A method and device for clustering intelligent devices based on adiabatic quantum computing

Abstract

This article relates to the financial field, and in particular to a method and device for clustering smart devices based on adiabatic quantum computing. It includes: obtaining feature data of multiple smart devices, the feature data is related to the application business of the smart devices and the environment in which the smart devices are located; using an adiabatic quantum computing clustering algorithm to perform unsupervised clustering on the feature data; according to the clustering results after unsupervised clustering, determining the class core features corresponding to each clustering result; according to the class core features, determining the type of smart device. This article uses an improved clustering algorithm of adiabatic quantum computing to perform clustering analysis on the operation of smart devices in commercial bank branches. Compared with the traditional branch network operation and management methods of smart devices, this method uses clustering analysis methods to achieve efficient operation and precise deployment of smart devices; combined with adiabatic quantum computing capabilities, it effectively improves algorithm efficiency, reduces computational complexity, and improves clustering accuracy and stability.

Description

Technical Field

[0001] This article relates to the financial field, and in particular to a smart device clustering method and apparatus based on adiabatic quantum computing. Background Art

[0002] As an important place for banks to serve customers, the optimization of the efficiency of their operation and management is of great significance to commercial banks. With the popularization of mobile payment, the off-counter handling of non-cash business has promoted the increase in the use of smart devices. Cabinets with more functions - smart counters have also emerged in the context of branch transformation and integration into the financial development of the new era. The difficulty of managing bank smart counter equipment lies in its wide distribution range, large number, and complex customer and environmental conditions in the region.

[0003] In the prior art, the management of smart devices is usually based on experience or random deployment, which easily leads to waste of resources and fails to make reasonable and effective use of the devices. Alternatively, the prior art usually analyzes based on basic clustering algorithms, but clustering algorithms are inefficient, computationally complex, and have poor computational stability for large amounts of data, and cannot solve the problems of time and space complexity.

[0004] In view of the problems of resource waste and low computing efficiency in current technologies, a method and apparatus for clustering intelligent devices based on adiabatic quantum computing is needed. Summary of the invention

[0005] To solve the above problems of the prior art, the embodiments of this document provide a method, apparatus, computer equipment and storage medium for clustering smart devices based on adiabatic quantum computing.

[0006] The embodiment of the present invention provides a smart device clustering method based on adiabatic quantum computing, comprising: obtaining feature data of multiple smart devices, wherein the feature data is related to the application business of the smart devices and the environment in which the smart devices are located; performing unsupervised clustering on the feature data using an adiabatic quantum computing clustering algorithm; determining a class core feature corresponding to each clustering result based on the clustering results after the unsupervised clustering; and determining the type of the smart device based on the class core feature.

[0007] According to one aspect of the embodiments of this invention, unsupervised clustering of feature data using an adiabatic computational clustering algorithm includes: preprocessing the feature data of the smart device to obtain sample data of the smart device; selecting a random number of sample data as initial cluster centers; determining a cost function using the following formula to calculate the minimum distance between the sample data of the smart device and the initial cluster center: Among them, μ k represents the kth initial cluster center, x irepresents the sample data calculated according to the characteristic data of the i-th smart device, and M represents the total number of sample data; according to the cost function and the adiabatic quantum system, the final state Hamiltonian is evolved; according to the ground state of the final state Hamiltonian, the current cluster center is determined.

[0008] According to one aspect of the embodiments of this document, preprocessing the feature data of the smart device further includes: setting different weights for feature data of different dimensions in the smart device respectively; and determining sample data of each smart device based on the feature data and the weights.

[0009] According to one aspect of the embodiments of this document, the characteristic data includes at least one of a device number, a customer ratio, a device usage frequency, and a device life cycle, or a combination thereof.

[0010] According to one aspect of the embodiments of this document, determining the class core features corresponding to each clustering result based on the clustering results after unsupervised clustering includes: adding labels to each class of clustering results based on feature data of the current cluster center calculated after unsupervised clustering, wherein the labels are class core features.

[0011] According to one aspect of the embodiments of this document, adding labels for each type of clustering results includes: when the average value of the distance between a smart device and the current cluster center is less than a first preset threshold, adding a first label to the clustering result of the cluster center; when the average value of the distance between a smart device and the current cluster center is greater than or equal to the first preset threshold and less than a second preset threshold, adding a second label to the clustering result of the cluster center; when the average value of the distance between a smart device and the current cluster center is greater than or equal to the second preset threshold and less than a third preset threshold, adding a third label to the clustering result of the cluster center.

[0012] According to one aspect of the embodiments of this document, the method includes: determining the smart device with the minimum distance to the cluster center corresponding to the first label as the first device, and the first device is a device that does not need to be improved; determining the smart device with the minimum distance to the cluster center corresponding to the second label as the second device, and the second device is a smart device to be maintained; determining the smart device with the minimum distance to the cluster center corresponding to the third label as the third device, and the third device is a smart device to be eliminated.

[0013] The embodiment of the present invention provides a smart device clustering device based on adiabatic quantum computing, and the device includes: a feature data acquisition unit, used to acquire feature data of multiple smart devices, wherein the feature data is related to the usage business of the smart devices and the environment in which the smart devices are located; a clustering unit, used to perform unsupervised clustering on the feature data using an adiabatic quantum computing clustering algorithm; a class core feature determination unit, used to determine the class core feature corresponding to each clustering result according to the clustering result after unsupervised clustering; and a device determination unit, used to determine the type of the smart device according to the class core feature.

[0014] The embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the smart device clustering method based on adiabatic quantum computing when executing the computer program.

[0015] The embodiments of the present invention further provide a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the smart device clustering method based on adiabatic quantum computing is implemented.

[0016] This solution uses an improved clustering algorithm of adiabatic quantum computing to perform cluster analysis on the operation of smart devices in commercial bank branch outlets. Compared with the traditional operation and management methods of smart devices in branch outlets, this method uses cluster analysis methods to achieve efficient operation and precise deployment of smart devices; in addition, combined with adiabatic quantum computing capabilities, it effectively improves algorithm efficiency, reduces computational complexity, improves clustering accuracy and stability, narrows the scope of attention on branch smart device business, and directly focuses on branch clusters with excellent performance and branch clusters with unsatisfactory performance, which helps to regularly evaluate and grasp the operation and management of smart cabinet equipment in each branch, and provide decision-making basis for further adjustment of resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of this article or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of this article. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flowchart of a smart device clustering method based on adiabatic quantum computing in an embodiment of this article;

[0019] Figure 2 It is a flowchart of a method for performing unsupervised clustering on feature data of smart devices according to an embodiment of this invention;

[0020] Figure 3 The figure is a flow chart of a method for preprocessing feature data of a smart device according to an embodiment of the present invention;

[0021] Figure 4 The figure shows a flow chart of a method for adding labels to each type of clustering results in an embodiment of this article;

[0022] Figure 5 The figure is a flow chart of a method for determining the type of a smart device according to an embodiment of this invention;

[0023] Figure 6 The figure shows a schematic diagram of the structure of a smart device clustering device based on adiabatic quantum computing according to an embodiment of this invention;

[0024] Figure 7 The figure shows a specific structural schematic diagram of the intelligent device clustering device based on adiabatic quantum computing in the embodiment of this article;

[0025] Figure 8 Shown is a schematic diagram of the structure of a computer device according to an embodiment of this invention.

[0026] Description of the accompanying symbols:

[0027] 601, feature data acquisition unit;

[0028] 6011. Sample data determination module;

[0029] 602, unsupervised clustering unit;

[0030] 6021, initial cluster center determination module;

[0031] 603, class core feature determination unit;

[0032] 604. Device determination unit;

[0033] 802. Computer equipment;

[0034] 804, processor;

[0035] 806. Memory;

[0036] 808, driving mechanism;

[0037] 810, input / output module;

[0038] 812. Input device;

[0039] 814. Output device;

[0040] 816. Presentation equipment;

[0041] 818. Graphical user interface;

[0042] 820, network interface;

[0043] 822, communication link;

[0044] 824. Communication bus. DETAILED DESCRIPTION

[0045] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.

[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of this article and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this article described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0047] This specification provides method operation steps as described in the embodiments or flow charts, but more or fewer operation steps may be included based on routine or non-creative work. The order of steps listed in the embodiments is only one way of executing the order of many steps and does not represent the only execution order. When the system or device product is executed in practice, it can be executed in the order of the method shown in the embodiments or the drawings or in parallel.

[0048] It should be noted that the method in this article can be used in the financial field as well as in the data processing field. This article does not limit the application field of the intelligent device clustering method and apparatus based on adiabatic quantum computing.

[0049] Figure 1 The flowchart of a smart device clustering method based on adiabatic quantum computing in an embodiment of this invention is shown, and includes the following steps:

[0050] Step 101, obtain characteristic data of multiple smart devices, the characteristic data is related to the application business of the smart device and the environment in which the smart device is located. The smart device in this step is the smart device of the bank branch. The smart counter will hand over most of the non-cash business to the customer for self-service, which can realize business migration, relieve the pressure of counter service, divert customers, and eliminate resource occupation. In practical applications, the difficulty of managing bank smart counter equipment lies in its wide distribution range, large number, and complex customer and environmental conditions in the area. For example, there are many smart devices in first-tier cities, large business volume, and high frequency of use; there are few smart devices in areas with a small population and underdeveloped economy, small business volume, and low frequency of use. Therefore, in some embodiments of this specification, the characteristic data of the smart device includes but is not limited to: the service life of the smart device, the frequency of use of the smart device, the maintenance frequency, the business volume of the branch, the number of customers, and other characteristic data that affect the use of the smart device. The characteristic data of the smart device can reflect the branch clusters with excellent performance and the branch clusters with unsatisfactory performance, which is helpful to regularly evaluate and grasp the operation and management of the smart cabinet equipment of each branch, and provide a decision-making basis for further adjusting resource allocation.

[0051] Step 102, using an adiabatic quantum computing clustering algorithm, unsupervised clustering of the feature data. In the embodiments of the present specification, the principle of adiabatic quantum computing is to transform the computational problem into a problem of seeking the ground state of a quantum system. In some embodiments of the present specification, adiabatic quantum computing first determines the initial ground state and final state physical environment of a closed physical system. According to the adiabatic quantum principle, a reasonable evolution time and final state are set, so that the system starts from the initial ground state, strives to maintain the lowest evolution energy level (in the ground state), so that its Hamilton quantity evolves slowly enough to the final state of the system, and the system will automatically complete the adiabatic evolution with minimum energy dissipation.

[0052] The Hamiltonian of the physical system in which the distance calculation problem is set is set as H f At the beginning of the operation, at time t = 0, the state of the quantum system is described by the initial Hamiltonian H0, and the system slowly evolves in an adiabatic manner to the state described by the Hamiltonian H f A final state described by , the ground state of which is the solution to the distance minimization problem. Let H corresponding to the function you want to minimize f The energy of the eigenstate is λ, and the domain of the function f is {0, 1} n , then the Hamiltonian of the final state can be written as:

[0053] H f :=∑ k∈{0，1}n f(k)|k> <k|,

[0054] Among them, H f$H$ represents the Hamiltonian in the process of time evolution, $f(k)$ represents the cost function, $k$ is the number of initial cluster centers, and $|k\rangle$ and $\langle k|$ represent the quantum ground states. For the distance description of this step, please refer to Figure 2 the description.

[0055] In some embodiments of this specification, adiabatic quantum computing sets the adiabatic quantum evolution conditions and allows the system to perform time evolution by itself under the domination of physical laws. This method avoids the currently difficult problems such as writing quantum algorithms and physical implementation in the gate circuit quantum computing model. Compared with traditional quantum algorithms, adiabatic quantum algorithms have improvements in the initialization, evolution, and measurement of quantum computing. The initial value of adiabatic quantum computing is globally correlated with the system at the beginning of the calculation; the structure of the adiabatic quantum algorithm is independent of time, and its calculation process depends on the structured properties of the adiabatic system; traditional quantum computing performs a single projection measurement on the results of a series of quantum gate evolutions, and the obtained result is a probabilistic representation of the required result, while adiabatic quantum computing measures the energy corresponding to each eigenvalue of the final state Hamiltonian of the system. By selecting the eigenvalue corresponding to the local minimum energy, the local optimal solution of the problem to be solved can be obtained.

[0056] In this step, through unsupervised clustering of adiabatic quantum, each intelligent device sample is assigned to the cluster center closest to it.

[0057] Step 103: Determine the class core features corresponding to each clustering result according to the clustering result after unsupervised clustering. According to the content in Step 102, the clustering result after unsupervised clustering is multiple clustering centers or cluster centers. Specifically, the clustering centers represented by one or more intelligent devices have obvious differences between different clustering centers. In this specification, the class core features are the features of each clustering center or cluster center obtained after clustering. Subsequently, the class core features of each category are further defined by labeling. For the specific description of determining the class core features of each clustering result, please refer to Figure 4 。

[0058] Step 104: Determine the types of intelligent devices according to the class core features. In some embodiments of this specification, the finally determined types of intelligent devices are related to the operating states of the devices. The types of intelligent devices include but are not limited to at least one of the first device, the second device, and the third device. Among them, the first device is an intelligent device with good equipment maintenance and excellent business performance that does not require improvement; the second device is an intelligent device with poor equipment maintenance, average performance, and requires enhanced manual maintenance and improvement; the third device is an intelligent device with poor performance and is recommended to be eliminated. In this step, for the description of determining the types of intelligent devices, please refer to Figure 4 。

[0059] Figure 2The following is a flowchart of an unsupervised clustering method for feature data of intelligent devices according to an embodiment of the present invention, which specifically includes the following steps:

[0060] Step 201: Preprocess the feature data of the intelligent device to obtain sample data of the intelligent device. In this step, preprocessing the feature data includes: setting different weights for feature data in different dimensions in the intelligent device. In some embodiments of the present specification, data in different dimensions of the intelligent device include, but are not limited to: data in the device usage frequency dimension, data in the traffic volume dimension. Among them, data in the device usage frequency dimension includes: device usage years, device usage frequency, device repair times, device repair frequency; data in the traffic volume dimension includes: data on the number of customers received by the device, traffic volume of the network point where the device is located, etc. Among them, the device usage frequency can be the usage frequency data within one month, one quarter, half a year, or one year; the device repair times can be the usage frequency data within one month, one quarter, half a year, or one year. In this specification, different weights are assigned to data in different dimensions.

[0061] Step 202: Select a random number of sample data as the initial cluster centers. In this step, randomly select sample data corresponding to any number of intelligent devices in the intelligent device as the initial cluster centers for clustering. For example, if there are 100 intelligent devices, randomly select sample data of 3 or 5 intelligent devices as the initial cluster centers.

[0062] Step 203: Determine the cost function and calculate the minimum value of the distance between the sample data of the intelligent device and the initial cluster centers. The cost function includes: where f(k) is the cost function, μ k represents the k-th initial cluster center, x i represents the sample data calculated according to the feature data of the i-th intelligent device, and M represents the total number of sample data. In some embodiments of the present specification, the cost function calculates the minimum value of the distance between the sample data of all other intelligent devices except the initial cluster centers and the initial cluster centers.

[0063] Step 204: According to the cost function and the Hamiltonian of the adiabatic quantum system evolving to the final state. In some embodiments of the present specification, the calculation formula of the Hamiltonian of the final state is as follows: and <k|, where k represents the number of initial cluster centers, |k> and <k| represent the quantum ground state, f(k) represents the cost function, and H f represents the Hamiltonian in the time evolution process. During the evolution of the adiabatic quantum system, the initial cluster centers determined in step 203 will change, and finally, several optimal center points and the best classification results are determined.

[0064] Calculate the distance between the sample data of all smart devices and k center points. The sample data of the smart device with the smallest distance from the center point is determined as the classification. Specifically, the closer the distance between the sample data of the smart device and the center of a cluster, the more similar the smart device is to the center of the cluster. The greater the similarity within the cluster and the greater the difference between clusters, the better the clustering effect.

[0065] Step 205, determine the current cluster center according to the ground state of the final Hamiltonian. In some embodiments of this specification, the physical evolution of adiabatic quantum computing is used to obtain the final cluster center, thereby determining which cluster center each smart device is closest to, and further completing the classification of each smart device.

[0066] In some embodiments of the present specification, according to the quantum mechanics postulate and adiabatic quantum computing theory, the state of the Hamiltonian in the time evolution process at any time t in the evolution time domain can be written as a linear superposition of its initial state and final state:

[0067] Where 0≤t≤T, t is time, T is H0->H f Crucial hysteresis factor of the evolution process, H0 represents the Hamiltonian at t = 0, H f represents the final state Hamiltonian. If T is set large enough, the ground state of the system initial state Thus, after the evolution begins, H(t) changes slowly as the system evolves according to the adiabatic theory, and eventually reaches the final state |φ(T)>. The final state of adiabatic evolution H f The ground state is close to the solution of the problem.

[0068] Figure 3 The flowchart of a method for preprocessing feature data of a smart device according to an embodiment of the present invention is shown, which specifically includes the following steps:

[0069] Step 301, respectively set different weights for the feature data of different dimensions in the smart device. As mentioned above, the data of different dimensions of the smart device include but are not limited to: data of the device usage frequency dimension and data of the business volume dimension. Among them, the data of each dimension and the data of the subdivided type in each dimension can be embodied in the form of data. And these data are normalized and outliers are processed. For example, there are 1,000 smart devices in a certain area of ​​East China, and the device usage frequency of each device is: 100 times, 50 times, 88 times, 30 times... and other frequency data within 1 month. The business volume data of each device is: 150 transactions, 200 transactions, 120 transactions, 340 transactions, etc. within 1 month..., and these data of different dimensions are normalized, and the data of different dimensions are processed into data of the same unit. For example, the device usage frequency data is uniformly divided by 100, 200, etc., and the device business volume data is uniformly divided by 100, 200, etc. data. In this way, the data is normalized.

[0070] Furthermore, different weights are set for the feature data of different dimensions after normalization. For example, a first weight is assigned to the data of the frequency of equipment use, and a second weight is assigned to the data of the business volume dimension. Furthermore, this specification may also assign a weight to the equipment age in the equipment frequency data, another weight to the equipment frequency, and different weights to different types of data in the equipment frequency data; this specification may assign a weight to the data of the number of customers received by the equipment in the data of the business volume dimension of the equipment, and another weight to the business volume data of the outlet where the equipment is located.

[0071] Step 302, determine the sample data of each smart device according to the feature data and the weight. In the step, further determine the sample data of each smart device according to the feature data and the weight. In this step, all feature data are multiplied by the weight of the feature data, and the sum is calculated to obtain the product sum of all feature data and weights. For example, the normalized usage frequency data of smart device a is 0.15, and the weight of the usage frequency data is 2; the normalized business volume dimension data of smart device a is 0.2, and the weight of the business volume dimension data is 3. The feature data of smart device a is multiplied by the weight, and the product sum of the feature data and the weight is obtained, and the sample data of smart device a is 0.15*2+0.2*3=0.9.

[0072] In some embodiments of the present specification, the clustering results include but are not limited to: cluster categories, the proportion of each cluster category to the number of all smart devices, cluster category differences (for example, cluster category variance), etc.

[0073] In some embodiments of the present specification, according to the clustering results after unsupervised clustering, determining the class core features corresponding to each clustering result includes: adding a label to each class of clustering results according to the feature data of the current cluster center calculated after unsupervised clustering, and the label is the class core feature. Specifically, the clustering result of the current cluster center is determined by analyzing the feature data of the current cluster center obtained by clustering, and the clustering result can also be explained by business logic. For example, according to the feature data of the current cluster center obtained by clustering, the feature data of the frequency of use of smart devices is low, and the feature data of the business volume carried by the smart devices is high, then the clustering result of the current cluster center can be interpreted as: the cluster center of this class is an excellent performing smart device. For another example, according to the feature data of another current cluster center, it is found that the feature data of the frequency of use of smart devices is high, and the feature data of the business volume carried by the smart devices is low, then the clustering result of the current cluster center can be interpreted as: the cluster center of this class is a generally performing smart device.

[0074] According to the above explanation of clustering results, a label can be added to each clustering result, and the label can be used as the core feature of each class.

[0075] Figure 4 The flowchart of a method for adding labels to each type of clustering result in an embodiment of this invention is shown, and includes the following steps:

[0076] Step 401: When the average value of the distance between the smart device with the minimum distance from the current cluster center and the current cluster center is less than a first preset threshold, a first label is added to the clustering result of the cluster center.

[0077] In some embodiments of the present specification, the distance between all smart devices and the current cluster center is calculated. For example, there are 100 smart devices, and three current cluster centers are determined by the above method, and the distance between the remaining 97 smart device sample data outside the current cluster center and the three current cluster center devices is calculated. For example, the distances between the first smart device and the three current cluster centers are calculated to be d11, d12, and d13, respectively, where d11 represents the distance between the first smart device and the first current cluster center; d12 represents the distance between the first device and the second current cluster center; d13 represents the distance between the first smart device and the third current cluster center, and it is determined that among the three distance values, d11 is the minimum distance value, then it can be determined that the first smart device is most similar to the first cluster center device. Similarly, the distances between the second device and the three current cluster centers are further determined to be d21, d22, d23, etc., where d21 is the minimum distance value, and it can be determined that the second smart device is most similar to the second cluster center device. The distances between the 97th device and the three current cluster centers are d971, d972, and d973, where d971 is the minimum distance. It can be determined that the 97th smart device is most similar to the center device of the first cluster.

[0078] In this step, if there are multiple smart devices with minimum distances to the current cluster center, the average distance between each smart device and the current cluster center is calculated. The average value is then compared with the preset threshold to add a label to the clustering result of the cluster center. For example, among 100 smart devices, the distances between 97 smart devices and the 3 current cluster centers are calculated. Among them, the smart devices with minimum distances to the 3 initial cluster centers and the minimum distances between each smart device and the current cluster center can be determined. For example, 32 smart devices have minimum distances to the first current cluster center; 52 smart devices have minimum distances to the second current cluster center; and 13 smart devices have minimum distances to the third current cluster center.

[0079] The average distances between the 32 smart devices with the minimum distance to the center of the first current cluster, the average distances between the 52 smart devices with the minimum distance to the center of the second current cluster, and the average distances between the 13 smart devices with the minimum distance to the center of the third current cluster are calculated.

[0080] Furthermore, the average value of the distances between 32 smart devices and the first current cluster center is calculated to be 0.3, the average value of the distances between 52 smart devices and the second current cluster center is calculated to be 0.2, and the average value of the distances between 13 smart devices and the third current cluster center is calculated to be 0.4.

[0081] Correspondingly, if the first preset threshold is 0.25, it can be determined that the 52 smart devices with the minimum distance to the second current cluster center have an average distance to the second cluster center that is less than the first preset threshold, and a first label is added to the clustering result of the second current cluster center. In some embodiments of the present specification, the first label may be: large business volume and excellent performance. In the present specification, the value of the first preset threshold can be adjusted according to actual conditions and is not limited to the value described in this step; the definition of the first label can also be adjusted according to actual conditions and is not limited to the description in this step.

[0082] Step 402, when the average value of the distance between the smart device with the minimum distance to the current cluster center is greater than or equal to the first preset threshold and less than the second preset threshold, a second label is added to the clustering result of the cluster center. Correspondingly, if the second preset threshold is 0.35, it can be determined that the 32 smart devices with the minimum distance to the first current cluster center have an average distance of 0.3 greater than the first preset threshold of 0.25 and less than the second preset threshold of 0.35. Then a second label is added to the clustering result of the first current cluster center. In some embodiments of the present specification, the second label may be: large business volume, general performance and need for improvement and strengthening. In the present specification, the value of the second preset threshold can be adjusted according to actual conditions, and is not limited to the value described in this step; the definition of the second label can also be adjusted according to actual conditions, and is not limited to the description in this step.

[0083] Step 403: When the average value of the distance between the smart device with the minimum distance from the current cluster center and the current cluster center is greater than or equal to the second preset threshold and less than the third preset threshold, a third label is added to the clustering result of the cluster center.

[0084] Correspondingly, if the third preset threshold is 0.45, it can be determined that the 13 smart devices with the minimum distance to the third current cluster center have an average distance of 0.4 greater than the second preset threshold of 0.35 and less than the third preset threshold of 0.45. A third label is added to the clustering result of the third current cluster center. In some embodiments of the present specification, the third label may be: small business volume, recommended to be abolished. In the present specification, the value of the third preset threshold can be adjusted according to actual conditions, and is not limited to the value described in this step; the definition of the third label can also be adjusted according to actual conditions, and is not limited to the description in this step.

[0085] Figure 5 The flowchart of a method for determining the type of a smart device according to an embodiment of the present invention is shown, which specifically includes the following steps:

[0086] Step 501: The smart device with the minimum distance to the cluster center corresponding to the first tag is determined as the first device, and the first device is a device that does not need to be improved. Figure 4 The three types of cluster centers corresponding to the first label, the second label, and the third label determined in the data are used to calculate the distances between multiple smart devices and each cluster center, and then obtain three distance values ​​between each smart device and the three types of cluster centers. Selecting the smallest distance value from the three distance values ​​of each device can determine which type of cluster center the smart device is closest to, and further determine that the smart device belongs to that type of cluster center. Specifically, when the smart device has the minimum distance value with the cluster center corresponding to the first label, the smart device is determined as the first device. Among them, the first device is a smart device with large business volume, good equipment maintenance, excellent business performance, and no need for improvement.

[0087] Step 502: The smart device with the minimum distance to the cluster center corresponding to the second tag is determined as the second device, and the second device is the smart device to be maintained. As described in step 501, when the smart device has the minimum distance to the cluster center corresponding to the second tag, the smart device is determined as the second device. The second device is a smart device with a large business volume but poor equipment maintenance, average performance, and the need for enhanced manual maintenance and improvement.

[0088] Step 503: The smart device with the minimum distance to the cluster center corresponding to the third tag is determined as the third device, and the third device is the smart device to be eliminated. As described above, when the smart device has the minimum distance to the cluster center corresponding to the third tag, the smart device is determined as the third device, and the third device is a smart device with a small business volume and to be eliminated.

[0089] like Figure 6 FIG. 1 is a schematic diagram of a structure of a smart device clustering device based on adiabatic quantum computing in an embodiment of the present invention. In this figure, the basic structure of the smart device clustering device based on adiabatic quantum computing is described. The functional units and modules therein can be implemented in software or in general chips or specific chips to achieve estimated interaction. The device specifically includes:

[0090] The characteristic data acquisition unit 601 is used to acquire characteristic data of multiple smart devices, where the characteristic data is related to the usage services of the smart devices and the environments in which the smart devices are located;

[0091] An unsupervised clustering unit 602, used to perform unsupervised clustering on the feature data using an adiabatic quantum computing clustering algorithm;

[0092] The class core feature determination unit 603 is used to determine the class core feature corresponding to each clustering result according to the clustering result after unsupervised clustering;

[0093] The device determining unit 604 is used to determine the type of the smart device according to the core features of the class.

[0094] This manual uses an improved clustering algorithm of adiabatic quantum computing to perform cluster analysis on the operation of smart devices in commercial bank branch outlets. Compared with the traditional branch outlets' operation and management methods for smart devices, this method uses cluster analysis methods to achieve efficient operation and precise deployment of smart devices; in addition, combined with adiabatic quantum computing capabilities, it effectively improves algorithm efficiency, reduces computational complexity, improves clustering accuracy and stability, narrows the scope of attention on branch smart device business, and directly focuses on branch clusters with excellent performance and branch clusters with unsatisfactory performance, which helps to regularly evaluate and grasp the operation and management of each branch's smart cabinet equipment, and provide a decision-making basis for further adjustment of resource allocation.

[0095] As an embodiment of this article, you can also refer to Figure 7 The figure shows a specific structural diagram of the smart device clustering device based on adiabatic quantum computing in this embodiment.

[0096] As an embodiment of this invention, the characteristic data acquisition unit 601 further includes: a sample data determination module 6011, which is used to pre-process the characteristic data of the smart device to obtain sample data of the smart device;

[0097] As an embodiment of this article, the unsupervised clustering unit 602 further includes:

[0098] The initial cluster center determination module 6021 is used to select a random number of sample data as initial cluster centers.

[0099] like Figure 8As shown, a computer device provided in an embodiment of this invention, the apparatus described in an embodiment of this invention can be a computer device in this embodiment, and the above method of this invention is executed. The computer device 802 may include one or more processors 804, such as one or more central processing units (CPUs), and each processing unit may implement one or more hardware threads. The computer device 802 may also include any memory 806, which is used to store any kind of information such as code, settings, data, etc. Non-limiting, for example, the memory 806 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any memory may use any technology to store information. Further, any memory may provide volatile or non-volatile retention of information. Further, any memory may represent a fixed or removable component of the computer device 802. In one case, when the processor 804 executes an associated instruction stored in any memory or a combination of memories, the computer device 802 may perform any operation of the associated instruction. The computer device 802 also includes one or more drive mechanisms 808 for interacting with any memory, such as a hard disk drive mechanism, an optical disk drive mechanism, etc.

[0100] The computer device 802 may also include an input / output module 810 (I / O) for receiving various inputs (via input devices 812) and for providing various outputs (via output devices 814). A specific output mechanism may include a presentation device 816 and an associated graphical user interface (GUI) 818. In other embodiments, the input / output module 810 (I / O), the input device 812, and the output device 814 may not be included, and the computer device 802 may be used as a computer device in a network. The computer device 802 may also include one or more network interfaces 820 for exchanging data with other devices via one or more communication links 822. One or more communication buses 824 couple the components described above together.

[0101] The communication link 822 may be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 822 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc. governed by any protocol or combination of protocols.

[0102] Corresponds to Figures 1 to 5 The method in this article also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are executed.

[0103] The embodiment of the present invention also provides a computer readable instruction, wherein when the processor executes the instruction, the program therein causes the processor to execute the following Figures 1 to 5 The method shown.

[0104] It should be understood that in the various embodiments of this document, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this document.

[0105] It should also be understood that in the embodiments of this article, the term "and / or" is only a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0106] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this article.

[0107] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0108] In the several embodiments provided herein, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.

[0109] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of this article.

[0110] In addition, each functional unit in each embodiment of this invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional unit.

[0111] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this article is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of this article. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0112] Specific embodiments are used in this article to illustrate the principles and implementation methods of this article. The description of the above embodiments is only used to help understand the methods and core ideas of this article. At the same time, for general technicians in this field, according to the ideas of this article, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on this article.

Claims

1. A smart device clustering method based on adiabatic quantum computing, characterized in that: The method comprises: Acquire characteristic data of a plurality of smart devices, wherein the characteristic data is related to application services of the smart devices and environments in which the smart devices are located, wherein the characteristic data at least includes any one of device number, customer ratio, device usage frequency, and device life cycle, or a combination thereof; The feature data is clustered unsupervisedly using an adiabatic quantum computing clustering algorithm, which includes: preprocessing the feature data of the smart device to obtain sample data of the smart device; selecting a random number of sample data as the initial cluster center; determining the cost function using the following formula to calculate the minimum distance between the sample data of the smart device and the initial cluster center: Among them, μ k represents the kth initial cluster center, x i represents sample data calculated according to the characteristic data of the i-th smart device, and M represents the total number of sample data; evolving to a final-state Hamiltonian according to the cost function and the adiabatic quantum system; determining the current cluster center according to the ground state of the final-state Hamiltonian; According to the clustering results after unsupervised clustering, determine the core features of each clustering result; According to the core features of the class, the type of smart device is determined, and the smart device includes: a first device, a second device and a third device, the first device is a device that does not need to be improved; the second device is a smart device to be maintained; and the third device is a smart device to be eliminated.

2. The intelligent device clustering method based on adiabatic quantum computing according to claim 1 is characterized in that: Preprocessing the characteristic data of the smart device further includes: Setting different weights for feature data of different dimensions in the smart device respectively; The sample data of each smart device is determined according to the feature data and the weight.

3. The intelligent device clustering method based on adiabatic quantum computing according to claim 2 is characterized in that: Determining the class core features corresponding to each clustering result according to the clustering results after unsupervised clustering includes: According to the characteristic data of the current cluster center calculated after unsupervised clustering, a label is added to each clustering result, and the label is the core feature of the class.

4. The intelligent device clustering method based on adiabatic quantum computing according to claim 3 is characterized in that: Adding labels to each clustering result includes: When the average value of the distance between the smart device with the minimum distance from the current cluster center and the current cluster center is less than a first preset threshold, a first label is added to the clustering result of the cluster center; When the average value of the distance between the smart device with the minimum distance from the current cluster center and the current cluster center is greater than or equal to the first preset threshold and less than the second preset threshold, a second label is added to the clustering result of the cluster center; When the average value of the distance between the smart device with the minimum distance from the current cluster center and the current cluster center is greater than or equal to the second preset threshold and less than the third preset threshold, a third label is added to the clustering result of the cluster center.

5. A smart device clustering device based on adiabatic quantum computing, characterized in that: The device comprises: A feature data acquisition unit, used to acquire feature data of a plurality of smart devices, wherein the feature data is related to the usage business of the smart devices and the environment in which the smart devices are located, wherein the feature data at least includes any one of device number, customer ratio, device usage frequency, and device life cycle, or a combination thereof; The clustering unit is used to perform unsupervised clustering on the feature data using an adiabatic quantum computing clustering algorithm, which includes: preprocessing the feature data of the smart device to obtain sample data of the smart device; selecting a random number of sample data as initial cluster centers; determining a cost function using the following formula to calculate the minimum distance between the sample data of the smart device and the initial cluster center: Among them, μ k represents the kth initial cluster center, x i represents sample data calculated according to the characteristic data of the i-th smart device, and M represents the total number of sample data; evolving to a final-state Hamiltonian according to the cost function and the adiabatic quantum system; determining the current cluster center according to the ground state of the final-state Hamiltonian; A class core feature determination unit, used to determine the class core feature corresponding to each clustering result according to the clustering result after unsupervised clustering; The device determination unit is used to determine the type of smart device according to the core characteristics of the class, and the smart device includes: a first device, a second device and a third device, the first device is a device that does not need to be improved; the second device is a smart device to be maintained; and the third device is a smart device to be eliminated.

6. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images