CN114900443B - Method and device for establishing incoming line traffic prediction model - Google Patents
Method and device for establishing incoming line traffic prediction model Download PDFInfo
- Publication number
- CN114900443B CN114900443B CN202210686679.9A CN202210686679A CN114900443B CN 114900443 B CN114900443 B CN 114900443B CN 202210686679 A CN202210686679 A CN 202210686679A CN 114900443 B CN114900443 B CN 114900443B
- Authority
- CN
- China
- Prior art keywords
- traffic
- data
- incoming line
- initial value
- correction
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 92
- 230000009467 reduction Effects 0.000 claims abstract description 62
- 238000012549 training Methods 0.000 claims abstract description 41
- 238000010801 machine learning Methods 0.000 claims abstract description 38
- 238000007781 pre-processing Methods 0.000 claims abstract description 19
- 238000012937 correction Methods 0.000 claims description 63
- 238000012545 processing Methods 0.000 claims description 30
- 238000004364 calculation method Methods 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 15
- 230000002159 abnormal effect Effects 0.000 claims description 12
- 238000004140 cleaning Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 8
- 238000009499 grossing Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 14
- 238000004422 calculation algorithm Methods 0.000 description 11
- 230000002354 daily effect Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000009826 distribution Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000013501 data transformation Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012896 Statistical algorithm Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 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
- 230000008439 repair process Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
- H04M3/5166—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing in combination with interactive voice response systems or voice portals, e.g. as front-ends
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention provides a method and a device for establishing an incoming line traffic prediction model, relates to the technical field of modeling, and can be used in the financial field or other technical fields. The method comprises the following steps: acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value; extracting features of the time sequence data to obtain sample data for training a model; and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model. The apparatus performs the above method. The method and the device for establishing the incoming line traffic prediction model provided by the embodiment of the invention can rapidly and accurately predict the incoming line traffic.
Description
Technical Field
The invention relates to the technical field of modeling, in particular to a method and a device for establishing an incoming line traffic prediction model.
Background
The telephone banking center is a channel for accessing customer incoming call consultation or complaint business by a bank, the types of incoming call lines can be divided according to regions, channels, business and the like, the bank needs to estimate incoming call line traffic of each line for a period of time (usually one month) next, and special line personnel are arranged for each incoming call line in advance. Because the traffic volume of each incoming line is affected by the external environment and fluctuates, and the accuracy is reduced due to the fact that the traffic volume is estimated in advance for a period of time, the traffic volume is accurately estimated by relying on manual experience or adopting a plurality of new technologies, and the existing estimation method comprises the following steps:
(1) Empirical estimation method
The method is widely adopted in the traditional banking business, mainly relies on the accumulation of business expert experience to comprehensively judge the business characteristics of each incoming line of the telephone banking, historical transaction amount and other factors, and reasonable experience values are set to guide the business amount distribution.
(2) Data modeling method
The method is currently adopted in large commercial banks, and by self-researching products or purchasing products, the historical traffic of incoming lines of telephone banks is analyzed and modeled by using a traditional statistical algorithm, so that the traffic of incoming lines in a certain period in the future is predicted.
The incoming line traffic is affected by the external environment and becomes relatively complex, and the accuracy of calculating the traffic for a long period of time in the future is insufficient by relying on a manual experience estimation or statistical method, so that the incoming line traffic cannot be accurately estimated, and the problem of personnel idling or personnel shortage is caused.
Disclosure of Invention
Aiming at the problems in the prior art, the embodiment of the invention provides a method and a device for establishing an incoming line traffic prediction model, which can at least partially solve the problems in the prior art.
In one aspect, the present invention provides a method for establishing an incoming line traffic prediction model, including:
acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
extracting features of the time sequence data to obtain sample data for training a model;
and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model.
Wherein the historical traffic data comprises a date and a current day traffic corresponding to the date; correspondingly, the preprocessing the historical traffic data to obtain time sequence data based on the deformation parameters comprises the following steps:
determining an initial value of the future number and an initial value of the reduction ratio parameter;
determining a contemporaneous average value corresponding to the date according to the initial value of the period number and the service volume of each day, and calculating a correction service volume corresponding to each service volume of each day according to the contemporaneous average value, the service volume of each day and the initial value of the reduction ratio parameter;
if the correction traffic is determined to be greater than zero, recording a current future number, a current reduction ratio parameter and a current correction traffic, adding 1 to an initial value of the future number, and continuously executing the initial value of the future number, the initial value of the reduction ratio parameter and the subsequent steps until the initial value of the future number reaches a preset future number threshold;
after the preset term threshold is reached, acquiring recorded correction traffic, performing mean square error calculation, and taking a deformation parameter with the minimum mean square error calculation result as a target deformation parameter;
And carrying out data correction on the historical traffic data according to the target deformation parameters and the target correction traffic corresponding to the target deformation parameters to obtain the time sequence data.
Wherein, before the step of determining the initial value of the future number and the initial value of the reduction ratio parameter, the method for establishing the incoming line traffic prediction model further comprises:
and cleaning the data of the historical traffic data.
After the step of performing data correction on the historical traffic data according to the target deformation parameter and the target correction traffic corresponding to the target deformation parameter, the method for establishing the incoming line traffic prediction model further comprises the following steps:
and determining abnormal data in the historical traffic data after data correction, and performing smoothing processing on the abnormal data to obtain the time sequence data.
