Enterprise management cost optimization method and system based on artificial intelligence
By obtaining enterprise management cost change data, using artificial intelligence models to determine abnormal probability and optimize change data, the problem of poor application of enterprise management cost optimization in the existing technology is solved, and a more scientific and reasonable cost optimization effect is achieved.
Patent Information
- Application Number
- CN202510123544.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-26
- Publication Date
- 2025-05-23
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the management of enterprise management costs, the existing technology generates enterprise management cost data based solely on demand, which is poor in application and is difficult to achieve scientific and reasonable cost optimization.
By obtaining the enterprise management cost change data within the preset time period, using the pre-trained first model to determine the abnormal probability, and then using the second model to determine the optimization change data based on the target change data and the target time point, to achieve scientific and reasonable optimization of enterprise management costs.
By analyzing the real data on changes in enterprise management costs, optimizing the matching of costs with the real situation of enterprise management, improving the realization of cost optimization and improving the optimization effect of enterprise management costs.
Smart Images
Figure CN120031197A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular, to an enterprise management cost optimization method and system based on artificial intelligence. Background Art
[0002] With the development of artificial intelligence technology, artificial intelligence technology is being applied to more and more scenarios. For example, artificial intelligence technology can be applied to enterprise management scenarios to achieve automated and intelligent data processing and analysis.
[0003] In the related art, for the management of enterprise management costs, corresponding enterprise management cost data can be generated according to the needs of enterprise managers to assist enterprise managers in formulating enterprise management costs. This method only generates enterprise management cost data based on needs, and has poor applicability. Summary of the invention
[0004] The purpose of the present invention is to provide an enterprise management cost optimization method and system based on artificial intelligence to achieve more scientific and reasonable enterprise management cost optimization and improve the optimization effect of enterprise management cost.
[0005] In order to achieve the above-mentioned objectives, in a first aspect, the present disclosure provides an enterprise management cost optimization method based on artificial intelligence, including: obtaining enterprise management cost change data within a preset time period, the enterprise management cost change data including change data corresponding to multiple types of enterprise management costs; determining the abnormal probability of the enterprise management cost according to the enterprise management cost change data through a pre-trained first model; determining target change data corresponding to the target type of enterprise management cost from the enterprise management cost change data according to the abnormal probability, and determining the target time point within the preset time period; determining the optimized change data corresponding to the target type of enterprise management cost according to the target change data and the target time point through a pre-trained second model, the starting time point corresponding to the optimized change data being the target time point.
[0006] Optionally, the pre-trained first model determines the abnormal probability of enterprise management costs based on the enterprise management cost change data, including: determining first change data corresponding to a first type of enterprise management costs from the change data corresponding to the multiple types of enterprise management costs, the change rate of the first type of enterprise management costs being less than a preset change rate; determining second change data corresponding to a second type of enterprise management costs from the change data corresponding to the multiple types of enterprise management costs, the change rate of the second type of enterprise management costs being greater than or equal to the preset change rate; extracting a first data feature based on the first change data; extracting a second data feature based on the second change data; determining the abnormal probability of enterprise management costs based on the first type, the first data feature, the second type and the second data feature through the pre-trained first model.
[0007] Optionally, the pre-trained first model determines the abnormal probability of enterprise management cost according to the first type, the first data feature, the second type and the second data feature, including: determining a first type weight according to the first type, and determining a second type weight according to the second type; inputting the first type weight, the first data feature, the second type weight and the second data feature into the pre-trained first model to obtain a first abnormal probability output by the pre-trained first model; inputting the first data feature and the second data feature into the pre-trained first model to obtain a second abnormal probability output by the pre-trained first model; and determining the abnormal probability of enterprise management cost according to the first abnormal probability and the second abnormal probability.
[0008] Optionally, the enterprise management cost optimization method also includes: obtaining a first training data set, the first training data set includes multiple first training samples, the multiple first training samples include training samples corresponding to different economic indexes and training samples corresponding to different enterprise management cost types, each first training sample includes: sample data characteristics and abnormal probability labels, the economic index is used to characterize the trend of economic changes; training the first model to be trained according to the first training data set to obtain the pre-trained first model.
[0009] Optionally, determining target change data corresponding to the target type of enterprise management cost from the enterprise management cost change data based on the abnormal probability, and determining the target time point within the preset time period include: determining a first type of feature corresponding to the enterprise management cost type that needs to be optimized based on the abnormal probability; determining target change data corresponding to the target type of enterprise management cost from the enterprise management cost change data based on the first type of feature; and determining the target time point from the preset time period based on the abnormal probability and the length of the preset time period.
