Trading methods and systems based on artificial intelligence

Abstract

The present invention discloses an artificial intelligence-based transaction method and system, which relates to the field of transaction management technology, and solves the problem that the data features of transaction data stored in a certain space are not close, the data clustering is poor, and the security of the data cannot be guaranteed. The present invention arranges several groups of transaction data generated within a certain period, arranges the several groups of transaction data based on the numerical values ​​of the internal transaction amounts, and then determines the relevant arrangement curve. Based on the construction of relevant correlation lines, the arrangement curve is divided into several correlation line segments, and then based on the numerical display between each different correlation line segment, the relevant variance is confirmed, so as to select a string of data columns with the closest numerical features, so as to achieve a better feature differentiation result, so that the transaction data stored subsequently can achieve a better synchronous storage effect, and the features of the related transaction data stored in the same group of storage spaces are closest to each other.

Description

Technical Field

[0001] The present invention relates to the technical field of transaction management, and in particular to a transaction method and system based on artificial intelligence. Background Art

[0002] Transaction data management is a series of standardized activities around the collection, processing, storage and exchange of data assets, aiming to ensure the compliance and efficient circulation of data. In the era of digital economy, data has become a core strategic resource, and its superimposed, multiplied and amplified effects on production are becoming increasingly prominent, and has become a key element in the development of modern economy.

[0003] The transaction data generated by its related programs and related objects are generally too messy and too large in capacity. When its artificial intelligence manages such transaction data, it generally distinguishes and organizes the transaction data before and after different times based on the time relationship generated by the corresponding transaction data, so that related transaction data with closer timelines are stored in the same storage space. However, with this original processing method, the data characteristics of the transaction data within a certain space stored are not close, the data clustering is poor, and the security of the data cannot be guaranteed. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides an artificial intelligence-based trading method and system, which solves the problems that the data features of the transaction data stored within a certain space are not similar, the data clustering is poor, and the security of the data cannot be guaranteed.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a trading method based on artificial intelligence, comprising the following steps:

[0006] S1) Based on the preset artificial intelligence management rules, the transaction data generated by the specified object within the specified period is sorted, and the transaction data is classified by associated features based on the preset classification number G of artificial intelligence. The specific method is as follows:

[0007] S11. Confirm the transaction data generated by the specified object within the specified period T one by one, where T is a preset value, sort the transaction amounts within the confirmed transaction data from small to large, generate a data sequence, and randomly sort the same transaction data from front to back;

[0008] S12. Generate a transaction amount change curve according to different transaction amounts corresponding to different rankings in the data sequence, wherein the vertical coordinate axis of the change curve is the transaction amount and the horizontal coordinate axis is the relevant ranking;

[0009] S13. According to the set number of categories G, where G is a positive integer, the transaction amount change curve is divided into G associated line segments with equal characteristics, specifically in the following manner:

[0010] S131. Select a minimum amount and a maximum amount from the transaction amount change curve, and construct a boundary line of the transaction amount change curve based on the confirmed relevant amounts, wherein the boundary line is perpendicular to the vertical coordinate axis and intersects with the point where the maximum amount and the point where the minimum amount are located, wherein the minimum amount corresponds to the minimum boundary line, and the maximum amount corresponds to the maximum boundary line;

[0011] S132, constructing (G-1) association lines between the maximum boundary line and the minimum boundary line, and the (G-1) association lines are all parallel to the boundary lines, and the (G-1) association lines are arranged in sequence within the interval range of the two groups of boundary lines and are synchronously perpendicular to the vertical coordinate axis, so that the (G-1) association lines are sorted from bottom to top, and the (G-1) association lines are moved up and down, and the sorting position of the corresponding association lines remains unchanged during the movement;

[0012] S133, based on different moving processes, mark the segments between each group of correlation lines and between the correlation line and the maximum boundary line or the minimum boundary line as pending segments, determine the correlation variances of several groups of different transaction amounts associated with the pending segments, and mark the determined correlation variances as GF i-k , where i represents different pending line segments, and k represents different moving processes;

[0013] Then the G correlation variances GF generated by the G undetermined line segments in different moving processes are i-k Perform variance processing again to identify the total variance FF k , based on several different FFs generated by several different moving processes k , select the minimum value, calibrate the moving process corresponding to the minimum value as the standard process, and simultaneously calibrate the position of the correlation line;

[0014] S134, confirming the transaction data associated with the segments divided by the association lines one by one, and marking the confirmed transaction data belonging to the same segment as transaction data with the same characteristics, wherein the transaction data with the same characteristics includes the data crossed by the lower association lines, but does not include the data crossed by the upper association lines. The transaction data with the same characteristics corresponding to the uppermost association line includes not only the data crossed by the lower association lines but also the data crossed by the uppermost boundary lines. In such classification process, the maximum boundary line and the minimum boundary line are also regarded as association lines.

[0015] S2) Based on the confirmed G transaction data with the same characteristics, G groups of sub-spaces are constructed in the main space managed by artificial intelligence, and based on the constructed sub-spaces, the related transaction data with the same characteristics are stored;

[0016] S3) Based on the transaction data with the same characteristics stored in the relevant subspace, according to the space capacity of the relevant subspace, determine the characteristic value from the values ​​related to the space capacity, and then determine the hiding rule based on the characteristic value, and then hide the transaction amounts in the subspace one by one based on the determined hiding rule.

[0017] Preferably, in step S2, the specific storage capacity of each group of sub-spaces is determined by the relevant capacity of the transaction data with the same characteristics, and the relevant capacity of the confirmed transaction data with the same characteristics is T t , then the confirmed capacity-related data satisfies: R t ≥T t +Y1, where Y1 is the preset value, where R t is the storage capacity of the corresponding subspace, where t represents the relevant subspace.

[0018] Preferably, in step S3, the specific method of determining the characteristic value is:

[0019] Determine the space capacity of the relevant subspace and mark it as K t , where t represents different subspaces;

[0020] A group of related parameters ranked at the last position is selected from the confirmed space capacity of the corresponding subspace, and the related parameters are calibrated as characteristic values ​​of the corresponding subspace.

[0021] Preferably, in step S3, after the characteristic value is determined, the transaction data sequence with the same characteristic needs to be confirmed:

[0022] The different characteristic transaction data in the same space are sorted. The corresponding characteristic transaction data have different corresponding related times. The same characteristic transaction data with earlier related times are arranged in front, and the same characteristic transaction data with later related times are arranged in the back. If the related times of two groups of same characteristic transaction data are the same, they are randomly sorted front and back. After several groups of characteristic transaction data are sorted, the sequence of the same characteristic transaction data is confirmed.

[0023] Preferably, in step S3, the specific method of hiding the transaction amounts in the secondary space one by one is:

[0024] If the characteristic value confirmed by the relevant subspace is an even number, the transaction data sequence with the same characteristic in the relevant subspace is hidden front and back, and the transaction amount in the previous group of transaction data in the transaction data sequence is hidden to the position of the relevant transaction amount in the next group of transaction data, so that the transaction amounts in the adjacent transaction data are swapped, and the transaction amount of the last group of transaction data is swapped to the position of the transaction amount of the first group of transaction data, and so on, to complete the hiding process of the overall transaction amount of the transaction data sequence with the same characteristic;

[0025] If the characteristic value confirmed by the relevant subspace is an odd number, the transaction data sequence with the same characteristics within the relevant subspace is intermittently hidden, and the transaction amount in the h group of transaction data in the current transaction data sequence is hidden to the position of the transaction amount in the (h+2) group of transaction data, where h=1, 2, ..., n, where n represents the total number of transaction data in the transaction data sequence. When h is n-1, the transaction amount in the transaction data at position (n-1) is hidden to the position of the transaction amount in the first group of transaction data. When h is n, the transaction amount in the transaction data at position n is hidden to the position of the transaction amount in the second group of transaction data.

[0026] Preferably, the method further comprises the following steps:

[0027] S4), when the relevant operator extracts the transaction data with the same characteristics stored in the current space, the space capacity of the relevant secondary space is confirmed first, and the characteristic value is determined in the same way as in step S3, and then based on the characteristic value, the transaction data with the same characteristics in the current space is exchanged to obtain the original transaction data with the same characteristics; the specific method is:

[0028] S41, based on the space capacity of the sub-space to be extracted, determining the feature value ranked last among the relevant parameters within the space capacity;

[0029] S42. If the characteristic value is an even number, the transaction amount in the latter group of transaction data is transferred to the location of the transaction amount in the former group of transaction data, and the transaction amount in the first group of transaction data is transferred to the location of the transaction amount in the last group of transaction data, and so on, completing the transfer process of the entire transaction amount of the entire sequence of transaction data with the same characteristic;

[0030] S43. If the characteristic value is an odd number, determine the current group of transaction data, transfer the transaction amount in the second group of transaction data behind the current group of transaction data to the location of the transaction amount in the current group of transaction data, transfer the transaction amount in the first group of transaction data to the second to last group of transaction data, and so on, to complete the transfer process of the entire transaction amount of the entire sequence of transaction data with the same characteristic.

[0031] Preferably, the trading system based on artificial intelligence includes:

[0032] The data feature classification end confirms and integrates the transaction data generated by the specified object within the specified period, and then classifies the transaction data based on the preset number of classifications. After sorting the transaction data, the transaction amount change curve is confirmed, and then the transaction amount change curve is related to the construction of the correlation line. Based on its specific construction process, several correlation line segments with the same characteristics are confirmed, and then the related data of the correlation line segments are confirmed to lock the transaction data with the same characteristics;

[0033] The sub-space construction end stores the determined transaction data with the same characteristics, and the capacity of the corresponding sub-space during the storage process meets the required capacity of the transaction data with the same characteristics;

[0034] The sub-space data hiding end determines the characteristic value based on the space capacity of the relevant sub-space, and then hides the transaction amounts in the transaction data at different positions one by one based on the parity characteristics of the characteristic value, and replaces the transaction amounts of different transaction data within the sub-space to complete the hiding;

[0035] The spatial data extraction end is operated by relevant operators to extract the transaction data with the same characteristics stored in the secondary space, and decrypt the transaction data with the same characteristics one by one according to the reverse decryption method of the encryption method to obtain the corresponding transaction data with the same characteristics.

[0036] The present invention provides a trading method and system based on artificial intelligence. Compared with the prior art, it has the following beneficial effects:

[0037] The present invention arranges several groups of transaction data generated within a certain period, arranges the several groups of transaction data based on the numerical value of the internal transaction amount, and then determines the relevant arrangement curve. Subsequently, based on the construction of the relevant correlation line, the arrangement curve is divided into several correlation line segments, and then based on the numerical display between each different correlation line segment, the relevant variance is confirmed, so as to select a string of data columns with the closest numerical characteristics, so as to achieve a better feature differentiation result, so that the transaction data stored subsequently can achieve a better synchronous storage effect, and the characteristics of the related transaction data stored in the same group of storage space are closest to each other, so as to achieve a better storage management effect;

[0038] For the relevant transaction data stored in different storage sub-spaces, based on the confirmed characteristic values, the transaction amounts of the transaction data in different sub-spaces are hidden. By adopting a specific method of hiding one by one, the different transaction amounts in different storage sub-spaces can be fully hidden, so as to achieve a better encryption processing effect of the transaction data and fully guarantee the security of the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] See also Figure 1 , the present application provides a trading method based on artificial intelligence, comprising the following steps:

[0042] S1), based on the preset artificial intelligence management rules, the transaction data generated by the specified object within the specified period is sorted, and the transaction data is classified according to the associated features based on the preset number of categories of artificial intelligence. Specifically, the preset management rules include management periods and management objects, that is, the specified period is a set of preset periods, and its management object can be understood as a managed transaction APP or transaction party. The specific method of such artificial intelligence management is bound in advance by the relevant operators, and the transaction data generated by its designated objects are managed accordingly. The number of categories preset by artificial intelligence is the specified number, that is, its artificial intelligence needs to classify the transaction data to be managed into several categories based on the determined relevant number, and the number of categories needs to be input in advance by the relevant operators, and the number of categories is determined based on the relevant features of the transaction data generated by the associated objects. The specific method of classifying associated features is:

[0043] S11. Confirm the transaction data generated by the specified object within the specified period T one by one, where T is a preset value, generally 30 days, and its specific value is determined by the operator based on experience. Sort the transaction amounts within the confirmed transaction data from small to large to generate a data sequence, and randomly sort the same transaction data from front to back;

[0044] S12. Generate a transaction amount change curve based on different transaction amounts corresponding to different rankings in the data sequence, where the vertical coordinate axis of the change curve is the transaction amount and the horizontal coordinate axis is the related ranking. For example, if the data sequence is {5, 15, 20, 21, 54, 76}, then the amount corresponding to the first group of rankings is 5. Then, six groups of related coordinates can be confirmed by sequentially confirming them, namely: (1, 5), (2, 15), (3, 20), (4, 21), (5, 54) and (6, 76).

[0045] S13. According to the set number of categories G, where G is a positive integer and its specific value is determined in advance by the relevant operator, the transaction amount change curve is divided into G related line segments with equal characteristics (the equal characteristics here can be understood as having roughly the same characteristics and little difference, that is, the corresponding transaction amounts are relatively clustered relative to the overall transaction data):

[0046] S131. Select a minimum amount and a maximum amount from the transaction amount change curve, and construct a boundary line of the transaction amount change curve based on the confirmed relevant amounts, wherein the boundary line is perpendicular to the vertical coordinate axis and intersects with the point where the maximum amount and the point where the minimum amount are located, wherein the minimum amount corresponds to the minimum boundary line, and the maximum amount corresponds to the maximum boundary line;

[0047] S132. Construct (G-1) association lines between the maximum boundary line and the minimum boundary line, and the (G-1) association lines are all parallel to the boundary lines, and the (G-1) association lines are arranged in sequence within the interval range of the two groups of boundary lines and are synchronously perpendicular to the vertical coordinate axis, so that the (G-1) association lines are sorted from bottom to top, and the (G-1) association lines are moved up and down, and the sorting position of the corresponding association lines remains unchanged during the movement (for example: based on the determined boundary line, if G is 5, then it is necessary to divide the curve into five groups of related line segments with the same characteristics, so four groups of association lines are required, from bottom to top, the position of each association line is the first position, the second position, the third position and the fourth position, so when the association line moves up and down, its sorting position cannot change, in order to identify some line segments with similar amount characteristics);

[0048] S133, based on different moving processes, mark the segments between each group of correlation lines and between the correlation line and the maximum boundary line or the minimum boundary line as pending segments, determine the correlation variances of several groups of different transaction amounts associated with the pending segments, and mark the determined correlation variances as GF i-k , where i represents different pending line segments, and k represents different moving processes;

[0049] Then the G correlation variances GF generated by the G undetermined line segments in different moving processes are i-k Perform variance processing again to identify the total variance FF k , based on several different FFs generated by several different moving processes k, select the minimum value, calibrate the moving process corresponding to this minimum value as the standard process, and synchronously calibrate the position of the associated line. For example, if the associated line is moving, then during the moving process, the pending line segments between the corresponding associated lines will also change. Then, for each different change process, the variance of the corresponding pending line segments will also change. In order to obtain the best feature division method for several transaction data, the generated variance is subjected to data discrete analysis again (that is, variance confirmation). When the degree of data discrete analysis reaches the lowest (that is, when the total variance is the smallest), the classification effect is the most accurate, and a better classification processing effect can be achieved. The variance processing method is: the pending line segments are associated with several groups of transaction data, and the several groups of transaction data are subjected to mean processing to determine the relevant mean, and then the associated variances of several transaction data are confirmed based on the corresponding variance processing formula. Then, the G associated variances can be subjected to variance processing again to determine the subsequent total variance again.

[0050] S134, confirming the transaction data associated with the segments divided by the association lines one by one, and marking the confirmed transaction data belonging to the same segment as transaction data with the same characteristics, wherein the transaction data with the same characteristics includes the data crossed by the lower association lines, but does not include the data crossed by the upper association lines. The transaction data with the same characteristics corresponding to the uppermost association line includes not only the data crossed by the lower association lines but also the data crossed by the uppermost boundary lines. In such classification process, the maximum boundary line and the minimum boundary line are also regarded as association lines.

[0051] Specifically, for the transaction data between the associated line segments, assuming that there are two groups of associated line segments, plus the maximum boundary line and the minimum boundary line, there are four groups of associated lines. Then the transaction feature data between the lowest associated lines can be expressed as: [x1, x2), where x1 is the data intersected by the lowest associated lines, and x2 is the data intersected by the uppermost associated lines. Then, by analogy, the intervals of the four groups of transaction feature data generated can be expressed as: [x2, x3), [x3, x4) and [x4, x5]. The positions represented by the data do not represent the value range, that is, the data between the corresponding positions belong to transaction data of the same type of characteristics. For each type of related feature data of different characteristics, different subspaces are set to store such similar feature data in a related manner, and the specific feature classification process of the corresponding transaction data is completed, so that the feature data stored in the related classification interval is more clustered and the features are closer;

[0052] S2) Based on the confirmed G transaction data with the same characteristics, G groups of sub-spaces are constructed in the main space managed by artificial intelligence, and the related transaction data with the same characteristics are stored based on the constructed sub-spaces, and the specific storage capacity of each group of sub-spaces is determined by the related capacity of the transaction data with the same characteristics. The related capacity based on the confirmed transaction data with the same characteristics is Tt , then the confirmed capacity-related data satisfies: R t ≥T t +Y1, where Y1 is the preset value, where R t is the storage capacity of the corresponding subspace, where t represents the relevant subspace, and its specific value is determined by the operator based on experience;

[0053] S3), based on the transaction data with the same characteristics stored in the relevant subspace, according to the space capacity of the relevant subspace, determine the characteristic value from the values ​​related to the space capacity, then determine the hiding rule based on the characteristic value, and then hide the transaction amounts in the subspace one by one based on the determined hiding rule. Specifically, different subspaces have stored different transaction data with the same characteristics, so there are different space capacities, and their space capacities are the storage capacity of the corresponding subspace;

[0054] The specific method of determining the characteristic value is as follows: confirm the spatial capacity of the relevant subspace and mark it as K t , where t represents different subspaces;

[0055] A set of related parameters ranked at the last position is selected from the confirmed space capacity of the corresponding subspace, and the related parameters are calibrated as the characteristic values ​​of the corresponding subspace. For example, if the space capacity of the proposed corresponding subspace is 545M, and the related parameter ranked at the last position is 5, then the related parameter 5 is the characteristic value of this space;

[0056] Among them, different characteristic transaction data in the same space are sorted, and the corresponding characteristic transaction data have different corresponding relevant times. The same characteristic transaction data with earlier relevant times are arranged in front, and the same characteristic transaction data with later relevant times are arranged in the back. If the relevant time of two groups of same characteristic transaction data is the same, they are randomly sorted in front and back. After several groups of characteristic transaction data are sorted, the sequence of the same characteristic transaction data is confirmed. Specifically, the so-called relevant time is the generation time of the corresponding characteristic transaction data. During the generation process of the related characteristic transaction data, the corresponding transaction amount will be generated within the specified generation time. The specific relevant methods for hiding one by one are:

[0057] If the characteristic value confirmed by the relevant subspace is an even number, the transaction data sequence with the same characteristic in the relevant subspace is hidden front and back, and the transaction amount in the previous group of transaction data in the transaction data sequence is hidden to the position of the relevant transaction amount in the next group of transaction data, so that the transaction amounts in the adjacent transaction data are exchanged, and the transaction amount of the last group of transaction data is exchanged to the position of the transaction amount of the first group of transaction data, and so on, to complete the hiding process of the overall transaction amount of the transaction data sequence with the same characteristic. Specifically, assuming that the determined transaction data sequence with the same characteristic is {A, B, C, D, E, F}, where A is a group of transaction data and there is a group of related transaction amounts in the transaction data, the transaction amount in A is transferred to B data, the transaction amount of B data is transferred to C data, the transaction amount of C data is transferred to D data, the transaction amount of D data is transferred to E data, the transaction amount of E data is transferred to F data, and finally the transaction amount of F data is transferred to A data, to complete the overall hiding of the transaction amount of the transaction data sequence with the same characteristic;

[0058] If the characteristic value confirmed by the relevant subspace is an odd number, the transaction data sequence with the same characteristic in the relevant subspace is intermittently hidden, and the transaction amount in the h group of transaction data in the transaction data sequence is hidden to the position of the transaction amount in the (h+2) group of transaction data, where h=1, 2, ..., n, where n represents the total number of transaction data in the transaction data sequence. When h is n-1, the transaction amount in the transaction data at the position (n-1) is hidden to the position of the transaction amount in the first group of transaction data. When h is n, the transaction amount in the transaction data at the position n is hidden to the position of the transaction amount in the second group of transaction data. For example: assuming that the determined transaction data sequence with the same characteristic is {A, B, C, D, E, F}, then the relevant transaction amount in A is hidden in C, the transaction amount in B is hidden in D, the relevant transaction amount in C is hidden in E, the relevant transaction amount in D is hidden in F, the relevant transaction amount in E is hidden in A, and the relevant transaction amount in F is hidden in B;

[0059] Specifically, after each different sub-space hides the transaction amounts of different locations one by one, the relevant transaction amounts of each different transaction data in the sub-space will be changed accordingly. Then the relevant transaction data within this space is inaccurate. Subsequently, to extract such transaction data, it is necessary to perform the relevant decryption process to decrypt such transaction data.

[0060] S4), when the relevant operator extracts the transaction data with the same characteristics stored in the current space, the space capacity of the relevant secondary space is confirmed first, and the characteristic value is determined in the same way as in step S3, and then based on the characteristic value, the transaction data with the same characteristics in the current space is exchanged to obtain the original transaction data with the same characteristics, wherein the specific method of data exchange is:

[0061] S41, based on the space capacity of the sub-space to be extracted, determining the feature value ranked last among the relevant parameters within the space capacity;

[0062] S42. If the characteristic value is an even number, the transaction amount in the latter group of transaction data is transferred to the location of the transaction amount in the former group of transaction data, and the transaction amount in the first group of transaction data is transferred to the location of the transaction amount in the last group of transaction data, and so on, completing the transfer process of the entire transaction amount of the entire sequence of transaction data with the same characteristic;

[0063] S43. If the characteristic value is an odd number, determine the current group of transaction data, transfer the transaction amount in the second group of transaction data behind the current group of transaction data to the location of the transaction amount in the current group of transaction data, transfer the transaction amount in the first group of transaction data to the second to last group of transaction data, and so on, to complete the entire transaction amount transfer process of the entire sequence of transaction data with the same characteristic;

[0064] Specifically, the decryption process is the reverse decryption method of the above encryption process, so it will not be elaborated here. When the transaction amount is moving and changing, there is a change mark at the relevant position, which means that the transaction amount here has undergone relevant changes. The transfer position of the transaction amount can be determined based on the change mark, which can better transfer and decrypt the transaction data sequence.

[0065] Example 2

[0066] In the specific implementation process, the present application also includes an artificial intelligence-based trading system for executing the corresponding trading method, and the trading system includes:

[0067] The data feature classification end confirms and integrates the transaction data generated by the specified object within the specified period, and then classifies the transaction data based on the preset number of classifications. After sorting the transaction data, the transaction amount change curve is confirmed, and then the transaction amount change curve is related to the construction of the correlation line. Based on its specific construction process, several correlation line segments with the same characteristics are confirmed, and then the related data of the correlation line segments are confirmed to lock the transaction data with the same characteristics;

[0068] The sub-space construction end stores the determined transaction data with the same characteristics, and the capacity of the corresponding sub-space during the storage process meets the required capacity of the transaction data with the same characteristics;

[0069] The sub-space data hiding end determines the characteristic value based on the space capacity of the relevant sub-space, and then hides the transaction amounts in the transaction data at different positions one by one based on the parity characteristics of the characteristic value, and replaces the transaction amounts of different transaction data within the sub-space to complete the hiding;

[0070] The spatial data extraction end is operated by relevant operators to extract the transaction data with the same characteristics stored in the secondary space, and decrypt the transaction data with the same characteristics one by one according to the reverse decryption method of the encryption method to obtain the corresponding transaction data with the same characteristics.

[0071] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0072] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A trading method based on artificial intelligence, characterized in that: The following steps are involved: S1) Based on the preset artificial intelligence management rules, the transaction data generated by the specified object within the specified period is sorted, and the transaction data is classified by associated features based on the preset classification number G of artificial intelligence. The specific method is as follows: S11. Confirm the transaction data generated by the specified object within the specified period T one by one, where T is a preset value, sort the transaction amounts within the confirmed transaction data from small to large, generate a data sequence, and randomly sort the same transaction data from front to back; S12. Generate a transaction amount change curve according to different transaction amounts corresponding to different rankings in the data sequence, wherein the vertical coordinate axis of the change curve is the transaction amount and the horizontal coordinate axis is the relevant ranking; S13. According to the set number of categories G, where G is a positive integer, the transaction amount change curve is divided into G associated line segments with equal characteristics, specifically in the following manner: S131. Select a minimum amount and a maximum amount from the transaction amount change curve, and construct a boundary line of the transaction amount change curve based on the confirmed relevant amounts, wherein the boundary line is perpendicular to the vertical coordinate axis and intersects with the point where the maximum amount and the point where the minimum amount are located, wherein the minimum amount corresponds to the minimum boundary line, and the maximum amount corresponds to the maximum boundary line; S132. Construct (G-1) association lines between the maximum boundary line and the minimum boundary line, and the (G-1) association lines are all parallel to the boundary lines, and the (G-1) association lines are arranged in sequence within the interval range of the two groups of boundary lines and are synchronously perpendicular to the vertical coordinate axis, so that the (G-1) association lines are sorted from bottom to top, and the (G-1) association lines are moved up and down, and the sorting position of the corresponding association lines remains unchanged during the movement; S133, based on different moving processes, mark the segments between each group of correlation lines and between the correlation line and the maximum boundary line or the minimum boundary line as pending segments, determine the correlation variances of several groups of different transaction amounts associated with the pending segments, and mark the determined correlation variances as GF i-k , where i represents different pending line segments, and k represents different moving processes; Then the G correlation variances GF generated by the G undetermined line segments in different moving processes are i-k Perform variance processing again to identify the total variance FF k , based on several different FFs generated by several different moving processes k , select the minimum value, calibrate the moving process corresponding to the minimum value as the standard process, and simultaneously calibrate the position of the correlation line; S134, confirming the transaction data associated with the segments divided by the association lines one by one, and marking the confirmed transaction data belonging to the same segment as transaction data with the same characteristics, wherein the transaction data with the same characteristics includes the data crossed by the lower association lines, but does not include the data crossed by the upper association lines. The transaction data with the same characteristics corresponding to the uppermost association line includes not only the data crossed by the lower association lines but also the data crossed by the uppermost boundary lines. In such classification process, the maximum boundary line and the minimum boundary line are also regarded as association lines. S2) Based on the confirmed G transaction data with the same characteristics, G groups of sub-spaces are constructed in the main space managed by artificial intelligence, and based on the constructed sub-spaces, the related transaction data with the same characteristics are stored; S3) Based on the transaction data with the same characteristics stored in the relevant subspace, according to the space capacity of the relevant subspace, determine the characteristic value from the values ​​related to the space capacity, and then determine the hiding rule based on the characteristic value, and then hide the transaction amounts in the subspace one by one based on the determined hiding rule.

2. The artificial intelligence-based trading method according to claim 1, characterized in that: In step S2, the specific storage capacity of each group of sub-spaces is determined by the relevant capacity of the transaction data with the same characteristics. The relevant capacity of the confirmed transaction data with the same characteristics is T t , then the confirmed capacity-related data satisfies: R t ≥T t +Y1, where Y1 is the preset value, where R t is the storage capacity of the corresponding subspace, where t represents the relevant subspace.

3. The artificial intelligence-based trading method according to claim 1, characterized in that: In step S3, the specific method of determining the characteristic value is: Determine the space capacity of the relevant subspace and mark it as K t , where t represents different subspaces; A group of related parameters ranked at the last position is selected from the confirmed space capacity of the corresponding subspace, and the related parameters are calibrated as characteristic values ​​of the corresponding subspace.

4. The artificial intelligence-based trading method according to claim 3, characterized in that: In step S3, after the characteristic value is determined, the transaction data sequence with the same characteristic needs to be confirmed: The different characteristic transaction data in the same space are sorted. The corresponding characteristic transaction data have different corresponding related times. The same characteristic transaction data with earlier related times are arranged in front, and the same characteristic transaction data with later related times are arranged in the back. If the related times of two groups of same characteristic transaction data are the same, they are randomly sorted front and back. After several groups of characteristic transaction data are sorted, the sequence of the same characteristic transaction data is confirmed.

5. The artificial intelligence-based trading method according to claim 4, characterized in that: In step S3, the specific method of hiding the transaction amounts in the secondary space one by one is: If the characteristic value confirmed by the relevant subspace is an even number, the transaction data sequence with the same characteristic in the relevant subspace is hidden front and back, and the transaction amount in the previous group of transaction data in the transaction data sequence is hidden to the position of the relevant transaction amount in the next group of transaction data, so that the transaction amounts in the adjacent transaction data are swapped, and the transaction amount of the last group of transaction data is swapped to the position of the transaction amount of the first group of transaction data, and so on, to complete the hiding process of the overall transaction amount of the transaction data sequence with the same characteristic; If the characteristic value confirmed by the relevant subspace is an odd number, the transaction data sequence with the same characteristic within the relevant subspace is hidden discontinuously, and the transaction amount in the h group of transaction data in the current transaction data sequence is hidden to the position of the transaction amount in the (h+2) group of transaction data, where h=1, 2, ..., n, where n represents the total number of transaction data in the transaction data sequence. When h is n-1, the transaction amount in the transaction data at position (n-1) is hidden to the position of the transaction amount in the first group of transaction data. When h is n, the transaction amount in the transaction data at position n is hidden to the position of the transaction amount in the second group of transaction data.

6. The artificial intelligence-based trading method according to claim 5, characterized in that: The following steps are also included: S4), when the relevant operators extract the transaction data with the same characteristics stored in this space, they first confirm the space capacity of the relevant sub-space, determine the characteristic value in the same way as in step S3, and then based on the characteristic value, exchange the transaction data with the same characteristics in this space to obtain the original transaction data with the same characteristics.

7. The artificial intelligence-based trading method according to claim 6, characterized in that: In step S4, the specific method of performing data exchange is: S41, based on the space capacity of the sub-space to be extracted, determining the feature value ranked last among the relevant parameters within the space capacity; S42. If the characteristic value is an even number, the transaction amount in the latter group of transaction data is transferred to the location of the transaction amount in the former group of transaction data, and the transaction amount in the first group of transaction data is transferred to the location of the transaction amount in the last group of transaction data, and so on, completing the transfer process of the entire transaction amount of the entire sequence of transaction data with the same characteristic; S43. If the characteristic value is an odd number, determine the current group of transaction data, transfer the transaction amount in the second group of transaction data behind the current group of transaction data to the location of the transaction amount in the current group of transaction data, transfer the transaction amount in the first group of transaction data to the second to last group of transaction data, and so on, to complete the transfer process of the entire transaction amount of the entire sequence of transaction data with the same characteristic.

8. A trading system based on artificial intelligence, which operates according to the trading method according to any one of claims 1 to 7, characterized in that: include: The data feature classification end confirms and integrates the transaction data generated by the specified object within the specified period, and then classifies the transaction data based on the preset number of classifications. After sorting the transaction data, the transaction amount change curve is confirmed, and then the transaction amount change curve is related to the construction of the correlation line. Based on its specific construction process, several correlation line segments with the same characteristics are confirmed, and then the related data of the correlation line segments are confirmed to lock the transaction data with the same characteristics; The sub-space construction end stores the determined transaction data with the same characteristics, and the capacity of the corresponding sub-space during the storage process meets the required capacity of the transaction data with the same characteristics; The sub-space data hiding end determines the characteristic value based on the space capacity of the relevant sub-space, and then hides the transaction amounts in the transaction data at different positions one by one based on the parity characteristics of the characteristic value, and replaces the transaction amounts of different transaction data within the sub-space to complete the hiding; The spatial data extraction end is operated by relevant operators to extract the transaction data with the same characteristics stored in the secondary space, and decrypt the transaction data with the same characteristics one by one according to the reverse decryption method of the encryption method to obtain the corresponding transaction data with the same characteristics.

Images