The method for establishing the incoming line traffic prediction model further comprises the following steps:
if the corrected traffic is determined to be less than or equal to zero, reducing the initial value of the reduction ratio parameter by a preset ratio, and continuously executing the calculation of the corrected traffic corresponding to each current day traffic and the subsequent steps according to the contemporaneous average value, each current day traffic and the initial value of the reduction ratio parameter.
The method for establishing the incoming line traffic prediction model further comprises the following steps:
if the date is determined to be the designated holiday, determining the contemporaneous average value by taking the year as granularity.
The method for establishing the incoming line traffic prediction model further comprises the following steps:
if the date is determined not to be a designated holiday, the contemporaneous average is determined with month as granularity and according to the week the date is on and the days of the week.
The invention provides a method for predicting traffic data of an incoming line based on the method for establishing an incoming line traffic prediction model, which comprises the following steps:
acquiring traffic data of an incoming line to be predicted;
and predicting the traffic data based on the incoming line traffic prediction model to obtain a traffic prediction result.
The traffic data prediction method of the incoming line comprises the following steps:
and correcting the traffic prediction result according to the target deformation parameter and the target correction traffic corresponding to the target deformation parameter to obtain a traffic prediction correction result.
In one aspect, the present invention proposes an apparatus for establishing an incoming line traffic prediction model, including:
The acquisition unit is used for acquiring historical traffic data of the incoming line, preprocessing the historical traffic data and acquiring time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
the extraction unit is used for extracting the characteristics of the time sequence data to obtain sample data for training a model;
and the combining unit is used for training at least two machine learning models through the sample data and combining the at least two machine learning models to obtain an incoming line traffic prediction model.
In still another aspect, an embodiment of the present invention provides an electronic device, including: a processor, a memory, and a bus, wherein,
the processor and the memory complete communication with each other through the bus;
the memory stores program instructions executable by the processor, the processor invoking the program instructions capable of performing the method of:
acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
Extracting features of the time sequence data to obtain sample data for training a model;
and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model.
Embodiments of the present invention provide a non-transitory computer readable storage medium comprising:
the non-transitory computer readable storage medium stores computer instructions that cause the computer to perform the method of:
acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
extracting features of the time sequence data to obtain sample data for training a model;
and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model.
The method and the device for establishing the incoming line traffic prediction model provided by the embodiment of the invention acquire the historical traffic data of the incoming line, preprocess the historical traffic data and acquire time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value; extracting features of the time sequence data to obtain sample data for training a model; and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model, so that incoming line traffic can be rapidly and accurately predicted.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
fig. 1 is a flowchart of a method for establishing an incoming line traffic prediction model according to an embodiment of the present invention.
Fig. 2 is a flowchart of a method for establishing an incoming line traffic prediction model according to another embodiment of the present invention.
Fig. 3 is a flowchart of a method for establishing an incoming line traffic prediction model according to another embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a method for establishing an incoming line traffic prediction model according to an embodiment of the present invention.
Fig. 5 is a schematic block diagram of a method for establishing an incoming line traffic prediction model according to an embodiment of the present invention.
Fig. 6 is a schematic structural diagram of a method for establishing an incoming line traffic prediction model according to an embodiment of the present invention.
Fig. 7 is a schematic block diagram of a method for establishing an incoming line traffic prediction model according to an embodiment of the present invention.
Fig. 8 is a schematic structural diagram of an apparatus for establishing an incoming line traffic prediction model according to an embodiment of the present invention.
Fig. 9 is a schematic diagram of an entity structure of an electronic device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention. It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be arbitrarily combined with each other.
Fig. 1 is a flowchart of a method for establishing an incoming line traffic prediction model according to an embodiment of the present invention, and as shown in fig. 1, the method for establishing an incoming line traffic prediction model according to an embodiment of the present invention includes:
step S1: acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the distortion parameters include a period number of a contemporaneous average of the historical traffic data and a reduction ratio parameter of traffic corresponding to the contemporaneous average.
Step S2: and extracting the characteristics of the time sequence data to obtain sample data for training a model.
Step S3: and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model.
In the step S1, the device acquires historical traffic data of the incoming line, performs preprocessing on the historical traffic data, and obtains time series data based on deformation parameters; the distortion parameters include a period number of a contemporaneous average of the historical traffic data and a reduction ratio parameter of traffic corresponding to the contemporaneous average. The apparatus may be a computer device or the like that performs the method, and may include a server. It should be noted that, the relevant data of the training model according to the embodiment of the present invention is authorized by the user. Historical traffic data of each incoming line of a telephone bank for nearly two years can be obtained through an in-line system, and specifically the type, date and current-day traffic of the incoming line can be included; the types of the incoming lines can be obtained by dividing according to regions, channels, business and the like.
The contemporaneous average of historical traffic data is illustrated as follows: if the date of the historical traffic data is 12 months 15 days, the contemporaneous average of the historical traffic data may be 1 month 15 days, 2 months 15 days … months 15 days, and the average of the traffic data for 12 days in total, and the corresponding period number of the contemporaneous average of the historical traffic data is 12.
As shown in fig. 2, that is, in the iterative calculation process of each period number, the value of the corrected traffic is zero or less every time it occurs, a reduction calculation is performed on the reduction ratio parameter of the traffic, for example, to reduce to 5%.
