CN117376982A - Service node capacity control method, device, equipment and storage medium - Google Patents
Service node capacity control method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN117376982A CN117376982A CN202311386682.XA CN202311386682A CN117376982A CN 117376982 A CN117376982 A CN 117376982A CN 202311386682 A CN202311386682 A CN 202311386682A CN 117376982 A CN117376982 A CN 117376982A
- Authority
- CN
- China
- Prior art keywords
- service
- data
- demand prediction
- influence
- hot spot
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000008602 contraction Effects 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims description 67
- 230000003993 interaction Effects 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 13
- 230000006399 behavior Effects 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 10
- 238000007621 cluster analysis Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 230000008451 emotion Effects 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 4
- 235000019633 pungent taste Nutrition 0.000 claims 1
- 230000006870 function Effects 0.000 description 13
- 238000004891 communication Methods 0.000 description 9
- 238000013528 artificial neural network Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 239000002699 waste material Substances 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000001105 regulatory effect Effects 0.000 description 3
- 230000001276 controlling effect Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000035755 proliferation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/08—Load balancing or load distribution
- H04W28/09—Management thereof
- H04W28/0925—Management thereof using policies
- H04W28/0933—Management thereof using policies based on load-splitting ratios
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/08—Load balancing or load distribution
- H04W28/09—Management thereof
- H04W28/0992—Management thereof based on the type of application
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
- H04W28/24—Negotiating SLA [Service Level Agreement]; Negotiating QoS [Quality of Service]
Abstract
The embodiment of the invention discloses a service node capacity control method, a device, equipment and a storage medium, which comprise the following steps: acquiring network hot spot data corresponding to an application to be controlled; the network hot spot data are non-standard format data of natural language types; inputting network hot spot data into a pre-trained service demand prediction model, and determining service demand prediction results of service nodes of each region corresponding to an application to be controlled according to output results of the service demand prediction model; aiming at each regional service node, acquiring the current service load of the regional service node, and performing expansion and contraction control on the regional service node according to the service demand prediction result and the current service load corresponding to the regional service node; the service demand prediction model comprises a semantic clustering module, a heat influence determining module and a service demand prediction module. The accuracy and timeliness of capacity expansion adjustment aiming at different service nodes are improved, so that the application can run more accurately, intelligently and stably during running.
Description
Technical Field
The present invention relates to the field of mobile communications technologies, and in particular, to a method, an apparatus, a device, and a storage medium for controlling a service node capacity.
Background
With the rapid development of communication technology, networks can carry multiple types of traffic, and for one network application, there is also a higher demand for required network traffic due to the different functions. Therefore, with the proliferation of various network service flows, the expansion and contraction capacity management of different service services is realized through flow prediction, and a particularly key effect is played on service bearing.
However, the existing scheme for traffic prediction management is often based on application level, combines short-term related characteristics of network traffic, completes prediction of application-ready demand traffic according to average traffic applied in a history process, and the considered data is often obtained from a single source, has a uniform structure and is difficult to process complex and dynamic network traffic data. In addition, in the aspect of prediction, only time-related information is often considered, hot spot information generated in real time by combining a network is difficult to provide more accurate, intelligent and stable flow demand prediction for service nodes for providing different functions in one application, and the increasingly improved demands of enterprises on the aspects of service demand prediction accuracy and system stability are difficult to meet.
Disclosure of Invention
The invention provides a service node capacity control method, a device, equipment and a storage medium, which are used for determining flow changes caused by the influence of network hot spots of different services in application by analyzing and predicting network hot spot data with wide sources, so as to timely perform expansion and contraction capacity adjustment on service nodes supporting different services, and improve the system stability during the operation of the application.
In a first aspect, an embodiment of the present invention provides a service node capacity control method, including:
acquiring network hot spot data corresponding to an application to be controlled; the network hot spot data are non-standard format data of natural language types;
inputting network hot spot data into a pre-trained service demand prediction model, and determining service demand prediction results of service nodes of each region corresponding to an application to be controlled according to output results of the service demand prediction model;
aiming at each regional service node, acquiring the current service load of the regional service node, and performing expansion and contraction control on the regional service node according to the service demand prediction result and the current service load corresponding to the regional service node;
the service demand prediction model at least comprises a semantic clustering module, a heat influence determining module and a service demand prediction module.
In a second aspect, an embodiment of the present invention further provides a service node capacity control device, including:
the data acquisition module is used for acquiring network hot spot data corresponding to the application to be controlled; the network hot spot data are non-standard format data of natural language types;
the service demand prediction module is used for inputting network hot spot data into the pre-trained service demand prediction model, and determining service demand prediction results of the service nodes of each region corresponding to the application to be controlled according to the output results of the service demand prediction model;
the node capacity control module is used for acquiring the current service load of the regional service node aiming at each regional service node, and performing expansion and contraction control on the regional service node according to the service demand prediction result and the current service load corresponding to the regional service node;
the service demand prediction model at least comprises a semantic clustering module, a heat influence determining module and a service demand prediction module.
In a third aspect, an embodiment of the present invention further provides a service node capacity control device, including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the service node capacity control method provided by the embodiment of the present invention.
