Technical field]

[0001] The present invention belongs to the technical field of big data processing, and in particular relates to an intelligent analysis method and system for environmental behavior data. [Background technology]

[0002] With the advancement of informatization, all types of entities have massive business data, which contains a lot of unknown and potential information. Intelligent analysis based on big data has been widely used in intelligent manufacturing, services, banking, telecommunications, insurance, transportation, retail, environmental protection and other fields. It can extract auxiliary information to guide the correct and critical decision-making.

[0003] At the same time, while the Industrial Revolution is changing people's lifestyles, environmental protection issues involved in daily operations of enterprises have become particularly important. With the increase of environmental awareness among people around the world, environmental protection in corporate operations has become an issue that cannot be ignored. Environmental monitoring is an important foundation for environmental protection and management. With the rapid development of the economy, it is difficult for third parties or enterprises to scientifically control environmental problems arising from internal production and operations. There is a lack of real-time monitoring platforms. When environmental problems are discovered, they often cause significant losses to enterprises, including profits and reputation. On the other hand, there is a certain connection between the need for environmental protection and the dissemination of environmental protection. When enterprises and other users engage in environmental protection, it is inevitable that it will have a certain impact on environmental behaviors involving information acquisition and dissemination. Especially when the amount of data is aggregated to form the so-called big data, it is possible to discover usable information from the impact through intelligent analysis.

[0004] In summary, obtaining environmental data during environmental monitoring is an important basis for environmental protection and management. However, for enterprises or third parties, there is a practical difficulty in the process of environmental protection. It is often difficult to obtain environmental data, which requires a lot of manpower and material resources. There may even be passive protection situations where environmental protection should be initiated for environmental monitoring.

[0005] Therefore, we construct an intelligent analysis method based on environmental data and environmental communication behavior data, and then build an environmental protection control platform, so as to provide real-time scientific control for enterprise production and operation, provide scientific basis for enterprise managers to control environmental pollution, and timely feedback on whether the solutions to environmental problems are real and effective, thus becoming a support platform for controlling environmental pollution that serves enterprises, and also creates favorable conditions for enterprises to seek new environmental solutions, which can promote the sustainable development of enterprises. In addition, it can also help third parties discover problems in the first place, so as to conduct third-party monitoring and reminders, and then initiate offline monitoring and acquisition of environmental data, and realize a prevention mechanism that combines real-time online information monitoring with offline data-based early warning monitoring;

[0006] Based on the above problems, the present invention changes the traditional environmental monitoring methods through the application of information technology, uses communication network technology to implement long-term, continuous and effective monitoring of environmental protection behaviors, and scientifically, accurately, comprehensively and efficiently monitors and manages the environmental conditions of the area under its jurisdiction, so that the environmental management work of the environmental protection department can achieve scientific monitoring and efficient management; through loose propagation node selection and setting, it is not necessary for every resonant propagation node to be accurate and effective. Through the voting-based judgment mechanism, it supports effective environmental data deviation warning in a wide range of propagation nodes, thereby improving the efficiency of environmental protection work. [Summary of the invention]

[0007] In order to solve the above problems in the prior art, the present invention proposes an intelligent analysis method and system for environmental behavior data, the method comprising:

[0008] Step S1: Obtain an intelligent analysis request for a first object, determine a second object based on the first object, and obtain environmental data and environmental behavior data of the second object; wherein: the environmental data is process data generated by the object during the environmental protection process; the environmental behavior data is propagation behavior data of the environmental data;

[0009] Step S2: obtaining a data propagation graph including the first object and the second object; wherein: the data propagation graph is composed of propagation nodes and their data propagation relationships; taking some propagation nodes in the propagation graph as nodes to be determined; locating resonant nodes from the nodes to be determined, and dividing the resonant nodes into a first resonant node set involving a positive resonant relationship and a second resonant node set involving a reverse resonant relationship according to the consistency relationship between the propagation behavior data;

[0010] Step S3: Obtain historical propagation behavior data of the resonance node; determine the resonance relationship y=fi(x) between the propagation behavior data related to the resonance node and the second object environment data y, including: a forward resonance relationship y=f.i1(x) for the first resonance node set, and a reverse resonance relationship y=f.i2(x) for the first resonance node set; wherein: x is the propagation data type or a combination of propagation data types of the resonance node, referred to as propagation behavior data input x; i=i1 or i2; i1∈1~I1, i2∈1~I2; i1, i2 are element numbers in the first and second resonance node sets respectively;