The deformation parameters are a set of parameters including an epoch N and a reduction ratio parameter M, denoted as N, M.
The historical traffic data comprises a date and the current day traffic corresponding to the date; correspondingly, the preprocessing the historical traffic data to obtain time sequence data based on the deformation parameters comprises the following steps:
determining an initial value of the future number and an initial value of the reduction ratio parameter; as shown in fig. 2, the initial value of the period number may be selected to be 2, and the initial value of the reduction ratio parameter may be selected to be 90%.
Before the step of determining the initial value of the future number and the initial value of the reduction ratio parameter, the method of establishing an incoming line traffic prediction model further includes:
and cleaning the data of the historical traffic data.
May include: (1) And eliminating the repeated value, namely if a plurality of records exist on the same date, reserving the latest record.
(2) Repair missing data. The date of the missing record may be filled with the missing value by taking the near 4 contemporaneous averages, and for a given holiday, the first four days of the mean during the given holiday, or the last four days of the mean may be filled with the missing value.
Determining a contemporaneous average value corresponding to the date according to the initial value of the period number and the service volume of each day, and calculating a correction service volume corresponding to each service volume of each day according to the contemporaneous average value, the service volume of each day and the initial value of the reduction ratio parameter; first, a past N contemporaneous averages of historical traffic data are calculated, assuming a historical traffic data sequence [ (T1, D1), (T2, D2),..times (TN, DN) ], where TX is the date and DX is the current day traffic.
The TX synchronization average value calculating method comprises the following steps: (1) If TX is not designated holidays (e.g., spring festival, national celebration, etc.), then its synchronization is Zhou Tianshu T1 at the same week as the last month (i.e., if 15 days of 12 months is the fourth week of 12 months, then T1 is the fourth week of 11 months, 17 days of 11 months), the synchronization of T1 is Zhou Tianshu T2 at the same week as the last month, and the N synchronization average value of TX is the synchronization average value (d1+d2.+ DN)/N of N months of reverse push.
(2) If TX is the designated holiday, then its contemporaneous average is taken as the same day of the last year traffic.
The corrected traffic DX' corresponding to each current day traffic is calculated according to the following calculation formula:
DX' =dx- [ (d1+d2+) +dn)/N ] ×m, where N is the average value of the N contemporaneous periods indicated above, the initial value is 2, M is the reduction ratio parameter, and the initial value is 90%.
If the correction traffic is determined to be greater than zero, recording a current future number, a current reduction ratio parameter and a current correction traffic, adding 1 to an initial value of the future number, and continuously executing the initial value of the future number, the initial value of the reduction ratio parameter and the subsequent steps until the initial value of the future number reaches a preset future number threshold; if DX ' is greater than zero, the sequence after primary deformation is [ (T1 ', D1 '), (T2 ', D2 '), ], the deformation parameters are [ N1, M1], and the sequence after primary deformation and the deformation parameters are recorded.
The preset period number threshold value can be set autonomously according to actual conditions, and is optionally 12. Finally, the calculation [ N, M ] is iterated until n=12.
After the preset term threshold is reached, acquiring recorded correction traffic, performing mean square error calculation, and taking a deformation parameter with the minimum mean square error calculation result as a target deformation parameter; referring to the above example, the post-deformation sequence corresponding to the deformation parameter [ N1, M1] is denoted as L1, the post-deformation sequence corresponding to the deformation parameter [ N2, M2] is denoted as L2 …, and the post-deformation sequence corresponding to the deformation parameter [ N11, M11] is denoted as L11. And (3) carrying out mean square error calculation on 2 DX ' corresponding to L1 to obtain a mean square error D1, carrying out mean square error calculation on 3 DX ' corresponding to L2 to obtain a mean square error D2 …, carrying out mean square error calculation on 12 DX ' corresponding to L11 to obtain a mean square error D11, and taking the deformation parameters [ N10, M10] as target deformation parameters if the mean square error D10 is minimum.
And carrying out data correction on the historical traffic data according to the target deformation parameters and the target correction traffic corresponding to the target deformation parameters to obtain the time sequence data. And (3) carrying out deformation processing on the historical traffic data according to the parameters [ N10, M10] by using the target correction traffic corresponding to the target deformation parameters, wherein for the historical traffic data, assuming that the data of any day is [ T, D ], and finding 11 synchronous average values D 'of the historical traffic data according to the date T, the date data is deformed into [ T, D-D' ×M10], and the data of each date in the historical traffic data sequence is processed in the same way, so as to finally obtain a new deformed sequence and deformation parameters [ N10, M10]. By subtracting the weighted average of the dates similar to the law of the traffic on a certain day, the method can obtain smaller fluctuation amplitude, more stable sequence, more obvious waveform periodicity law than the original sequence and better predictable effect.
Meanwhile, by carrying out reduction processing on the reduction ratio parameter M, the generated new sequence values can be guaranteed to be positive values, so that the algorithm applicable to the new sequence is wider.
After the step of performing data correction on the historical traffic data according to the target deformation parameter and the target correction traffic corresponding to the target deformation parameter, the method for establishing the incoming line traffic prediction model further includes:
And determining abnormal data in the historical traffic data after data correction, and performing smoothing processing on the abnormal data to obtain the time sequence data.
And (3) carrying out abnormal data processing on the deformed data sequence, judging abnormal data by adopting a normal distribution 3 sigma principle, setting a point which is three times of the standard deviation of the data set as noise data, and carrying out smoothing processing on the noise data by adopting a smoothdata method.