In a fourth aspect, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform a service node capacity control method provided by embodiments of the present invention.
The embodiment of the invention provides a service node capacity control method, a device, equipment and a storage medium, which are used for acquiring network hot spot data corresponding to an application to be controlled; the network hot spot data are non-standard format data of natural language types; inputting network hot spot data into a pre-trained service demand prediction model, and determining service demand prediction results of service nodes of each region corresponding to an application to be controlled according to output results of the service demand prediction model; aiming at each regional service node, acquiring the current service load of the regional service node, and performing expansion and contraction control on the regional service node according to the service demand prediction result and the current service load corresponding to the regional service node; the service demand prediction model at least comprises a semantic clustering module, a heat influence determining module and a service demand prediction module. According to the technical scheme, the service demand prediction model integrated with the semantic clustering function, the heat influence determining function and the service demand prediction function is utilized to capture non-standard format network hot spot data related to the application to be controlled in the network, such as semantic analysis, feature extraction, heat determination, influence determination and the like are carried out, further, the possible service demands of different services in the application to be controlled in a future period are predicted and determined based on the processed information such as the feature tag, the heat value, the influence and the like, and as different services provided in the application to be controlled can be supported by different regional service nodes in operation, the service flow required by the different regional service nodes in the future period can be determined according to the prediction result, further, the expansion and contraction control of the regional service nodes can be realized based on the determined service flow and the current service load of the regional service node at the current moment, the dynamic control of the service capacity of the different regional service nodes corresponding to the application to be controlled is realized according to the network hot spot condition, the service capacity of the different regional service nodes to be controlled is prevented from being down due to a large amount of access, the service capacity of the regional service nodes to be controlled is reduced, the waste is improved, the service nodes to be used in the service nodes can be accurately regulated in running in time, and the intelligent operation can be more stable.
It should be understood that the description of this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a service node capacity control method according to a first embodiment of the present invention;
fig. 2 is a flowchart of a service node capacity control method according to a second embodiment of the present invention;
fig. 3 is a flowchart illustrating a procedure of inputting each feature classification label to a heat influence determining module to determine a heat value and an influence value corresponding to network hotspot data according to a second embodiment of the present invention;
FIG. 4 is a training flowchart of a service demand prediction model according to a second embodiment of the present invention;
Fig. 5 is a schematic structural diagram of a service node capacity control device according to a third embodiment of the present invention;
fig. 6 is a schematic structural diagram of a service node capacity control device according to a fourth embodiment of the present invention.
Detailed Description
In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a flowchart of a service node capacity control method according to an embodiment of the present invention, where the embodiment of the present invention is applicable to a case of performing service capacity adjustment on an area service node supporting different services. The method may be performed by a serving node capacity control apparatus, which may be configured in a serving node capacity control device. Alternatively, the service node capacity control device may be a notebook, a desktop computer, an intelligent tablet, or the like, which is not limited in the embodiment of the present invention.
As shown in fig. 1, the method for controlling the capacity of a service node provided by the embodiment of the invention specifically includes the following steps:
s101, acquiring network hot spot data corresponding to an application to be controlled.
The network hot spot data is non-standard format data of natural language type.
In this embodiment, the application to be controlled may be specifically understood as an application that performs dynamic control on service capacity of different service nodes according to a preset requirement of an actual situation. Network hotspot data can be understood in particular as data information which is acquired by the network in real time and which relates to the application to be controlled. Optionally, the network hotspot data may be news and comments with more clicks, more forwarding or higher discussion on the network, or may be other data aiming at user behavior, user feedback of an application to be controlled and influence of a user in a social network, which is not limited in the embodiment of the present invention.
Specifically, when it is desired to dynamically regulate the service capacities of different service nodes in an application to be controlled, different operations and feedback behaviors of a user for the application to be controlled can be obtained in real time in a network through a web crawler, and the user determines the collected behavior information, the published content and other information as network hot spot data corresponding to the application to be controlled according to the content, such as news articles, social media posts, comment replies and the like, which are published in real time in the network by the application to be controlled, wherein the obtained network hot spot data is data directly extracted from the network, and is mostly natural language type data instead of standard format data which can be directly used for service capacity prediction.
Optionally, when the web crawler acquires the web hotspot data, the web crawler may screen the data acquired by the web crawler, and only the data related to the application to be controlled and having the click rate or influence exceeding the preset threshold is determined as the web hotspot data corresponding to the application to be controlled.
S102, inputting network hot spot data into a pre-trained service demand prediction model, and determining service demand prediction results of the regional service nodes corresponding to the application to be controlled according to the output results of the service demand prediction model.
The service demand prediction model at least comprises a semantic clustering module, a heat influence determining module and a service demand prediction module.
In this embodiment, the service demand prediction model may be specifically understood as a neural network model for extracting and clustering features such as semantic features of the network hotspot data of the natural sentence type input therein, determining indexes related to heat and influence contained therein, and predicting different service demands in a targeted manner according to the extracted features and the determined indexes. It can be understood that each service demand prediction model is a model based on service specific training supported by the corresponding application, and can output service demands of service nodes of different areas corresponding to various services supported by the application respectively. A regional service node is to be understood as a service node in particular for supporting different services in an application to be controlled. The service demand prediction result can be specifically understood as a service capacity required by a regional service node for an application to be controlled in a future period of time, which is output by the service demand prediction model.