[0011] Step S4: determining the resonance amplitude of the resonance node; and setting a discrimination weight based on the resonance amplitude; associating the resonance amplitude to indicate the degree of positive amplification or reverse amplification of the propagation behavior data input by the resonance node; when the bidirectional judgment target is sensitivity, setting a larger discrimination weight for the one with a higher resonance amplitude, and when the bidirectional judgment target is stability, setting a smaller discrimination weight for the one with a higher resonance amplitude;

[0012] Step S5: monitoring the i1th resonance node and the i2th resonance node to obtain the first object-related environmental behavior data in real time, and bidirectionally judging whether the environmental data of the first object has effectively changed based on the real-time environmental behavior data, the forward resonance relationship and the reverse resonance relationship. If so, proceed to the next step; otherwise, repeat the step;

[0013] The step S5 specifically includes the following steps:

[0014] Step S51: Monitor the i-th resonance node to obtain its real-time environmental behavior data input xi=(x.i1, ..., xi k ,...,xi K );

[0015] Step S52: based on the resonance relationship yi=fi(xi) of the i-th resonance node, the corresponding yi value is obtained, where the yi value is the y.i1 value obtained according to the forward resonance relationship, or the y.i2 value obtained according to the reverse resonance relationship;

[0016] Step S53: Use the positive discriminant value w i1 and the reverse discriminant w i2 Perform a two-way judgment. If the y.i1 value enters the default valid range of environmental data, the positive judgment value w is obtained. i1 ; If the y.i2 value enters the effective range of environmental data, the reverse discrimination value w is obtained i2 ; If the total positive discriminant value ∑ i1 w i1 Exceeding the positive discrimination threshold and the total reverse discrimination value ∑ i2 w i2 If the reverse discrimination threshold is exceeded, the process proceeds to step S6; otherwise, step S5 is repeated;

[0017] Step S6: Pushing environmental data interference information based on the effective changes.

[0018] Furthermore, the intelligent analysis request comes from the first object or a third party.

[0019] Furthermore, an intelligent analysis request is sent through a mobile terminal held by the first object or an early warning terminal held by a third party.

[0020] Furthermore, the first object and the second object are the same or different.

[0021] Furthermore, there are multiple valid changes, and the pushed environmental data interference information is related to the valid changes.

[0022] Furthermore, the positive discrimination threshold is greater than the reverse discrimination threshold.

[0023] An intelligent analysis system for environmental behavior data includes: the system is used to implement the above-mentioned intelligent analysis method for environmental behavior data.

[0024] An intelligent analysis platform for environmental behavior data, characterized in that it includes a processor, the processor is coupled to a memory, the memory stores program instructions, and when the program instructions stored in the memory are executed by the processor, the intelligent analysis method for environmental behavior data is implemented.

[0025] A computer-readable storage medium includes a program, which, when running on a computer, enables the computer to execute the intelligent analysis method for environmental behavior data.

[0026] An intelligent analysis server, characterized in that the intelligent analysis server is configured to execute the intelligent analysis method for environmental behavior data.

[0027] The beneficial effects of the present invention include:

[0028] (2) Discover the resonant correlation relationship in the correlation relationship, and then discover the possible changes in environmental data through the resonance relationship between the data of multiple propagation nodes and objects; improve the stability of the judgment through the collaborative judgment of multiple resonance nodes; then distinguish the relationship between positive and negative correlations, and perform discovery verification while discovering, so as to form a two-way prediction and judgment, and enhance the accuracy of the judgment of effective data changes; through loose resonance nodes and their intelligent analysis process, it is not necessary for each resonance node to be accurate and effective. Through the voting-based judgment mechanism, it supports effective environmental data deviation warning in a wide range of propagation nodes;

[0029] (2) While using the one-way data prediction brought by the resonance relationship based on the time axis to make judgments, the credibility of the judgment value is verified horizontally through the constraint relationship between the resonance nodes. The stability of intelligent analysis is further improved through the combination of horizontal and vertical, and the availability of communication behavior data based on big data is improved;

[0030] (3) The resonance amplitude supports the flexible and effective trade-off between the sensitivity and stability of the resonance node. On the basis of quantitative calculation, the discrimination value and the effective range are adjusted to support both discrete and continuous discrimination modes, and effective warning information can be obtained based on different trade-off factors.

Brief Description of the Drawings

[0031] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present application, but do not constitute an improper limitation of the present invention. In the drawings:

[0032] Figure 1 Schematic diagram of the intelligent analysis method for environmental behavior data provided by the present invention. [Specific implementation method]

[0033] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments, wherein the illustrative embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.