[0010] Optionally, the target change data corresponding to the target type of enterprise management cost is determined from the enterprise management cost change data based on the abnormal probability, and the target time point within the preset time period is determined, including: determining the target time point from the preset time period based on the abnormal probability and the time length of the preset time period; determining a second type of feature of the starting optimization time that can meet the target time point based on the position of the target time point in the preset time period; and determining the target change data corresponding to the target type of enterprise management cost from the enterprise management cost change data based on the second type of feature.
[0011] Optionally, the pre-trained second model determines the optimization change data corresponding to the enterprise management cost of the target type according to the target change data and the target time point, including: generating a cost sequence to be optimized according to the target change data and the target time point, the cost sequence to be optimized including multiple sequence elements, and the sequence elements corresponding to the target time point correspond to a start time identifier; inputting the cost sequence to be optimized into the pre-trained second model to obtain the optimization cost sequence output by the pre-trained second model; obtaining a preset cost optimization duration; and determining the optimization change data corresponding to the enterprise management cost of the target type according to the optimization cost sequence and the preset cost optimization duration.
[0012] Optionally, the enterprise management cost optimization method also includes: obtaining a second training data set, the second training data set includes multiple second training samples, the multiple second training samples include training samples corresponding to different economic indexes and training samples corresponding to different enterprise management cost types, each second training sample includes: a sample cost sequence and a label cost sequence, the sample cost sequence includes sequence elements corresponding to a start time identifier; training the second model to be trained according to the second training data set to obtain the pre-trained second model.
[0013] Optionally, the enterprise management cost optimization method also includes: displaying the optimization change data; in response to receiving an optimization adjustment request triggered by a user, adjusting the target time point according to the optimization adjustment request to obtain an adjusted starting time point, and adjusting the target change data to obtain adjusted change data; generating an adjusted cost sequence to be optimized according to the adjusted change data and the adjusted starting time point, the adjusted cost sequence to be optimized including multiple sequence elements, and the sequence element corresponding to the adjusted starting time point corresponds to a starting time identifier; inputting the adjusted cost sequence to be optimized into the pre-trained second model to obtain the adjusted optimization cost sequence output by the pre-trained second model; and determining the adjusted optimization change data according to the adjusted optimization cost sequence.
[0014] In a second aspect, the present disclosure provides an enterprise management cost optimization system based on artificial intelligence, including: an acquisition module, used to acquire enterprise management cost change data within a preset time period, the enterprise management cost change data including change data corresponding to multiple types of enterprise management costs; a determination module, used to: determine the abnormal probability of the enterprise management cost according to the enterprise management cost change data through a pre-trained first model; determine the target change data corresponding to the target type of enterprise management cost from the enterprise management cost change data according to the abnormal probability, and determine the target time point within the preset time period; determine the optimized change data corresponding to the target type of enterprise management cost according to the target change data and the target time point through a pre-trained second model, and the starting time point corresponding to the optimized change data is the target time point.
[0015] Through the above technical solution, the change data corresponding to various types of enterprise management costs are obtained. Based on the change data, the abnormal probability of the enterprise management cost is first determined by the first model, and then the target change data and target time point of the target type of enterprise management cost are determined based on the abnormal probability. Then, the optimized change data is determined according to the target change data and the target time point through the second model, and the starting time point of the optimized change data is the target time point. On the one hand, since the abnormal probability is analyzed based on the real enterprise management cost change data, the enterprise management cost optimization can be matched with the actual situation of enterprise management; on the other hand, since the optimized change data is generated based on the target time point within the preset time period as the starting point, the feasibility of enterprise management cost optimization is higher. Therefore, this technical solution can achieve more scientific and reasonable enterprise management cost optimization and improve the optimization effect of enterprise management cost.
[0016] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure, but do not constitute a limitation to the present disclosure. In the accompanying drawings:
[0018] Figure 1 It is a schematic diagram of an application scenario shown according to an exemplary embodiment.
[0019] Figure 2 It is a flowchart of a method for optimizing enterprise management costs based on artificial intelligence shown according to an exemplary embodiment.
[0020] Figure 3 It is an application example diagram of a first model shown according to an exemplary embodiment.
[0021] Figure 4 It is an application example diagram of a second model shown according to an exemplary embodiment.
[0022] Figure 5 It is a block diagram of a system for optimizing enterprise management costs based on artificial intelligence shown according to an exemplary embodiment.
[0023] Figure 6 It is a block diagram of an electronic device shown according to an exemplary embodiment. Specific Embodiments
[0024] The following details the specific embodiments of the present disclosure in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only for explaining and understanding the present disclosure, and are not used to limit the present disclosure.