In this embodiment, the semantic clustering module may be specifically understood as a set of neural network layers for performing natural semantic extraction, feature extraction, and clustering on the input of the natural language type data therein. Alternatively, the semantic clustering module may be a neural network sub-model with natural semantic extraction and feature cluster analysis, which is obtained by combining and adjusting a natural semantic model and a cyclic neural network or a convolutional neural network. The heat influence determination module may be understood in particular as determining a set of neural network layers corresponding to heat values and influences based on inputting features therein related to heat and positive and negative influences. The service demand prediction module may be specifically understood as a set of neural network layers that predicts service capacities required by service nodes in different areas corresponding to the application to be controlled according to data input therein and having heat, influence and feature labels.
Specifically, network hot spot data are input into a pre-trained service demand prediction model, natural semantic extraction and cluster analysis of the network hot spot data are completed through a semantic clustering module in the service demand prediction model, various labels of different categories which are consistent with the network hot spot data are obtained, then the heat and influence of the network hot spot data are determined for each determined label through a heat influence determining module, the determined labels, heat and influence are integrated as intermediate results and then are input into the service demand prediction module, and the output results are determined to be service demand prediction results of service nodes of each area corresponding to an application to be controlled.
S103, aiming at each regional service node, acquiring the current service load of the regional service node, and performing expansion and contraction control on the regional service node according to the service demand prediction result and the current service load corresponding to the regional service node.
In this embodiment, the current service load may be specifically understood as the service capacity given by the regional service node at the current moment.
Specifically, for each regional service node, the corresponding service demand prediction result is a predicted service capacity which may be needed by the regional service node in a future period, and the obtained current service load is the service capacity of the regional service node at the current moment, so that the regional service node can meet the service demand of an application to be controlled in the future period, the service capacity of the regional service node needs to be adjusted from the current service load to the service capacity corresponding to the corresponding service demand prediction result, and the process is the capacity expansion control for the regional service node.
According to the technical scheme, network hot spot data corresponding to the application to be controlled are obtained; the network hot spot data are non-standard format data of natural language types; inputting network hot spot data into a pre-trained service demand prediction model, and determining service demand prediction results of service nodes of each region corresponding to an application to be controlled according to output results of the service demand prediction model; aiming at each regional service node, acquiring the current service load of the regional service node, and performing expansion and contraction control on the regional service node according to the service demand prediction result and the current service load corresponding to the regional service node; the service demand prediction model at least comprises a semantic clustering module, a heat influence determining module and a service demand prediction module. According to the technical scheme, the service demand prediction model integrated with the semantic clustering function, the heat influence determining function and the service demand prediction function is utilized to capture non-standard format network hot spot data related to the application to be controlled in the network, such as semantic analysis, feature extraction, heat determination, influence determination and the like are carried out, further, the possible service demands of different services in the application to be controlled in a future period are predicted and determined based on the processed information such as the feature tag, the heat value, the influence and the like, and as different services provided in the application to be controlled can be supported by different regional service nodes in operation, the service flow required by the different regional service nodes in the future period can be determined according to the prediction result, further, the expansion and contraction control of the regional service nodes can be realized based on the determined service flow and the current service load of the regional service node at the current moment, the dynamic control of the service capacity of the different regional service nodes corresponding to the application to be controlled is realized according to the network hot spot condition, the service capacity of the different regional service nodes to be controlled is prevented from being down due to a large amount of access, the service capacity of the regional service nodes to be controlled is reduced, the waste is improved, the service nodes to be used in the service nodes can be accurately regulated in running in time, and the intelligent operation can be more stable.
Example two
Fig. 2 is a flowchart of a service node capacity control method according to the second embodiment of the present invention, where the technical solution of the present invention is further optimized based on the foregoing alternative technical solutions, and the obtained network hot spot data is input to a semantic clustering module to obtain a plurality of feature classification labels, so that the service capacity of the low-access area service node is reduced and the waste is avoided based on the data source label, the user interaction label, the content quality label and the positive and negative direction influence feature label in each feature classification label, the heat value and the influence value corresponding to the network hot spot data are determined in a heat influence determining module, and finally, each feature classification label, the heat value and the influence value are input to a trained service demand prediction module, so as to obtain a service demand prediction result of each area service node corresponding to an application to be controlled, and perform expansion and contraction control on the area service node based on each service demand prediction result and the current service load of the corresponding area service node, thereby ensuring that the area service node to be impacted is not down due to a large amount of access, and the service capacity of the low-access area service node is avoided, and the accuracy and the performance of expansion and the capacity adjustment for different service nodes are improved, so that the intelligent and stable operation can be performed in time.
As shown in fig. 2, a service node capacity control method provided in a second embodiment of the present invention specifically includes the following steps:
s201, acquiring network hot spot data corresponding to an application to be controlled.
The network hot spot data is non-standard format data of natural language type.
S202, preprocessing the network hot spot data.
The data preprocessing comprises at least one of word segmentation, stop word removal and word stem processing.