[0034] The present invention proposes an intelligent analysis method and system for environmental behavior data. Figure 1 As shown, the method comprises the following steps:

[0035] Step S1: obtaining an intelligent analysis request for a first object, determining a second object based on the first object, and obtaining environmental data and environmental behavior data of the second object; specifically: extracting the first object from the intelligent analysis request for the first object and obtaining the object attribute of the first object, obtaining the object closest to the object attribute of the first object as the second object; wherein: the environmental data is process data generated by the object during the environmental protection process; the environmental behavior data is propagation behavior data of the environmental data;

[0036] Preferably: the intelligent analysis request comes from the first object or a third party;

[0037] Preferably: sending an intelligent analysis request through a mobile terminal held by the first object or an early warning terminal held by a third party;

[0038] Preferably: the object is a subject with environmental execution capability such as an enterprise or an individual; the object attributes include one or more of object nature, number of people, composition, environmental execution capability, number of devices and their types, and hardware and software device composition;

[0039] Preferably: the first object and the second object are the same or different; when the environmental data reserve for the first object is insufficient, the first object and the second object are different; when the environmental data reserve for the first object is sufficient, the first object and the second object are the same;

[0040] Preferably: the environmental data includes online environmental data and / or offline environmental data; wherein: the online environmental data is environmental data of the first object collected by an environmental collection device; the environmental collection device includes: manual input device, network camera, environmental protection wearable device, environmental detection device; the offline environmental data includes the publicity tasks carried out by the object for environmental protection or the key focus tasks being guided, and the relevant data generated based on these tasks; for example: environmental information data, environmental risk treatment plan, external environmental responsibility report, distribution of release channels for environmental protection data, media distribution, regional distribution, user portraits, release cycle of various media, circulation, audience range, average daily visits of online users, degree of interaction of netizens, online user stickiness, etc.;

[0041] Alternatively: obtaining the environmental data of the second object through an environmental database; the environmental database is set in a big data server or an early warning terminal held by a third party;

[0042] Preferred: Environmental behavior data is defined as the communication data generated by the pro-environmental behavior of the communication subject; including: click-through rate, reprint rate, reading data, hot topic duration, number of readings, number of likes, number of comments, distribution of publishing channels, media distribution, regional distribution, user portrait and other data types;

[0043] Preferably: the communication data is data generated by the communication subject in new media and traditional media communication;

[0044] Step S2: obtaining a data propagation graph including the first object and the second object; taking some propagation nodes in the data propagation graph as nodes to be determined; locating resonant nodes from the nodes to be determined, and dividing the resonant nodes into a first resonant node set involving a positive resonant relationship, including the 1st to I1th resonant nodes; and a second resonant node set involving a reverse resonant relationship, including the 1st to I2th resonant nodes according to the consistency relationship between the propagation behavior data; wherein: I1 and I2 are the number of elements in the first resonant node set and the second resonant node set respectively; the data propagation graph is composed of propagation nodes and their data propagation relationships;

[0045] Preferably: the points in the propagation graph are propagation nodes, and the edges are propagation behavior relationships between propagation nodes; the propagation graph contains data source nodes for environmental data propagation;

[0046] Preferably: the data source node is one or more; the data source node may also be called an object node; and is also a member of the propagation chain;

[0047] The method of taking some propagation nodes in the propagation graph as nodes to be determined is as follows: if there is a node in the propagation graph, and the number of propagation paths (number of data propagation times) from the node to the first object and the second object per unit time is greater than a preset value (is equivalent), then the node is taken as a node to be determined;

[0048] Preferably: the unit time is 1d, 1m, etc.;

[0049] The method of locating the resonant node from the to-be-determined node is as follows: processing each to-be-determined node in turn, if there is a propagation behavior data type x in the to-be-determined node k Or a combination of multiple propagation behavior data types (x k1 , x k2 , ...) and the environmental data y of the second object have a positive correlation, then the resonance node and its corresponding propagation behavior data type or combination are placed in the first resonance node set; conversely, if the change relationship has an anti-correlation relationship, then the resonance node and its corresponding propagation behavior data type or combination are placed in the second resonance node set; where: k is the type number of the propagation behavior data;

[0050] Preferably: the positive correlation is a fuzzy positive correlation; that is, in most of the time (90% of the time), there is a positive correlation between the two, and it is considered that there is a positive correlation between the two; in this way, the correlation condition can be fuzzy, and more selectable resonance nodes and their propagation data types can be introduced; thereby supporting subsequent fuzzy judgment and early warning; the anti-correlation is similar; in the process of correlation judgment, the number of tuples of data in the propagation behavior data input can be increased to increase the stability of the correlation;

[0051] Step S3: Obtain historical propagation behavior data of the resonance node; determine the resonance relationship y=fi(x) between the propagation behavior data related to the resonance node and the second object and the second object environment data y; wherein: x is the propagation data type or a combination of propagation data types of the resonance node, referred to as propagation behavior data input x;

[0052] Preferably: the environmental behavior data is all or part of the environmental behavior data of the resonance node; further: the part of the environmental behavior data is environmental behavior data related to the first object and / or the second object;

[0053] Wherein: the environmental data dissemination behavior of the first object includes active dissemination behavior and passive dissemination behavior;

[0054] The step S3 specifically includes the following steps:

[0055] Step S31: sequentially obtain an unprocessed resonance node i, wherein: i=i1 or i2; i1∈1~I1, i2∈1~I2; i1 is the element number in the first resonance node set, and i2 is the element number in the second resonance node set;

[0056] Step S32: the propagation behavior data input x is a tuple or a multitudinous group about the propagation behavior data type; the resonance node and its propagation behavior data type or combination are screened for stability; different propagation data types or combinations of propagation data types are tried, and the propagation data type or combination of propagation data types with the strongest stability is selected as the determined propagation behavior data input x; obviously, for different resonance nodes, their resonance relationships and propagation behavior data input x are the same or different;

[0057] The stability screening is specifically as follows: in the process of judging the stability, each propagation behavior data type or type combination is judged respectively; firstly, the propagation behavior data type or type combination is segmented, and the correlation direction of each data segment is judged. The more data segments have the same correlation direction, the stronger the stability is; otherwise, the weaker the stability is; when judging the correlation direction of each data segment, the relationship between the mean of the data segment and the mean of the environmental data is used to reduce the complexity of the judgment;

[0058] Preferably: during the trial process, the propagation data and the environment data of the second object are sparsely sampled or intercepted to reduce the amount of trial calculation; of course, it can also be reduced by increasing the length of the data segment, but this may result in an inability to obtain consistent correlation conclusions in the same data segment;

[0059] Preferably: for different resonance nodes, the types of propagation behavior data involved are the same or different;

[0060] Step S33: sequentially process each resonance node i, and fit its propagation behavior data input x and the environmental data of the second object to obtain a resonance relationship y=fi(x) between the propagation behavior data input x and the environmental data y of the i-th resonance node; wherein the resonance relationship y=f.i1(x) for the first resonance node set and the reverse resonance relationship y=f.i2(x) for the first resonance node set are specifically subdivided into: a forward resonance relationship y=f.i1(x) for the first resonance node set;

[0061] Step S34: determine whether all resonant nodes have been processed, if yes, then end, otherwise, return to step S31;

[0062] Step S4: determining the resonance amplitude of the resonance node; and setting a discrimination weight based on the resonance amplitude; associating the resonance amplitude to indicate the degree of positive or negative amplification of the propagation behavior data input by the resonance node;

[0063] Preferably: when the resonance amplitude is larger, the associated amplification degree or amplification capacity is larger, and vice versa, it is smaller; however, it should be noted that this amplification is unstable, so it is necessary to perform collaborative discrimination of multiple resonance nodes;

[0064] The determining of the resonance amplitude of the resonance node specifically comprises: determining the resonance amplitude of the resonance node in a discrete manner or a continuous manner;

[0065] The discrete method for determining the resonance amplitude Ai of the resonance node i specifically includes the following steps:

[0066] Step S4D1: input the propagation behavior data and perform data segmentation to obtain one or more data segments ph; ph=1-PH; PH is the number of data segments; this segmentation is established on the time axis;

[0067] Step S4D2: For each data segment ph, find the y value of the phth data segment, i.e., y ph Mean of the values And find the increment Δx of the propagation behavior data input ph ; Δx ph.k =x ph.k -x k0 ; Where: k is the kth element in the propagation behavior data input x, K is the number of elements; k = 1~K, that is, the data of the kth propagation type; x k0 It is the minimum value of the change range of the data type of the kth propagation behavior, that is, the initial change value;

[0068] Step S4D3: Calculate the resonance amplitude Ai of the data segment ph ph ;

[0069] Alternative: y0 is the initial value of the environmental data, that is, the minimum value of its variation range;

[0070] Step S4D4: Obtain the resonance amplitude Ai of the i-th resonance node; Ai=(∑ ph Ai ph ) / PH;

[0071] The method of determining the resonance amplitude of the resonance node in a continuous manner specifically includes the following steps:

[0072] Step S4A1: Obtain the resonance relationship y=fi(x); wherein: i=i1 or i2; f=f.i1 or f.i2;

[0073] Step S4A2: Input the propagation behavior data of the resonance relationship x=(xk ) so that it changes from the initial value x within the range corresponding to the propagation behavior data type k k0 Starting from, change in increments of unit change, that is, each time the element value x is set after the change k =x k +Δx k , Δx k is 1 or multiple times of the unit change; the corresponding y value is obtained after the change according to the resonance relationship, y=y+Δy; the change of each element is simultaneous or non-simultaneous, that is, there may be multiple element values ​​corresponding to k values ​​changing at the same time; repeat the above process until the number of changes var covering the propagation behavior data input reaches the preset quantity value VAR; where: k is the kth tuple in the propagation behavior data input x, k=1~K, that is, the data of the kth propagation type;