[0025] With the development of artificial intelligence technology, artificial intelligence technology is applied to more and more scenarios. For example, artificial intelligence technology can be applied to enterprise management scenarios to achieve automated and intelligent data processing and analysis, etc.
[0026] In related technologies, for the management of enterprise management costs, corresponding enterprise management cost data can be generated according to the needs of enterprise managers to assist enterprise managers in formulating enterprise management costs. In this way, only generating enterprise management cost data based on needs has poor applicability.
[0027] Based on this, an embodiment of the present disclosure provides a technical solution, which obtains change data corresponding to multiple types of enterprise management costs, and based on the change data, first determines the abnormal probability of the enterprise management cost through a first model, and then determines the target change data and target time point of the enterprise management cost of the target type based on the abnormal probability, and then determines the optimized change data according to the target change data and the target time point through a second model, and the starting time point of the optimized change data is the target time point.
[0028] On the one hand, since the abnormal probability is analyzed based on the real enterprise management cost change data, the enterprise management cost optimization can match the actual situation of enterprise management; on the other hand, since the optimization change data is generated based on the target time point within the preset time period as the starting point, the feasibility of enterprise management cost optimization is higher.
[0029] Therefore, this technical solution can achieve more scientific and reasonable enterprise management cost optimization and improve the optimization effect of enterprise management costs.
[0030] Figure 1 is a schematic diagram showing an application scenario according to an exemplary embodiment. Figure 1 As shown, in this application scenario, a client and a server are involved, and the client and the server are connected in communication.
[0031] The client can be used to implement various human-computer interaction functions; the server can be used as a data processing terminal. Through data interaction between the client and the server, the cost optimization of enterprise management can be achieved together.
[0032] The client and server in this application scenario may be part of an enterprise management system.
[0033] Regarding the client side, it can be a mini-program, an application, etc. Regarding the server side, it can be a server, a cloud, etc.
[0034] Figure 2 is a flowchart of an enterprise management cost optimization method based on artificial intelligence according to an exemplary embodiment. The method can be applied to Figure 1 The application scenarios shown are as follows: Figure 2 As shown, the method comprises the following steps:
[0035] Step S21, obtaining enterprise management cost change data within a preset time period, wherein the enterprise management cost change data includes change data corresponding to various types of enterprise management costs.
[0036] Step S22, determining the abnormal probability of the enterprise management cost according to the enterprise management cost change data through the pre-trained first model.
[0037] Step S23, according to the abnormal probability, determine the target change data corresponding to the enterprise management cost of the target type from the enterprise management cost change data, and determine the target time point within the preset time period.
[0038] Step S24, determining the optimized change data corresponding to the enterprise management cost of the target type according to the target change data and the target time point through the pre-trained second model, and the starting time point corresponding to the optimized change data is the target time point.
[0039] In step S21, a user (e.g., an enterprise manager) may initiate an enterprise management cost optimization request on the client, and the client may send the request to the server, which may respond based on the request. Furthermore, the user may further upload enterprise management cost change data through the client, which may then be synchronized to the server by the client to obtain enterprise management cost change data.
[0040] In some embodiments, enterprise management cost optimization requires a corresponding data basis, so change data within a period of time is required, and the period of time can be a preset time period.
[0041] In some embodiments, a preset time period range may be pre-set, and the user may upload the change data based on the preset time period range. For example, the preset time period may be three months, six months, or one year.
[0042] In some embodiments, the enterprise management cost change data involves change data corresponding to multiple types of enterprise management costs. The multiple types can be understood as types of enterprise management costs.
[0043] By way of example, the various types may include, but are not limited to: human resource costs, administrative costs, financial costs, marketing costs, research and development costs, logistics and supply chain costs, information technology costs, risk management costs, and management decision costs.
[0044] It is understandable that users can upload corresponding types of enterprise management cost change data according to their own needs.
[0045] Regarding the change data, it may involve the enterprise management costs corresponding to multiple time points. For example, assuming that the preset time period is three months, the change data may include the enterprise management costs corresponding to multiple time points within these three months. Each month in the three months may involve one or more time points, which is not limited here.
[0046] In step S22, the abnormal probability of the enterprise management cost can be determined according to the enterprise management cost change data through the pre-trained first model.
[0047] In some embodiments, the first model may be a large model or other neural network models, which is not limited herein.
[0048] The first model can be pre-trained to have the ability to predict abnormality probabilities.
[0049] In some embodiments, the abnormal probability can be determined directly based on various types of enterprise management costs, or after performing corresponding data analysis on various types of enterprise management costs, corresponding data features can be extracted and the abnormal probability can be determined based on the data features.
[0050] Therefore, as an optional implementation, step S22 includes: determining first change data corresponding to a first type of enterprise management cost from the change data corresponding to multiple types of enterprise management costs, and the change rate of the first type of enterprise management cost is less than a preset change rate; determining second change data corresponding to a second type of enterprise management cost from the change data corresponding to multiple types of enterprise management costs, and the change rate of the second type of enterprise management cost is greater than or equal to a preset change rate; extracting first data features based on the first change data; extracting second data features based on the second change data; determining the abnormal probability of enterprise management costs based on the first type, the first data features, the second type and the second data features through a pre-trained first model.
[0051] In some embodiments, the change rate of the enterprise management cost can be determined according to the specific enterprise management cost type. Therefore, it is not necessary to determine the change rate, and the change data can be directly classified based on the type of the enterprise management cost to obtain the first change data and the second change data.
[0052] The preset change rate may be a change rate within the range of 0 to 10%, which indicates that the enterprise management cost remains basically unchanged or changes very little.
[0053] Therefore, the management cost of the first type of enterprise has a smaller rate of change, while the management cost of the second type of enterprise has a higher rate of change.
[0054] For example, human resource costs, administrative costs and financial costs will not change much, so they can be classified as the first type of enterprise management costs. However, marketing costs, R&D costs, logistics and supply chain costs, information technology costs, risk management costs and management decision costs will change due to various circumstances and can be classified as the second type of enterprise management costs.
[0055] Furthermore, different data features may be extracted from the first change data and the second change data.
[0056] For example, for the first change data, since the change rate is small, the average value can be used as the data feature. For the second change data, since the change rate is high, data features of multiple dimensions such as the change rate, standard deviation, mean square error, maximum value, minimum value, etc. can be determined as corresponding data features.
[0057] Furthermore, the pre-trained first model determines the abnormal probability of the enterprise management cost based on the first type, the first data feature, the second type and the second data feature.
[0058] As an optional implementation, the abnormal probability of enterprise management cost is determined according to the first type, the first data feature, the second type and the second data feature through a pre-trained first model, including: determining the first type weight according to the first type, and determining the second type weight according to the second type; inputting the first type weight, the first data feature, the second type weight and the second data feature into the pre-trained first model to obtain the first abnormal probability output by the pre-trained first model; inputting the first data feature and the second data feature into the pre-trained first model to obtain the second abnormal probability output by the pre-trained first model; and determining the abnormal probability of enterprise management cost according to the first abnormal probability and the second abnormal probability.
[0059] In this implementation, corresponding type weights may be determined for specific enterprise management cost types.
[0060] In some embodiments, the higher the importance of enterprise management costs, the higher the corresponding type weight. For example, the type weight of human resource costs is higher, the type weight of administrative costs is lower, the type weight of financial costs is lower, and the type weights of marketing costs, R&D costs, logistics and supply chain costs, information technology costs, risk management costs, and management decision costs are higher. In different scenarios, different type weights can be configured according to actual needs, which is not limited here.
[0061] Furthermore, the first type weight, the first data feature, the second type weight, and the second data feature can all be input into the pre-trained first model to obtain the first abnormality probability output by the model. Then, the first data feature and the second data feature can be input into the first model to obtain the second abnormality probability.
[0062] It can be understood that the determination of the first abnormal probability is affected by weight, while the determination of the second abnormal probability is not affected by weight. The two abnormal probabilities can be integrated to obtain the final abnormal probability.
[0063] In some embodiments, the first abnormal probability and the second abnormal probability can be compared, and if the two abnormal probabilities differ greatly, the average value is used as the final abnormal probability. If the two abnormal probabilities differ slightly, the first abnormal probability can be determined as the final abnormal probability. The difference being large can be a difference greater than 10%.
[0064] It can be seen from the above implementation method of determining the abnormal probability that the first model can output the abnormal probability according to the data characteristics.
[0065] Therefore, as an optional implementation, the training process of the first model includes: obtaining a first training data set, the first training data set includes multiple first training samples, the multiple first training samples include training samples corresponding to different economic indexes and training samples corresponding to different enterprise management cost types, each first training sample includes: sample data characteristics and abnormal probability labels, and the economic index is used to characterize the trend of economic changes; training the first model to be trained according to the first training data set to obtain a pre-trained first model.
[0066] In this implementation, some training samples are collected under different economic indexes. Also, some training samples are collected under different enterprise management cost types. This allows the training samples to have differences in economic indexes and enterprise management cost types. Thus, the first model is trained using the training samples, so that the generalization performance of the first model is stronger.
[0067] In some embodiments, the method for extracting the sample data feature may refer to the method for extracting the first data feature and the second data feature described above.
[0068] In some embodiments, the abnormal probability labels may be manually labeled or labeled in other ways.
[0069] In some embodiments, when training is performed based on the first training data set, training samples corresponding to different economic indexes can be used for training first, and then training samples corresponding to different types of enterprise management costs can be used for training. Finally, training samples corresponding to different economic indexes and training samples corresponding to different types of enterprise management costs can be mixed for training.
[0070] Figure 3 is an example diagram of an application of a first model according to an exemplary embodiment. Figure 3 As shown, the training samples corresponding to different economic indexes and the training samples corresponding to different types of enterprise management costs are first used to implement the training of the first model to obtain the pre-trained first model.
[0071] Next, when applying the first model, the data to be optimized (i.e., the change data corresponding to various types of enterprise management costs) is divided by type. Based on the division results, data features are extracted and type weights are determined. Finally, the data features and type weights are combined to make predictions through the first model to obtain the abnormal probability.
[0072] In step S23, according to the abnormal probability, the target change data corresponding to the enterprise management cost of the target type is determined from the enterprise management cost change data, and the target time point within the preset time period is determined.
[0073] In this embodiment, the target type may be one or more types, and the target time point may be a specified time point within a preset time period.
[0074] As a first optional implementation, step S23 includes: determining the first type of characteristics corresponding to the enterprise management cost type that needs to be optimized based on the abnormal probability; determining the target change data corresponding to the target type of enterprise management cost from the enterprise management cost change data based on the first type of characteristics; determining the target time point from the preset time period based on the abnormal probability and the length of the preset time period.
[0075] In some embodiments, type features corresponding to different abnormal probability ranges may be pre-configured, and the corresponding type features may be determined according to the abnormal probability range matched by the current abnormal probability.
[0076] In some embodiments, the type feature corresponding to the high abnormal probability range can be the frequency of occurrence of high enterprise management costs, that is, if the abnormal probability is large, the enterprise management cost type with high enterprise management costs can be determined as the target type, and accordingly, the corresponding change data can be used as the target change data. For example, if the abnormal probability is large, human resource costs and management decision costs can be used as target types.
[0077] The type feature corresponding to the low abnormal probability range can be the importance of the enterprise management cost type. Furthermore, if the abnormal probability is small, the enterprise management cost type with a low importance can be determined as the target type, and accordingly, the corresponding change data can be used as the target change data. For example, if the abnormal probability is small, administrative costs and financial costs can be used as target types.
[0078] In some embodiments, the higher the abnormal probability, the longer the length of the preset time period, and the later the target time point is in the preset time period. The lower the abnormal probability, the shorter the length of the preset time period, and the earlier the target time point is in the preset time period.
[0079] As a second optional implementation, step S23 includes: determining a target time point from a preset time period based on the abnormal probability and the length of the preset time period; determining, based on the position of the target time point in the preset time period, that the starting optimization time can meet the second type of characteristics of the target time point; and determining, based on the second type of characteristics, target change data corresponding to the target type of enterprise management cost from the enterprise management cost change data.
[0080] In this embodiment, the target time point may be determined according to the above embodiment, and then, based on the position of the target time point in the preset time period, the starting optimization time may be determined to satisfy the second type of characteristics of the target time point.
[0081] In some embodiments, the earlier the target time point is in the preset time period, the faster the cost adjustment speed of the enterprise management cost type is. The later the target time point is in the preset time period, the slower the cost adjustment speed of the enterprise management cost type is.
[0082] Regarding the speed of cost adjustment, for example: the adjustment speed of human resource costs is faster, while the adjustment speed of R&D costs is slower, etc., or, in different application scenarios, there may be different implementation methods.
[0083] Therefore, based on the position of the target time point in the preset time period, the target change data corresponding to the enterprise management cost of the target type can be determined.
[0084] In different application scenarios, the above two implementation methods can be selected according to needs.
[0085] In step S24, the optimized change data corresponding to the enterprise management cost of the target type is determined according to the target change data and the target time point through the pre-trained second model.
[0086] In some embodiments, the second model can be a large model or other neural network model, which is not limited here.
[0087] As an optional implementation, step S24 includes: generating a cost sequence to be optimized based on the target change data and the target time point, the cost sequence to be optimized including multiple sequence elements, and the sequence elements corresponding to the target time point correspond to a start time identifier; inputting the cost sequence to be optimized into a pre-trained second model to obtain an optimized cost sequence output by the pre-trained second model; obtaining a preset cost optimization duration; and determining the optimized change data corresponding to the enterprise management cost of the target type based on the optimized cost sequence and the preset cost optimization duration.
[0088] In this implementation, the target change data may be converted into a sequence, and the starting time point may be marked in the sequence according to the target time point. The multiple sequence elements may be enterprise management costs arranged in chronological order.
[0089] Furthermore, the cost sequence to be optimized is input into the second model, and the optimized cost sequence output by the second model can be obtained.
[0090] In some embodiments, the length of the optimized cost sequence is not limited, so the optimized cost sequence can be further adjusted in combination with the preset cost optimization duration to convert and obtain the final optimized change data.
[0091] For example, if the optimized cost sequence length is greater than the cost optimization duration, some sequence elements in the middle of the optimized cost sequence can be omitted to meet the cost optimization duration. If the optimized cost sequence length is less than the cost optimization duration, some sequence elements can be added to the optimized cost sequence, and the added sequence elements need to be close to the existing sequence elements.
[0092] Then, the adjusted optimized cost sequence is converted into the enterprise management cost corresponding to each time point, and the optimized change data can be obtained. It can be understood that some of the time points in each time point may include time points that overlap with time points in the preset time period.
[0093] In some embodiments, the intervals between the various time points may be equal, so that based on the target time point, the optimized change data corresponding to each time point can be converted.
[0094] It can be seen that the second model can output an optimized cost sequence based on the input cost sequence to be optimized.
[0095] Therefore, as an optional implementation, the training process of the second model includes: obtaining a second training data set, the second training data set includes multiple second training samples, the multiple second training samples include training samples corresponding to different economic indexes and training samples corresponding to different enterprise management cost types, each second training sample includes: a sample cost sequence and a label cost sequence, the sample cost sequence includes sequence elements corresponding to a start time identifier; training the second model to be trained according to the second training data set to obtain a pre-trained second model.
[0096] In this implementation, some training samples are collected under different economic indexes. Also, some training samples are collected under different enterprise management cost types. The training samples are made to have differences in economic indexes and enterprise management cost types. Thus, the second model is trained using the training samples, so that the generalization performance of the second model is stronger.
[0097] In some embodiments, the sample cost series and the label cost series can be acquired based on real enterprise management cost change data and configured with the assistance of manual labeling.
[0098] When training based on the second training data set, you can first use training samples corresponding to different economic indices for training, then use training samples corresponding to different types of enterprise management costs for training, and finally mix training samples corresponding to different economic indices and training samples corresponding to different types of enterprise management costs for training.
[0099] Figure 4 is an example diagram of an application of the second model according to an exemplary embodiment. Figure 4 As shown, when conducting model training, model training is performed based on training samples under the dimensions of economic index and enterprise management cost type.
[0100] When the model is applied, a cost sequence to be optimized is generated based on the target time point and target change data, and the second model can output an optimized cost sequence based on the inputted cost sequence to be optimized. Adjustments are made based on the optimized cost sequence to obtain the final optimized change data.
[0101] After the optimization change data is obtained in step S24, it can be directly fed back to the client, and then fed back to the enterprise manager by the client.
[0102] For business managers, they can make requests for the feedback optimization change data to further adjust the data.
[0103] Therefore, as an optional implementation, the method also includes: displaying optimization change data; in response to receiving an optimization adjustment request triggered by a user, adjusting the target time point according to the optimization adjustment request to obtain an adjusted starting time point, and adjusting the target change data to obtain adjusted change data; generating an adjusted cost sequence to be optimized according to the adjusted change data and the adjusted starting time point, the adjusted cost sequence to be optimized including multiple sequence elements, and the sequence elements corresponding to the adjusted starting time point have corresponding starting time identifiers; inputting the adjusted cost sequence to be optimized into the pre-trained second model to obtain the adjusted optimization cost sequence output by the pre-trained second model; and determining the adjusted optimization change data according to the adjusted optimization cost sequence.
[0104] In this implementation, the optimization change data may be displayed through the client, and if the user believes that adjustment is required, an optimization adjustment request may be triggered.
[0105] In the optimization adjustment request, the adjustment method of the target time point and the adjustment method of the target type can be indicated. Therefore, based on the optimization adjustment request, the target time point can be adjusted to obtain the adjusted start time point, and the target change data can be adjusted to obtain the adjusted change data.
[0106] Furthermore, based on the adjusted change data and the adjusted starting time point, an adjusted cost sequence to be optimized can be generated. The adjusted cost sequence to be optimized is input into the second model to obtain the adjusted optimized cost sequence output by the second model.
[0107] Since the adjusted optimization cost sequence is already the result of adjustment based on user needs, there is no need to adjust it based on the preset optimization duration. The adjusted optimization cost sequence can be directly converted into adjusted optimization change data.
[0108] Through the introduction of the disclosed embodiments, it can be seen that, on the one hand, since the abnormal probability is analyzed based on the real enterprise management cost change data, the enterprise management cost optimization can be matched with the real situation of enterprise management; on the other hand, since the optimization change data is generated based on the target time point within the preset time period as the starting point, the feasibility of enterprise management cost optimization is higher. Therefore, this technical solution can achieve more scientific and reasonable enterprise management cost optimization and improve the optimization effect of enterprise management cost.
[0109] Figure 5 is a block diagram of an enterprise management cost optimization system based on artificial intelligence according to an exemplary embodiment. Figure 5 As shown, the system includes:
[0110] The acquisition module 501 is used to acquire enterprise management cost change data within a preset time period, wherein the enterprise management cost change data includes change data corresponding to multiple types of enterprise management costs.
[0111] Determination module 502 is used to: determine the abnormal probability of enterprise management cost according to the enterprise management cost change data through a pre-trained first model; determine the target change data corresponding to the enterprise management cost of the target type from the enterprise management cost change data according to the abnormal probability, and determine the target time point within the preset time period; determine the optimized change data corresponding to the enterprise management cost of the target type according to the target change data and the target time point through a pre-trained second model, and the starting time point corresponding to the optimized change data is the target time point.
[0112] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.
[0113] Figure 6 FIG. 6 is a block diagram of an electronic device 600 according to an exemplary embodiment. Figure 6 As shown, the electronic device 600 may include: a processor 601 and a memory 602. The electronic device 600 may also include one or more of a multimedia component 603, an input / output (I / O) interface 604, and a communication component 605.
[0114] The processor 601 is used to control the overall operation of the electronic device 600 to complete all or part of the steps in the above-mentioned enterprise management cost optimization method based on artificial intelligence. The memory 602 is used to store various types of data to support the operation of the electronic device 600, which may include, for example, instructions for any application or method used to operate on the electronic device 600, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 602 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 603 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, which is used to receive external audio signals. The received audio signal may be further stored in the memory 602 or sent through the communication component 605. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 604 provides an interface between the processor 601 and other interface modules, and the above-mentioned other interface modules may be keyboards, mice, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 605 is used for wired or wireless communication between the electronic device 600 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 605 may include: Wi-Fi module, Bluetooth module, NFC module.
[0115] In an exemplary embodiment, the electronic device 600 may be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned XXXX method.
[0116] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned enterprise management cost optimization method based on artificial intelligence are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 602 including program instructions, and the above-mentioned program instructions can be executed by the processor 601 of the electronic device 600 to complete the above-mentioned enterprise management cost optimization method based on artificial intelligence.
[0117] In another exemplary embodiment, a computer program product is also provided, which includes a computer program that can be executed by a processor, and when the computer program is executed by the processor, the steps of the above-mentioned enterprise management cost optimization method based on artificial intelligence are implemented.
[0118] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned enterprise management cost optimization method based on artificial intelligence are implemented.
[0119] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.
[0120] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.
[0121] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.
Claims
1. An enterprise management cost optimization method based on artificial intelligence, characterized in that: include: Acquire enterprise management cost change data within a preset time period, wherein the enterprise management cost change data includes change data corresponding to multiple types of enterprise management costs; Determining the abnormal probability of the enterprise management cost according to the enterprise management cost change data through the pre-trained first model; According to the abnormal probability, determining the target change data corresponding to the enterprise management cost of the target type from the enterprise management cost change data, and determining the target time point within the preset time period; The optimized change data corresponding to the enterprise management cost of the target type is determined by the pre-trained second model according to the target change data and the target time point, and the starting time point corresponding to the optimized change data is the target time point.
2. The enterprise management cost optimization method according to claim 1, characterized in that: The first pre-trained model determines the abnormal probability of the enterprise management cost according to the enterprise management cost change data, including: Determining first change data corresponding to a first type of enterprise management cost from the change data corresponding to the plurality of types of enterprise management costs, wherein a change rate of the first type of enterprise management cost is less than a preset change rate; Determining second change data corresponding to a second type of enterprise management cost from the change data corresponding to the plurality of types of enterprise management costs, wherein a change rate of the second type of enterprise management cost is greater than or equal to the preset change rate; Extracting a first data feature according to the first change data; extracting a second data feature according to the second change data; The abnormal probability of enterprise management cost is determined by a pre-trained first model according to the first type, the first data feature, the second type and the second data feature.
3. The enterprise management cost optimization method according to claim 2, characterized in that: The pre-trained first model determines the abnormal probability of the enterprise management cost according to the first type, the first data feature, the second type, and the second data feature, including: determining a first type weight according to the first type, and determining a second type weight according to the second type; Inputting the first type weight, the first data feature, the second type weight, and the second data feature into the pre-trained first model to obtain a first abnormality probability output by the pre-trained first model; Inputting the first data feature and the second data feature into the pre-trained first model to obtain a second abnormality probability output by the pre-trained first model; The abnormal probability of the enterprise management cost is determined according to the first abnormal probability and the second abnormal probability.
4. The enterprise management cost optimization method according to claim 3, characterized in that: The enterprise management cost optimization method further includes: Acquire a first training data set, the first training data set comprising a plurality of first training samples, the plurality of first training samples comprising training samples corresponding to different economic indexes and training samples corresponding to different types of enterprise management costs, each first training sample comprising: a sample data feature and an abnormal probability label, the economic index being used to characterize an economic change trend; The first model to be trained is trained according to the first training data set to obtain the pre-trained first model.
5. The enterprise management cost optimization method according to claim 1, characterized in that: The step of determining target change data corresponding to the enterprise management cost of the target type from the enterprise management cost change data according to the abnormal probability, and determining the target time point within the preset time period, comprises: Determine, according to the abnormal probability, a first type of feature corresponding to the enterprise management cost type that needs to be optimized; According to the first type of features, determining target change data corresponding to the enterprise management cost of the target type from the enterprise management cost change data; The target time point is determined from the preset time period according to the abnormal probability and the length of the preset time period.
6. The enterprise management cost optimization method according to claim 1, characterized in that: The step of determining target change data corresponding to the enterprise management cost of the target type from the enterprise management cost change data according to the abnormal probability, and determining the target time point within the preset time period, comprises: Determining the target time point from the preset time period according to the abnormal probability and the length of the preset time period; According to the position of the target time point in the preset time period, determining that the starting optimization time can satisfy the second type of characteristics of the target time point; According to the second type feature, target change data corresponding to the enterprise management cost of the target type is determined from the enterprise management cost change data.
7. The enterprise management cost optimization method according to claim 1, characterized in that: The second model pre-trained to determine the optimized change data corresponding to the enterprise management cost of the target type according to the target change data and the target time point includes: Generate a cost sequence to be optimized according to the target change data and the target time point, wherein the cost sequence to be optimized includes a plurality of sequence elements, and the sequence element corresponding to the target time point has a corresponding start time identifier; Inputting the cost sequence to be optimized into the pre-trained second model to obtain an optimized cost sequence output by the pre-trained second model; Get the preset cost optimization duration; According to the optimized cost sequence and the preset cost optimization duration, the optimized change data corresponding to the enterprise management cost of the target type is determined.
8. The enterprise management cost optimization method according to claim 7, characterized in that: The enterprise management cost optimization method further includes: Acquire a second training data set, the second training data set includes a plurality of second training samples, the plurality of second training samples include training samples corresponding to different economic indexes and training samples corresponding to different enterprise management cost types, each second training sample includes: a sample cost sequence and a label cost sequence, the sample cost sequence includes a sequence element corresponding to a start time identifier; The second model to be trained is trained according to the second training data set to obtain the pre-trained second model.
9. The enterprise management cost optimization method according to claim 1, characterized in that: The enterprise management cost optimization method further includes: Displaying the optimization change data; In response to receiving an optimization adjustment request triggered by a user, adjusting the target time point according to the optimization adjustment request to obtain an adjusted start time point, and adjusting the target change data to obtain adjusted change data; Generate an adjusted cost sequence to be optimized according to the adjusted change data and the adjusted starting time point, wherein the adjusted cost sequence to be optimized includes a plurality of sequence elements, and the sequence element corresponding to the adjusted starting time point has a corresponding starting time identifier; Inputting the adjusted cost sequence to be optimized into the pre-trained second model to obtain the adjusted optimized cost sequence output by the pre-trained second model; According to the adjusted optimization cost sequence, the adjusted optimization change data is determined.
10. An enterprise management cost optimization system based on artificial intelligence, characterized in that: include: An acquisition module, used to acquire enterprise management cost change data within a preset time period, wherein the enterprise management cost change data includes change data corresponding to various types of enterprise management costs; Identify modules for: Determining the abnormal probability of the enterprise management cost according to the enterprise management cost change data through the pre-trained first model; According to the abnormal probability, determining the target change data corresponding to the enterprise management cost of the target type from the enterprise management cost change data, and determining the target time point within the preset time period; The optimized change data corresponding to the enterprise management cost of the target type is determined by the pre-trained second model according to the target change data and the target time point, and the starting time point corresponding to the optimized change data is the target time point.