Specifically, in order to ensure that the network hotspot data can be directly used for semantic feature extraction and cluster analysis, some information which may affect semantic recognition and feature extraction and is contained in the directly acquired network hotspot data needs to be filtered, such as operations of removing stop words and the like, and in order to ensure the correctness of semantic recognition, preprocessing such as word segmentation and word stem operation and the like needs to be performed on the acquired network hotspot data.
S203, determining the data and the processed network hot spot data as new network hot spot data, and inputting the new network hot spot data into a semantic clustering module for feature extraction and cluster analysis.
S204, inputting the network hotspot data into a semantic clustering module, extracting features of the network hotspot data, performing cluster analysis on the extracted semantic features and data features, and determining a plurality of feature classification labels corresponding to the network hotspot data.
Specifically, inputting network hotspot data into a semantic clustering module, extracting semantic features of the network hotspot data by utilizing a neural network layer related to a natural semantic model, extracting features of data information except natural language in the network hotspot data, such as praise amount, data sources, behavior types and the like by utilizing a neural network layer related to a cyclic neural network or a convolutional neural network, finally extracting to obtain semantic features and data features, and clustering each semantic feature and data feature according to a cluster center which is preset to obtain a plurality of feature classification labels which correspond to the network hotspot data and contain different types of information.
S205, inputting the feature classification labels into a heat influence determining module, and determining a heat value and an influence value corresponding to the network hot spot data.
Specifically, each feature classification label is input to a heat influence determining module, and is respectively substituted into a determining formula of a heat value and an influence value according to different information types contained in the feature classification label, and finally the heat value and the influence value corresponding to network hot spot data are output.
Optionally, fig. 3 is a flowchart illustrating a procedure of inputting each feature classification label to a heat influence determining module to determine a heat value and an influence value corresponding to network hotspot data, and as shown in fig. 3, the method specifically includes the following steps:
S2051, determining a data source label related to the data source, a user interaction label related to the user interaction and a content quality label related to the content quality from the feature classification labels.
The data source tag may be a tag determined according to the access amount of the network hotspot data corresponding to the data source, the user interaction tag may be a tag extracted based on information related to the number of points or comments interacted with by the user in the network hotspot data, and the content quality tag may be specifically understood as a tag extracted according to information related to the content quality accuracy and credibility in the network hotspot data.
S2052, summing the product of the data source label and the preset data source weight, the product of the user interaction label and the preset user interaction weight, and the product of the content quality label and the preset content quality weight to determine a heat value corresponding to the network hot spot data.
In this embodiment, the data source weight, the user interaction weight and the content quality weight may be specifically understood as weight coefficients preset according to practical situations, and may be determined according to the reliability of different data sources, the activity level of user interaction and the accuracy of content quality. Optionally, the data source weight, the user interaction weight and the content quality weight may be manually determined in advance based on historical statistics, or may be used as an adjustable weight parameter in the model training process, and may be determined and adjusted in the training process, which is not limited in the embodiment of the present invention.
The determination formula of the corresponding heat value of the network hot spot data is shown as follows:
heat value = data source weight + data source tag + user interaction weight + user interaction tag + content quality weight + content quality tag
It can be understood that when there are multiple tags of the same type, normalization processing can be performed on the products of the tags and the corresponding weights, so as to complete the unification of the heat value determination of different network hot spot data.
S2053, determining positive influence feature labels with positive influence and negative influence feature labels with negative influence from the feature classification labels.
Optionally, determining an influence type of the feature classification tag according to at least one of emotion tendencies, user behaviors, user feedback and social influence contained in each feature classification tag; and determining the positive influence weight or the negative influence weight corresponding to each feature classification label according to the influence type of each feature classification label.
For example, when feature extraction is performed on network hotspot data by using an emotion analysis technology, particularly, emotion tendencies of a user are extracted, and then feature classification labels containing the emotion tendencies are output. The labels including the purchasing behavior, the sharing behavior, the evaluating behavior, and the like of the user in each feature classification label may be determined as positively influencing feature labels, and the labels including the unsubscribing behavior, the complaint behavior, and the like of the user in each feature classification label may be determined as negatively influencing feature labels. The positive and negative surface effects of each feature classification label can be determined according to feedback information such as user scores, comments, complaints and the like contained in each feature classification label. The positive and negative classifications of different feature classification tags may be determined based on the contribution of the influence of network hotspot data in the social network to the positive and negative influence. For example, the number of fans, forwarding numbers and the like of the users on the social media can be used as indexes of positive influence, and the number of negative comments of the users on the social media can be used as indexes of negative influence.
S2054, multiplying each positive influence characteristic label with the corresponding positive influence weight respectively, accumulating and summing the positive influence weights, and determining the positive influence value.
Specifically, each positive influence characteristic label is multiplied by a corresponding positive influence weight, products obtained after multiplication are accumulated and summed, and the sum is determined as a positive influence value. Optionally, the products may be normalized before being cumulatively summed to unify the dimensions of the positive and negative impact values.
S2055, multiplying each negative influence characteristic label with the corresponding negative influence weight respectively, accumulating and summing the negative influence weights to determine a negative influence value.