[0074] Preferably, the preset number value is a preset value, for example, 1000 to 10000; it can be dynamically set according to the size of K value;

[0075] Alternatively: repeat the above process until the propagation behavior data input space is evenly covered; of course, the purpose of the above process is also to make the environmental data input evenly cover the data input space;

[0076] Step S4A3: taking the average value of the change in the y value as the resonance amplitude Ai of the resonance node i; specifically: calculating the resonance amplitude Ai based on the following formula (1); wherein: var=1~VAR is the number of changes in the propagation behavior data input;

[0077]

[0078] Alternative: Setting

[0079] Alternatively: Calculate the resonance amplitude Ai based on the following formula (2); where: α k is the normalization coefficient of the propagation behavior data type k;

[0080]

[0081] Preferably: the normalization coefficient is a preset value; and k α k =1; For example: set the weight coefficients of click rate and number of comments to 0.4 and 0.6 respectively;

[0082] The discrimination weight w of the i-th resonance node is set based on the resonance amplitude Ai iSpecifically, the same discrimination weight is set for each resonance node; the higher the sensitivity of the resonance node, the worse its stability. Therefore, when it is impossible to make an effective trade-off between sensitivity and stability, consider setting the same discrimination weight for these resonance nodes, and correspondingly, set a looser judgment range for balance in subsequent effective changes;

[0083] Alternatively: adopt method 1, set the discrimination weight w of the i-th resonance node based on the resonance amplitude Ai i , specifically: Set the resonance amplitude to the discrimination weight w i ; At this time, the advantage of considering sensitivity is weighed. When there are fewer resonant nodes and the range of variation of propagation behavior data is small, it is preferred to use this method to set the discrimination weight;

[0084] Alternatively: adopt method 2, set the discrimination weight w of the i-th resonance node based on the resonance amplitude Ai i Specifically: Set the value (1 / resonance amplitude) as the discrimination weight w i ; At this time, the advantage of stability is weighed. When there are many resonant nodes and the range of variation of the propagation behavior data is large, this method is used to set the discrimination weight;

[0085] Alternatively: the discrimination weight w of the i-th resonance node is set based on the resonance amplitude Ai i Specifically, when the mean of the resonance amplitude Ai is greater than the mean of the resonance amplitude, method 2 is adopted; and when the mean of the resonance amplitude Ai is less than or equal to the mean of the resonance amplitude, method 1 is adopted; wherein: the mean of the resonance amplitude is the historical mean obtained from its historical data, not the real-time mean;

[0086] Alternatively: Calculate the discrimination weight w based on the following formula (3): i ; It can be seen that at this time, the larger the resonance amplitude, the stronger the sensitivity of the resonance node, and a larger discrimination weight is set for it; otherwise, a smaller discrimination weight is set; where: c i is the weight conversion coefficient of the resonance node i; the weight conversion coefficient is a preset value and can be set based on the common consideration of other attributes of the resonance node;

[0087]

[0088] Alternatively: Calculate the discrimination weight w based on the following formula (4) or (5): i ;

[0089]

[0090] Furthermore, the discrimination weight w is calculated based on the following formula (6) or (7): i ;

[0091]

[0092] The above method of determining the resonance amplitude is based on a fitting function, and there is no additional restriction on whether the data is set centrally. The resonance amplitude can also be calculated based on historical propagation data and historical propagation behavior data;

[0093] Preferably: when the bidirectional judgment target is sensitivity, a larger discrimination weight is set for the one with a higher resonance amplitude, and when the bidirectional judgment target is stability, a smaller discrimination weight is set for the one with a higher resonance amplitude;

[0094] Step S5: monitoring the i1th resonance node and the i2th resonance node to obtain the first object-related environmental behavior data in real time, and bidirectionally judging whether the environmental data of the first object has effectively changed based on the real-time environmental behavior data, the forward resonance relationship and the reverse resonance relationship. If so, proceed to the next step; otherwise, repeat the step;

[0095] The step S5 specifically includes the following steps:

[0096] Step S51: Monitor the i-th resonance node to obtain the first object-related real-time environmental behavior data input xi=(x.i1, ..., xi k ,...,xi K );

[0097] Step S52: based on the resonance relationship yi=fi(xi) of the i-th resonance node, the corresponding yi value is obtained, where the yi value is the y.i1 value obtained according to the forward resonance relationship, or the y.i2 value obtained according to the reverse resonance relationship;

[0098] Step S53: If the value of y.i1 enters the default valid range of environmental data, the positive discrimination value w is obtained. i1 ; If the y.i2 value enters the effective range of environmental data, the reverse discrimination value w is obtained i2 ; If the total positive discriminant value ∑ i1 w i1 Exceeding the positive discrimination threshold and the total reverse discrimination value ∑ i2 w i2 If the reverse discrimination threshold is exceeded, the process proceeds to step S6; otherwise, step S5 is repeated. Obviously, when a default effective change occurs, it indicates that the environmental data may not meet expectations.

[0099] Preferably, before executing step S53, execute the following steps:

[0100] Step S531: Select the resonance amplitude Ai of the resonance node i determined in a discrete manner, and set the discrimination weights w of all resonance nodes.i1 =w i2 =1;

[0101] Step S532: Obtaining a default effective variation range of the environmental data;

[0102] Step S533: the resonance nodes are divided into 1st to I1st level resonance nodes and 1st to I2nd level resonance nodes; wherein: the 1st to I1st level resonance nodes and the 1st to I2nd level resonance nodes have different and non-opposite performances in the consistency of propagation behavior data, while the 1st to I1st level resonance nodes and the 1st to I2nd level resonance nodes have different and opposite performances in the consistency of propagation behavior data; the resonance amplitude gradually decreases from the 1st to I1st level and the 1st to I2nd level; the larger the resonance amplitude, the stronger the sensitivity, the worse the stability, and the lower the credibility;

[0103] Step S534: adjusting the default effective variation range according to the resonance amplitude Ai of each resonance node i; so that the adjustment degree gradually increases from level 1 to level I1 and level 1 to level I2; that is, the maximum adjustment degree can be set for level I1 and level I2, so that they correspond to a wider effective variation range on the basis of the default effective variation range;

[0104] Preferably: the forward discrimination threshold and the reverse discrimination threshold are both preset values; further: the forward discrimination threshold is greater than the reverse discrimination threshold; for example: the two are equal to 0.8 and 0.2 respectively;

[0105] Alternatively: based on the resonance relationship yi=fi(xi) of the i-th resonance node, obtain its corresponding yi value; if the yi value enters the effective variation range of the environmental data, obtain the discrimination value w i ; If the total discriminant value ∑ i w i If the voting threshold is exceeded, then go to step S6; otherwise, repeat step S5;

[0106] Preferably: the voting threshold is a preset value, such as if there are 10 resonance nodes, the voting threshold is set to 9;

[0107] Alternatively: based on the resonance relationship yi=fi(xi) of the i-th resonance node, obtain its corresponding yi value; if the yi value enters the effective variation range of the environmental data, obtain the discrimination value w i ; If the total discriminant value ∑ i w i If the voting threshold is exceeded, then go to step S6; otherwise, repeat step S5;

[0108] Furthermore, if the y.i1 value enters the effective range of the environmental data after adjustment, the positive discriminant value w is obtained. i1; If the y.i2 value enters the effective range of the environmental data after adjustment, the reverse discrimination value w is obtained i2 ; If the total positive discriminant value ∑ i1 w i1 Exceeding the positive discrimination threshold and the total reverse discrimination value ∑ i2 w i2 If the reverse discrimination threshold is exceeded, the process proceeds to step S6; otherwise, step S5 is repeated;

[0109] Preferably: before entering the next step S6, the constraint verification between the resonance nodes is first performed; the constraint verification is used to verify the constraint relationship between the environmental behavior data of the resonance nodes; this is actually to verify the credibility of the discrimination value horizontally through the constraint relationship between the resonance nodes before using the one-way data prediction of the resonance node for discrimination; when the constraint verification passes, enter the next step, otherwise, repeat step S5 or perform manual feedback reminder;

[0110] The execution of constraint verification between resonance nodes specifically includes the following steps:

[0111] Step S5A1: for the first resonance node, the ratio relationship between (1 / A.1i) is used as the first constraint relationship between nodes 1 to I1; for the second resonance node, the ratio relationship between (1 / A.12) is used as the second constraint relationship between nodes 1 to I2;

[0112] Step S5A2: If the ratio relationship between the real-time environmental behavior data x.i1 of the first resonance node satisfies the first constraint relationship and the ratio relationship between the real-time environmental behavior data x.i2 of the second resonance node satisfies the second constraint relationship, the constraint verification passes;

[0113] Alternatively, the step S5A2 is specifically as follows:

[0114] Step S5A21: For xi=(x.i1,...,xi k ,...,xi K ); for each environmental behavior data type k, make the following judgments in turn;

[0115] Judgment 1: x.1 k , ..., x.i1 k , ..., x.I1 k , the ratio relationship between them satisfies the first constraint relationship;

[0116] Judgment 2: x.1 k , ..., x.i2 k , ..., x.I2 k , the ratio relationship between them satisfies the second constraint relationship;

[0117] If judgment 1 and / or judgment 2 are satisfied, it is determined that the vertical constraint verification for the environmental behavior data type k is satisfied;

[0118] Step S5A22: After all environmental behavior data k are judged, further determine whether the constraint verification condition is satisfied. If yes, determine that the constraint verification is passed; otherwise, determine that the constraint verification is not passed.

[0119] The constraint verification condition is one or more of the following conditions:

[0120] Condition 1: The vertical constraint verification for specific environmental behavior data k is satisfied; for example, the constraint verification involving click rate and forwarding rate is satisfied;

[0121] Condition 2: The vertical constraint verification of specific environmental behavior data k exceeding the first agreed proportion (30%) is satisfied;

[0122] Condition 3: The vertical constraint verification of the specific environmental behavior data k is satisfied, and the constraint verification of the specific environmental behavior data k exceeding the second agreed proportion (50%) is satisfied;

[0123] Step S6: Pushing environmental data interference information based on the effective change; specifically: pushing the change in environmental data to the first object / the second object; if the default effective change of the environmental data of the first object exceeds the environmental standard parameter, issuing an alarm reminder;

[0124] Preferably: the effective change is a default effective change range;

[0125] Preferably, the environmental data interference information includes pushing environmental data changes to the first object / second object and an alarm reminder to a third party, reminding the third party to interfere with or pay attention to the environmental protection behavior of the first object;

[0126] Preferably: the default effective change is one or more; when the default effective change is more than one, a step setting is performed for them, and the default effective changes of the step setting are overlapping or non-overlapping;

[0127] Initiate acquisition of real-time environmental data of the first object based on the alarm reminder, and determine whether an actual effective change exceeds environmental standard parameters based on the real-time environmental data;

[0128] Preferably: the environmental standard parameters are ideal index values ​​of relevant environmental protection indicators, relevant environmental protection policies and guidelines at home and abroad, green procurement material information, physical conditions of workers, daily work-related injuries, wastewater discharge, waste gas discharge, particulate matter discharge, and environmental indicators of waste and recyclables;

[0129] Preferably: the environmental standard parameters are derived from domestic and foreign environmental protection policy standards;

[0130] Based on the same inventive concept, the present invention also provides an intelligent analysis system for environmental behavior data, the system comprising: an intelligent analysis server, a propagation node, a first object, a second object and a collection device;

[0131] Preferably: the first object and the second object both belong to propagation nodes;

[0132] The intelligent analysis server is used to implement the above-mentioned intelligent analysis method for environmental behavior data; the collection device is used to collect environmental data of the first object and the second object;

[0133] A computer program (also referred to as a program, software, software application, script, or code) can be written in any form of a programming language, including an assembled or interpreted language, a declarative or procedural language, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program, or in multiple collaborative files (e.g., files storing one or more modules, subroutines, or code portions). A computer program can be deployed to execute on one computer or on multiple computers located at one site or distributed across multiple sites and interconnected by a communication network.

[0134] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0135] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0136] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An intelligent analysis method for environmental behavior data, characterized in that: The method comprises: Step S1: Obtain an intelligent analysis request for a first object, determine a second object based on the first object, and obtain environmental data and environmental behavior data of the second object; wherein: the environmental data is process data generated by the object during the environmental protection process; the environmental behavior data is propagation behavior data of the environmental data; The step of obtaining a smart analysis request for the first object and determining the second object based on the first object includes: extracting the first object from the smart analysis request for the first object and obtaining an object attribute of the first object, and obtaining an object that is closest to the object attribute of the first object as the second object; Step S2: obtaining a data propagation graph including the first object and the second object; wherein: the data propagation graph is composed of propagation nodes and their data propagation relationships; taking some propagation nodes in the propagation graph as nodes to be determined; locating resonant nodes from the nodes to be determined, and dividing the resonant nodes into a first resonant node set involving a positive resonant relationship and a second resonant node set involving a reverse resonant relationship according to the consistency relationship between the propagation behavior data; The method of taking some of the propagation nodes in the propagation graph as nodes to be determined is as follows: if there is a node in the propagation graph, and the number of propagation paths from the node to the first object and the second object per unit time is greater than a preset value, then the node is taken as a node to be determined; The method of locating the resonant node from the to-be-determined node is as follows: processing each to-be-determined node in turn, if there is a propagation behavior data type x in the to-be-determined node k Or a combination of multiple propagation behavior data types (x k1 , x k2 , ...) and the environmental data y of the second object have a positive correlation, then the resonance node and its corresponding propagation behavior data type or combination are placed in the first resonance node set; conversely, if the change relationship has an anti-correlation relationship, then the resonance node and its corresponding propagation behavior data type or combination are placed in the second resonance node set; where: k is the type number of the propagation behavior data; Step S3: Obtain historical propagation behavior data of the resonance node; determine the resonance relationship y=fi(x) between the propagation behavior data related to the resonance node and the second object environment data y, including: a forward resonance relationship y=f.i1(x) for the first resonance node set, and a reverse resonance relationship y=f.i2(x) for the second resonance node set; wherein: x is the propagation behavior data type or a combination of propagation behavior data types of the resonance node, referred to as propagation behavior data input x; i=i1 or i2; i1∈1~I1, i2∈1~I2; i1 is the element number in the first resonance node set, and i2 is the element number in the second resonance node set; Step S4: determining the resonance amplitude of the resonance node; and setting a discrimination weight based on the resonance amplitude; associating the resonance amplitude to indicate the degree of positive or negative amplification of the propagation behavior data input by the resonance node; Step S5: monitoring the i1th resonance node and the i2th resonance node to obtain the first object-related environmental behavior data in real time, and bidirectionally judging whether the environmental data of the first object has changed effectively based on the environmental behavior data, the forward resonance relationship and the reverse resonance relationship, if yes, proceeding to the next step; otherwise, repeating this step; The step S5 specifically includes the following steps: Step S51: Monitor the i-th resonance node to obtain its first object-related environmental behavior data input xi=(x.i1, ..., xi k ,…,xi K ); Step S52: based on the resonance relationship yi=fi(xi) of the i-th resonance node, the corresponding yi value is obtained, where the yi value is the y.i1 value obtained according to the forward resonance relationship, or the y.i2 value obtained according to the reverse resonance relationship; Step S53: Use the positive discriminant value w i1 and the reverse discriminant value w i2 Perform a two-way judgment. If the y.i1 value enters the default valid range of environmental data, the positive judgment value w is obtained. i1 ; If the y.i2 value enters the effective range of environmental data, the reverse discrimination value w is obtained i2 ; If the total positive discriminant value ∑ i1 w i1 Exceeding the positive discrimination threshold and the total reverse discrimination value ∑ i2 w i2 If the reverse discrimination threshold is exceeded, the process proceeds to step S6; otherwise, step S5 is repeated; Before entering the next step S6, first perform constraint verification between the resonance nodes; the constraint verification is used to verify the constraint relationship between the environmental behavior data of the resonance nodes; when the constraint verification passes, enter the next step, otherwise, repeat step S5 or perform manual feedback reminder; The execution of constraint verification between resonance nodes specifically includes the following steps: Step S5A1: for the first resonance node, the ratio relationship between 1 / A.i1 is used as the first constraint relationship between nodes 1 to I1; for the second resonance node, the ratio relationship between 1 / A.i2 is used as the second constraint relationship between nodes 1 to I2, wherein A.i1 represents the resonance amplitude of node i1, and A.i2 represents the resonance amplitude of node i2; Step S5A2: If the ratio relationship between the real-time environmental behavior data x.i1 of the first resonance node satisfies the first constraint relationship and the ratio relationship between the real-time environmental behavior data x.i2 of the second resonance node satisfies the second constraint relationship, the constraint verification passes; Step S6: Pushing environmental data interference information based on the effective changes.

2. The intelligent analysis method for environmental behavior data according to claim 1 is characterized in that: The intelligent analysis request comes from the first object or a third party.

3. The intelligent analysis method for environmental behavior data according to claim 2 is characterized in that: An intelligent analysis request is sent through a mobile terminal held by the first object or an early warning terminal held by a third party.

4. The intelligent analysis method for environmental behavior data according to claim 3 is characterized in that: The first object and the second object are the same as or different from each other.

5. The intelligent analysis method for environmental behavior data according to claim 4 is characterized in that: There are multiple valid changes, and the pushed environmental data interference information is related to the valid changes.

6. The intelligent analysis method for environmental behavior data according to claim 5, characterized in that: The positive discrimination threshold is greater than the negative discrimination threshold.

7. An intelligent analysis platform for environmental behavior data, characterized in that: The method comprises a processor, wherein the processor is coupled to a memory, wherein the memory stores program instructions, and when the program instructions stored in the memory are executed by the processor, the intelligent analysis method for environmental behavior data described in any one of claims 1 to 6 is implemented.

8. A computer-readable storage medium, characterized in that: The invention comprises a program which, when running on a computer, enables the computer to execute the intelligent analysis method for environmental behavior data as described in any one of claims 1 to 6.

Images