In the step S2, the device performs feature extraction on the time-series data to obtain sample data for training a model. The features may include three general classes of features, each described as follows:
(1) Time period characteristics. Whether or not the year is at the end of the beginning of the year, the month is at the end of the month, the month is at the week, the day is the day of the year, the day is the legal holiday whether the day is legal holidays, whether the last three days are legal holidays, and the remaining legal holidays.
(2) Sequence association features. A daily value for the last 30 days, a daily value for the same day of the first three weeks, and a daily value for each of the first three month contemporaneous dates.
(3) Sequence statistics. The average value of the past 30 days, the average value of the same week as the month, the average value of the same period as the month, the average value of the last week, the average value of the last month, the cycle ratio of yesterday and the same week as the last week, the cycle ratio of yesterday and the last month, and the cycle ratio of yesterday and the last year. After the feature data is generated, the data set is subjected to reduction processing, the simplified representation of the data is obtained through a reduction technology, the occupied space of the simplified data can be reduced, but the nearly same analysis result can be generated, and the efficiency of the whole system can be improved.
The above feature data may be organized into feature data sets and model training may be performed by the feature data sets.
In the step S3, the device trains at least two machine learning models through the sample data, and combines the at least two machine learning models to obtain an incoming line traffic prediction model. Taking training three machine learning models as an example, the following is explained:
the deepforest, xgboost, randomforesta algorithm is used to obtain a model of each algorithm, and a predicted value can be generated by using each single algorithm prediction model alone.
The single algorithm prediction model predicts the values of each day for a future period of time, the default period being 30 days, i.e. three prediction results d1, d2, d3 are obtained each day for 30 days. Model integration, i.e., combining models, after prediction is completed, the combining strategy may include:
(1) The bisection method is to average the prediction results of three models.
(2) And (5) measuring an algorithm. The random operation produces 100 sets of weight combinations [ (a 1, b1, c 1), (a 2, b2, c 2),. A.100, b100, c 100) ], where ax+bx+cx=1. Substituting each weight combination into the predicted data of the test set to calculate, calculating a new predicted result d=d1×a1+d2×b1+d3×c1, then adopting a Root Mean Square Error (RMSE) method to respectively calculate the predicted value and the true value of the predicted result generated by each weight combination on the test set, and evaluating the best weight combination as a final model combination weight parameter. And after the weight combination parameters are determined, the combination model can be used for carrying out combination prediction on the predicted value of each day of a period in the future.
The method for establishing the incoming line traffic prediction model provided by the embodiment of the invention obtains the historical traffic data of the incoming line, carries out preprocessing on the historical traffic data, and obtains time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value; extracting features of the time sequence data to obtain sample data for training a model; and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model, so that incoming line traffic can be rapidly and accurately predicted.
Further, the historical traffic data includes a date and a current day traffic corresponding to the date; correspondingly, the preprocessing the historical traffic data to obtain time sequence data based on the deformation parameters comprises the following steps:
determining an initial value of the future number and an initial value of the reduction ratio parameter; reference is made to the above description and will not be repeated.
Determining a contemporaneous average value corresponding to the date according to the initial value of the period number and the service volume of each day, and calculating a correction service volume corresponding to each service volume of each day according to the contemporaneous average value, the service volume of each day and the initial value of the reduction ratio parameter; reference is made to the above description and will not be repeated.
If the correction traffic is determined to be greater than zero, recording a current future number, a current reduction ratio parameter and a current correction traffic, adding 1 to an initial value of the future number, and continuously executing the initial value of the future number, the initial value of the reduction ratio parameter and the subsequent steps until the initial value of the future number reaches a preset future number threshold; reference is made to the above description and will not be repeated.
After the preset term threshold is reached, acquiring recorded correction traffic, performing mean square error calculation, and taking a deformation parameter with the minimum mean square error calculation result as a target deformation parameter; reference is made to the above description and will not be repeated.
And carrying out data correction on the historical traffic data according to the target deformation parameters and the target correction traffic corresponding to the target deformation parameters to obtain the time sequence data. Reference is made to the above description and will not be repeated.
The method for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values, and further can rapidly and accurately predict the incoming line traffic.
Further, before the step of determining the initial value of the future number and the initial value of the reduction ratio parameter, the method of establishing an incoming line traffic prediction model further includes:
And cleaning the data of the historical traffic data. Reference is made to the above description and will not be repeated.
The method for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values, and further can rapidly and accurately predict the incoming line traffic.
Further, after the step of performing data correction on the historical traffic data according to the target deformation parameter and the target correction traffic corresponding to the target deformation parameter, the method for establishing the incoming line traffic prediction model further includes:
and determining abnormal data in the historical traffic data after data correction, and performing smoothing processing on the abnormal data to obtain the time sequence data. Reference is made to the above description and will not be repeated.
The method for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values, and further can rapidly and accurately predict the incoming line traffic.
Further, the method for establishing the incoming line traffic prediction model further comprises the following steps:
if the corrected traffic is determined to be less than or equal to zero, reducing the initial value of the reduction ratio parameter by a preset ratio, and continuously executing the calculation of the corrected traffic corresponding to each current day traffic and the subsequent steps according to the contemporaneous average value, each current day traffic and the initial value of the reduction ratio parameter. As shown in fig. 2, the preset proportion may be set autonomously according to the actual situation, and may be selected to be 5%.