Specifically, each negative influence characteristic label is multiplied by its corresponding negative influence weight, and the products obtained after the multiplication are accumulated and summed, and the sum is determined as a negative influence value. Optionally, the products may be normalized before being cumulatively summed to unify the negative impact value and the dimension of the negative impact value.
S2056, determining the difference between the positive influence value and the negative influence value as the influence value corresponding to the network hot spot data.
S206, writing the feature classification labels, the heat value and the influence value which are related to the service influence area in the feature classification labels into a preset standard format template, inputting the written standard format template into a service demand prediction module, and determining the output of the service demand prediction module as the output result of the service demand prediction model.
In this embodiment, the standard format template may be specifically understood as a template preset according to the actual situation and used to uniformly input the data format of the service requirement prediction module.
Specifically, feature classification labels related to the service influence area in each feature classification label, namely feature classification labels of service nodes of the influence determination area, and the heat value and the influence value are written into a preset standard format template and input into a service demand prediction module for prediction, so that service demand prediction results of service nodes of areas corresponding to different services in the application to be controlled can be obtained.
S207, determining a capacity adjustment value of the regional service node according to the service demand prediction result corresponding to the regional service node and the difference value of the current service load.
Specifically, the service capacity value in the service demand prediction result corresponding to the regional service node is differenced from the current service load, the difference is the capacity adjustment value which needs to be adjusted by the regional service node, when the difference is positive, the regional service node can be considered to be expanded, and when the difference is negative, the regional service node can be considered to be contracted.
And S208, performing expansion and contraction control on the regional service node on the basis of the current service load according to the capacity adjustment value.
Optionally, before acquiring network hotspot data corresponding to an application to be controlled, training of a service demand prediction model is further required to be completed, and fig. 4 is a training flow chart of the service demand prediction model provided by the second embodiment of the present invention, as shown in fig. 4, and specifically includes the following steps:
s301, acquiring a historical network hotspot data set and historical regional node service requirements corresponding to each historical network hotspot data set.
In this embodiment, the historical network hotspot data set may be specifically understood as data information related to the application to be controlled, which is acquired by the network during a period of history time. The historical regional node service requirement can be specifically understood as a regional service requirement which is finally generated by responding to historical network hot spot data by the regional service node corresponding to each service in the application to be controlled.
S302, determining a historical feature classification label, a historical heat value and a historical influence value of each historical network hot spot data.
Specifically, the corresponding historical feature classification label, the historical heat value and the historical influence value are determined through manual labeling or analysis aiming at historical network hot spot data.
S303, constructing training samples according to each historical network hot spot data, and historical feature classification labels, historical heat values, historical influence values and historical area node service requirements corresponding to the historical network hot spot data, and constructing a training sample set according to each training sample.
Specifically, for each historical network hotspot data, a corresponding historical feature classification label is used as a label of the historical network hotspot data to construct a first training sub-sample, and a set of each first training sub-sample is determined to be a first training sample subset. And constructing a second training subsamples by taking the historical heat value and the historical influence value corresponding to each historical network hot spot data as the labels of the corresponding historical feature classification labels, and determining the set of each second training subsamples as a second training sample subset. And constructing a third training subsamples by taking the history area node service requirements as corresponding history feature classification labels, history heat values and history influence values, and determining a set of each third training subsamples as a third training sample subset. And determining a set constructed by the first training sample subset, the second training sample subset and the third training sample subset together as a training sample set.
S304, training an initial semantic clustering module, an initial heat influence determining module and an initial service demand predicting module in the initial service demand predicting model through a first training sample subset, a second training sample subset and a third training sample subset which are contained in the training sample set.
The first training sample subset is a training sample subset formed by each historical network hot spot data and a corresponding historical characteristic classification label; the second training sample subset is a training sample subset formed by each historical feature classification label, a corresponding historical heat value and a corresponding historical image value; the third training sample subset is a training sample subset formed by each historical feature classification label, a historical heat value, a historical image value and corresponding historical area node service requirements.
Specifically, an initial semantic clustering module in an untrained initial service demand prediction model is trained through a first training sample subset in the training sample subsets, so that the trained semantic clustering module has the capability of extracting and clustering semantic features and data features of natural language paragraphs input into the semantic clustering module. And training an initial heat influence determining module in the initial service demand prediction model through the second training sample subset to adjust parameters of weights related to heat and influence in the initial heat influence determining module, so that the trained heat influence determining module has the capability of correctly outputting corresponding heat values and influence of network hot spot data. And training the initial service demand prediction module in the initial service demand prediction model through the third training sample subset, so that the trained service demand prediction module has the capability of outputting service capacity prediction results required by different services in corresponding applications according to the heat value, the influence and the characteristic classification labels.
According to the technical scheme, the acquired network hot spot data are input into the semantic clustering module to obtain a plurality of output feature classification labels, and further, based on the data source labels, the user interaction labels, the content quality labels and the positive and negative influence feature labels in the feature classification labels, the heat value and the influence value corresponding to the network hot spot data are determined in the heat influence determination module, and finally, the feature classification labels, the heat value and the influence value are input into the trained service demand prediction module to obtain service demand prediction results of the application to be controlled corresponding to the regional service nodes, expansion and contraction control of the regional service nodes is completed based on the service demand prediction results and the current service loads of the corresponding regional service nodes, so that the service nodes of the area to be impacted are prevented from being down due to a large number of accesses, the service capacity of the regional service nodes with low access is reduced, waste is avoided, the accuracy and timeliness of capacity adjustment for different service nodes are improved, and the application can run more accurately, intelligently and stably during running.
Example III
Fig. 5 is a schematic structural diagram of a service node capacity control device according to a third embodiment of the present invention, and as shown in fig. 5, the service node capacity control device may include a data acquisition module 31, a service demand prediction module 32, and a node capacity control module 33.
The data acquisition module 31 is configured to acquire network hotspot data corresponding to an application to be controlled; the network hot spot data are non-standard format data of natural language types; the service demand prediction module 32 is configured to input network hotspot data into a pre-trained service demand prediction model, and determine a service demand prediction result of each regional service node corresponding to the application to be controlled according to an output result of the service demand prediction model; the node capacity control module 33 is configured to obtain, for each regional service node, a current service load of the regional service node, and perform expansion and contraction control on the regional service node according to a service demand prediction result and the current service load corresponding to the regional service node; the service demand prediction model at least comprises a semantic clustering module, a heat influence determining module and a service demand prediction module.
According to the technical scheme, the service demand prediction model integrated with the semantic clustering function, the heat influence determining function and the service demand prediction function is utilized to capture non-standard format network hot spot data related to the application to be controlled in the network, such as semantic analysis, feature extraction, heat determination, influence determination and the like are carried out, further, the possible service demands of different services in the application to be controlled are predicted and determined in a future period based on the processed information such as the feature tag, the heat value and the influence, and as different services provided in the application to be controlled can be supported by different regional service nodes in operation, the service flow required by the different regional service nodes in the future period can be determined according to the prediction result, further, the expansion and contraction control of the regional service nodes can be realized based on the determined service flow and the current service load of the regional service node at the current moment, the dynamic control of the service capacity of the regional service node corresponding to the application to be controlled is realized according to the network hot spot condition, the service capacity of the different regional service nodes to be controlled is not down due to a large amount of access, the service capacity of the regional service node to be controlled is reduced, the waste is improved, the service node to be accurately regulated in time, and the intelligent operation can be carried out more stably in time.
Optionally, the service demand prediction module 32 includes:
the feature analysis unit is used for inputting the network hot spot data into the semantic clustering module, extracting features of the network hot spot data, carrying out cluster analysis on the extracted semantic features and data features, and determining a plurality of feature classification labels corresponding to the network hot spot data;
the heat influence determining unit is used for inputting the feature classification labels into the heat influence determining module and determining heat values and influence values corresponding to the network hot spot data;
the demand prediction unit is used for writing the feature classification labels, the heat value and the influence value which are related to the service influence area in the feature classification labels into a preset standard format template, inputting the written standard format template into the service demand prediction module, and determining the output of the service demand prediction module as the output result of the service demand prediction model.
Optionally, the heat influence determining unit is specifically configured to:
determining a data source label related to the data source, a user interaction label related to the user interaction and a content quality label related to the content quality from the feature classification labels;
summing the product of the data source label and the preset data source weight, the product of the user interaction label and the preset user interaction weight and the product of the content quality label and the preset content quality weight to determine a heat value corresponding to the network hot spot data;
Determining positively influencing feature tags with positive influences and negatively influencing feature tags with negative influences from the feature classification tags;
multiplying each positive influence characteristic label with the corresponding positive influence weight respectively, accumulating and summing, and determining the positive influence value;
multiplying each negative influence characteristic label with the corresponding negative influence weight respectively, accumulating and summing, and determining the negative influence value;
and determining the difference between the positive influence value and the negative influence value as the influence value corresponding to the network hot spot data.
Optionally, determining a positively influencing feature tag having a positive influence and a negatively influencing feature tag having a negative influence from the feature classification tags includes:
determining the influence type of the feature classification labels according to at least one of emotion tendencies, user behaviors, user feedback and social influence contained in the feature classification labels;
and determining the positive influence weight or the negative influence weight corresponding to each feature classification label according to the influence type of each feature classification label.
Optionally, the node capacity control module 33 includes:
the adjustment value determining unit is used for determining a capacity adjustment value of the regional service node according to the service demand prediction result corresponding to the regional service node and the difference value of the current service load;
And the expansion and contraction capacity control unit is used for carrying out expansion and contraction capacity control on the regional service node on the basis of the current service load according to the capacity adjustment value.
Optionally, the service node capacity control device further includes: and a data preprocessing module.
The data preprocessing module is used for preprocessing the network hot spot data before inputting the network hot spot data into the semantic clustering module; determining the data and the processed network hot spot data as new network hot spot data, and inputting the new network hot spot data into a semantic clustering module for feature extraction and cluster analysis; the data preprocessing comprises at least one of word segmentation, stop word removal and word stem processing.
Optionally, before acquiring the network hotspot data corresponding to the application to be controlled, the method further includes:
acquiring a historical network hotspot data set and historical regional node service requirements corresponding to each historical network hotspot data set;
determining a historical feature classification label, a historical heat value and a historical influence value of each historical network hot spot data;
constructing training samples according to each historical network hot spot data, and historical feature classification labels, historical heat values, historical influence values and historical area node service requirements corresponding to the historical network hot spot data, and constructing a training sample set according to each training sample;
Training an initial semantic clustering module, an initial heat influence determining module and an initial service demand predicting module in an initial service demand predicting model through a first training sample subset, a second training sample subset and a third training sample subset which are contained in a training sample set;
the first training sample subset is a training sample subset formed by each historical network hot spot data and a corresponding historical characteristic classification label; the second training sample subset is a training sample subset formed by each historical feature classification label, a corresponding historical heat value and a corresponding historical image value; the third training sample subset is a training sample subset formed by each historical feature classification label, a historical heat value, a historical image value and corresponding historical area node service requirements.
The service node capacity control device provided by the embodiment of the invention can execute the service node capacity control method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example IV
Fig. 6 is a schematic structural diagram of a service node capacity control device according to a fourth embodiment of the present invention. Service node capacity control device 40 may be an electronic device intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 6, the service node capacity control device 40 includes at least one processor 41, and a memory such as a Read Only Memory (ROM) 42, a Random Access Memory (RAM) 43, etc. communicatively connected to the at least one processor 41, wherein the memory stores a computer program executable by the at least one processor, and the processor 41 can perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 42 or the computer program loaded from the storage unit 48 into the Random Access Memory (RAM) 43. In the RAM43, various programs and data required for the operation of the service node capacity control device 40 can also be stored. The processor 41, the ROM 42 and the RAM43 are connected to each other via a bus 44. An input/output (I/O) interface 45 is also connected to bus 44.
A plurality of components in the service node capacity control device 40 are connected to the I/O interface 45, including: an input unit 46 such as a keyboard, a mouse, etc.; an output unit 47 such as various types of displays, speakers, and the like; a storage unit 48 such as a magnetic disk, an optical disk, or the like; and a communication unit 49 such as a network card, modem, wireless communication transceiver, etc. The communication unit 49 allows the service node capacity control device 40 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.
The processor 41 may be various general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 41 performs the various methods and processes described above, such as the serving node capacity control method.
In some embodiments, the service node capacity control method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program may be loaded and/or installed onto the service node capacity control device 40 via the ROM 42 and/or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the service node capacity control method described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to perform the service node capacity control method in any other suitable way (e.g. by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A service node capacity control method, comprising:
acquiring network hot spot data corresponding to an application to be controlled; wherein, the network hot spot data is non-standard format data of natural language type;
inputting the network hot spot data into a pre-trained service demand prediction model, and determining a service demand prediction result of each regional service node corresponding to the application to be controlled according to an output result of the service demand prediction model;
Aiming at each regional service node, acquiring the current service load of the regional service node, and performing expansion and contraction control on the regional service node according to the service demand prediction result corresponding to the regional service node and the current service load;
the service demand prediction model at least comprises a semantic clustering module, a heat influence determining module and a service demand prediction module.
2. The method of claim 1, wherein said inputting the network hotspot data into a pre-trained service demand prediction model comprises:
inputting the network hot spot data to the semantic clustering module, extracting features of the network hot spot data, performing cluster analysis on the extracted semantic features and data features, and determining a plurality of feature classification labels corresponding to the network hot spot data;
inputting each characteristic classification label into the heat influence determining module to determine a heat value and an influence value corresponding to the network hot spot data;
and writing the feature classification labels, the heat value and the influence value which are related to the service influence area in the feature classification labels into a preset standard format template, inputting the written standard format template into the service demand prediction module, and determining the output of the service demand prediction module as the output result of the service demand prediction model.
3. The method of claim 2, wherein said inputting each of the feature classification tags to the hotness-impact determination module determines a hotness value and an impact value corresponding to the network hotspot data, comprising:
determining a data source label related to the data source, a user interaction label related to the user interaction and a content quality label related to the content quality from the feature classification labels;
summing the product of the data source label and the preset data source weight, the product of the user interaction label and the preset user interaction weight and the product of the content quality label and the preset content quality weight to determine a heat value corresponding to the network hot spot data;
determining positively influencing feature tags with positive influence and negatively influencing feature tags with negative influence from the feature classification tags;
multiplying each positive influence characteristic label with a corresponding positive influence weight respectively, accumulating and summing the positive influence weights to determine a positive influence value;
multiplying each negative influence characteristic label with a corresponding negative influence weight respectively, accumulating and summing, and determining the negative influence value;
And determining the difference between the positive influence value and the negative influence value as the influence value corresponding to the network hot spot data.
4. A method according to claim 3, wherein said determining positively influencing feature tags having a positive influence and negatively influencing feature tags having a negative influence from each of said feature class tags comprises:
determining the influence type of the feature classification labels according to at least one of emotion tendencies, user behaviors, user feedback and social influence contained in each feature classification label;
and determining the positive influence weight or the negative influence weight corresponding to each feature classification label according to the influence type of each feature classification label.
5. The method of claim 2, further comprising, prior to said entering the network hotspot data into the semantic clustering module:
performing data preprocessing on the network hot spot data;
determining the data and the processed network hot spot data as new network hot spot data, and inputting the new network hot spot data into the semantic clustering module for feature extraction and cluster analysis;
the data preprocessing comprises at least one of word segmentation, stop word removal and word stem processing.
6. The method of claim 1, wherein performing capacity expansion and contraction control on the regional service node according to the service demand prediction result corresponding to the regional service node and the current service load comprises:
determining a capacity adjustment value of the regional service node according to a service demand prediction result corresponding to the regional service node and the difference value of the current service load;
and performing expansion and contraction capacity control on the regional service node on the basis of the current service load according to the capacity adjustment value.
7. The method according to any one of claims 1-6, further comprising, prior to said obtaining network hotspot data corresponding to the application to be controlled:
acquiring a historical network hotspot data set and historical regional node service requirements corresponding to each historical network hotspot data set;
determining a historical feature classification label, a historical heat value and a historical influence value of each historical network hot spot data;
constructing training samples according to each historical network hotspot data, the historical feature classification labels, the historical heat values, the historical influence values and the historical area node service requirements corresponding to the historical network hotspot data, and constructing a training sample set according to each training sample;
Training an initial semantic clustering module, an initial heat influence determining module and an initial service demand predicting module in an initial service demand predicting model through a first training sample subset, a second training sample subset and a third training sample subset which are contained in the training sample set;
the first training sample subset is a training sample subset formed by each historical network hot spot data and a corresponding historical characteristic classification label; the second training sample subset is a training sample subset formed by each historical feature classification label, a corresponding historical heat value and a corresponding historical image value; the third training sample subset is a training sample subset formed by each historical feature classification label, the historical heat value, the historical image value and the corresponding historical area node service requirement.
8. A service node capacity control apparatus, comprising:
the data acquisition module is used for acquiring network hot spot data corresponding to the application to be controlled; wherein, the network hot spot data is non-standard format data of natural language type;
the service demand prediction module is used for inputting the network hot spot data into a pre-trained service demand prediction model, and determining a service demand prediction result of each regional service node corresponding to the application to be controlled according to an output result of the service demand prediction model;
The node capacity control module is used for acquiring the current service load of the regional service node for each regional service node, and performing expansion and contraction capacity control on the regional service node according to the service demand prediction result corresponding to the regional service node and the current service load;
the service demand prediction model at least comprises a semantic clustering module, a heat influence determining module and a service demand prediction module.
9. A service node capacity control apparatus, characterized by comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the serving node capacity control method of any one of claims 1-7.
10. A storage medium containing computer executable instructions which, when executed by a computer processor, are for performing the serving node capacity control method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311386682.XA CN117376982A (en) | 2023-10-24 | 2023-10-24 | Service node capacity control method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311386682.XA CN117376982A (en) | 2023-10-24 | 2023-10-24 | Service node capacity control method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117376982A true CN117376982A (en) | 2024-01-09 |
Family
ID=89400033
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311386682.XA Pending CN117376982A (en) | 2023-10-24 | 2023-10-24 | Service node capacity control method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117376982A (en) |
-
2023
- 2023-10-24 CN CN202311386682.XA patent/CN117376982A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11809505B2 (en) | Method for pushing information, electronic device | |
| CN114494776A (en) | Model training method, device, equipment and storage medium | |
| CN116684330A (en) | Traffic prediction method, device, equipment and storage medium based on artificial intelligence | |
| CN113392920B (en) | Method, apparatus, device, medium, and program product for generating cheating prediction model | |
| CN114037059A (en) | Pre-training model, model generation method, data processing method and data processing device | |
| CN114090601B (en) | Data screening method, device, equipment and storage medium | |
| CN113051911B (en) | Method, apparatus, device, medium and program product for extracting sensitive words | |
| CN113361621B (en) | Method and device for training model | |
| CN117376982A (en) | Service node capacity control method, device, equipment and storage medium | |
| CN114610953A (en) | Data classification method, device, equipment and storage medium | |
| CN112560437B (en) | Text smoothness determining method, target model training method and device | |
| CN113612777A (en) | Training method, traffic classification method, device, electronic device and storage medium | |
| CN112633381A (en) | Audio recognition method and training method of audio recognition model | |
| CN113010782A (en) | Demand amount acquisition method and device, electronic equipment and computer readable medium | |
| CN117131197B (en) | Method, device, equipment and storage medium for processing demand category of bidding document | |
| CN116431809A (en) | Text labeling method, device and storage medium based on bank customer service scene | |
| CN117633226A (en) | Classification method and device, storage medium and electronic equipment | |
| CN115859152A (en) | Model training data generation method and device and electronic equipment | |
| CN115563390A (en) | Method, device, equipment and storage medium for determining course recommendation list | |
| CN117952709A (en) | Data processing method, device, equipment and storage medium | |
| CN115329897A (en) | Feature extraction model training method and device and electronic equipment | |
| CN114925202A (en) | Text classification method and device, computer equipment and storage medium | |
| CN114547417A (en) | Media resource ordering method and electronic equipment | |
| CN114880599A (en) | Information sorting method and device, electronic equipment and storage medium | |
| CN117574168A (en) | Information report generation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |