CN116502107A - Data intelligent operation platform supply chain data processing system based on artificial intelligence - Google Patents
Data intelligent operation platform supply chain data processing system based on artificial intelligence Download PDFInfo
- Publication number
- CN116502107A CN116502107A CN202310742637.7A CN202310742637A CN116502107A CN 116502107 A CN116502107 A CN 116502107A CN 202310742637 A CN202310742637 A CN 202310742637A CN 116502107 A CN116502107 A CN 116502107A
- Authority
- CN
- China
- Prior art keywords
- data
- supply chain
- clustering
- sparse
- cluster
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012545 processing Methods 0.000 title claims abstract description 37
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 44
- 230000000875 corresponding effect Effects 0.000 claims description 118
- 230000010354 integration Effects 0.000 claims description 83
- 238000009826 distribution Methods 0.000 claims description 42
- 230000002596 correlated effect Effects 0.000 claims description 30
- 238000013507 mapping Methods 0.000 claims description 5
- 238000007621 cluster analysis Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 17
- 239000004973 liquid crystal related substance Substances 0.000 description 6
- 238000003860 storage Methods 0.000 description 6
- 238000010606 normalization Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000006835 compression Effects 0.000 description 3
- 238000007906 compression Methods 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 239000002994 raw material Substances 0.000 description 3
- 238000013519 translation Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000013068 supply chain management Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention relates to the technical field of data processing, in particular to a data intelligent operation platform supply chain data processing system based on artificial intelligence. The encryption method and the encryption device have the advantage that the encryption effect on the supply chain data is better in the process of processing the supply chain data by integrating the data clustering set.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a data processing system of a data intelligent operation platform supply chain based on artificial intelligence.
Background
Along with the continuous development of the artificial intelligence technical field, the existing artificial intelligence technology can analyze and process data, and assist enterprises to make decisions, such as supply chain data, according to analysis and processing results, the intelligent data operation platform based on artificial intelligence analyzes supply and demand conditions according to the supply chain data, and production is reasonably arranged according to the supply and demand conditions so as to improve production planning efficiency. However, due to the continuous expansion of the data scale of the data platform based on artificial intelligence and the diversification of analysis scenes, the requirements on data processing and storage are higher, for example, encryption processing is required in the process of storing the supply chain data reflecting private information such as supply requirements.
In the prior art, the encryption processing method for data is to replace or scramble the data to hide the original data information, which is essentially to perform indifferent processing on the data according to the rule of the algorithm itself, but the indifferent processing has different encryption effects on the data represented by different information, for example, the privacy information reflected by the supply chain data is mainly represented by the time distribution relationship of the supply chain data, if the supply chain data is replaced or scrambled only according to the encryption processing method in the prior art, the hiding effect on the time distribution relationship of the supply chain data is more random, and the encryption effect on the supply chain data cannot be ensured.
Disclosure of Invention
In order to solve the technical problem that the encryption effect on the supply chain data cannot be ensured because the generated hiding effect of the data time distribution relation is random when the encryption processing method in the prior art is used for encrypting the supply chain data, the invention aims to provide a supply chain data processing system of a data intelligent operation platform based on artificial intelligence, which adopts the following technical scheme:
the invention provides a data intelligent operation platform supply chain data processing system based on artificial intelligence, which comprises:
the system comprises a supply chain data acquisition module, a data intelligent operation platform and a control module, wherein the supply chain data acquisition module is used for acquiring at least two kinds of supply chain data in a preset time range;
the integration necessity acquisition module is used for carrying out cluster analysis according to the time distribution density characteristics corresponding to the supply chain data and dividing all the supply chain data into at least two sparse data cluster sets; obtaining the concentration degree of each supply chain data in each sparse data clustering set according to the integral time difference characteristic, the time difference change characteristic and the density of each supply chain data in the sparse data clustering set; obtaining integration necessity between any two sparse data clustering sets according to the variety and quantity difference, the time length difference and the concentration degree difference of each supply chain data in any two sparse data clustering sets;
The supply chain data processing module is used for integrating the sparse data clustering sets according to the integration necessity to obtain at least two integrated data clustering sets; and carrying out encryption processing on the supply chain data according to the integrated data clustering set.
Further, the method for acquiring the sparse data clustering set comprises the following steps:
mapping the supply chain data in a preset time range onto a time axis to obtain a supply chain data time axis; and taking a starting point of a supply chain data time axis as a starting center point, obtaining a drift vector of each drift window according to a mean shift clustering algorithm, and carrying out drift clustering according to the inverted drift vector based on the mean shift clustering algorithm to obtain at least two sparse data clustering sets.
Further, the method for acquiring the concentration degree comprises the following steps:
acquiring a corresponding time length of each sparse data clustering set, selecting one sparse data clustering set as a target sparse data clustering set, taking a clustering center of the target sparse data clustering set as a target clustering center, and obtaining the density of each supply chain data in the target sparse data clustering set according to the data quantity of each supply chain data in the target coefficient data clustering set and the corresponding time length, wherein the data quantity is positively correlated with the density, and the time length is negatively correlated with the density;
In the target sparse data clustering set, taking the average value of the distance between each data in each supply chain data and the target clustering center as the integral time difference characteristic of each supply chain data in the target data set;
in the target sparse data clustering set, taking the standard deviation of the distance between each data in each supply chain data and the target clustering center as the time difference change characteristic of each supply chain data in the target data set;
and obtaining the concentration degree of each supply chain data in the target data set according to the density, the integral time difference characteristic and the time difference change characteristic, wherein the density and the concentration degree are positively correlated, the integral time difference characteristic and the concentration degree are negatively correlated, and the time difference change characteristic and the concentration degree are negatively correlated.
Further, the method for acquiring the integration necessity comprises the following steps:
selecting any one sparse data clustering set as a target sparse data clustering set, and taking any other sparse data clustering set as a comparison sparse data clustering set; taking the difference of the category number of the supply chain data between the target sparse data clustering set and the comparison sparse data clustering set as a category number difference characteristic;
Acquiring the time length of each sparse data clustering set; taking the difference of the time length between the target sparse data clustering set and the comparison sparse data clustering set as a time length difference characteristic;
calculating the concentration degree difference of each supply chain data between the target sparse data clustering set and the comparison sparse data clustering set, and taking the accumulated sum of all the concentration degree differences corresponding to the target sparse data clustering set and the comparison sparse data clustering set as the corresponding concentration degree integral difference characteristic;
and obtaining integration necessity between a target sparse data clustering set and a comparison sparse data clustering set according to the category number difference feature, the time length difference feature and the concentration degree overall difference feature, wherein the category number difference feature is positively correlated with the integration necessity, the time length difference feature is positively correlated with the integration necessity, and the concentration degree overall difference feature is positively correlated with the integration necessity.
Further, the method for acquiring the integrated data cluster set comprises the following steps:
counting all integration necessity corresponding to all sparse data cluster sets, taking two sparse data cluster sets with the largest integration necessity as a first cluster set to be integrated, and taking other sparse data cluster sets except the first cluster set to be integrated as a first non-integrated cluster set; integrating two sparse data cluster sets corresponding to the two first cluster sets to be integrated into one integrated data cluster set;
Taking the two first unconformity cluster sets with the largest integration necessity as a second cluster set to be integrated, and taking the other first unconformity cluster sets except the second sparse data cluster set as a second unconformity cluster set; integrating two sparse data cluster sets corresponding to the two second cluster sets to be integrated into one integrated data cluster set;
and analogizing to obtain at least two integrated data clustering sets;
when only one sparse data cluster set is left in the final non-integrated sparse data cluster set, the rest sparse data cluster set is used as an integrated data cluster set.
Further, the integrating the two sparse data cluster sets into one integrated data cluster set includes:
the supply chain data with earliest corresponding moment in each sparse data clustering set is used as starting point data corresponding to each sparse data clustering set; taking the time distance between the two to-be-integrated cluster sets and the corresponding starting point data as the characteristic distance;
on a time axis corresponding to the preset time range of all the supply chain data, when the time lengths corresponding to the two to-be-integrated clustering sets are equal, shifting any one to-be-integrated clustering set data to the other to-be-integrated clustering set direction by a corresponding characteristic distance length to obtain an integrated data clustering set; when the time lengths corresponding to the two to-be-integrated cluster sets are unequal, shifting the to-be-integrated cluster set with the shortest time length to the to-be-integrated cluster set with the longest time length by the corresponding characteristic distance length to obtain an integrated data cluster set.
Further, the method for acquiring the time length of the sparse data clustering set comprises the following steps:
and taking the extremely poor time corresponding to the supply chain data in each sparse data clustering set as the time length of each sparse data clustering set.
The invention has the following beneficial effects:
in consideration of the fact that privacy data in supply chain data show a time distribution relationship, the method integrates the sparse data clustering set representing time sequence continuous supply chain data to obtain the integrated data clustering set, the distribution relationship of original data on a time sequence is destroyed through the integration process, so that the hiding effect of the time distribution relationship of the supply chain data is better, the time interval corresponding to the integrated data is shorter, the encryption effect on the supply chain data is better, and meanwhile data storage is facilitated. In consideration of the fact that when the difference between the sparse data clustering sets is larger, the degree of damage of the integrated data clustering sets to the distribution relation on the time sequence corresponding to the original data is larger, therefore, the integration necessity is obtained by analyzing the distribution difference between any two sparse data clustering sets, the sparse data clustering sets are subjected to integration judgment through the integration necessity, the two sparse data clustering sets corresponding to the obtained integrated data clustering sets are the set with the largest distribution difference, and the subsequent encryption effect on the supply chain data is improved. The invention is based on the time distribution density characteristic corresponding to the supply chain data when acquiring the sparse data clustering sets, so that the data time aggregation difference in each sparse data clustering set is larger, the expression degree of the original supply chain data is reduced, the hiding effect on the time distribution relation of the supply chain data is improved, and the encryption effect on the supply chain data is better. In summary, the encryption effect on the supply chain data is better in the process of integrating the data clustering set to process the supply chain data.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of an artificial intelligence based data processing system for a data intelligent operation platform supply chain according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description refers to the specific implementation, structure, characteristics and effects of an artificial intelligence-based data intelligent operation platform supply chain data processing system according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a specific scheme of a data intelligent operation platform supply chain data processing system based on artificial intelligence, which is specifically described below with reference to the accompanying drawings.
Referring now to FIG. 1, shown is a block diagram illustrating a system architecture for providing an artificial intelligence based data intelligence operation platform supply chain data processing system in accordance with one embodiment of the present invention, the system comprising: a supply chain data acquisition module 101, an integration necessity acquisition module 102, and a supply chain data processing module 103.
The supply chain data acquisition module 101 is configured to acquire at least two kinds of supply chain data within a preset time range on the data intelligent operation platform.
The embodiment of the invention aims to provide a supply chain data processing system of a data intelligent operation platform based on artificial intelligence, which utilizes a clustering method based on sparse distribution to acquire a sparse data clustering set of supply chain data on time sequence, further analyzes the sparse data clustering set, integrates the two-by-two sparse data clustering sets to obtain an integrated data clustering set, thereby destroying the time distribution relation of the supply chain data, and further carries out encryption processing on the supply chain data according to the integrated data clustering set to realize efficient encryption of the supply chain data. It is first necessary to acquire a data processing object corresponding to an embodiment of the present invention.
The embodiment of the invention firstly obtains at least two kinds of supply chain data within a preset time range from a data intelligent operation platform. In the embodiment of the present invention, the preset time range is set to 1 day. Since supply chain data can reflect the supply of actual production demands, there are typically a variety of supply chain data that need to be encrypted, such as: in the supply chain data corresponding to the supply amounts or output amounts of different raw materials, the supply amount of each raw material corresponds to one type of supply chain data, and similarly, the output rate of each raw material corresponds to one type of supply chain data, and when the industrial scale is larger, the corresponding supply chain data is more in variety. It should be noted that, when the supply chain data is encrypted later, the type tag of the supply chain data needs to be combined; when only one type of supply chain data exists in the preset time range set in the embodiment of the present invention, the preset time range needs to be further expanded until at least two types of acquired supply chain data exist, and the implementer can set the size of the preset time range according to the specific implementation environment, which is not further described herein.
The integration necessity obtaining module 102 is configured to perform cluster analysis according to the time distribution density features corresponding to the supply chain data, and divide all the supply chain data into at least two sparse data cluster sets; obtaining the concentration degree of each supply chain data in each sparse data clustering set according to the integral time difference characteristic, the time difference change characteristic and the density of each supply chain data in the sparse data clustering set; and obtaining the integration necessity between any two sparse data clustering sets according to the difference of the variety and the number of the supply chain data, the difference of the time length and the difference of the concentration degree of each supply chain data in any two sparse data clustering sets.
For an industrial chain, supply demand information is very important, and the whole operation state of the industrial chain can be reflected, so that strict protection is required. The supply chain data includes supply demand information, which is information reflected by the supply chain data, and therefore, the supply chain data needs to be stored in an encrypted manner. The supply chain data are obtained in time sequence, namely, each supply chain data corresponds to one time data, namely, the time sequence is formed by the supply chain data in time sequence, if the supply chain data are formed into the time sequence in time sequence, the time relation among the supply chain data can reflect the time distribution relation of the original supply chain data in the corresponding time sequence, so that if the supply chain data need to be encrypted, the time sequence distribution relation of the supply chain data needs to be destroyed, and the more the supply chain data are distributed in the time sequence, the better the corresponding encryption effect is.
It is contemplated that in actual supply chain management, the time and size of supply and production can reflect the supply chain relationship. Because the time relation between the corresponding supply and output data of the supply chain can be expressed in time sequence, the embodiment of the invention only encrypts the supply chain data according to the time sequence relation of the data in the supply chain, and the more similar the corresponding supply chain data in time sequence, the more outstanding the corresponding supply chain relation expression, so that the further analysis can be carried out according to the time distance distribution condition of the supply chain data in time sequence. According to the embodiment of the invention, cluster analysis is carried out according to the time distribution density characteristics corresponding to the supply chain data, and all the supply chain data are divided into at least two sparse data cluster sets. The sparse data clustering set is obtained to weaken the representation of the supply chain data on the supply chain relation, and the data adjacent in time sequence are integrated together through the sparse data clustering set, so that the subsequent storage of the supply chain data is more convenient.
Preferably, the method for acquiring the sparse data clustering set comprises the following steps:
mapping the supply chain data in a preset time range onto a time axis to obtain a supply chain data time axis; and taking a starting point of a supply chain data time axis as a starting center point, obtaining a drift vector of each drift window according to a mean shift clustering algorithm, and carrying out drift clustering according to the inverted drift vector based on the mean shift clustering algorithm to obtain at least two sparse data clustering sets. In the embodiment of the invention, the size of the drift window is set to 2h. It should be noted that, the implementer can adjust the size of the drift window according to the implementation environment, and when selecting the initial center point of the mean shift clustering algorithm, the implementer can select other time nodes on the supply chain data time axis according to the implementation environment, which will not be further described herein.
For the mean shift clustering algorithm, the direction corresponding to the shift vector, namely the direction with larger data density, is the same, and the direction with smaller data density can be obtained by reversing the direction of the shift vector. Therefore, the drift clustering is further carried out according to the opposite direction of the drift vector, so that the drift window can shift to the direction of data sparseness, and a sparse clustering set, namely a sparse data clustering set, is obtained. It should be noted that, the drift window, the initial center point, and the drift vector are all terms in the mean shift clustering algorithm, and the mean shift clustering algorithm is a prior art well known to those skilled in the art, and is not further defined and described herein. The time correlation between the supply chain data in the obtained sparse data clustering set is smaller, and the degree of expression of the supply chain relationship is also smaller, so that the hiding effect of the corresponding supply chain information is better when the supply chain data is encrypted by analyzing according to the sparse data clustering set.
The sparse data clustering set is only divided into sets according to the time distribution condition corresponding to the supply chain data, and the time sequence of the supply chain data is not changed, so that if encryption of the supply chain data is required to be further realized, the time sequence of the supply chain data is required to be changed. In consideration of the fact that the sparse data clustering set can represent a plurality of data which are continuous in time sequence, and the time ranges corresponding to the sparse data clustering set are not greatly different, the data can be conveniently stored and the change of the supply chain data time sequence can be realized by integrating or mapping the sparse data clustering sets. However, when two different sparse data cluster sets are integrated, the corresponding integrated data sets have different damage degrees to the time distribution relation of the supply chain data, so when the sparse data cluster sets are selected for integration, the sparse data cluster sets need to be further analyzed, and the damage degree of the data set integrated according to the sparse data cluster sets to the time distribution relation of the supply chain data is higher.
Considering that there are typically a variety of supply chain data, and that different kinds of supply chain data are typically staggered across a properly functioning industrial chain, there are typically a variety of supply chain data per sparse data cluster set. Considering that the time distribution and the number of different kinds of supply chain data in the sparse data clustering sets are generally different, further analysis can be performed according to the time and the number distribution of each kind of supply chain data in each sparse data clustering set. According to the embodiment of the invention, the concentration degree of each supply chain data in each sparse data clustering set is obtained according to the integral time difference characteristic, the time difference change characteristic and the density of each supply chain data in the sparse data clustering set. The distribution of each supply chain data in each sparse data cluster set is characterized by the degree of concentration.
Preferably, the method for acquiring the concentration degree includes:
and obtaining the corresponding time length of each sparse data clustering set, selecting one sparse data clustering set as a target sparse data clustering set, taking the clustering center of the target sparse data clustering set as a target clustering center, and obtaining the density of each supply chain data in the target sparse data clustering set according to the data quantity and the corresponding time length of each supply chain data in the target coefficient data clustering set, wherein the data quantity and the density are positively correlated, and the time length and the density are negatively correlated.
Preferably, the method for acquiring the time length of the sparse data clustering set comprises the following steps:
and taking the extremely poor time corresponding to the supply chain data in each sparse data clustering set as the time length of each sparse data clustering set.
The time range of the sparse data clustering set obtained by the embodiment of the invention is related to the size of the drift window in the mean shift clustering algorithm, so that the density can be represented by the quantity of each supply chain data in the target sparse data clustering set, but the time range corresponding to the corresponding supply chain data is different when the time sequence distribution of the supply chain data of different sparse data clustering sets is different, so that the density is calculated by the time length, the quantity of each supply chain data is combined, and the time distribution condition of the supply chain data of the target sparse data clustering set is combined, so that the subsequent characterization of the concentration degree is more accurate.
In the target sparse data clustering set, the average value of the distance between each data in each supply chain data and the target clustering center is taken as the integral time difference characteristic of each supply chain data in the target data set. In the embodiment of the invention, since the sparse data clustering set is obtained according to the mean shift clustering algorithm, the clustering center is usually the center of the time range corresponding to the sparse data clustering set, so that the overall time distribution condition of each supply chain data, namely the overall time difference characteristic, can be represented by calculating the average value of the distance between each data in each supply chain data and the clustering center in time.
In the target sparse data clustering set, the standard deviation of the distance between each data in each supply chain data and the target clustering center is used as a time difference change characteristic of each supply chain data in the target data set. And calculating the standard deviation of the time distance between all data in each supply chain data and the clustering center, and reflecting the time difference change of each data in each supply chain data relative to the clustering center, namely the time difference change characteristic. It should be noted that, the practitioner may also replace the standard deviation by the variance as the time difference variation feature according to the specific implementation environment, which is not further described herein.
And obtaining the concentration degree of each supply chain data in the target data set according to the density, the integral time difference characteristic and the time difference change characteristic, wherein the density and the concentration degree are positively correlated, the integral time difference characteristic and the concentration degree are negatively correlated, and the time difference change characteristic and the concentration degree are negatively correlated. According to the distribution of each supply chain data in time sequence, when each whole data in each supply chain data is closer to the clustering center, the distance between each data and the clustering center is smaller, the density is larger, the distribution of the corresponding type of supply chain data in the target data set is more concentrated, and the concentration degree is larger; that is, the smaller the overall time difference characteristic of each supply chain data in the target data set, the smaller the time difference change characteristic, and the greater the density, the greater the corresponding concentration degree.
In the embodiment of the invention, the product of the integral time difference characteristic and the time difference change characteristic of each supply chain data in the target data set is calculated, the sum value of the product and the preset adjusting parameter is calculated, and the normalized value of the ratio between the density and the sum value is used as the corresponding concentration degree. It should be noted that, the implementer may also obtain the concentration degree according to the density, the time difference change feature and the overall time difference feature by other methods according to the specific implementation situation, but it needs to ensure that the density is positively correlated with the concentration degree, the overall time difference feature is negatively correlated with the concentration degree, and the time difference change feature is negatively correlated with the concentration degree, which is not further described herein.
In an embodiment of the invention, the firstThe seed supply chain data is at->Acquisition of concentration levels in individual sparse data cluster setsThe method is expressed as:
wherein, the liquid crystal display device comprises a liquid crystal display device,is->The seed supply chain data is at->Concentration degree in the individual sparse data cluster set, +.>Is->The seed supply chain data is at->Data amount in the sparse data cluster set, +.>Is->The time length of each sparse data cluster set; />Is->The seed supply chain data is at->The distance average value between each data in the sparse data clustering set and the clustering center is the integral time difference characteristic; />Is->The seed supply chain data is at->The standard deviation of the distance between each data in the sparse data clustering set and the clustering center, namely the time difference change characteristic,/I>For presetting the adjustment parameters for preventing the denominator from being 0, in the embodiment of the present invention the adjustment parameters are preset +.>The setting of the preset adjustment parameter is 0.01, and it should be noted that the operator can adjust the preset adjustment parameter according to the specific implementation environment, which is not described herein. />Is->The seed supply chain data is at->Density in a set of sparse data clusters, +.>As the normalization function, in the embodiment of the present invention, the normalization function adopts linear normalization; it should be noted that, the practitioner may use other normalization methods according to the specific implementation environment, and the linear normalization is a prior art well known to those skilled in the art, which is not further limited and described herein. Further according to- >The seed supply chain data is at->Acquiring the concentration degree in each sparse data clustering set to obtain each supply chain data in each sparse data clustering setThe degree of concentration in the data cluster set.
In addition, the practitioner can also obtain the first through other forms of formulasThe seed supply chain data is at->Concentration in a collection of sparse data clusters, for example:
wherein, the liquid crystal display device comprises a liquid crystal display device,for the exponential function based on the natural constant e, the rest parameters have the same meaning as the parameters in the formula corresponding to the method for obtaining the concentration degree in the embodiment of the present invention, and will not be further described herein.
In order to make the damage degree of the integrated data set to the time distribution relation of the supply chain data higher, the embodiment of the invention obtains the integration necessity between any two sparse data clustering sets according to the difference of the variety number and the time length of the supply chain data in any two sparse data clustering sets and the difference of the concentration degree of each supply chain data. And (3) representing the damage degree of the data set integrated by the two sparse data clustering sets to the time distribution relation of the supply chain through the integration necessity.
Preferably, the method for acquiring the integration necessity includes:
selecting any one sparse data clustering set as a target sparse data clustering set, and taking any other sparse data clustering set as a comparison sparse data clustering set; taking the difference of the category number of the supply chain data between the target sparse data clustering set and the comparison sparse data clustering set as a category number difference characteristic; firstly, comparing the difference of data distribution between a target sparse data clustering set and a comparison sparse data clustering set through category data difference characteristics, wherein the larger the corresponding category number difference is, the weaker the representation of the relation of supply chain data between the two sparse data clustering sets to the original supply chain is after combination, and the larger the corresponding integration necessity is.
Acquiring the time length of each sparse data clustering set; and taking the difference of the time length between the target sparse data clustering set and the comparison sparse data clustering set as a time length difference characteristic. The time length is the extremely poor time corresponding to the supply chain data in the sparse data clustering set. Although the time range of each sparse data clustering set should be the same in theory in the process of acquiring the sparse data clustering sets, the supply chain data distribution relationship of different sparse data clustering sets can be represented by the extremely poor time, the larger the corresponding time length standard deviation is, the larger the distribution relationship difference between the corresponding sparse data clustering sets is, which means that the weaker the supply chain data between the two sparse data clustering sets after combination is in the representation of the relationship of the original supply chain is, and the larger the corresponding integration necessity is.
And calculating the concentration degree difference of each supply chain data between the target sparse data clustering set and the comparison sparse data clustering set, and taking the accumulated sum of all the concentration degree differences corresponding between the target sparse data clustering set and the comparison sparse data clustering set as the corresponding concentration degree integral difference characteristic. The concentration degree can reflect the distribution relation of each supply chain data in each sparse data clustering set, and is mainly expressed as the time distribution relation between a plurality of data corresponding to each supply chain data and a clustering center, and considering that a plurality of supply chain data generally exist in each sparse data clustering set, namely each sparse data clustering set corresponds to a plurality of concentration degrees, in order to represent the difference between a target sparse data clustering set and a comparison sparse data clustering set in concentration degree, the embodiment of the invention represents the integral difference characteristic of concentration degree through the accumulation sum of the concentration degree differences of all kinds of supply chain data in the two sparse data clustering sets. The larger the corresponding concentration degree difference is, the larger the concentration degree overall difference characteristic is, which means that the weaker the supply chain data between two sparse data clustering sets is in the expression of the relation of the original supply chain after combination, and the larger the corresponding integration necessity is.
And obtaining integration necessity between the target sparse data clustering set and the comparison sparse data clustering set according to the category number difference feature, the time length difference feature and the concentration degree overall difference feature, wherein the category number difference feature and the integration necessity are positively correlated, the time length difference feature and the integration necessity are positively correlated, and the concentration degree overall difference feature and the integration necessity are positively correlated. The variety and quantity difference features, the time length difference features and the concentrated range overall difference features between two sparse data clustering sets can be further combined to obtain corresponding integration necessity. The larger the difference feature of the category number is, the larger the difference feature of the time length is, and the larger the difference of the integral feature of the concentration degree is, the larger the corresponding integration necessity is, so that the difference feature of the category number is positively correlated with the integration necessity, the difference feature of the time length is positively correlated with the integration necessity, and the integral difference feature of the concentration degree is positively correlated with the integration necessity.
In the embodiment of the invention, the corresponding integration necessity is represented by the product among the category number difference feature, the time length difference feature and the concentration degree overall difference feature. It should be noted that, the implementer may also characterize the integration necessity according to the category number difference feature, the time length difference feature and the concentration degree overall difference feature by other methods, for example, the category number difference feature, the time length difference feature and the concentration degree overall difference feature are weighted and summed to obtain the corresponding integration necessity, which is not further described herein.
In the embodiment of the invention, the following is the firstThe sparse data clustering set is taken as a target sparse data clustering set, and the third is taken asThe sparse data clustering set is taken as a comparison sparse data clustering set corresponding to the target sparse data clustering set, and the acquisition method of integration necessity between the target sparse data clustering set and the corresponding comparison sparse data clustering set is expressed as follows in terms of a formula:
wherein, the liquid crystal display device comprises a liquid crystal display device,is->Set of sparse data clusters and +.>Integration necessity between the individual sparse data cluster sets,/->Is->Supply chain data category number corresponding to each sparse data cluster set, +.>Is->Supply chain data category number corresponding to each sparse data cluster set, +.>Is->Time length corresponding to each sparse data cluster set, < +.>Is the firstTime length corresponding to each sparse data cluster set, < +.>Is->The seed supply chain data is at->Concentration degree in the individual sparse data cluster set, +.>Is->The seed supply chain data is at->Concentration degree in the individual sparse data cluster set, +.>For the total number of categories corresponding to all supply chain data, < + >>Is->Set of sparse data clusters and +.>Species number difference features between the individual sparse data clusters, +. >Is->Set of sparse data clusters and +.>Sparse data aggregationTime length difference feature between class sets, +.>Is->Set of sparse data clusters and +.>The concentration degree among the sparse data clustering sets is characterized by overall difference. Further according to->Set of sparse data clusters and +.>And obtaining the integration necessity between every two other sparse data clustering sets by using the acquisition method of the integration necessity between every two sparse data clustering sets.
In addition, the implementer may also obtain the integration necessity between the target sparse data cluster set and the corresponding contrast sparse data cluster set through other forms of formulas, for example:
wherein, the parameters in the formula and the embodiment of the invention are the firstSet of sparse data clusters and +.>The formulas corresponding to the acquisition methods for the integration necessity among the sparse data cluster sets are the same, and further description is omitted here.
The supply chain data processing module 103 is configured to integrate the two sparse data cluster sets according to the integration necessity to obtain at least two integrated data cluster sets; and carrying out encryption processing on the supply chain data according to the integrated data clustering set.
The integration necessity between any two sparse data clustering sets is obtained through the integration necessity obtaining module, and all sparse data clustering sets can be further integrated according to the integration necessity. According to the embodiment of the invention, the sparse data clustering sets are integrated according to the integration necessity to obtain at least two integrated data clustering sets.
Preferably, the method for acquiring the integrated data cluster set comprises the following steps:
counting all integration necessity corresponding to all sparse data cluster sets, taking two sparse data cluster sets with the largest integration necessity as a first cluster set to be integrated, and taking other sparse data cluster sets except the first cluster set to be integrated as a first non-integrated cluster set; integrating two sparse data cluster sets corresponding to the two first cluster sets to be integrated into one integrated data cluster set; taking the two first unconformity cluster sets with the largest integration necessity as a second cluster set to be integrated, and taking the other first unconformity cluster sets except the second sparse data cluster set as a second unconformity cluster set; integrating two sparse data cluster sets corresponding to the two second cluster sets to be integrated into one integrated data cluster set; and analogically, obtaining at least two integrated data clustering sets.
Considering that for any one sparse data cluster set, another sparse data cluster set with the highest integration necessity corresponding to the sparse data cluster set can be obtained, but the sparse data cluster set with the highest integration necessity corresponding to the other sparse data cluster set may not be the sparse data cluster set. Therefore, the embodiment of the invention analyzes the integration necessity whole, and selects all the sparse data clustering sets with highest integration necessity for integration, so that the damage degree of the obtained integration data clustering set to the time distribution relation of the supply chain data is maximized. After each integrated data cluster set is obtained, in order to prevent data repetition caused by integration of a subsequent non-integrated sparse data cluster set and an integrated sparse data cluster set, two sparse data cluster sets with the largest integration necessity are continuously selected for integration on the basis of the remaining non-integrated sparse data cluster sets, and the process is iterated, so that each sparse data cluster set can obtain an integrated data cluster set with the optimal whole, and the damage degree of the whole obtained integrated data cluster set on the time distribution relation of the supply chain data is maximized.
When the number of the sparse data clustering sets is even, all the sparse data clustering sets can be integrated through iteration. However, considering that the number of the sparse data cluster sets may not be even, that is, when the two sparse data cluster sets are integrated, one single sparse data cluster set may remain, and for the case that one single sparse data cluster set remains last, when one sparse data cluster set remains in the non-integrated cluster set, the embodiment of the invention takes the remaining one sparse data cluster set as one integrated data cluster set.
For example, if all the supply chain data corresponds to seven sparse data cluster sets A, B, C, D, E, F, G, and two sparse data cluster sets can all obtain the corresponding integration necessity. If the integration necessity between A and B is the largest in all integration necessity, integrating A and B to obtain a corresponding integrated data cluster set AB, namely, A and B are a first cluster set to be integrated, and the rest C, D, E, F, G is a first non-integrated cluster set; if the integration necessity between C and D is the largest in all the remaining C, D, E, F, G corresponding integration necessity, integrating C and D to obtain a corresponding integrated data cluster set CD, namely, C and D are second cluster sets to be integrated, and the remaining E, F, G is second non-integrated cluster set; and continuing to analyze all integration necessity corresponding to the residual E, F, G, if the integration necessity between E and F is the largest in all integration necessity corresponding to E, F, G, integrating E and F to obtain a corresponding integrated data cluster set EF, wherein E and F are a third cluster set to be integrated, G is a third non-integrated cluster set, at this time, a single sparse data cluster set G remains in the non-integrated cluster set, and the sparse data cluster set G is used as a single integrated data cluster set G. Namely, seven sparse data cluster sets A, B, C, D, E, F, G are integrated to obtain four integrated data cluster sets AB, CD, EF and G.
Preferably, the integrating of the two sparse data cluster sets into one integrated data cluster set comprises:
the supply chain data with earliest corresponding moment in each sparse data clustering set is used as starting point data corresponding to each sparse data clustering set; taking the time distance between the two to-be-integrated cluster sets and the corresponding starting point data as the characteristic distance; on a time axis corresponding to a preset time range of all the supply chain data, when the time lengths corresponding to the two to-be-integrated clustering sets are equal, shifting any one to-be-integrated clustering set data to the other to-be-integrated clustering set direction by a corresponding characteristic distance length to obtain an integrated data clustering set; when the time lengths corresponding to the two to-be-integrated cluster sets are unequal, shifting the to-be-integrated cluster set with the shortest time length to the to-be-integrated cluster set with the longest time length by the corresponding characteristic distance length to obtain an integrated data cluster set.
Considering that different sparse data clustering sets have different corresponding local time ranges within a preset time range, namely that the time ranges corresponding to different sparse data clustering sets are not overlapped, if two sparse data clustering sets are integrated, one sparse data clustering set is required to be mapped or translated to the other sparse data clustering set, so that overlapping or overlapping of the time ranges is realized, and an integrated data clustering set is obtained. Since the purpose of the embodiment of the invention is to encrypt the supply chain data, the encrypted supply chain data can be restored according to the secret key for convenience, and the mapping or translation method corresponding to each integration process is unified by calculating the characteristic distance. In order to facilitate separation of the sparse data cluster set from the integrated data cluster set, the embodiment of the invention represents the translation attribute of the supply chain data corresponding to the sparse data cluster set, which is translated by time, as 1, and represents the translation attribute of the supply chain data corresponding to the sparse data cluster set, which is unchanged in time, as 0, thereby providing a basis for subsequent recovery of the supply chain data by the integrated data cluster set.
So far, all integrated data clustering sets corresponding to all supply chain data are obtained. The supply chain data encryption process may further be performed based on the aggregate data cluster set. Although the integrated data clustering sets are encrypted in the integrated process, in order to facilitate storage and protection of data, in the embodiment of the invention, each integrated data clustering set is further encrypted by a block encryption method to obtain encrypted supply chain ciphertext data, and the supply chain ciphertext data and a corresponding encryption key are further transmitted to a data intelligent operation platform for compression storage. It should be noted that, the packet encryption is a prior art well known to those skilled in the art, and the data transmission and the compression storage are technical means well known to those skilled in the art, which are not important matters of the embodiment of the present invention, and the implementer may select the data transmission method and the compression storage method according to the specific implementation environment, which is not further limited and described herein.
In summary, the method and the system analyze according to the time distribution density corresponding to the supply chain data, divide all the supply chain data into at least two sparse data clustering sets, obtain corresponding concentration degrees according to the quantity and time distribution characteristics of different types of supply chain data in the sparse data clustering sets, obtain corresponding integration necessity according to the difference of the quantity, the time length and the concentration degrees of the corresponding types of any two sparse data clustering sets, integrate every two sparse data clustering sets to obtain an integrated data clustering set through the integration necessity, and encrypt the supply chain data through the integrated data clustering set. The encryption method and the encryption device have the advantage that the encryption effect on the supply chain data is better in the process of processing the supply chain data by integrating the data clustering set.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (7)
1. An artificial intelligence based data intelligent operation platform supply chain data processing system, characterized in that the system comprises:
the system comprises a supply chain data acquisition module, a data intelligent operation platform and a control module, wherein the supply chain data acquisition module is used for acquiring at least two kinds of supply chain data in a preset time range;
the integration necessity acquisition module is used for carrying out cluster analysis according to the time distribution density characteristics corresponding to the supply chain data and dividing all the supply chain data into at least two sparse data cluster sets; obtaining the concentration degree of each supply chain data in each sparse data clustering set according to the integral time difference characteristic, the time difference change characteristic and the density of each supply chain data in the sparse data clustering set; obtaining integration necessity between any two sparse data clustering sets according to the variety and quantity difference, the time length difference and the concentration degree difference of each supply chain data in any two sparse data clustering sets;
The supply chain data processing module is used for integrating the sparse data clustering sets according to the integration necessity to obtain at least two integrated data clustering sets; and carrying out encryption processing on the supply chain data according to the integrated data clustering set.
2. The artificial intelligence based data intelligent operation platform supply chain data processing system according to claim 1, wherein the method for acquiring the sparse data cluster set comprises:
mapping the supply chain data in a preset time range onto a time axis to obtain a supply chain data time axis; and taking a starting point of a supply chain data time axis as a starting center point, obtaining a drift vector of each drift window according to a mean shift clustering algorithm, and carrying out drift clustering according to the inverted drift vector based on the mean shift clustering algorithm to obtain at least two sparse data clustering sets.
3. The artificial intelligence based data intelligent operation platform supply chain data processing system according to claim 1, wherein the method for acquiring the concentration degree comprises the following steps:
acquiring a corresponding time length of each sparse data clustering set, selecting one sparse data clustering set as a target sparse data clustering set, taking a clustering center of the target sparse data clustering set as a target clustering center, and obtaining the density of each supply chain data in the target sparse data clustering set according to the data quantity of each supply chain data in the target coefficient data clustering set and the corresponding time length, wherein the data quantity is positively correlated with the density, and the time length is negatively correlated with the density;
In the target sparse data clustering set, taking the average value of the distance between each data in each supply chain data and the target clustering center as the integral time difference characteristic of each supply chain data in the target data set;
in the target sparse data clustering set, taking the standard deviation of the distance between each data in each supply chain data and the target clustering center as the time difference change characteristic of each supply chain data in the target data set;
and obtaining the concentration degree of each supply chain data in the target data set according to the density, the integral time difference characteristic and the time difference change characteristic, wherein the density and the concentration degree are positively correlated, the integral time difference characteristic and the concentration degree are negatively correlated, and the time difference change characteristic and the concentration degree are negatively correlated.
4. The artificial intelligence based data intelligent operation platform supply chain data processing system according to claim 1, wherein the method for acquiring the integration necessity comprises:
selecting any one sparse data clustering set as a target sparse data clustering set, and taking any other sparse data clustering set as a comparison sparse data clustering set; taking the difference of the category number of the supply chain data between the target sparse data clustering set and the comparison sparse data clustering set as a category number difference characteristic;
Acquiring the time length of each sparse data clustering set; taking the difference of the time length between the target sparse data clustering set and the comparison sparse data clustering set as a time length difference characteristic;
calculating the concentration degree difference of each supply chain data between the target sparse data clustering set and the comparison sparse data clustering set, and taking the accumulated sum of all the concentration degree differences corresponding to the target sparse data clustering set and the comparison sparse data clustering set as the corresponding concentration degree integral difference characteristic;
and obtaining integration necessity between a target sparse data clustering set and a comparison sparse data clustering set according to the category number difference feature, the time length difference feature and the concentration degree overall difference feature, wherein the category number difference feature is positively correlated with the integration necessity, the time length difference feature is positively correlated with the integration necessity, and the concentration degree overall difference feature is positively correlated with the integration necessity.
5. The artificial intelligence based data intelligent operation platform supply chain data processing system according to claim 4, wherein the method for acquiring the integrated data cluster set comprises the following steps:
Counting all integration necessity corresponding to all sparse data cluster sets, taking two sparse data cluster sets with the largest integration necessity as a first cluster set to be integrated, and taking other sparse data cluster sets except the first cluster set to be integrated as a first non-integrated cluster set; integrating two sparse data cluster sets corresponding to the two first cluster sets to be integrated into one integrated data cluster set;
taking the two first unconformity cluster sets with the largest integration necessity as a second cluster set to be integrated, and taking the other first unconformity cluster sets except the second sparse data cluster set as a second unconformity cluster set; integrating two sparse data cluster sets corresponding to the two second cluster sets to be integrated into one integrated data cluster set;
and analogizing to obtain at least two integrated data clustering sets;
when one sparse data cluster set is remained in the non-integrated cluster set, the remained sparse data cluster set is used as an integrated data cluster set.
6. The artificial intelligence based data intelligent operation platform supply chain data processing system according to claim 5, wherein the integration of the two sparse data cluster sets into one integrated data cluster set comprises:
The supply chain data with earliest corresponding moment in each sparse data clustering set is used as starting point data corresponding to each sparse data clustering set; taking the time distance between the two to-be-integrated cluster sets and the corresponding starting point data as the characteristic distance;
on a time axis corresponding to the preset time range of all the supply chain data, when the time lengths corresponding to the two to-be-integrated clustering sets are equal, shifting any one to-be-integrated clustering set data to the other to-be-integrated clustering set direction by a corresponding characteristic distance length to obtain an integrated data clustering set; when the time lengths corresponding to the two to-be-integrated cluster sets are unequal, shifting the to-be-integrated cluster set with the shortest time length to the to-be-integrated cluster set with the longest time length by the corresponding characteristic distance length to obtain an integrated data cluster set.