The method for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values, and further can rapidly and accurately predict the incoming line traffic.
Further, the method for establishing the incoming line traffic prediction model further comprises the following steps:
if the date is determined to be the designated holiday, determining the contemporaneous average value by taking the year as granularity. Reference is made to the above description and will not be repeated.
The method for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time series data with reasonable values aiming at the designated holiday, and further can rapidly and accurately predict the incoming line traffic.
Further, the method for establishing the incoming line traffic prediction model further comprises the following steps:
if the date is determined not to be a designated holiday, the contemporaneous average is determined with month as granularity and according to the week the date is on and the days of the week. Reference is made to the above description and will not be repeated.
The method for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time series data with reasonable values aiming at unspecified holidays, and further can rapidly and accurately predict incoming line traffic.
Further, as shown in fig. 3, the incoming line traffic data prediction method based on the method for establishing an incoming line traffic prediction model includes the following steps:
r1: acquiring traffic data of an incoming line to be predicted;
r2: and predicting the traffic data based on the incoming line traffic prediction model to obtain a traffic prediction result. And inputting the traffic data of the incoming line to be predicted to the incoming line traffic prediction model, and taking the output result of the incoming line traffic prediction model as a traffic prediction result.
Further, the traffic data prediction method of the incoming line includes:
and correcting the traffic prediction result according to the target deformation parameter and the target correction traffic corresponding to the target deformation parameter to obtain a traffic prediction correction result.
The specific method comprises the following steps: for the traffic prediction result [ T, D ], the N-period synchronization average before the T-day is D ', the T-day traffic prediction result is d1=d+d' ×m. And correcting the prediction result of each day of a period in the future according to the method, wherein the obtained result is the prediction of the daily traffic in the period in the future.
It should be noted that, the method for establishing the incoming line traffic prediction model provided by the embodiment of the invention can be used in the financial field, and also can be used in any technical field except the financial field.
As shown in fig. 4, the method of the embodiment of the present invention may be implemented in a modular manner, including:
the data transformation device 1 is responsible for acquiring the history incoming call data of each line of the processed telephone bank, and performs data cleaning, data analysis, data transformation, noise processing and the like after acquiring the required data, and comprises the following components: the incoming line historical traffic acquiring unit 11, the historical traffic data cleaning unit 12, the data deformation strategy analyzing unit 13, the data deformation processing unit 14 and the noise data processing unit 15 can acquire telephone banking line incoming time series data with smaller fluctuation range and relatively stable trend after passing through the data deformation device 1.
The model prediction device 2 is responsible for modeling and predicting telephone banking line incoming time series data with relatively stable trend after deformation processing, and obtaining accurate prediction results of the time series every day within a period in the future. Comprising the following steps: the data acquisition unit 21, the data processing unit 22, the model training unit 23 and the combined prediction unit 24 can obtain accurate prediction data of each day in a period of the deformed time series after being processed by the model prediction device 2.
The processing and correcting device 3 is responsible for correcting the predicted result of the model predicting device, correcting the predicted data daily in a future period predicted by the model predicting device 2 according to the deformation parameters generated by the data deforming device 1, and obtaining the accurate predicted result of daily traffic in the future period of the telephone bank incoming line.
As shown in fig. 5, the data morphing device 1 specifically includes:
an incoming line history traffic acquiring unit 11 for acquiring history traffic data of each incoming line of a telephone bank, comprising: incoming line type, incoming date, incoming traffic on the day.
A historical traffic data cleansing unit 12 for cleansing acquired telephone banking incoming line historical traffic data, comprising: and removing repeated data and repairing missing data, and obtaining continuous and unrepeated historical traffic time sequence data after data cleaning.
The data deformation policy analysis unit 13 is configured to calculate and analyze a data deformation policy, determine deformation parameters, and include a period number N for taking a contemporaneous average value, and a proportion parameter M for reducing an amplitude of a sequence average value, where the period number N and the proportion parameter M are calculated and determined, and then calculate a new sequence as deformed time sequence data.
The data morphing processing unit 14 morphs the historical traffic data sequence according to the morphing parameters [ N, M ] analyzed by the data morphing policy analysis unit 13, generating new time series data.
The noise data processing unit 15 performs noise analysis on the new time-series data, sets a point three times the standard deviation of the data set as noise data by using a normal distribution 3 principle, and performs smoothing processing on the noise data by using a smoothdata method.
As shown in fig. 6, the model prediction apparatus 2 includes:
the data acquisition unit 21 is responsible for acquiring time series data and holiday data after the application incoming line history traffic data sequence is subjected to deformation processing.
The data processing unit 22 is responsible for carrying out feature processing and normalization processing on the data to obtain a data format which can be used for training a machine learning algorithm, wherein the features comprise three main types of features of time period features, sequence association features and sequence statistical features. The normalization processing is carried out after the feature generation so as to facilitate the training of the input machine learning algorithm.
The model training unit 23 is responsible for performing model training on the feature data set generated by the data processing unit 22 by using the deepforest, xgboost, randomforesta three machine learning algorithms, so as to obtain a plurality of single algorithm training models.
The combined prediction unit 24 is responsible for carrying out combined prediction on the single algorithm model generated by the model training unit 23 to obtain a final prediction result of the prediction data of each day in a period of the future.
As shown in fig. 7, the process correction device 3 includes:
the deformation parameter acquiring unit 31 is responsible for acquiring the deformation parameters [ N, M ] generated by the data deforming device for later data correction.
The data correction unit 32 is responsible for calculating the result predicted by the correction model prediction device according to the obtained deformation parameters, and obtaining final traffic prediction data.
Fig. 8 is a schematic structural diagram of an apparatus for establishing an incoming line traffic prediction model according to an embodiment of the present invention, as shown in fig. 8, where the apparatus for establishing an incoming line traffic prediction model according to an embodiment of the present invention includes an obtaining unit 801, an extracting unit 802, and a combining unit 803, where:
the acquiring unit 801 is configured to acquire historical traffic data of an incoming line, pre-process the historical traffic data, and obtain time series data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value; the extracting unit 802 is configured to perform feature extraction on the time series data to obtain sample data for training a model; the combining unit 803 is configured to train at least two machine learning models according to the sample data, and combine the at least two machine learning models to obtain an incoming line traffic prediction model.
Specifically, an acquiring unit 801 in the device is configured to acquire historical traffic data of an incoming line, perform preprocessing on the historical traffic data, and obtain time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value; the extracting unit 802 is configured to perform feature extraction on the time series data to obtain sample data for training a model; the combining unit 803 is configured to train at least two machine learning models according to the sample data, and combine the at least two machine learning models to obtain an incoming line traffic prediction model.
The device for establishing the incoming line traffic prediction model provided by the embodiment of the invention acquires the historical traffic data of the incoming line, performs preprocessing on the historical traffic data, and obtains time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value; extracting features of the time sequence data to obtain sample data for training a model; and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model, so that incoming line traffic can be rapidly and accurately predicted.
Further, the acquiring unit 801 is specifically configured to:
determining an initial value of the future number and an initial value of the reduction ratio parameter; reference is made to the above description and will not be repeated.
Determining a contemporaneous average value corresponding to the date according to the initial value of the period number and the service volume of each day, and calculating a correction service volume corresponding to each service volume of each day according to the contemporaneous average value, the service volume of each day and the initial value of the reduction ratio parameter; reference is made to the above description and will not be repeated.
If the correction traffic is determined to be greater than zero, recording a current future number, a current reduction ratio parameter and a current correction traffic, adding 1 to an initial value of the future number, and continuously executing the initial value of the future number, the initial value of the reduction ratio parameter and the subsequent steps until the initial value of the future number reaches a preset future number threshold; reference is made to the above description and will not be repeated.
After the preset term threshold is reached, acquiring recorded correction traffic, performing mean square error calculation, and taking a deformation parameter with the minimum mean square error calculation result as a target deformation parameter; reference is made to the above description and will not be repeated.
And carrying out data correction on the historical traffic data according to the target deformation parameters and the target correction traffic corresponding to the target deformation parameters to obtain the time sequence data. Reference is made to the above description and will not be repeated.
The device for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values, and further can rapidly and accurately predict the incoming line traffic.
Further, before the step of determining the initial value of the future number and the initial value of the reduction ratio parameter, the means for establishing an incoming line traffic prediction model is further configured to:
and cleaning the data of the historical traffic data. Reference is made to the above description and will not be repeated.
The device for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values, and further can rapidly and accurately predict the incoming line traffic.
Further, after the step of performing data correction on the historical traffic data according to the target deformation parameter and the target correction traffic corresponding to the target deformation parameter, the apparatus for establishing the incoming line traffic prediction model is further configured to:
And determining abnormal data in the historical traffic data after data correction, and performing smoothing processing on the abnormal data to obtain the time sequence data. Reference is made to the above description and will not be repeated.
The device for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values, and further can rapidly and accurately predict the incoming line traffic.
Further, the device for establishing the incoming line traffic prediction model is further used for:
if the corrected traffic is determined to be less than or equal to zero, reducing the initial value of the reduction ratio parameter by a preset ratio, and continuously executing the calculation of the corrected traffic corresponding to each current day traffic and the subsequent steps according to the contemporaneous average value, each current day traffic and the initial value of the reduction ratio parameter. As shown in fig. 2, the preset proportion may be set autonomously according to the actual situation, and may be selected to be 5%.
The device for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values, and further can rapidly and accurately predict the incoming line traffic.
Further, the device for establishing the incoming line traffic prediction model is further used for:
If the date is determined to be the designated holiday, determining the contemporaneous average value by taking the year as granularity. Reference is made to the above description and will not be repeated.
The device for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time sequence data with reasonable values aiming at the designated holiday, and further can rapidly and accurately predict the incoming line traffic.
Further, the device for establishing the incoming line traffic prediction model is further used for:
if the date is determined not to be a designated holiday, the contemporaneous average is determined with month as granularity and according to the week the date is on and the days of the week. Reference is made to the above description and will not be repeated.
The device for establishing the incoming line traffic prediction model provided by the embodiment of the invention can obtain time series data with reasonable values aiming at unspecified holidays, and further can rapidly and accurately predict incoming line traffic.
Further, the traffic data prediction device of the incoming line based on the device for establishing the traffic prediction model of the incoming line is specifically configured to:
acquiring traffic data of an incoming line to be predicted; reference is made to the above description and will not be repeated.
And predicting the traffic data based on the incoming line traffic prediction model to obtain a traffic prediction result. Reference is made to the above description and will not be repeated.
The traffic data prediction device of the incoming line provided by the embodiment of the invention can rapidly and accurately predict the traffic of the incoming line.
Further, the traffic data prediction device of the incoming line is specifically configured to:
and correcting the traffic prediction result according to the target deformation parameter and the target correction traffic corresponding to the target deformation parameter to obtain a traffic prediction correction result. Reference is made to the above description and will not be repeated.
The traffic data prediction device of the incoming line provided by the embodiment of the invention can further rapidly and accurately predict the traffic of the incoming line.
The embodiment of the device for establishing the incoming line traffic prediction model provided in the embodiment of the present invention may be specifically used to execute the processing flow of each method embodiment, and the functions thereof are not described herein in detail, and reference may be made to the detailed description of the method embodiments.
Fig. 9 is a schematic diagram of an entity structure of an electronic device according to an embodiment of the present invention, as shown in fig. 9, where the electronic device includes: a processor (processor) 901, a memory (memory) 902, and a bus 903;
Wherein, the processor 901 and the memory 902 complete communication with each other through the bus 903;
the processor 901 is configured to call the program instructions in the memory 902 to perform the methods provided in the above method embodiments, for example, including:
acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
extracting features of the time sequence data to obtain sample data for training a model;
and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model.
The present embodiment discloses a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, are capable of performing the methods provided by the above-described method embodiments, for example comprising:
Acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
extracting features of the time sequence data to obtain sample data for training a model;
and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model.
The present embodiment provides a computer-readable storage medium storing a computer program that causes the computer to execute the methods provided by the above-described method embodiments, for example, including:
acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
Extracting features of the time sequence data to obtain sample data for training a model;
and training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In the description of the present specification, reference to the terms "one embodiment," "one particular embodiment," "some embodiments," "for example," "an example," "a particular example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (11)
1. A method of establishing an incoming line traffic prediction model, comprising:
acquiring historical traffic data of an incoming line, preprocessing the historical traffic data, and obtaining time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
extracting features of the time sequence data to obtain sample data for training a model;
training at least two machine learning models through the sample data, and combining the at least two machine learning models to obtain an incoming line traffic prediction model;
the historical traffic data comprises a date and the current day traffic corresponding to the date; correspondingly, the preprocessing the historical traffic data to obtain time sequence data based on the deformation parameters comprises the following steps:
Determining an initial value of the future number and an initial value of the reduction ratio parameter;
determining a contemporaneous average value corresponding to the date according to the initial value of the period number and the service volume of each day, and calculating a correction service volume corresponding to each service volume of each day according to the contemporaneous average value, the service volume of each day and the initial value of the reduction ratio parameter;
if the correction traffic is determined to be greater than zero, recording a current future number, a current reduction ratio parameter and a current correction traffic, adding 1 to an initial value of the future number, and continuously executing the initial value of the future number, the initial value of the reduction ratio parameter and the subsequent steps until the initial value of the future number reaches a preset future number threshold;
after the preset term threshold is reached, acquiring recorded correction traffic, performing mean square error calculation, and taking a deformation parameter with the minimum mean square error calculation result as a target deformation parameter;
and carrying out data correction on the historical traffic data according to the target deformation parameters and the target correction traffic corresponding to the target deformation parameters to obtain the time sequence data.
2. The method of modeling incoming line traffic as defined in claim 1, wherein prior to said step of determining an initial value of said future number and an initial value of said downscaling parameter, said method of modeling incoming line traffic further comprises:
And cleaning the data of the historical traffic data.
3. The method of building an incoming line traffic prediction model according to claim 1, wherein after said step of data correcting historical traffic data according to said target deformation parameter and a target correction traffic corresponding to said target deformation parameter, said method of building an incoming line traffic prediction model further comprises:
and determining abnormal data in the historical traffic data after data correction, and performing smoothing processing on the abnormal data to obtain the time sequence data.
4. The method of modeling incoming line traffic as defined in claim 1, wherein said method of modeling incoming line traffic further comprises:
if the corrected traffic is determined to be less than or equal to zero, reducing the initial value of the reduction ratio parameter by a preset ratio, and continuously executing the calculation of the corrected traffic corresponding to each current day traffic and the subsequent steps according to the contemporaneous average value, each current day traffic and the initial value of the reduction ratio parameter.
5. The method of modeling incoming line traffic as defined in claim 1, wherein said method of modeling incoming line traffic further comprises:
If the date is determined to be the designated holiday, determining the contemporaneous average value by taking the year as granularity.
6. The method of modeling incoming line traffic as defined in claim 1, wherein said method of modeling incoming line traffic further comprises:
if the date is determined not to be a designated holiday, the contemporaneous average is determined with month as granularity and according to the week the date is on and the days of the week.
7. A traffic data prediction method of an incoming line based on the method of building an incoming line traffic prediction model according to claim 1, comprising:
acquiring traffic data of an incoming line to be predicted;
and predicting the traffic data based on the incoming line traffic prediction model to obtain a traffic prediction result.
8. The traffic data prediction method of an incoming line according to claim 7, comprising:
and correcting the traffic prediction result according to the target deformation parameter and the target correction traffic corresponding to the target deformation parameter to obtain a traffic prediction correction result.
9. An apparatus for establishing an incoming line traffic prediction model, comprising:
The acquisition unit is used for acquiring historical traffic data of the incoming line, preprocessing the historical traffic data and acquiring time sequence data based on deformation parameters; the deformation parameters comprise the period number of the contemporaneous average value of the historical traffic data and the reduction ratio parameter of the traffic corresponding to the contemporaneous average value;
the extraction unit is used for extracting the characteristics of the time sequence data to obtain sample data for training a model;
the combining unit is used for training at least two machine learning models through the sample data and combining the at least two machine learning models to obtain an incoming line traffic prediction model;
the historical traffic data comprises a date and the current day traffic corresponding to the date; correspondingly, the acquisition unit is specifically configured to:
determining an initial value of the future number and an initial value of the reduction ratio parameter;
determining a contemporaneous average value corresponding to the date according to the initial value of the period number and the service volume of each day, and calculating a correction service volume corresponding to each service volume of each day according to the contemporaneous average value, the service volume of each day and the initial value of the reduction ratio parameter;
If the correction traffic is determined to be greater than zero, recording a current future number, a current reduction ratio parameter and a current correction traffic, adding 1 to an initial value of the future number, and continuously executing the initial value of the future number, the initial value of the reduction ratio parameter and the subsequent steps until the initial value of the future number reaches a preset future number threshold;
after the preset term threshold is reached, acquiring recorded correction traffic, performing mean square error calculation, and taking a deformation parameter with the minimum mean square error calculation result as a target deformation parameter;
and carrying out data correction on the historical traffic data according to the target deformation parameters and the target correction traffic corresponding to the target deformation parameters to obtain the time sequence data.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of any one of claims 1 to 8 when the computer program is executed by the processor.
11. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210686679.9A CN114900443B (en) | 2022-06-17 | 2022-06-17 | Method and device for establishing incoming line traffic prediction model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210686679.9A CN114900443B (en) | 2022-06-17 | 2022-06-17 | Method and device for establishing incoming line traffic prediction model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114900443A CN114900443A (en) | 2022-08-12 |
| CN114900443B true CN114900443B (en) | 2024-01-26 |
Family
ID=82729150
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210686679.9A Active CN114900443B (en) | 2022-06-17 | 2022-06-17 | Method and device for establishing incoming line traffic prediction model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114900443B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105847598A (en) * | 2016-04-05 | 2016-08-10 | 浙江远传信息技术股份有限公司 | Method and device for call center multifactorial telephone traffic prediction |
| CN110163417A (en) * | 2019-04-26 | 2019-08-23 | 阿里巴巴集团控股有限公司 | A kind of prediction technique of portfolio, device and equipment |
| CN113918884A (en) * | 2020-07-09 | 2022-01-11 | 顺丰科技有限公司 | Traffic prediction model construction method and traffic prediction method |
-
2022
- 2022-06-17 CN CN202210686679.9A patent/CN114900443B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105847598A (en) * | 2016-04-05 | 2016-08-10 | 浙江远传信息技术股份有限公司 | Method and device for call center multifactorial telephone traffic prediction |
| CN110163417A (en) * | 2019-04-26 | 2019-08-23 | 阿里巴巴集团控股有限公司 | A kind of prediction technique of portfolio, device and equipment |
| CN113918884A (en) * | 2020-07-09 | 2022-01-11 | 顺丰科技有限公司 | Traffic prediction model construction method and traffic prediction method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114900443A (en) | 2022-08-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104424598A (en) | Cash demand quantity predicating device and method | |
| CN111210093B (en) | Daily water consumption prediction method based on big data | |
| CN104811344A (en) | Network dynamic service monitoring method and apparatus | |
| CN110852476A (en) | Passenger flow prediction method and device, computer equipment and storage medium | |
| CN111932292A (en) | Cigarette product sales prediction method and device based on deep neural network | |
| CN111062602A (en) | Method, device and storage medium for enterprise credit risk assessment | |
| CN114900443B (en) | Method and device for establishing incoming line traffic prediction model | |
| CN111738507A (en) | Bank clearing position fund payment amount prediction method, device, equipment and medium | |
| CN110992189A (en) | Resource data estimation method, resource data estimation device, computer equipment and storage medium | |
| CN112165412A (en) | Flow prediction method based on time series model | |
| CN109242578A (en) | Value of house prediction technique, device, computer equipment and storage medium | |
| CN111583010A (en) | Data processing method, device, equipment and storage medium | |
| CN112529682B (en) | Cash distribution method based on data modeling, intelligent terminal and storage medium | |
| CN114399345A (en) | Financial power transmission right price prediction method and device | |
| CN113487110A (en) | Spare payment management method and device | |
| CN115238940A (en) | Training method and device of power grid load prediction model | |
| CN112861064A (en) | Social credit evaluation source data processing method, system, terminal and medium | |
| CN113962741B (en) | Coal sales data prediction method, equipment and medium | |
| CN111737240A (en) | Method and device for judging energy consumption attribution date and storage medium | |
| CN111008721A (en) | Self-adaptive prediction method and system based on time series prediction model | |
| CN101669137A (en) | Processing device for normalizing bars representative of weighted variable quantities | |
| CN113743532B (en) | Abnormality detection method, abnormality detection device, abnormality detection apparatus, and computer storage medium | |
| CN112766535B (en) | Building load prediction method and system considering load curve characteristics | |
| CN110688373A (en) | OFFSET method based on logistic regression | |
| CN117011063B (en) | Customer transaction risk prediction processing 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |