CN116843368A - Marketing data processing method based on ARMA model - Google Patents
Marketing data processing method based on ARMA model Download PDFInfo
- Publication number
- CN116843368A CN116843368A CN202310871130.1A CN202310871130A CN116843368A CN 116843368 A CN116843368 A CN 116843368A CN 202310871130 A CN202310871130 A CN 202310871130A CN 116843368 A CN116843368 A CN 116843368A
- Authority
- CN
- China
- Prior art keywords
- time period
- last time
- arma model
- optimal
- obtaining
- 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.)
- Granted
Links
- 241001123248 Arma Species 0.000 title claims abstract description 61
- 238000003672 processing method Methods 0.000 title claims abstract description 16
- 239000011159 matrix material Substances 0.000 claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000000034 method Methods 0.000 claims description 21
- 238000007781 pre-processing Methods 0.000 claims description 2
- 230000011218 segmentation Effects 0.000 claims 1
- 230000000694 effects Effects 0.000 description 8
- 238000007405 data analysis Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data prediction, in particular to a marketing data processing method based on an ARMA model, which comprises the steps of segmenting a time sequence to obtain a plurality of time periods, and obtaining the prediction error quantity of the ARMA model corresponding to each time period according to an information quantity criterion; obtaining a distance matrix of each time period except the last time period and the last time period, obtaining an optimal path of the distance matrix of each time period except the last time period and the last time period, and obtaining the matching degree of each time period except the last time period and the last time period according to the optimal path of the distance matrix and the distance matrix; obtaining a target time period according to the matching degree; and obtaining an ARMA model with the optimal last time period according to the target time period, and carrying out marketing data processing according to the ARMA model with the optimal last time period. The invention improves the accuracy of local prediction and realizes more accurate local data prediction.
Description
Technical Field
The invention relates to the technical field of data prediction, in particular to a marketing data processing method based on an ARMA model.
Background
With the aggravation of market competition, the establishment of marketing strategies and the evaluation of the release effect of each enterprise are more and more emphasized, and the analysis of marketing data can help the enterprise to better know the advertisement release effect, the client behavior trend and the market change, so that more effective marketing strategies are established, and therefore, the marketing data analysis is needed. Traditional marketing data analysis is mainly used for analyzing dynamic changes and potential rules of time series data through an ARMA model.
Because the traditional ARMA model frequently adopts unified order p and order q for data prediction on the whole time sequence, if the stability of the whole time sequence is poor, larger interference exists on a prediction result, so that the prediction result is inaccurate; in order to obtain a more accurate prediction effect, the invention provides an ARMA model-based marketing data processing method for carrying out data prediction by segmenting a time sequence, obtaining the most suitable order p and the most suitable order q of different time periods, obtaining the matching degree of each time period and the last time period, and obtaining the optimal ARMA model of the last time period according to the matching degree.
Disclosure of Invention
The invention provides a marketing data processing method based on an ARMA model, which aims to solve the existing problems.
The marketing data processing method based on the ARMA model adopts the following technical scheme:
one embodiment of the present invention provides a marketing data processing method based on an ARMA model, the method comprising the steps of:
collecting marketing data, and preprocessing the marketing data to obtain a time sequence;
segmenting the time sequence to obtain a plurality of time periods, and obtaining the prediction error quantity of the ARMA model corresponding to each time period according to the information quantity criterion;
obtaining a distance matrix of each time period except the last time period and the last time period, obtaining an optimal path of the distance matrix of each time period except the last time period and the last time period, and obtaining the matching degree of each time period except the last time period and the last time period according to the optimal path of the distance matrix and the predicted error amount; obtaining a target time period according to the matching degree of each time period except the last time period and the last time period; and obtaining an ARMA model with the optimal last time period according to the target time period, and carrying out marketing data processing according to the ARMA model with the optimal last time period.
Preferably, the time sequence is segmented to obtain a plurality of time periods, and the specific method comprises the following steps:
starting from the last time sequence data, recording each m time sequence data as a time period to obtain a plurality of time periods, wherein if the residual time sequence data is less than m, the actual time sequence data quantity is taken as a time period, and m is the preset quantity.
Preferably, the method obtains the prediction error amount of the ARMA model corresponding to each time period according to the information amount criterion, including
The body method comprises the following steps:
processing time sequence data corresponding to a reference time period according to a red pool information amount criterion by taking any time period as the reference time period to obtain an optimal order p and an optimal order q, and constructing a corresponding ARMA model according to the time sequence data corresponding to the next time period of the reference time period and the optimal order p and the optimal order q of the reference time period to obtain a plurality of predicted values of the next time period of the reference time period; and carrying out difference between a plurality of predicted values of the next time period of the reference time period and corresponding time sequence data, taking absolute values, and recording the accumulated values of all the absolute values in the next time period of the reference time period as the predicted error amount of the corresponding ARMA model of the reference time period.
Preferably, the optimal path according to the distance matrix and the prediction error amount are obtained when dividing the last
The matching degree of each time period and the last time period outside the interval comprises the following specific methods:
for any time period except the last time period, the time period and the most time period are acquiredThe distance matrix and the optimal path of the latter time period are recorded as a first data value by the numerical value of each data point in the distance matrix, wherein P represents the matching degree of the time period and the last time period; n represents the number of data points on the best path in the distance matrix; l (L) t A Euclidean distance value representing the t data point on the optimal path in the distance matrix to an ideal diagonal in the distance matrix; d (D) t A first data value representing a t-th data point on the best path in the distance matrix; a represents the prediction error amount of the ARMA model corresponding to the time zone.
Preferably, the matching degree between each time period except the last time period and the last time period is obtained
The target time period comprises the following specific methods:
the time period with the smallest matching degree with the last time period is recorded as the target time period.
Preferably, the method for obtaining the ARMA model with the optimal last time period according to the target time period includes the following specific steps:
taking the optimal orders p and q corresponding to the target time period as the new optimal orders p and q of the last time period, and obtaining the ARMA model with the optimal time period according to the new optimal orders p and q of the last time period.
The technical scheme of the invention has the beneficial effects that: the traditional ARMA model frequently adopts unified order p and order q for data prediction on the whole time sequence, but if the stability of the whole time sequence is poor, larger interference exists on a prediction result, so that the prediction result is inaccurate; in order to obtain a more accurate prediction effect, compared with the prior art, the method and the device improve the accuracy of local prediction by segmenting time sequence data, further obtain the matching degree of different time periods by utilizing the local prediction error amount and the stability approximation degree value between different time periods, further obtain an optimal local ARMA model and realize more accurate local data prediction.
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.
FIG. 1 is a flow chart of the steps of the ARMA model-based marketing data processing method of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description is given below of the specific implementation, structure, characteristics and effects of the ARMA model-based marketing data processing method according to the present invention in combination with the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the ARMA model-based marketing data processing method provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of a method for processing marketing data based on an ARMA model according to an embodiment of the invention is shown, the method includes the steps of:
step S001: and acquiring marketing data to obtain a time sequence.
It should be noted that, in the traditional marketing data analysis, dynamic change and potential law analysis are mainly performed on time series data through an ARMA model, but the prediction effect of the ARMA model is different due to different p and q values in the ARMA model, so that the data is affected; in order to obtain a more accurate prediction effect, the embodiment provides a marketing data processing method based on an ARMA model.
Specifically, in order to implement the marketing data processing method based on the ARMA model provided in the present embodiment, marketing data needs to be collected first, and the specific process is as follows: and acquiring business balance data of the marketing platform in three days, wherein the business balance data is recorded once per hour, and a sequence formed by sorting all the business balance data according to time sequence is recorded as a time sequence, wherein each business balance data corresponds to one time sequence data.
So far, the time sequence is obtained by the method.
Step S002: and segmenting the time sequence to obtain a plurality of time periods, and obtaining the prediction error quantity of the time periods according to the red pool information quantity criterion.
It should be noted that, because the conventional ARMA model often adopts a unified order p and an order q to perform data prediction on the overall time sequence, if the stability of the overall time sequence is poor, a large interference exists on the prediction result, so that the prediction result is inaccurate; therefore, the time sequence is segmented, and the optimal orders p and q of each time period are respectively analyzed, so that the stability of the local time sequence data can be increased, and the accuracy of local prediction is improved.
It should be further noted that, when the optimal orders p and q are estimated only by means of local existing time sequence data, a certain error exists in the obtained prediction result, and in order to make the prediction result more accurate, the order corresponding to the time when the prediction result error is the smallest may be adopted as the local optimal order to establish the ARMA model.
Specifically, segmenting the time sequence, starting from the last time sequence data, recording every 100 time sequence data as a time period to obtain a plurality of time periods, wherein each time period has a plurality of time sequence data, and if the residual time sequence data is less than 100, taking the actual time sequence data as a time period;
in this embodiment, an arbitrary time period is taken as an example to describe, according to the red pool information quantity criterion, the time sequence data corresponding to the time period is processed to obtain an optimal order p and an optimal order q, and according to the optimal order p and the optimal order q of the time period, a corresponding ARMA model is constructed for the time sequence data corresponding to the next time period of the time period, so as to obtain a plurality of predicted values of the next time period of the time period, wherein each predicted value corresponds to one time sequence data in the next time period of the time period; taking absolute values by making differences between a plurality of predicted values of the next time period of the time period and corresponding time sequence data, and recording the accumulated values of all the absolute values in the corresponding time period as the predicted error amount of the corresponding ARMA model of the time period; the business data size corresponding to the time sequence data of the next time period of the last time period is 0, and the red pool information amount criterion is a known technique, which is not described in this embodiment.
So far, the prediction error amounts of the ARMA model corresponding to all the time periods are obtained through the method.
Step S003: obtaining the matching degree of the time period and the last time period according to the distance matrix between the time periods and the prediction error quantity of the time period, obtaining the target time period according to the matching degree, obtaining the optimal ARMA model according to the target time period, and predicting marketing data.
It should be noted that, when predicting the future marketing data, the future marketing data may be predicted according to the time sequence data corresponding to the last time period; since the prediction error amount of the last time period is not necessarily actually minimum, the accuracy of the prediction result is not high; in order to minimize the prediction error amount of the last time period, the approximation degree between the time sequence data of other time periods except the last time period and the time sequence data of the last time period can be calculated, the matching degree of the last time period and other time periods can be obtained according to the approximation degree, the time period which is most matched with the last time period can be obtained according to the matching degree, the optimal orders p and q corresponding to the time period which is most matched with the last time period are taken as the optimal orders p and q of the last time period, and the new ARMA model of the last time period can be obtained, so that the accuracy of the prediction result is improved.
It should be further noted that, when calculating the approximation degree between the time series data of each time period except the last time period and the time series data of the last time period, the approximation degree of the two time periods may be obtained by using a DTW matching algorithm; however, if the DTW matching algorithm is directly used to calculate the approximation degree of the two time periods, because the time sequence data corresponding to the two time periods are in parallel relationship, a larger distance value can be obtained, so that the DTW cannot calculate the effective approximation degree; the time sequence data in the two time periods in parallel relation have similar variation trend and similar stability, so that the optimal orders p and q corresponding to the two time periods are not different greatly, and the time period which is most similar to the stability of the last time period can be obtained as the time period which is most matched with the last time period according to the similarity of the stability of the time sequence data in the two time periods.
Further, when the similarity determination of the stability of the time series data in the two time periods is performed, if the optimal path of the time series data corresponding to the two time periods on the distance matrix of the DTW matching algorithm is closer to the ideal diagonal line from the lower left to the upper right in the distance matrix, the stability of the time series data corresponding to the two time periods is more similar; however, since the value corresponding to the approach position of the optimal path and the diagonal line in the distance matrix may be still large, the approximation judgment of the stability degree of the time series data corresponding to the two time periods may be interfered, and the anti-interference processing is required.
In addition, when the anti-interference processing is performed, the euclidean distance value from each data point on the optimal path to the ideal diagonal line in the distance matrix corresponding to the two time periods can be obtained, if the euclidean distance value is smaller, the time sequence data change representing the two time periods is more approximate, the prediction error amount corresponding to each other time period except the last time period is more suitable as the prediction error amount corresponding to the last time period, and even if the corresponding partial numerical value on the optimal path is larger, the stability degree is more approximate, so that the data of the original position is smaller, and the interference when the similarity judgment of the stability degree is performed between the time sequence data of each other time period except the last time period and the time sequence data of the last time period is eliminated.
Specifically, this embodiment is described by taking any time period other than the last time period as an example, a distance matrix of the time sequence data of the time period and the time sequence data of the last time period is obtained, an optimal path of the distance matrix in the DTW algorithm is obtained, and a diagonal line from bottom left to top right in the distance matrix is recorded as an ideal diagonal line, wherein the distance matrix and the optimal path corresponding to the distance matrix are known contents of the DTW algorithm, and this embodiment is not described.
Further, the numerical value of each data point in the distance matrix is recorded as a first data value, and the matching degree of the time period and the last time period is obtained according to the first data value, wherein the calculation formula of the matching degree of the time period and the last time period is as follows:
wherein, P represents the matching degree of the time period and the last time period, if the matching degree is smaller, the stability degree corresponding to the time period and the last time period is more approximate, and the prediction error is smaller; n represents the number of data points on the best path in the distance matrix; l (L) t The Euclidean distance value from the t data point on the optimal path in the distance matrix to the ideal diagonal line in the distance matrix is represented, if the Euclidean distance value is smaller, the change of the two time sequence data corresponding to the t data point in the time sequence data corresponding to the time period and the time sequence data corresponding to the last time period is more approximate, and the stability is more approximate; d (D) t The first data value representing the t data point on the optimal path in the distance matrix, wherein the larger the first data value is, the larger the absolute value of the difference value of the two time sequence data corresponding to the t data point is in the time sequence data corresponding to the time period and the time sequence data corresponding to the last time period; a represents the prediction error amount of the ARMA model corresponding to the time zone.
And acquiring the matching degree of each time period except the last time period and the last time period.
Further, the time period with the minimum matching degree with the last time period is recorded as a target time period, the optimal orders p and q corresponding to the target time period are used as new optimal orders p and q of the last time period, and an ARMA model with the optimal last time period is obtained according to the new optimal orders p and q of the last time period; and carrying out a section of marketing data prediction under the non-occurrence period according to the ARMA model to complete the final marketing data prediction processing of the ARMA model, thereby obtaining a more accurate marketing data prediction result of the ARMA model.
This embodiment is completed.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (6)
1. A marketing data processing method based on an ARMA model, the method comprising the steps of:
collecting marketing data, and preprocessing the marketing data to obtain a time sequence;
segmenting the time sequence to obtain a plurality of time periods, and obtaining the prediction error quantity of the ARMA model corresponding to each time period according to the information quantity criterion;
obtaining a distance matrix of each time period except the last time period and the last time period, obtaining an optimal path of the distance matrix of each time period except the last time period and the last time period, and obtaining the matching degree of each time period except the last time period and the last time period according to the optimal path of the distance matrix and the predicted error amount; obtaining a target time period according to the matching degree of each time period except the last time period and the last time period; and obtaining an ARMA model with the optimal last time period according to the target time period, and carrying out marketing data processing according to the ARMA model with the optimal last time period.
2. The ARMA model-based marketing data processing method according to claim 1, wherein the time sequence segmentation is carried out to obtain a plurality of time periods, and the specific method comprises the following steps:
starting from the last time sequence data, recording each m time sequence data as a time period to obtain a plurality of time periods, wherein if the residual time sequence data is less than m, the actual time sequence data quantity is taken as a time period, and m is the preset quantity.
3. The method for processing the marketing data based on the ARMA model according to claim 1, wherein the method for obtaining the prediction error amount of the ARMA model corresponding to each time period according to the information amount criterion comprises the following specific steps:
processing time sequence data corresponding to a reference time period according to a red pool information amount criterion by taking any time period as the reference time period to obtain an optimal order p and an optimal order q, and constructing a corresponding ARMA model according to the time sequence data corresponding to the next time period of the reference time period and the optimal order p and the optimal order q of the reference time period to obtain a plurality of predicted values of the next time period of the reference time period; and carrying out difference between a plurality of predicted values of the next time period of the reference time period and corresponding time sequence data, taking absolute values, and recording the accumulated values of all the absolute values in the next time period of the reference time period as the predicted error amount of the corresponding ARMA model of the reference time period.
4. The method for processing marketing data based on the ARMA model according to claim 1, wherein the obtaining the matching degree between each time segment except the last time segment and the last time segment according to the optimal path between the distance matrix and the predicted error amount comprises the following specific steps:
for any time period except the last time period, acquiring a distance matrix and an optimal path of the time period and the last time period, and recording the numerical value of each data point in the distance matrix as a first data value, wherein P represents the matching degree of the time period and the last time period; n represents the number of best paths in the distance matrixNumber of points; l (L) t A Euclidean distance value representing the t data point on the optimal path in the distance matrix to an ideal diagonal in the distance matrix; d (D) t A first data value representing a t-th data point on the best path in the distance matrix; a represents the prediction error amount of the ARMA model corresponding to the time zone.
5. The method for processing the marketing data based on the ARMA model according to claim 1, wherein the target time period is obtained according to the matching degree of each time period except the last time period, comprising the following specific steps:
the time period with the smallest matching degree with the last time period is recorded as the target time period.
6. The ARMA model-based marketing data processing method according to claim 3, wherein the method for obtaining the ARMA model with the optimal last time period according to the target time period comprises the following specific steps:
taking the optimal orders p and q corresponding to the target time period as the new optimal orders p and q of the last time period, and obtaining the ARMA model with the optimal time period according to the new optimal orders p and q of the last time period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310871130.1A CN116843368B (en) | 2023-07-17 | 2023-07-17 | Marketing data processing method based on ARMA model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310871130.1A CN116843368B (en) | 2023-07-17 | 2023-07-17 | Marketing data processing method based on ARMA model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116843368A true CN116843368A (en) | 2023-10-03 |
| CN116843368B CN116843368B (en) | 2024-01-26 |
Family
ID=88174223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310871130.1A Active CN116843368B (en) | 2023-07-17 | 2023-07-17 | Marketing data processing method based on ARMA model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116843368B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105427360A (en) * | 2015-11-11 | 2016-03-23 | 华南理工大学 | Error-controllable CAGE sequence representation algorithm for dynamic grid |
| CN106411896A (en) * | 2016-09-30 | 2017-02-15 | 重庆邮电大学 | APDE-RBF neural network based network security situation prediction method |
| CN107346460A (en) * | 2017-07-18 | 2017-11-14 | 吉林大学 | Following operating mode Forecasting Methodology based on the lower front truck operation information of intelligent network contact system |
| CN109377752A (en) * | 2018-10-19 | 2019-02-22 | 桂林电子科技大学 | Short-term traffic flow variation prediction method, apparatus, computer equipment and storage medium |
| CN110163429A (en) * | 2019-05-10 | 2019-08-23 | 湖南大学 | A kind of short-term load forecasting method based on similar day optimal screening |
| CN111915061A (en) * | 2020-07-06 | 2020-11-10 | 同济大学 | Switch action current curve prediction method and fault discrimination method thereof |
| CN114202123A (en) * | 2021-12-14 | 2022-03-18 | 深圳壹账通智能科技有限公司 | Service data prediction method and device, electronic equipment and storage medium |
| CN115994137A (en) * | 2023-03-23 | 2023-04-21 | 无锡弘鼎软件科技有限公司 | Data management method based on application service system of Internet of things |
-
2023
- 2023-07-17 CN CN202310871130.1A patent/CN116843368B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105427360A (en) * | 2015-11-11 | 2016-03-23 | 华南理工大学 | Error-controllable CAGE sequence representation algorithm for dynamic grid |
| CN106411896A (en) * | 2016-09-30 | 2017-02-15 | 重庆邮电大学 | APDE-RBF neural network based network security situation prediction method |
| CN107346460A (en) * | 2017-07-18 | 2017-11-14 | 吉林大学 | Following operating mode Forecasting Methodology based on the lower front truck operation information of intelligent network contact system |
| CN109377752A (en) * | 2018-10-19 | 2019-02-22 | 桂林电子科技大学 | Short-term traffic flow variation prediction method, apparatus, computer equipment and storage medium |
| CN110163429A (en) * | 2019-05-10 | 2019-08-23 | 湖南大学 | A kind of short-term load forecasting method based on similar day optimal screening |
| CN111915061A (en) * | 2020-07-06 | 2020-11-10 | 同济大学 | Switch action current curve prediction method and fault discrimination method thereof |
| CN114202123A (en) * | 2021-12-14 | 2022-03-18 | 深圳壹账通智能科技有限公司 | Service data prediction method and device, electronic equipment and storage medium |
| CN115994137A (en) * | 2023-03-23 | 2023-04-21 | 无锡弘鼎软件科技有限公司 | Data management method based on application service system of Internet of things |
Non-Patent Citations (13)
| Title |
|---|
| J. KLEPSCH等: "Prediction of functional ARMA processes with an application to traffic data", 《ECONOMETRICS AND STATISTICS》, pages 128 - 149 * |
| SHUQI LI等: "Fractional frequency hybrid model based on EEMD for financial time series forecasting", 《COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION》, pages 1 - 14 * |
| Y. GRENIER: "Time-dependent ARMA modeling of nonstationary signals", 《IEEE TRANSACTIONS ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING》, vol. 31, no. 4, pages 899, XP055163657, DOI: 10.1109/TASSP.1983.1164152 * |
| 刘宜成等: "基于动态时间调整的时空图卷积路网交通流量预测", 《交通运输系统工程与信息》, vol. 22, no. 3, pages 147 - 152 * |
| 叶燕清;杨克巍;姜江;葛冰峰;豆亚杰;: "基于加权动态时间弯曲的多元时间序列相似性匹配方法", 模式识别与人工智能, no. 04 * |
| 叶燕清等: "基于加权动态时间弯曲的多元时间序列相似性匹配方法", 《模式识别与人工智能》, no. 04, pages 314 - 326 * |
| 司存友等: "基于相似性搜索的水文时间序列预测模型研究", 《信息技术》, no. 03, pages 19 - 24 * |
| 朱英伦等: "模式匹配在GDP时间序列预测中的应用", 《科技和产业》, vol. 16, no. 12, pages 80 - 82 * |
| 翁颖钧等: "基于分段线性动态时间弯曲的时间序列聚类算法研究", 《微型电脑应用》, no. 09, pages 19 - 23 * |
| 赵朋;王建伟;孙茂棚;周雅欣;: "高速公路突发事件恢复重建期交通量预测", 长安大学学报(自然科学版), no. 05 * |
| 邵翔;郭谋发;游林旭;: "基于改进DTW的接地故障波形互相关度聚类选线方法", 电力自动化设备, no. 11 * |
| 邵翔等: "基于改进DTW的接地故障波形互相关度聚类选线方法", 《电力自动化设备》, no. 11, pages 63 - 70 * |
| 郭敏;肖翔;蓝金辉;: "道路交通流短时预测方法综述", 自动化技术与应用, no. 06 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116843368B (en) | 2024-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2017143919A1 (en) | Method and apparatus for establishing data identification model | |
| CN108665120B (en) | Method and device for establishing scoring model and evaluating user credit | |
| CN109190442B (en) | Rapid face detection method based on deep cascade convolution neural network | |
| Wu et al. | A data mining approach for spatial modeling in small area load forecast | |
| WO2019165673A1 (en) | Reimbursement form risk prediction method, apparatus, terminal device, and storage medium | |
| WO2021238455A1 (en) | Data processing method and device, and computer-readable storage medium | |
| CN110648014B (en) | Regional wind power prediction method and system based on space-time quantile regression | |
| CN110444011B (en) | Traffic flow peak identification method and device, electronic equipment and storage medium | |
| CN111178623B (en) | Business process remaining time prediction method based on multilayer machine learning | |
| CN109727446A (en) | A kind of identification and processing method of electricity consumption data exceptional value | |
| CN109492748B (en) | Method for establishing medium-and-long-term load prediction model of power system based on convolutional neural network | |
| CN111008726B (en) | Class picture conversion method in power load prediction | |
| CN110263827A (en) | Abnormal transaction detection method and device based on transaction rule identification | |
| CN113569724B (en) | Road extraction method and system based on attention mechanism and dilation convolution | |
| CN111105241A (en) | Identification method for anti-fraud of credit card transaction | |
| CN104850868A (en) | Customer segmentation method based on k-means and neural network cluster | |
| CN108241872A (en) | The adaptive Prediction of Stock Index method of Hidden Markov Model based on the multiple features factor | |
| CN114048436A (en) | Construction method and construction device for forecasting enterprise financial data model | |
| CN116933946A (en) | Rail transit OD passenger flow prediction method and system based on passenger flow destination structure | |
| CN114912720A (en) | Memory network-based power load prediction method, device, terminal and storage medium | |
| CN111144462A (en) | Unknown individual identification method and device for radar signals | |
| CN110751400B (en) | Risk assessment method and device | |
| CN116843368B (en) | Marketing data processing method based on ARMA model | |
| CN115511012B (en) | Class soft label identification training method with maximum entropy constraint | |
| CN110032585B (en) | Time sequence double-layer symbolization 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 |