7. The artificial intelligence based data intelligent operation platform supply chain data processing system according to claim 3 or 4, wherein the method for obtaining the time length of the sparse data cluster set comprises:
and taking the extremely poor time corresponding to the supply chain data in each sparse data clustering set as the time length of each sparse data clustering set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310742637.7A CN116502107B (en) | 2023-06-21 | 2023-06-21 | Data intelligent operation platform supply chain data processing system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310742637.7A CN116502107B (en) | 2023-06-21 | 2023-06-21 | Data intelligent operation platform supply chain data processing system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116502107A true CN116502107A (en) | 2023-07-28 |
| CN116502107B CN116502107B (en) | 2023-09-01 |
Family
ID=87316829
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310742637.7A Active CN116502107B (en) | 2023-06-21 | 2023-06-21 | Data intelligent operation platform supply chain data processing system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116502107B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170124669A1 (en) * | 2014-06-09 | 2017-05-04 | Sicpa Holding Sa | Creating secure data in an oil and gas supply chain |
| US20200294067A1 (en) * | 2019-03-15 | 2020-09-17 | Target Brands, Inc. | Time series clustering analysis for forecasting demand |
| US20220358163A1 (en) * | 2021-05-06 | 2022-11-10 | NB Ventures, Inc., dba GEP | Data processing in enterprise application |
| CN115456541A (en) * | 2022-09-23 | 2022-12-09 | 浙大城市学院 | Supply chain management method and system for cross-border trade |
| CN115577380A (en) * | 2022-12-01 | 2023-01-06 | 武汉惠强新能源材料科技有限公司 | Material data management method and system based on MES system |
| CN116186634A (en) * | 2023-04-26 | 2023-05-30 | 青岛新航农高科产业发展有限公司 | Intelligent management system for construction data of building engineering |
| CN116226893A (en) * | 2023-05-09 | 2023-06-06 | 北京明苑风华文化传媒有限公司 | Client marketing information management system based on Internet of things |
-
2023
- 2023-06-21 CN CN202310742637.7A patent/CN116502107B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170124669A1 (en) * | 2014-06-09 | 2017-05-04 | Sicpa Holding Sa | Creating secure data in an oil and gas supply chain |
| US20200294067A1 (en) * | 2019-03-15 | 2020-09-17 | Target Brands, Inc. | Time series clustering analysis for forecasting demand |
| US20220358163A1 (en) * | 2021-05-06 | 2022-11-10 | NB Ventures, Inc., dba GEP | Data processing in enterprise application |
| CN115456541A (en) * | 2022-09-23 | 2022-12-09 | 浙大城市学院 | Supply chain management method and system for cross-border trade |
| CN115577380A (en) * | 2022-12-01 | 2023-01-06 | 武汉惠强新能源材料科技有限公司 | Material data management method and system based on MES system |
| CN116186634A (en) * | 2023-04-26 | 2023-05-30 | 青岛新航农高科产业发展有限公司 | Intelligent management system for construction data of building engineering |
| CN116226893A (en) * | 2023-05-09 | 2023-06-06 | 北京明苑风华文化传媒有限公司 | Client marketing information management system based on Internet of things |
Non-Patent Citations (3)
| Title |
|---|
| YU TU 等: "Online Segmentation Algorithm for Time Series Based on BIRCH Clustering Features", 《2010 INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND SECURITY》, pages 55 - 59 * |
| 武森 等: "基于稀疏指数排序的高维数据并行聚类算法", 《系统工程理论与实践》, vol. 31, no. 2, pages 13 - 18 * |
| 陈明豪: "基于深度学习的加密流量应用类型识别关键技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 03, pages 139 - 92 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116502107B (en) | 2023-09-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jian et al. | Measurement-induced criticality in random quantum circuits | |
| Jiang et al. | Flashe: Additively symmetric homomorphic encryption for cross-silo federated learning | |
| Shaukat et al. | Chaos theory and its application: an essential framework for image encryption | |
| CN104751403A (en) | Multi-chaos system based method of encrypting images related to plaintext | |
| Vatsalan et al. | Efficient two-party private blocking based on sorted nearest neighborhood clustering | |
| Meixell et al. | Scenario analysis of demand in a technology market using leading indicators | |
| CN113141355B (en) | Electricity consumption encrypted data anomaly detection method and system based on full homomorphism | |
| Romano et al. | Measuring dependency via intrinsic dimensionality | |
| CN110084326A (en) | A kind of industrial equipment method for detecting abnormality based on fuzzy set | |
| Jia et al. | Privacy-preserving data classification and similarity evaluation for distributed systems | |
| CN116502107B (en) | Data intelligent operation platform supply chain data processing system based on artificial intelligence | |
| CN116128164A (en) | Power load short-term prediction system and method based on privacy protection | |
| CN113051407B (en) | Network intelligent operation and maintenance knowledge graph collaborative construction and sharing method and device | |
| CN117171817B (en) | Electronic signature safety management method based on block chain | |
| Ferry et al. | Probabilistic dataset reconstruction from interpretable models | |
| Bertin et al. | Temperature in nonequilibrium systems with conserved energy | |
| Stan et al. | A new crypto-classifier service for energy efficiency in smart cities | |
| CN107886463B (en) | Digital image encryption method based on Chen system and cellular automaton | |
| Cui et al. | Enhancing Robustness of Gradient-Boosted Decision Trees through One-Hot Encoding and Regularization | |
| Baur et al. | Predicting high-dimensional heterogeneous time series employing generalized local states | |
| AT&T | ||
| CN108920501B (en) | Relational table non-key attribute feature extraction and data generation method | |
| Charron et al. | Regional analysis and modelling of water temperature metrics for Atlantic Salmon (Salmo salar) in Eastern Canada. Final report. | |
| Wittmann et al. | Event-driven anomalies in spatiotemporal taxi passseger demand | |
| CN109886825A (en) | A kind of agricultural internet of things data multi-angle of view projective clustering reconstructing method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |