CN113689020A - Service information prediction method, device, computer equipment and storage medium - Google Patents
Service information prediction method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN113689020A CN113689020A CN202010424124.8A CN202010424124A CN113689020A CN 113689020 A CN113689020 A CN 113689020A CN 202010424124 A CN202010424124 A CN 202010424124A CN 113689020 A CN113689020 A CN 113689020A
- Authority
- CN
- China
- Prior art keywords
- event
- data
- historical
- service information
- model
- 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 54
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 238000012549 training Methods 0.000 claims description 60
- 230000006870 function Effects 0.000 claims description 30
- 230000007704 transition Effects 0.000 claims description 28
- 238000012360 testing method Methods 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 19
- 238000012163 sequencing technique Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 9
- 238000010801 machine learning Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000000926 separation method Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 238000000692 Student's t-test Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012353 t test Methods 0.000 description 1
- 238000007794 visualization technique Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Accounting & Taxation (AREA)
- Entrepreneurship & Innovation (AREA)
- Finance (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a method and a device for predicting service information under the influence of an event, computer equipment and a storage medium. The method comprises the following steps: acquiring a service information prediction request; searching event data corresponding to the service information prediction request; extracting event size type data and event characteristic data corresponding to the event data; and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data. The method and the device can effectively improve the accuracy of business information prediction under the influence of events.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for predicting service information, a computer device, and a storage medium.
Background
With the development of computer technology and internet technology, big data technology has emerged. Big data is a data set which cannot be captured, managed and processed by a conventional software tool within a certain time range, and is a massive, high-growth-rate and diversified information asset which can have stronger decision-making power, insight discovery power and flow optimization capability only by a new processing mode. The market demand of the commodity is determined by predicting the relevant business information big data in the commodity circulation process, so that the service satisfaction rate in the commodity circulation process can be effectively improved, and the waste rate is reduced.
At present, a method for predicting relevant business information data in a commodity circulation process mainly comprises machine learning, wherein the machine learning can more flexibly perform characteristic engineering processing on the commodity circulation data, so that the market demand of a commodity is determined according to the characteristics of the business information data in the commodity circulation process, and the production flow of the commodity is guided.
However, when the machine learning method encounters an event in the commodity circulation process, the prediction accuracy is difficult to guarantee.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a method, an apparatus, a computer device, and a storage medium for predicting business information under the influence of events, which can improve the prediction accuracy.
A method for predicting business information under the influence of an event, the method comprising:
acquiring a service information prediction request;
searching event data corresponding to the service information prediction request;
extracting event size type data and event characteristic data corresponding to the event data;
and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
In one embodiment, before the inputting the event size type data and the event feature data into a preset information prediction model and obtaining a service information prediction result, the method further includes:
acquiring historical service information and historical event data;
obtaining model training data according to the historical service information and the historical event data;
training an initial information prediction model through the model training data to obtain a prediction transition model;
and adding an output result factor coefficient adjusting function to the prediction transition model according to the historical service information and the historical event data to obtain a preset information prediction model.
In one embodiment, the obtaining model training data and model test data according to the historical service information and the historical event data includes:
extracting event size type data and event characteristic data corresponding to the historical event data; and adding an event type label to the historical service information according to the event size type data, and adding an event characteristic label to the historical event data according to the event characteristic data to generate model training data.
In one embodiment, extracting event feature data corresponding to the historical event data includes:
extracting article association information corresponding to each event in the historical event data and service processing point categories corresponding to each event;
according to the extracted article correlation information, acquiring a binning characteristic and a sequencing characteristic corresponding to the historical event data, and acquiring a cross characteristic corresponding to the historical event data according to the service processing point category corresponding to each event;
and the sorting features are obtained based on the ranking features corresponding to the extracted item association information.
In one embodiment, the adding, according to the historical service information and the historical event data, an output result factor coefficient adjustment function to the prediction transition model to obtain a preset information prediction model includes:
acquiring factor coefficients corresponding to various types of historical events according to historical service information corresponding to various types of historical events and historical service information corresponding to historical events which do not occur;
generating an output result factor coefficient adjusting function according to the factor coefficient corresponding to each type of historical event;
and adding the output result factor coefficient adjusting function to the prediction transition model to obtain a preset information prediction model.
In one embodiment, the generating an output result factor coefficient adjustment function according to the factor coefficient corresponding to each type of historical event includes:
carrying out T test on model training data corresponding to various types of historical events;
and generating an output result factor coefficient adjusting function according to the T test result data of the model training data corresponding to each type of historical event.
An apparatus for predicting traffic information under the influence of an event, the apparatus comprising:
the request acquisition module is used for acquiring a service information prediction request;
the data searching module is used for searching the event data corresponding to the service information prediction request;
the characteristic extraction module is used for extracting event size type data and event characteristic data corresponding to the event data;
and the service information prediction module is used for inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, and the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
In one embodiment, the system further comprises a model building module, wherein the model building module is used for:
acquiring historical service information and historical event data;
obtaining model training data according to the historical service information and the historical event data;
training an initial information prediction model through the model training data to obtain a prediction transition model;
and adding an output result factor coefficient adjusting function to the prediction transition model according to the historical service information and the historical event data to obtain a preset information prediction model.
A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
acquiring a service information prediction request;
searching event data corresponding to the service information prediction request;
extracting event size type data and event characteristic data corresponding to the event data;
and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring a service information prediction request;
searching event data corresponding to the service information prediction request;
extracting event size type data and event characteristic data corresponding to the event data;
and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
The service information prediction method, the device, the computer equipment and the storage medium under the influence of the event obtain the service information prediction request; searching event data corresponding to the service information prediction request; extracting event size type data and event characteristic data corresponding to the event data; and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data. According to the method, based on historical service information and historical event data, a preset information prediction model is built through a factor coefficient adjustment method, then a service information prediction request is obtained, event size type data and event characteristic data corresponding to event data influencing service information are input into the preset information prediction model, the preset information prediction model is combined with specific event data to obtain a service information prediction result under the influence of a corresponding event, and the accuracy of service information prediction under the influence of the event can be effectively improved.
Drawings
FIG. 1 is a diagram of an exemplary implementation of a method for predicting business information under the influence of events;
FIG. 2 is a flowchart illustrating a method for predicting service information under the influence of an event according to an embodiment;
FIG. 3 is a schematic diagram illustrating a process of constructing a predictive model of predefined information according to an embodiment;
FIG. 4 is a schematic sub-flow chart of step 308 of FIG. 3 in one embodiment;
FIG. 5 is a schematic sub-flow chart of step 403 of FIG. 4 in one embodiment;
FIG. 6 is a block diagram of an embodiment of a device for predicting traffic information under the influence of an event;
FIG. 7 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The bin packing visualization method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. When an event is encountered, if a user of the terminal 102 needs to predict the service information of the event period, the terminal 102 may submit a service information prediction request to the server 104, and the server 104 obtains the service information prediction request; searching event data corresponding to the service information prediction request; extracting event size type data and event characteristic data corresponding to the event data; and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, a method for predicting service information under the influence of an event is provided, which is described by taking the method as an example applied to the server 104 in fig. 1, and includes the following steps:
The service information prediction request specifically refers to a service information prediction request under the influence of an event, and the purpose of the request is to obtain a corresponding service information prediction result. In one embodiment, the business information can be sales volume of commodities, the business information prediction method under the influence of events can be used for analyzing and predicting the sales volume of the commodities under the influence of the events, and after accurate sales volume is predicted, a factory can be further guided to carry out reasonable production, so that the production structure is optimized, and the overall efficiency of a supply chain is improved.
Specifically, when a service is affected by an event, in order to analyze and predict service information data during the event, a user may submit a service information prediction request to the server through the terminal 102, and then the server 104 may obtain the service information prediction request to perform service information prediction under the subsequent event.
The event data specifically refers to an event that will occur or data corresponding to an event that is occurring and affects a service. In one embodiment, the forecast objective of the business information forecast request is the sales volume of the goods to be sold during the event. The event may specifically be that a concert is held in the range of a store where the commodity to be sold is located, or that an outbreak of a disease epidemic situation occurs.
Specifically, after the server receives the service information prediction request, event data corresponding to the service information prediction request may be further searched for in order to analyze and predict the service information. The event data can be prestored in a preset event database, and when an event is happening or about to break out, a worker can record the event data related to the event into the preset event database. And assigns a corresponding number to each event data. The service information prediction request submitted by the terminal 102 may include the number of the event data in addition to the data of the service information. Therefore, the server 104 can find the event data corresponding to the service information prediction request under the influence of the event after receiving the service information prediction request under the influence of the event.
In particular, the event data may include size type data of the event and feature data of the event. The size type data of the event refers to the type of the event, and for example, whether the event belongs to a big event or a small event can be judged according to the influence of the event on the object to be analyzed. The characteristic data of the event refers to characteristic data obtained according to the attribute construction of the event, in order to increase the prediction accuracy of the preset data model, the event can be constructed into corresponding characteristics, and then the characteristics are added into the preset information prediction model, so that on one hand, the influence degree of the event on the prediction can be obtained according to the importance degree of the event in the model, and on the other hand, the robustness of the preset information prediction model can be increased. In one embodiment, the business information prediction method under the influence of an event is specifically used for predicting sales volume of a commodity to be sold in a store in the event period, and the constructed features may specifically include: category features, binning features, ranking features, and cross features.
And step 207, inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
The preset information prediction model corresponds to the service information to be predicted one by one and is specially used for analyzing the service information to be predicted so as to obtain a service information prediction result under the influence of an event. The preset information prediction model is constructed based on historical service information and historical event data, and is obtained after the post-processing is carried out through factor coefficient adjustment. When the preset information prediction model obtains the event size type data and the event characteristic data corresponding to the event data, the influence of the event on the service information can be analyzed based on the event size type data and the event characteristic data corresponding to the event data, so that a service information prediction result corresponding to the event information can be obtained. The event size type data corresponding to the event data is used for determining the size of a factor coefficient used for adjustment in a preset information prediction model, and the feature data corresponding to the event data is used for determining the influence of the event on a service information prediction result under the influence of the event. Since the overall prediction accuracy of the model is somewhat degraded due to the occurrence of events, but the prediction accuracy is not degraded for each event, the prediction result factor adjustment method is performed to better adjust the prediction accuracy. In one embodiment, the business information forecast request is directed to obtaining sales data for a particular product to be sold during the event. The preset information prediction model is specifically used for analyzing the strategy of realizing the sales data. The predetermined information prediction model may be a machine learning model.
The business information prediction method under the influence of the event obtains a business information prediction request; searching event data corresponding to the service information prediction request; extracting event size type data and event characteristic data corresponding to the event data; and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data. According to the method, based on historical service information and historical event data, a preset information prediction model is built through a factor coefficient adjustment method, then a service information prediction request is obtained, event size type data and event characteristic data corresponding to event data influencing service information are input into the preset information prediction model, the preset information prediction model is combined with specific event data to obtain a service information prediction result under the influence of a corresponding event, and the accuracy of service information prediction under the influence of the event can be effectively improved.
In one embodiment, before step 207, the method further includes:
And step 304, obtaining model training data according to the historical service information and the historical event data.
And step 306, training the initial information prediction model through the model training data to obtain a prediction transition model.
And 308, adding an output result factor coefficient adjusting function to the prediction transition model according to the historical service information and the historical event data to obtain a preset information prediction model.
The historical service information refers to historical data related to service information of a service information prediction method under the influence of events, and if sales of the to-be-analyzed object are predicted, the historical service information refers to sales of the to-be-analyzed object in the historical record in each period. And model training data and model testing data refer to data used for training and testing the initial machine learning model. The initial model can be trained based on the model training data, then an output result factor coefficient adjusting module is added to the prediction transition model obtained through training, and the model obtained through training is subjected to post-processing through factor coefficient adjustment to obtain the available preset information prediction model. In addition, model test data can be obtained according to historical service information and historical event data, the trained initial information prediction model is tested through the model test data, and the model passing the test is used as a prediction transition model. The training of the initial information prediction model by the model training data specifically means that the initial information prediction model is supervised trained by the model training data. In the embodiment, corresponding model training data are obtained based on historical service information and historical event data, then corresponding prediction models are obtained through the data, and factor coefficient adjusting modules are added to the trained models, so that the model training efficiency can be effectively improved, and the model prediction accuracy is ensured.
In one embodiment, step 304 includes: extracting event size type data and event characteristic data corresponding to historical event data; and adding an event type label to the historical service information according to the event size type data, and adding an event characteristic label to the historical event data according to the event characteristic data to generate model training data.
The event feature data can be added with corresponding event feature labels for historical event data. The method includes the steps that corresponding event type labels are added to historical service information, and specifically, event type labels of whether the target historical data are in an event period or not are added to the target historical data according to the date of occurrence of an event. Meanwhile, whether the event is a big event or a small event can be determined according to the sales volume change situation of the event period, and then an event type label of the big event or the small event is added to the target historical data. Firstly, analyzing the histogram distribution condition of the mean value of the sales volume of each event; then comparing the average value of the sales volume during the event with the average value of the sales volume on weekends; and dividing events with the average value of the event sales volume being more than 50% of the average value of the weekend sales volume into large events, and dividing the rest into small events. And the event characteristic data can be directly added into the historical event data as a characteristic tag. In order to increase the prediction accuracy of the business information prediction model under the influence of the event, the characteristic data corresponding to the event can be extracted and added into the machine learning model, so that on one hand, the influence degree of the event on the prediction can be obtained through the importance degree of the event in the model, and on the other hand, the robustness of the model can be increased. Specifically, adding the event feature tag to the event data may specifically include the following features. Class characteristics: directly marking an event with a 0-1 event characteristic label of whether the event is an event period; the box separation characteristic: in order to more carefully distinguish the influence of each event, the events are subjected to box separation according to the average value of the sales volume through the sales volume histogram of the events, and box separation feature labels are added; sequencing characteristics: sorting different event types according to the average sales volume, and adding corresponding sorting result labels; cross characteristics: in order to reflect the influence degree of the event on different commodities of different stores, the commodities and the event category are crossed to obtain corresponding cross characteristics. And then, the historical service information added with the event type label and the historical event data added with the event characteristic label can be used as model training data. In one embodiment, the target of business information prediction under the influence of an event is sales volume of store commodities, and at this time, historical sales volume data of the store commodities and event data of devices can be used as a sample, corresponding model training data and model test data are constructed, and data of a recent year is divided into a training set (10 months) and a test set (2 months). In the embodiment, the available model training data and model test data can be efficiently obtained by adding the corresponding data labels to the historical service information and the historical event data, so that the training efficiency of the model is improved, and the accuracy of model prediction is improved.
In one embodiment, article association information corresponding to each event in historical event data and a service processing point category corresponding to each event are extracted; according to the extracted article correlation information, acquiring a binning characteristic and a sequencing characteristic corresponding to historical event data, and acquiring a cross characteristic corresponding to the historical event data according to the service processing point category corresponding to each event; and the sorting features are obtained based on the ranking features corresponding to the extracted item association information.
Data binning, also known as discrete binning or segmentation, is a data preprocessing technique used to reduce the effects of minor observation errors, and is a method of grouping multiple continuous values into a smaller number of "bins". When the method is used for analyzing the commodity sales volume in the event period, the binning specifically refers to binning the event according to the average sales volume value through the sales volume histogram of the event period. The cross feature is to cross the service processing point category, the service and the event category to obtain the corresponding cross feature in order to reflect the degree of influence of the event on different services of different service processing point categories.
Specifically, feature data corresponding to each historical event may be constructed based on specific data of each event in the historical event data. These feature data may be appended to the corresponding historical event data. In the embodiment, the event is constructed into corresponding characteristics, and the degree of influence of the event on the prediction can be obtained according to the importance degree of the event in the service information prediction model under the influence of the event, so that the prediction accuracy of the service information prediction process under the influence of the event is improved, and the robustness of the service information prediction model under the influence of the event can be improved.
As shown in FIG. 4, in one embodiment, step 308 comprises:
and 405, adding an output result factor coefficient adjusting function to the prediction transition model to obtain a preset information prediction model.
Wherein, each type of historical event has a factor coefficient corresponding to the historical event. Therefore, before the model outputs the business information prediction result under the influence of the event, corresponding factor coefficients can be added based on the event data corresponding to the business information prediction request under the influence of the input event, and the business information prediction result under the influence of the last event can be determined. When an event occurs, the overall prediction accuracy of the business information prediction model under the influence of the event is reduced to a certain extent, but the prediction accuracy is not reduced when the event occurs for each object to be analyzed, so that a prediction result factor adjusting method is performed for better adjusting the prediction accuracy. The obtaining of the factor coefficient corresponding to each type of historical event according to the historical service information corresponding to each type of historical event and the historical service information corresponding to the historical event that does not occur means obtaining the factor coefficient for result adjustment based on whether the event occurs or not. As one embodiment, the present application is used for analyzing the commodity sales at an event time, where the factor coefficient corresponding to each type of historical event is the average of the sales occurred/the average of the sales not occurred. After the factor coefficients corresponding to various types of time are obtained, an output result factor coefficient adjusting module corresponding to the model can be correspondingly generated, and the module is used for carrying out factor adjustment on the original output result of the model to obtain a final service information prediction result under the influence of an event. And after an output result factor coefficient adjusting module is added to the prediction transition model, the preset information prediction model can be obtained. In the embodiment, the factor coefficients of various types of historical events are determined based on historical service information of event periods and non-event periods, and then an output result factor coefficient adjusting module is added to the prediction transition model, so that the model can be post-processed, and the prediction accuracy of the finally generated preset information prediction model can be effectively improved.
As shown in FIG. 5, in one embodiment, step 403 comprises:
And step 504, generating an output result factor coefficient adjusting function according to T test result data of the model training data corresponding to each type of historical event.
Among them, T test is also called student's T test. the t test is to use the t distribution theory to deduce the probability of occurrence of difference, so as to compare whether the difference between two averages is significant or not. In the application, the T test can be performed on the sample on the training set, so that whether the factor coefficient adjustment is performed on the result data of the business information prediction under the influence of the event is determined. For a sample of a certain type of historical event, if the final sample data is not significant, the factor coefficient adjustment processing is not carried out on the prediction result, and if the final sample data is significant, the factor coefficient adjustment processing is carried out. In this embodiment, whether the corresponding factor coefficient adjustment processing is performed on the prediction result is determined through T test, so that the accuracy of factor coefficient adjustment can be improved, and the prediction accuracy of the preset information prediction model can be improved.
It should be understood that although the various steps in the flow charts of fig. 2-5 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-5 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.
In one embodiment, as shown in fig. 6, there is provided a traffic information prediction apparatus under the influence of an event, including: a request obtaining module 601, a data searching module 603, a feature extracting module 605 and a service information predicting module 607, wherein:
a request obtaining module 601, configured to obtain a service information prediction request;
a data searching module 603, configured to search event data corresponding to the service information prediction request;
a feature extraction module 605, configured to extract event size type data and event feature data corresponding to the event data;
and the service information prediction module 607 is configured to input the event size type data and the event feature data into a preset information prediction model, and obtain a service information prediction result, where the preset information prediction model is adjusted and constructed by using a factor coefficient based on the historical service information and the historical event data.
In one embodiment, the system further comprises a model building module, wherein the model building module is used for: acquiring historical service information and historical event data; acquiring model training data according to historical service information and historical event data; training the initial information prediction model through model training data to obtain a prediction transition model; and adding an output result factor coefficient adjusting function to the prediction transition model according to the historical service information and the historical event data to obtain a preset information prediction model.
In one embodiment, the model building module is further configured to: extracting event size type data and event characteristic data corresponding to historical event data; and adding an event type label to the historical service information according to the event size type data, and adding an event characteristic label to the historical event data according to the event characteristic data to generate model training data.
In one embodiment, the model building module is further configured to: extracting article association information corresponding to each event in historical event data and service processing point categories corresponding to each event; according to the extracted article correlation information, acquiring a binning characteristic and a sequencing characteristic corresponding to historical event data, and acquiring a cross characteristic corresponding to the historical event data according to the service processing point category corresponding to each event; and the sorting features are obtained based on the ranking features corresponding to the extracted item association information.
In one embodiment, the model building module is further configured to: acquiring factor coefficients corresponding to various types of historical events according to historical service information corresponding to various types of historical events and historical service information corresponding to historical events which do not occur; generating an output result factor coefficient adjusting function according to the factor coefficient corresponding to each type of historical event; and adding an output result factor coefficient adjusting function for the prediction transition model to obtain a preset information prediction model.
In one embodiment, the model building module is further configured to: carrying out T test on model training data corresponding to various types of historical events; and generating an output result factor coefficient adjusting function according to the T test result data of the model training data corresponding to each type of historical event.
For specific limitations of the device for predicting service information under the influence of an event, reference may be made to the above limitations on the method for predicting service information under the influence of an event, which are not described herein again. All or part of each module in the service information prediction device under the influence of the event can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing business information prediction data under the influence of events. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of predicting business information under the influence of an event.
Those skilled in the art will appreciate that the architecture shown in fig. 7 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring a service information prediction request;
searching event data corresponding to the service information prediction request;
extracting event size type data and event characteristic data corresponding to the event data;
and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
In one embodiment, the processor, when executing the computer program, further performs the steps of: acquiring historical service information and historical event data; acquiring model training data according to historical service information and historical event data; training the initial information prediction model through model training data to obtain a prediction transition model; and adding an output result factor coefficient adjusting function to the prediction transition model according to the historical service information and the historical event data to obtain a preset information prediction model.
In one embodiment, the processor, when executing the computer program, further performs the steps of: extracting event size type data and event characteristic data corresponding to historical event data; and adding an event type label to the historical service information according to the event size type data, and adding an event characteristic label to the historical event data according to the event characteristic data to generate model training data.
In one embodiment, the processor, when executing the computer program, further performs the steps of: extracting article association information corresponding to each event in historical event data and service processing point categories corresponding to each event; according to the extracted article correlation information, acquiring a binning characteristic and a sequencing characteristic corresponding to historical event data, and acquiring a cross characteristic corresponding to the historical event data according to the service processing point category corresponding to each event; and the sorting features are obtained based on the ranking features corresponding to the extracted item association information.
In one embodiment, the processor, when executing the computer program, further performs the steps of: acquiring factor coefficients corresponding to various types of historical events according to historical service information corresponding to various types of historical events and historical service information corresponding to historical events which do not occur; generating an output result factor coefficient adjusting function according to the factor coefficient corresponding to each type of historical event; and adding an output result factor coefficient adjusting function for the prediction transition model to obtain a preset information prediction model.
In one embodiment, the processor, when executing the computer program, further performs the steps of: carrying out T test on model training data corresponding to various types of historical events; and generating an output result factor coefficient adjusting function according to the T test result data of the model training data corresponding to each type of historical event.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring a service information prediction request;
searching event data corresponding to the service information prediction request;
extracting event size type data and event characteristic data corresponding to the event data;
and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring historical service information and historical event data; acquiring model training data according to historical service information and historical event data; training the initial information prediction model through model training data to obtain a prediction transition model; and adding an output result factor coefficient adjusting function to the prediction transition model according to the historical service information and the historical event data to obtain a preset information prediction model.
In one embodiment, the computer program when executed by the processor further performs the steps of: extracting event size type data and event characteristic data corresponding to historical event data; and adding an event type label to the historical service information according to the event size type data, and adding an event characteristic label to the historical event data according to the event characteristic data to generate model training data.
In one embodiment, the computer program when executed by the processor further performs the steps of: extracting article association information corresponding to each event in historical event data and service processing point categories corresponding to each event; according to the extracted article correlation information, acquiring a binning characteristic and a sequencing characteristic corresponding to historical event data, and acquiring a cross characteristic corresponding to the historical event data according to the service processing point category corresponding to each event; and the sorting features are obtained based on the ranking features corresponding to the extracted item association information.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring factor coefficients corresponding to various types of historical events according to historical service information corresponding to various types of historical events and historical service information corresponding to historical events which do not occur; generating an output result factor coefficient adjusting function according to the factor coefficient corresponding to each type of historical event; and adding an output result factor coefficient adjusting function for the prediction transition model to obtain a preset information prediction model.
In one embodiment, the computer program when executed by the processor further performs the steps of: carrying out T test on model training data corresponding to various types of historical events; and generating an output result factor coefficient adjusting function according to the T test result data of the model training data corresponding to each type of historical event.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method for predicting business information under the influence of an event, the method comprising: acquiring a service information prediction request;
searching event data corresponding to the service information prediction request;
extracting event size type data and event characteristic data corresponding to the event data;
and inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, wherein the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
2. The method according to claim 1, wherein before inputting the event size type data and the event feature data into a preset information prediction model and obtaining a service information prediction result, the method further comprises:
acquiring historical service information and historical event data;
obtaining model training data according to the historical service information and the historical event data;
training an initial information prediction model through the model training data to obtain a prediction transition model;
and adding an output result factor coefficient adjusting function to the prediction transition model according to the historical service information and the historical event data to obtain a preset information prediction model.
3. The method of claim 2, wherein obtaining model training data and model test data based on the historical traffic information and historical event data comprises:
extracting event size type data and event characteristic data corresponding to the historical event data; and adding an event type label to the historical service information according to the event size type data, and adding an event characteristic label to the historical event data according to the event characteristic data to generate model training data.
4. The method of claim 3, wherein extracting event feature data corresponding to the historical event data comprises:
extracting article association information corresponding to each event in the historical event data and service processing point categories corresponding to each event;
according to the extracted article correlation information, acquiring a binning characteristic and a sequencing characteristic corresponding to the historical event data, and acquiring a cross characteristic corresponding to the historical event data according to the service processing point category corresponding to each event;
and the sorting features are obtained based on the ranking features corresponding to the extracted item association information.
5. The method according to claim 2, wherein the adding an output result factor coefficient adjustment function to the prediction transition model according to the historical service information and the historical event data, and obtaining a preset information prediction model comprises:
acquiring factor coefficients corresponding to various types of historical events according to historical service information corresponding to various types of historical events and historical service information corresponding to historical events which do not occur;
generating an output result factor coefficient adjusting function according to the factor coefficient corresponding to each type of historical event;
and adding the output result factor coefficient adjusting function to the prediction transition model to obtain a preset information prediction model.
6. The method of claim 5, wherein generating an output result factor coefficient adjustment function based on the factor coefficients corresponding to each type of historical event comprises:
carrying out T test on model training data corresponding to various types of historical events;
and generating an output result factor coefficient adjusting function according to the T test result data of the model training data corresponding to each type of historical event.
7. An apparatus for predicting traffic information under the influence of an event, the apparatus comprising:
the request acquisition module is used for acquiring a service information prediction request;
the data searching module is used for searching the event data corresponding to the service information prediction request;
the characteristic extraction module is used for extracting event size type data and event characteristic data corresponding to the event data;
and the service information prediction module is used for inputting the event size type data and the event characteristic data into a preset information prediction model to obtain a service information prediction result, and the preset information prediction model is adjusted and constructed by adopting a factor coefficient based on historical service information and historical event data.
8. The apparatus of claim 7, further comprising a model building module to:
acquiring historical service information and historical event data;
obtaining model training data according to the historical service information and the historical event data;
training an initial information prediction model through the model training data to obtain a prediction transition model;
and adding an output result factor coefficient adjusting function to the prediction transition model according to the historical service information and the historical event data to obtain a preset information prediction model.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 6 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010424124.8A CN113689020A (en) | 2020-05-19 | 2020-05-19 | Service information prediction method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010424124.8A CN113689020A (en) | 2020-05-19 | 2020-05-19 | Service information prediction method, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113689020A true CN113689020A (en) | 2021-11-23 |
Family
ID=78576253
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010424124.8A Pending CN113689020A (en) | 2020-05-19 | 2020-05-19 | Service information prediction method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113689020A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113887845A (en) * | 2021-12-07 | 2022-01-04 | 中国南方电网有限责任公司超高压输电公司广州局 | Extreme event prediction method, apparatus, device, medium, and computer program product |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10134027A (en) * | 1996-10-29 | 1998-05-22 | Toyota Motor Corp | Method and device for supporting prediction of sale |
| JP2003178356A (en) * | 2001-12-11 | 2003-06-27 | Japan Tobacco Inc | Sales prediction method |
| US8160917B1 (en) * | 2007-04-13 | 2012-04-17 | Sas Institute Inc. | Computer-implemented promotion optimization methods and systems |
| CN104484708A (en) * | 2014-11-12 | 2015-04-01 | 南京大学 | Unary linear regression and least square method-based commodity demand prediction method |
| US20170278053A1 (en) * | 2016-03-22 | 2017-09-28 | Wal-Mart Stores, Inc. | Event-based sales prediction |
| CN109272142A (en) * | 2018-08-14 | 2019-01-25 | 苏州朗动网络科技有限公司 | Data predication method, device, computer equipment and storage medium based on event |
| CN110555713A (en) * | 2018-05-31 | 2019-12-10 | 北京京东尚科信息技术有限公司 | method and device for determining sales prediction model |
-
2020
- 2020-05-19 CN CN202010424124.8A patent/CN113689020A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10134027A (en) * | 1996-10-29 | 1998-05-22 | Toyota Motor Corp | Method and device for supporting prediction of sale |
| JP2003178356A (en) * | 2001-12-11 | 2003-06-27 | Japan Tobacco Inc | Sales prediction method |
| US8160917B1 (en) * | 2007-04-13 | 2012-04-17 | Sas Institute Inc. | Computer-implemented promotion optimization methods and systems |
| CN104484708A (en) * | 2014-11-12 | 2015-04-01 | 南京大学 | Unary linear regression and least square method-based commodity demand prediction method |
| US20170278053A1 (en) * | 2016-03-22 | 2017-09-28 | Wal-Mart Stores, Inc. | Event-based sales prediction |
| CN110555713A (en) * | 2018-05-31 | 2019-12-10 | 北京京东尚科信息技术有限公司 | method and device for determining sales prediction model |
| CN109272142A (en) * | 2018-08-14 | 2019-01-25 | 苏州朗动网络科技有限公司 | Data predication method, device, computer equipment and storage medium based on event |
Non-Patent Citations (3)
| Title |
|---|
| 刘晓敏 等: "如何提高销售预测准确率", 《企业管理》, no. 03 * |
| 叶剑锋 等: "基于数据挖掘的快消行业销售预测", 《微型电脑应用》, vol. 35, no. 07 * |
| 杨金花;: "图书销量预测偏差与校正", 出版发行研究, no. 04 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113887845A (en) * | 2021-12-07 | 2022-01-04 | 中国南方电网有限责任公司超高压输电公司广州局 | Extreme event prediction method, apparatus, device, medium, and computer program product |
| CN113887845B (en) * | 2021-12-07 | 2022-04-08 | 中国南方电网有限责任公司超高压输电公司广州局 | Extreme event prediction method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110598845B (en) | Data processing method, data processing device, computer equipment and storage medium | |
| WO2022267735A1 (en) | Service data processing method and apparatus, computer device, and storage medium | |
| CN109543925B (en) | Risk prediction method and device based on machine learning, computer equipment and storage medium | |
| CN111080398B (en) | Commodity recommendation method, commodity recommendation device, computer equipment and storage medium | |
| JP4254623B2 (en) | Topic analysis method, apparatus thereof, and program | |
| US10324705B2 (en) | System and method for run-time update of predictive analytics system | |
| US20210103858A1 (en) | Method and system for model auto-selection using an ensemble of machine learning models | |
| CN111178949B (en) | Service resource matching reference data determining method, device, equipment and storage medium | |
| US11620474B2 (en) | Model reselection for accommodating unsatisfactory training data | |
| CN110674636B (en) | Power consumption behavior analysis method | |
| CN110888911A (en) | Sample data processing method and device, computer equipment and storage medium | |
| CN113537850A (en) | Storage optimization method and device, computer equipment and storage medium | |
| CN112328869A (en) | User loan willingness prediction method and device and computer system | |
| CN113110961B (en) | Equipment abnormality detection method and device, computer equipment and readable storage medium | |
| CN112101024B (en) | Target object identification system based on app information | |
| CN114692889A (en) | Meta-feature training model for machine learning algorithm | |
| CN113689020A (en) | Service information prediction method, device, computer equipment and storage medium | |
| CN110378481B (en) | Decision processing method and device based on rough set, computer and storage medium | |
| CN113610225A (en) | Quality evaluation model training method and device, electronic equipment and storage medium | |
| CN110852400A (en) | Classification model evaluation method and device, computer equipment and storage medium | |
| JP7401747B2 (en) | Sorting program, sorting device and sorting method | |
| CN112749821B (en) | Express delivery quantity prediction method, device, computer equipment and storage medium | |
| CN114372866A (en) | Enterprise data processing method and device, computer equipment and storage medium | |
| CN116644826A (en) | Method and system for predicting user consumption based on fusion model | |
| CN117541885A (en) | Sample data processing method, device, storage medium and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |