CN114627330A - Time sequence flow prediction method and device, storage medium and electronic equipment - Google Patents
Time sequence flow prediction method and device, storage medium and electronic equipmentDownload PDFInfo
- Publication number
- CN114627330A CN114627330ACN202210203124.4ACN202210203124ACN114627330ACN 114627330 ACN114627330 ACN 114627330ACN 202210203124 ACN202210203124 ACN 202210203124ACN 114627330 ACN114627330 ACN 114627330A
- Authority
- CN
- China
- Prior art keywords
- time series
- historical
- data
- heat map
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Translated fromChinese
Description
Translated fromChinese技术领域technical field
本发明涉及数据处理技术领域，特别涉及一种时序流量预测方法及装置、存储介质及电子设备。The present invention relates to the technical field of data processing, and in particular, to a method and device for predicting time series traffic, a storage medium and an electronic device.
背景技术Background technique
目前，随着互联网行业的高速发展，网络业务发展出了各种各样的形式，如即时通信、搜索引擎、社交娱乐、远程办公、在线交易和公共服务等，网络业务规模爆炸性增长，网络需求量也随之增长，然而网络资源是有限的，同一时间过多用户的访问点击必然会造成网络拥塞和服务质量降低，因此对用户的网络行为进行分析，通过对信息流进行预测，可以帮助企业对网络资源进行管理设计和规划，有效降低企业成本。近年来时序流量预测在行业内发展迅速。At present, with the rapid development of the Internet industry, various forms of network business have developed, such as instant messaging, search engines, social entertainment, telecommuting, online transactions and public services. However, network resources are limited, and too many users' access and clicks at the same time will inevitably cause network congestion and reduce service quality. Therefore, analyzing users' network behavior and predicting information flow can help enterprises. Manage, design and plan network resources to effectively reduce enterprise costs. Time series traffic forecasting has developed rapidly in the industry in recent years.
现有主流时序预测方法都是从时间序列本身入手，通常都是将时间序列作为预测模型的输入，基于时间序列的时域特征和频域特征预测出数据发展趋势，目前的时序预测方式在对并行时间序列预测时，难以将各时间序列之间的信息进行关联，降低了对并行时间序列预测的准确性。The existing mainstream time series forecasting methods all start with the time series itself, usually taking the time series as the input of the forecasting model, and predicting the data development trend based on the time domain and frequency domain characteristics of the time series. In parallel time series forecasting, it is difficult to correlate the information between each time series, which reduces the accuracy of parallel time series forecasting.
发明内容SUMMARY OF THE INVENTION
有鉴于此，本发明提供一种时序流量预测方法及装置、存储介质及电子设备，通过引入包含了高维度的特征的图像特征数据，将各时间序列之间的信息进行关联，提高对时间序列预测的准确性。In view of this, the present invention provides a time series traffic prediction method and device, a storage medium and an electronic device. By introducing image feature data including high-dimensional features, the information between each time series is correlated, and the time series is improved. Prediction accuracy.
为实现上述目的，本发明实施例提供如下技术方案：To achieve the above purpose, the embodiments of the present invention provide the following technical solutions:
本发明第一方面公开一种时序流量预测方法，包括：A first aspect of the present invention discloses a time series traffic prediction method, comprising:
获取各个历史业务时间序列；Obtain each historical business time series;
对各个所述历史业务时间序列进行处理，得到历史时序热力图；Process each of the historical business time series to obtain a historical time series heat map;
提取所述历史时序热力图中的图像特征数据；extracting image feature data in the historical time series heatmap;
对所述图像特征数据进行处理，得到每个所述历史业务时间序列的预测时序流量。The image feature data is processed to obtain the predicted time series traffic of each of the historical business time series.
上述的方法，可选的，所述对各个所述历史业务时间序列进行处理，得到历史时序热力图，包括：In the above method, optionally, the processing of each of the historical business time series to obtain a historical time series heatmap includes:
基于各个所述历史业务时间序列，构建数据矩阵；Building a data matrix based on each of the historical business time series;
将所述数据矩阵中的每个数值进行归一化处理，得到历史时序热力图。Normalize each value in the data matrix to obtain a historical time series heat map.
上述的方法，可选的，所述提取所述历史时序热力图中的图像特征数据，包括：In the above method, optionally, the extracting image feature data in the historical time series heatmap includes:
将所述历史时序热力图输入预先训练完成的特征提取模型中，使得所述特征提取模型从所述历史时序热力图中提取出各项图像高维度特征；Inputting the historical time series heat map into a pre-trained feature extraction model, so that the feature extraction model extracts various image high-dimensional features from the historical time series heat map;
将各项所述图像高维度特征作为所述历史时序热力图像的图像特征数据。Each item of the image high-dimensional features is used as the image feature data of the historical time series thermal image.
上述的方法，可选的，所述对所述图像特征数据进行处理，得到每个所述历史业务时间序列的预测时序流量，包括：In the above method, optionally, the processing of the image feature data to obtain the predicted time series traffic of each of the historical business time series includes:
将所述图像特征数据输入预先训练完成的分类神经网络中，使得所述分类神经网络输出所述历史时序热力图的图片分类数据，其中，所述图片分类数据中包含各项所述历史业务时间序列的时序流量的预测信息；Input the image feature data into the pre-trained classification neural network, so that the classification neural network outputs the image classification data of the historical time series heat map, wherein the image classification data includes the historical business time of each item Prediction information of the time series flow of the sequence;
调用预设的回归函数对所述图像分类数据进行回归处理，得到与所述图像分类数据对应的回归数据；Calling a preset regression function to perform regression processing on the image classification data to obtain regression data corresponding to the image classification data;
对所述回归数据进行逆归一化处理，得每个所述历史业务时间序列的预测时序流量。Perform inverse normalization processing on the regression data to obtain the predicted time series traffic of each of the historical business time series.
上述的方法，可选的，还包括：The above method, optionally, further includes:
基于每个所述历史业务时间序列的预测时序流量进行风险评估，以得到每个所述历史业务时间序列所对应的业务的风险评分。Risk assessment is performed based on the predicted time series traffic of each of the historical service time series, so as to obtain a risk score of the service corresponding to each of the historical service time series.
本发明第二方面公开一种时序流量预测装置，包括：A second aspect of the present invention discloses a time-series traffic prediction device, comprising:
获取单元，用于获取各个历史业务时间序列；The acquisition unit is used to acquire each historical business time series;
第一处理单元，用于对各个所述历史业务时间序列进行处理，得到历史时序热力图；a first processing unit, configured to process each of the historical business time series to obtain a historical time series heat map;
提取单元，用于提取所述历史时序热力图中的图像特征数据；an extraction unit for extracting image feature data in the historical time series heatmap;
第二处理单元，用于对所述图像特征数据进行处理，得到每个所述历史业务时间序列的预测时序流量。The second processing unit is configured to process the image feature data to obtain the predicted time series traffic of each of the historical business time series.
上述的装置，可选的，所述第一处理单元，包括：In the above-mentioned device, optionally, the first processing unit includes:
构建子单元，用于基于各个所述历史业务时间序列，构建数据矩阵；constructing a subunit for constructing a data matrix based on each of the historical business time series;
归一化处理子单元，用于将所述数据矩阵中的每个数值进行归一化处理，得到历史时序热力图。The normalization processing subunit is used for normalizing each value in the data matrix to obtain a historical time series heat map.
上述的装置，可选的，所述提取单元，包括：The above-mentioned device, optionally, the extraction unit includes:
输入子单元，用于将所述历史时序热力图输入预先训练完成的特征提取模型中，使得所述特征提取模型从所述历史时序热力图中提取出各项图像高维度特征；The input subunit is used to input the historical time series heat map into the pre-trained feature extraction model, so that the feature extraction model extracts various image high-dimensional features from the historical time series heat map;
确定子单元，用于将各项所述图像高维度特征作为所述历史时序热力图像的图像特征数据。A determination subunit, configured to use each item of the high-dimensional image features as image feature data of the historical time series thermal image.
上述的装置，可选的，所述第二处理单元，包括：For the above device, optionally, the second processing unit includes:
输出子单元，用于将所述图像特征数据输入预先训练完成的分类神经网络中，使得所述分类神经网络输出所述历史时序热力图的图片分类数据，其中，所述图片分类数据中包含各项所述历史业务时间序列的时序流量的预测信息；The output subunit is used to input the image feature data into the pre-trained classification neural network, so that the classification neural network outputs the image classification data of the historical time series heat map, wherein the image classification data includes each Prediction information of the time series traffic of the historical business time series described in item;
调用子单元，用于调用预设的回归函数对所述图像分类数据进行回归处理，得到与所述图像分类数据对应的回归数据；a calling subunit for calling a preset regression function to perform regression processing on the image classification data to obtain regression data corresponding to the image classification data;
逆归一化处理子单元，用于对所述回归数据进行逆归一化处理，得每个所述历史业务时间序列的预测时序流量。The inverse normalization processing subunit is configured to perform inverse normalization processing on the regression data to obtain the predicted time series traffic of each of the historical business time series.
上述的装置，可选的，还包括：The above-mentioned device, optionally, also includes:
风险评估单元，用于基于每个所述历史业务时间序列的预测时序流量进行风险评估，以得到每个所述历史业务时间序列所对应的业务的风险评分。A risk assessment unit, configured to perform risk assessment based on the predicted time series traffic of each of the historical service time series, so as to obtain a risk score of the service corresponding to each of the historical service time series.
本发明第三方面公开一种存储介质，所述存储介质包括存储的指令，其中，在所述指令运行时控制所述存储介质所在的设备执行上述的时序流量预测方法。A third aspect of the present invention discloses a storage medium, the storage medium includes stored instructions, wherein when the instructions are executed, a device where the storage medium is located is controlled to execute the above-mentioned method for predicting time series traffic.
本发明第四方面公开一种电子设备，包括存储器，以及一个或者一个以上的指令，其中一个或者一个以上指令存储于存储器中，且经配置以由一个或者一个以上处理器执行如上所述的时序流量预测方法。A fourth aspect of the present invention discloses an electronic device comprising a memory, and one or more instructions, wherein the one or more instructions are stored in the memory and configured to be executed by one or more processors in a sequence as described above Traffic forecast method.
本发明提供一种时序流量预测方法及装置、存储介质及电子设备，该方法包括：获取各个历史业务时间序列，对各个历史业务时间序列进行处理，得到历史时序热力图；提取历史时序热力图中的图像特征数据；对图像特征数据进行处理，得到每个历史业务时间序列的预测时序流量。通过将各个历史业务时间序列转换成历史时序热力图，并从历史时序热力图中提取到图像特征数据，该图像特征数据包含了时间序列各种高维度的特征，含盖了时间序列全局及局部的相关特征，通过对图像特征数据进行处理，可以得到各个历史业务时间序列的预测时序流量，引入涵盖了高维度的特征的图像特征数据，有效的提高了对时间序列的流量进行预测的准确性。The present invention provides a time series traffic prediction method and device, storage medium and electronic equipment. The method includes: acquiring each historical business time series, processing each historical business time series, and obtaining a historical time series heat map; extracting the historical time series heat map image feature data; process the image feature data to obtain the predicted time series traffic of each historical business time series. By converting each historical business time series into a historical time series heat map, and extracting image feature data from the historical time series heat map, the image feature data contains various high-dimensional features of the time series, covering the global and local time series. By processing the image feature data, the predicted time series traffic of each historical business time series can be obtained, and the image feature data covering high-dimensional features can be introduced, which effectively improves the accuracy of the time series traffic prediction. .
附图说明Description of drawings
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据提供的附图获得其他的附图。In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the provided drawings without creative work.
图1为本发明实施例提供的一种时序流量预测方法的方法流程图；FIG. 1 is a method flowchart of a time series traffic prediction method provided by an embodiment of the present invention;
图2为本发明实施例提供的一种历史时序热力图的示例图；FIG. 2 is an example diagram of a historical time series heat map provided by an embodiment of the present invention;
图3为本发明实施例提供的提取图像特征数据的一种方法流程图；3 is a flowchart of a method for extracting image feature data provided by an embodiment of the present invention;
图4为本发明实施例提供的获得历史业务时间序列的预测时序流量的一种方法流程图；FIG. 4 is a flowchart of a method for obtaining a predicted time sequence traffic of a historical service time series provided by an embodiment of the present invention;
图5为本发明实施例提供的一种时序流量预测方法的又一方法流程图；Fig. 5 is another method flow chart of a time series traffic prediction method provided by an embodiment of the present invention;
图6为本发明实施例提供的一种时序流量预测装置的结构示意图；FIG. 6 is a schematic structural diagram of an apparatus for predicting time-series traffic according to an embodiment of the present invention;
图7为本发明实施例提供的一种电子设备的结构示意图。FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
在本申请中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。In this application, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device comprising a list of elements includes not only those elements, but also no Other elements expressly listed, or which are also inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.
目前，工业界对于时序流量的应用需求在增加，尤其是多业务场景的时序数据并行预测。多个时间序列之间，保持有一定相关性，同时由彼此独立演进，每个序列即存在独立特性，又存在一致性。由于多业务场景的序列预测存在以上特点，因此对预测的算法的特征处理及模型结构，有一定要求。At present, the application requirements of time series traffic in the industry are increasing, especially the parallel prediction of time series data in multi-service scenarios. There is a certain correlation between multiple time series, and at the same time they evolve independently of each other. Each series has independent characteristics and consistency. Due to the above characteristics of sequence prediction in multi-service scenarios, there are certain requirements for the feature processing and model structure of the prediction algorithm.
现有的主流时序预测方法都是从时间序列本身入手，直接将时间序列作为模型训练的输入，特征的提取是以时域或者频域为特征，同时辅助以外部影响因子特征，采用卷积神经网络及线型机器学习回归模型进行拟合。现有的模型所应用的算法有时序算法、机器学习类算法、神经网络类算法等。时序算法类如hotwinter、ARIMA、MA、AR等算法，这类算法适用于规律简单、时序易拆解、外部影响因子较少的场景，通常用于不易受环境影响的数据预测中。机器学习类算法有gbm、xgboost，gbrt等，这类算法通常是通过随机森林回归加梯度提升的方法构造多个回归树，通过权重分配来达到最终的预测目标，适用于外部特征较多的场景，方便根据特征贡献程度选择合适的特征，剪裁模型。神经网络类模型最典型的就是lstm模型，基于长短期的时间记忆单元，可以对流量进行准确拟合，但是神经网络类模型，需要大量的数据做训练，而现实场景中的流量数据可能没有那么多，会对训练产生影响，使得预测结果不准确。The existing mainstream time series prediction methods all start from the time series itself, and directly use the time series as the input of model training. Network and linear machine learning regression models are fitted. The algorithms used in the existing models include time series algorithms, machine learning algorithms, and neural network algorithms. Time series algorithms such as hotwinter, ARIMA, MA, AR and other algorithms are suitable for scenarios with simple rules, easy disassembly of time series, and few external influence factors, and are usually used in data prediction that is not easily affected by the environment. Machine learning algorithms include gbm, xgboost, gbrt, etc. These algorithms usually construct multiple regression trees through random forest regression and gradient boosting, and achieve the final prediction goal through weight distribution, which is suitable for scenes with many external features , it is convenient to select appropriate features according to the degree of feature contribution and tailor the model. The most typical neural network model is the lstm model, which can accurately fit traffic based on long-term and short-term time memory units. However, neural network models require a lot of data for training, and the traffic data in real scenarios may not be so If it is too large, it will affect the training and make the prediction results inaccurate.
目前应用的模型的结构可以分为三种；单序列独立预测、层级递归预测、并行预测。单序列独立预测即每个序列单独预测，所用的特征仅来自于该序列本身，而与其他序列的信息无关，显然这种方式无法利用序列间的相关信息；层级递归预测，利用前一个序列的结果，去做后一个序列的输入特征，这种方法能够利用到其他序列的信息，但是特征获取方式单一，在多个序列存在复杂相关性的场景，信息利用不足；并行预测，在模型构建时，将所有的时序数据引入同一个隐层，输出为多输出，同时对多个序列进行特征的提取与预测，这种方式特征提取能力强，但是仍然利用的是时频域序列特征，容易倾向于广义特征或者局部特征，序列的高维度相关特性不容易被提取。The structure of the currently applied model can be divided into three types; single-sequence independent prediction, hierarchical recursive prediction, and parallel prediction. Single-sequence independent prediction means that each sequence is predicted independently, and the features used are only from the sequence itself, and have nothing to do with the information of other sequences. Obviously, this method cannot use the relevant information between sequences; hierarchical recursive prediction, using the previous sequence As a result, to do the input features of the latter sequence, this method can use the information of other sequences, but the feature acquisition method is single, and in the scene where multiple sequences have complex correlations, the information is insufficiently utilized; parallel prediction, when the model is constructed. , introduce all time series data into the same hidden layer, output multiple outputs, and perform feature extraction and prediction on multiple sequences at the same time. This method has strong feature extraction ability, but still uses time-frequency domain sequence features, which is easy to tend to For generalized features or local features, the high-dimensional correlation features of sequences are not easy to extract.
传统预测时序时，通常使用时间序列的时域特征或是频域特征，而在对并行时间序列进行预测时，使用时域特征或是频域特征难以将各时间序列之间的信息进行关联，降低了对并行时间序列预测的准确性。When traditionally predicting time series, time-domain features or frequency-domain features of time series are usually used. When predicting parallel time-series, it is difficult to correlate information between time series using time-domain features or frequency-domain features. Reduced accuracy of forecasting for parallel time series.
基于上述的问题，本发明提供一种时序流量预测方法，通过将时间序列处理成时序热力图，并从时序热力图中提取图像特征数据，使用图像特征数据预测各个时序序列的时序流量；图像特征数据中包含各时间序列的全局和局部的并行相关特征，通过使用图像特征数据预测个时间序列的时间流量，有效的提高了预测的准确性。Based on the above problems, the present invention provides a method for predicting time series traffic. By processing the time series into a time series heat map, and extracting image feature data from the time series heat map, the image feature data is used to predict the time series traffic of each time series sequence; The data contains the global and local parallel correlation features of each time series. By using the image feature data to predict the time flow of each time series, the accuracy of the prediction is effectively improved.
本发明可用于众多通用或专用的计算装置环境或配置中。例如：个人计算机、服务器计算机、手持设备或便携式设备、平板型设备、多处理器装置、包括以上任何装置或设备的分布式计算环境等等。The present invention may be used in numerous general purpose or special purpose computing device environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet-type devices, multi-processor devices, distributed computing environments including any of the above, and the like.
参照图1，为本发明实施例提供的一种时序流量预测方法的方法流程图，具体说明如下所述：Referring to FIG. 1, it is a method flow chart of a time series traffic prediction method provided by an embodiment of the present invention, and the specific description is as follows:
S101、获取各个历史业务时间序列。S101. Acquire each historical business time series.
在获取各个历史业务时间序列时，可以从时间序列数据库中获取，进一步的，可以对用户发送的预测指令进行解析，基于预测指令中的序列信息，在时间序列数据库中获取各个历史业务时间序列。When obtaining each historical business time series, it can be obtained from the time series database, and further, the forecast command sent by the user can be parsed, and each historical business time series can be obtained from the time series database based on the sequence information in the forecast command.
优选的，各个历史业务时间序列可以为并行的时间序列，每个历史业务时间序列可以为用户在办理业务时生成的时间序列。历史业务时间序列可以为不同业务的时间序列。Preferably, each historical business time series may be a parallel time series, and each historical business time series may be a time series generated by a user when handling a business. The historical business time series can be the time series of different businesses.
S102、对各个历史业务时间序列进行处理，得到历史时序热力图。S102. Process each historical business time series to obtain a historical time series heat map.
将各个历史业务时间序列转换成历史时序热力图，其中，历史时序热力图中包含了多项图像高维度特征。Convert each historical business time series into a historical time series heat map, in which the historical time series heat map contains a number of high-dimensional image features.
对各个历史业务时间序列进行处理，得到历史时序热力图的具体过程如下所述：The specific process of processing each historical business time series to obtain the historical time series heatmap is as follows:
基于各个历史业务时间序列，构建数据矩阵；Build a data matrix based on each historical business time series;
将数据矩阵中的每个数值进行归一化处理，得到历史时序热力图。Normalize each value in the data matrix to obtain a historical time series heatmap.
需要说明的是，在构建数据矩阵时，基于预设的历史数据窗口对各个历史业务时间序列进行截取，由此可以得到每个历史业务时间序列的截取时间序列，使用各个截取时间序列组成数据矩阵；进一步的，在对历史业务时间序列进行截取时，可以从历史业务时间序列的任意一点开始截取，也可以根据实际需求进行截取，例如从历史业务时间序列的开端开始截取。在历史业务时间序列的数据长度短于历史数据窗口的窗口长度时，可以及对历史时间序列进行补零操作，以便得到该历史时间序列的截取时间序列。It should be noted that, when constructing a data matrix, each historical business time series is intercepted based on a preset historical data window, so that the intercepted time series of each historical business time series can be obtained, and each intercepted time series is used to form a data matrix. ; Further, when intercepting the historical business time series, interception can be started from any point of the historical business time series, or interception can be performed according to actual needs, such as intercepting from the beginning of the historical business time series. When the data length of the historical business time series is shorter than the window length of the historical data window, a zero-padding operation can be performed on the historical time series, so as to obtain the intercepted time series of the historical time series.
在得到数据矩阵后，将数据矩阵中的每个数值进行归一化处理，进一步的，在将数据矩阵中的每个数值进行归一化处理后，可以将每个归一化后的数据乘以65536后转RGB，从而得到历史时序热力图，示例性的，参照图2，为本发明实施例提供的一种历史时序热力图的示例图，图2中时间5为待预测的时间点，序列1至序列4为构成历史序列热力图的历史时间序列，时间1至时间4为历史时间序列构建历史序列热力图的时间点。After the data matrix is obtained, each value in the data matrix is normalized. Further, after each value in the data matrix is normalized, each normalized data can be multiplied by 65536 is converted to RGB to obtain a historical time series heat map. For example, referring to FIG. 2 , it is an example diagram of a historical time series heat map provided by an embodiment of the present invention. Time 5 in FIG. 2 is the time point to be predicted. Sequences 1 to 4 are the historical time series that constitute the historical sequence heatmap, and time 1 to time4 are the time points at which the historical sequence heatmap is constructed.
本发明实施例提供的方法中，将各个历史时间序列转换成历史序列热力图，可以将各个历史时间序列进行关联，其中，历史序列热力图中包含了各个历史时间序列的高维度的信息，将各个历史时间序列转换成历史序列热力图后，便于后续从历史序列热力图中获取各个历史时间序列的高维度的信息。In the method provided by the embodiment of the present invention, each historical time series is converted into a historical sequence heat map, and each historical time series can be correlated, wherein the historical sequence heat map contains high-dimensional information of each historical time series, and the After each historical time series is converted into a historical series heat map, it is convenient to obtain high-dimensional information of each historical time series from the historical series heat map.
S103、提取历史时序热力图中的图像特征数据。S103, extract the image feature data in the historical time series heat map.
本发明实施例提供的方法中，在得到历史时序热力图后，需要从历史时序热力图中提取特向特征数据，参照图3，为本发明实施例提供的提取图像特征数据的流程示例图，具体说明如下所述：In the method provided by the embodiment of the present invention, after obtaining the historical time series heat map, it is necessary to extract characteristic feature data from the historical time series heat map. Referring to FIG. 3 , it is an exemplary flowchart for extracting image feature data provided by the embodiment of the present invention. The specific instructions are as follows:
S301、将历史时序热力图输入预先训练完成的特征提取模型中，使得特征提取模型从所述历史时序热力图中提取出各项图像高维度特征。S301. Input the historical time series heat map into a pre-trained feature extraction model, so that the feature extraction model extracts various image high-dimensional features from the historical time series heat map.
需要说明的是，特征提取模型使用可提取特征的神经网络构成，具体如深度残差网络、离散Hopfield网络等等。特征提取模型在投入使用之前先进行训练，在对特征提取模型训练完成后，将特征提取模型投入使用。It should be noted that the feature extraction model is composed of a neural network that can extract features, such as a deep residual network, a discrete Hopfield network, and the like. The feature extraction model is trained before being put into use, and after the feature extraction model is trained, the feature extraction model is put into use.
特征提取模型从历史时序热力图中提取出各项图像高维度特征，示例性的，图像高维度特征包含但不限于图像的颜色特征、纹理特征、形状特征、空间特征等等。不同的特征表征了时间序列的流量的不同特性，具体如，颜色表征流量的整体走向；纹理特征表征流量间的差异特性；形状特征表征流量间的局部相关性、相似性；空间特征表征并行多条流量在周期上的重叠与重复的特性。The feature extraction model extracts various high-dimensional features of the image from the historical time series heatmap. Exemplarily, the high-dimensional features of the image include but are not limited to color features, texture features, shape features, spatial features, and the like of the image. Different features characterize the different characteristics of the flow in time series, for example, color indicates the overall trend of flow; texture feature indicates the difference between flows; shape feature indicates local correlation and similarity between flows; The overlapping and repeating characteristics of the strip flow in the period.
S302、将各项图像高维度特征作为历史时序热力图像的图像特征数据。S302. Use the high-dimensional features of each image as the image feature data of the historical time series thermal image.
本发明实施例提供的方法中，使用训练完成的特征提取模型对历史时序热力图进行处理，进而可以从历史时序热力图中提取出表征了时间序列的不同特性的图像高维度特征，将各项图像高维度特征确定为图像特征数据，由此，图像特征数据包含了各项历史业务时间序列在不同方面的特性，有效的将历史业务时间序列的特征相互关联，使得各个历史业务时间序列之间的关系更加的紧密。In the method provided by the embodiment of the present invention, the trained feature extraction model is used to process the historical time series heat map, and then high-dimensional image features representing different characteristics of the time series can be extracted from the historical time series heat map, and each Image high-dimensional features are determined as image feature data. Therefore, image feature data includes the characteristics of various historical business time series in different aspects, effectively correlating the characteristics of historical business time series with each other, so that the relationship between each historical business time series relationship is closer.
S104、对图像特征数据进行处理，得到每个历史业务时间序列的预测时序流量。S104. Process the image feature data to obtain the predicted time series traffic of each historical service time series.
示例性的，参照图4，为本发明实施例提供的对图像特征数据进行处理，得到每个历史业务时间序列的预测时序流量的方法流程图，具体说明如下所述：Exemplarily, referring to FIG. 4 , it is a flowchart of a method for processing image feature data to obtain the predicted time sequence traffic of each historical service time series provided by an embodiment of the present invention, and the specific description is as follows:
S401、将图像特征数据输入预先训练完成的分类神经网络中，使得分类神经网络输出历史时序热力图的图片分类数据，其中，所述图片分类数据中包含各项历史业务时间序列的时序流量的预测信息。S401. Input the image feature data into the pre-trained classification neural network, so that the classification neural network outputs the image classification data of the historical time series heat map, wherein the image classification data includes the prediction of the time series traffic of each historical business time series information.
分类神经网络可以为BP神经网络、CNN神经网络，还可以是使用稠密连接机制构建的多分类稠密卷积神经网络，其中，稠密连接机制能够有效的缓解梯度问题，加强特征传播。优选的，卷积神经网络使用稠密连接机制进行构建时，有效的缓解梯度问题，加强特征传播，鼓励特征服用以极大地减少了参数数量，降低了网络训练过程中对训练样本的需求；进一步的，本发明在应用的过程中，分类数不得低于历史业务时间序列去重后的数值量。The classification neural network can be a BP neural network, a CNN neural network, or a multi-class dense convolutional neural network constructed using a dense connection mechanism. The dense connection mechanism can effectively alleviate the gradient problem and enhance feature propagation. Preferably, when the convolutional neural network is constructed using a dense connection mechanism, the gradient problem can be effectively alleviated, feature propagation is enhanced, and feature taking is encouraged to greatly reduce the number of parameters and reduce the demand for training samples during network training; further , in the process of application of the present invention, the number of classifications shall not be lower than the value of the historical business time series after deduplication.
使用分类神经网络对图像特征数据进行处理，输出图片分类数据，示例性的，输出的图片分类数据可以为由数字组成的字符串，进一步的，图片分类数据为离散数据。The image feature data is processed using a classification neural network, and the image classification data is output. Exemplarily, the output image classification data may be a string composed of numbers. Further, the image classification data is discrete data.
需要说明的是，使用分类神经网络对图像特征数据进行处理的过程，实质上也是将历史时序热力图进行分类的过程，通过将历史时序热力图进行分类，可以预测出构建历史时序热力图的各个历史业务时间序列的时序流量。It should be noted that the process of using the classification neural network to process the image feature data is also the process of classifying the historical time series heat map. Time series traffic of historical business time series.
S402、调用预设的回归函数对图像分类数据进行回归处理，得到与图像分类数据对应的回归数据。S402, calling a preset regression function to perform regression processing on the image classification data to obtain regression data corresponding to the image classification data.
优选的，回归函数可以为线性激活函数，使用线性激活函数对图像分类数据进行回归处理，以得到回归数据，需要说明的是，回归数据为线型连续数据。Preferably, the regression function may be a linear activation function, and the linear activation function is used to perform regression processing on the image classification data to obtain regression data. It should be noted that the regression data is linear continuous data.
S403、对回归数据进行逆归一化处理，得每个历史业务时间序列的预测时序流量。S403. Perform inverse normalization processing on the regression data to obtain the predicted time series traffic of each historical business time series.
本发明实施例提供的方法中，使用分类神经网络对图像特征数据进行处理，可以进行高精度的拟合，从而可以得到精确度很高的预测数据，可以提高对时间蓄力的预测结果。In the method provided by the embodiment of the present invention, the classification neural network is used to process the image feature data, so that high-precision fitting can be performed, so that prediction data with high accuracy can be obtained, and the prediction result of time accumulation can be improved.
优选的，在得到每个历史业务时间序列的预测时序流量后，可以使用预设的风险评估机制基于每个历史业务时间序列的预测时序流量进行风险评估操作，从而得到每个历史业务时间序列所对应的业务的风险评分，需要说明的是，风险评分为业务在预测时序流量所对应的时间点的评分，该风险评分可以用于表征业务在预测时序流量所对应的时间点的风险度，工作人员可以根据预测时间流量和风险评分安排该时间点的业务办理工作，以及对设备的维护工作等。对业务进行风险评估可以有利于工作人员安排各种工作，以便于规避掉各种风险，为客户提供良好的业务办理环境。Preferably, after obtaining the predicted time series traffic of each historical business time series, a risk assessment operation may be performed based on the predicted time series traffic of each historical business time series by using a preset risk assessment mechanism, so as to obtain the information of each historical business time series. The risk score of the corresponding business. It should be noted that the risk score is the score of the business at the time point corresponding to the predicted time series traffic. Personnel can arrange business processing work at this time point and equipment maintenance work according to the predicted time flow and risk score. Risk assessment of business can help staff to arrange various jobs, so as to avoid various risks and provide customers with a good business handling environment.
本发明实施例提供的方法中，获取各个历史业务时间序列，对各个历史业务时间序列进行处理，得到历史时序热力图；提取历史时序热力图中的图像特征数据；对图像特征数据进行处理，得到每个历史业务时间序列的预测时序流量。通过将各个历史业务时间序列转换成历史时序热力图，并从历史时序热力图中提取到图像特征数据，该图像特征数据包含了时间序列各种高维度的特征，含盖了时间序列全局及局部的相关特征，通过对图像特征数据进行处理，可以得到各个历史业务时间序列的预测时序流量，引入涵盖了高维度的特征的图像特征数据，有效的提高了对时间序列的流量进行预测的准确性。In the method provided by the embodiment of the present invention, each historical business time series is obtained, and each historical business time series is processed to obtain a historical time series heat map; the image feature data in the historical time series heat map is extracted; the image feature data is processed to obtain Forecast time series traffic for each historical business time series. By converting each historical business time series into a historical time series heat map, and extracting image feature data from the historical time series heat map, the image feature data contains various high-dimensional features of the time series, covering the global and local time series. By processing the image feature data, the predicted time series traffic of each historical business time series can be obtained, and the image feature data covering high-dimensional features can be introduced, which effectively improves the accuracy of the time series traffic prediction. .
参照图5，为本发明实施例提供的时序流量预测方法的其中一种应用示例图，具体说明如下所述：Referring to FIG. 5 , it is a diagram of an example application of the time series traffic prediction method provided by the embodiment of the present invention, and the specific description is as follows:
从历史数据中获取M条时间序列，其中，时间序列可以理解为上文中的历史业务时间序列；需要说明的是，每个时间序列的数据长度为N，对各个时间序列进行归一化处理，得到归一化的数据矩阵，进一步的，该数据矩阵还可以称为M*N的时序矩阵。对数据举证进行图像化处理，具体的，将数据矩阵中的每个数值乘以65536，并转为RGB，从而可以得到历史时序热力图，在对历史时序热力图进行特征处理时，可以使用深度残差网络对历史时序热力图进行处理，从而得到图像特征数据，基于图像特征数据进行分类处理，可以使用多分类稠密层进行分类处理，从而实现精细化分类，使用线性计划函数对分类后得到的数据进行回归处理，以便将离散分类的数据回归为线型连续数据；将进行回归处理得到的数据进行逆归一化处理，从而得到包含了每项时间序列的预测时序流量的预测数据。Obtain M time series from historical data, where the time series can be understood as the historical business time series above; it should be noted that the data length of each time series is N, and each time series is normalized. A normalized data matrix is obtained, and further, the data matrix may also be called an M*N time series matrix. Image processing of data evidence. Specifically, multiply each value in the data matrix by 65536 and convert it to RGB, so that the historical time series heat map can be obtained. When performing feature processing on the historical time series heat map, depth can be used. The residual network processes the historical time series heat map to obtain image feature data, and performs classification processing based on the image feature data. Multi-class dense layers can be used for classification processing to achieve refined classification, and the linear planning function is used to classify the obtained results. Perform regression processing on the data to regress the discrete classified data into linear continuous data; perform inverse normalization processing on the data obtained from the regression processing to obtain forecast data including the forecasted time series flow of each time series.
需要说明的是，将历史业务时间序列转换为历史时序热力图的过程中，将各个时间序列的值并行进行排列，并将各个值像素化，即根据每个序列不同时间点上的值对应图像上的每个像素，由此即可得到历史时序热力图，具体如图2所示，图2的历史时序热力图由多个图片方块组成，进一步的，每个图片方块存在对应的时间序列和时间点。It should be noted that in the process of converting historical business time series into historical time series heatmaps, the values of each time series are arranged in parallel, and each value is pixelized, that is, images corresponding to the values at different time points of each series For each pixel on , the historical time series heat map can be obtained, as shown in Figure 2. The historical time series heat map in Figure 2 is composed of multiple picture blocks. Further, each picture block has a corresponding time series and point in time.
使用残差神经网络从历史时序热力图中提取特征，可以从各个图片方块中提取整体特征，从历史时序热力图的像素分布中提取局部特征，从而可以得到包含了整体特征和局部特征的图像特征数据，进一步的，图像特征数据中包含但不限于历史时序热力图的颜色特征、纹理特征、形状特征以及空间特征等，具体的，颜色特征如直方图、颜色分布等全局特征，能够表征图像的区域的表面性质；纹理特征如灰度共生矩阵，作为全局特征，能够很好的抵抗噪声的影响；形状特征如轮廓、区域可以描述图像信息的局部特质；空间特征如区域的重叠、方位，能够区分不同区域的流动情况。进一步的，使用残差神经网络从历史时序热力图中提取特征时，可以通过从输入直接引入一个短连接到非线性层的输出上，以实现更好的拟合分类函数，获得更高的分类精度。Using the residual neural network to extract features from the historical time series heat map, the overall features can be extracted from each picture block, and the local features can be extracted from the pixel distribution of the historical time series heat map, so that the image features containing the global features and local features can be obtained. Data, further, the image feature data includes but is not limited to the color features, texture features, shape features and spatial features of the historical time series heatmap. Specifically, the color features such as histogram, color distribution and other global features can characterize the The surface properties of the region; texture features such as gray level co-occurrence matrix, as global features, can well resist the influence of noise; shape features such as contours and regions can describe the local characteristics of image information; spatial features such as region overlap and orientation can Distinguish flows in different regions. Further, when using the residual neural network to extract features from the historical time series heatmap, a short connection can be directly introduced from the input to the output of the nonlinear layer to achieve a better fit classification function and obtain higher classification. precision.
使用稠密连接机制构建的多分类稠密卷积神经网络对图像特征数据进行处理，以便得到包含每个时间序列的时序流量的预测信息的图片分类数据，多分类稠密卷积神经网络可以进行精细化分类，以便得到更加精确的预测结果。在得到图片分类数据后，使用线性激活函数对图片分类数据进行回归处理，从而将离散分类的图片分类数据回归为线型连续的回归数据，对回归数据进行逆归一化处理，从而得到每项时间序列的预测时序流量的预测数据。The multi-class dense convolutional neural network constructed using the dense connection mechanism processes the image feature data to obtain the image classification data containing the prediction information of the time series traffic of each time series. The multi-class dense convolutional neural network can perform refined classification. , in order to obtain more accurate prediction results. After obtaining the image classification data, use the linear activation function to perform regression processing on the image classification data, so as to regress the discretely classified image classification data into linear continuous regression data, and perform inverse normalization processing on the regression data to obtain each item. Forecast data for time series traffic for time series.
本发明通过将时间序列转换成热力图像，再从热力图像中进行高维度的相关特征提取，从而涵盖了时间序列的局部及全局的特征，并对未来时间点数据进行高精度的拟合预测，提高了对时间序列的时序流量的预测准确性；除此之外，还可以应用于多种场景进行预测，扩大了场景的适用性。By converting the time series into a thermal image, and extracting high-dimensional related features from the thermal image, the invention covers the local and global features of the time series, and performs high-precision fitting prediction on future time point data. The prediction accuracy of time series traffic is improved; in addition, it can also be applied to various scenarios for prediction, which expands the applicability of scenarios.
与图1相对应的，本发明实施例还提供一种时序流量预测装置，用于支持图1所示的方法在实际生活中的应用，该装置可设置于智能计算终端或是分布式计算终端中。参照图6，为本发明实施例提供的时序流量预测装置的结构示意图，具体说明如下所述：Corresponding to FIG. 1 , an embodiment of the present invention also provides a time series traffic prediction device, which is used to support the application of the method shown in FIG. 1 in real life. The device can be set in an intelligent computing terminal or a distributed computing terminal. middle. Referring to FIG. 6 , it is a schematic structural diagram of an apparatus for timing traffic prediction provided by an embodiment of the present invention, and the specific description is as follows:
获取单元601，用于获取各个历史业务时间序列；an obtaining
第一处理单元602，用于对各个所述历史业务时间序列进行处理，得到历史时序热力图；a
提取单元603，用于提取所述历史时序热力图中的图像特征数据；
第二处理单元604，用于对所述图像特征数据进行处理，得到每个所述历史业务时间序列的预测时序流量。The
本发明实施例提供的装置中，获取各个历史业务时间序列，对各个历史业务时间序列进行处理，得到历史时序热力图；提取历史时序热力图中的图像特征数据；对图像特征数据进行处理，得到每个历史业务时间序列的预测时序流量。通过将各个历史业务时间序列转换成历史时序热力图，并从历史时序热力图中提取到图像特征数据，该图像特征数据包含了时间序列各种高维度的特征，含盖了时间序列全局及局部的相关特征，通过对图像特征数据进行处理，可以得到各个历史业务时间序列的预测时序流量，引入涵盖了高维度的特征的图像特征数据，有效的提高了对时间序列的流量进行预测的准确性。In the device provided by the embodiment of the present invention, each historical business time series is obtained, and each historical business time series is processed to obtain a historical time series heat map; the image feature data in the historical time series heat map is extracted; the image feature data is processed to obtain Forecast time series traffic for each historical business time series. By converting each historical business time series into a historical time series heat map, and extracting image feature data from the historical time series heat map, the image feature data contains various high-dimensional features of the time series, covering the global and local time series. By processing the image feature data, the predicted time series traffic of each historical business time series can be obtained, and the image feature data covering high-dimensional features can be introduced, which effectively improves the accuracy of time series traffic prediction. .
本发明实施例提供的装置中，所述第一处理单元602，可以配置为：In the apparatus provided by the embodiment of the present invention, the
构建子单元，用于基于各个所述历史业务时间序列，构建数据矩阵；constructing a subunit for constructing a data matrix based on each of the historical business time series;
归一化处理子单元，用于将所述数据矩阵中的每个数值进行归一化处理，得到历史时序热力图。The normalization processing subunit is used for normalizing each value in the data matrix to obtain a historical time series heat map.
本发明实施例提供的装置中，所述提取单元603，可以配置为：In the apparatus provided by the embodiment of the present invention, the
输入子单元，用于将所述历史时序热力图输入预先训练完成的特征提取模型中，使得所述特征提取模型从所述历史时序热力图中提取出各项图像高维度特征；The input subunit is used to input the historical time series heat map into the feature extraction model that has been trained in advance, so that the feature extraction model extracts various image high-dimensional features from the historical time series heat map;
确定子单元，用于将各项所述图像高维度特征作为所述历史时序热力图像的图像特征数据。A determination subunit, configured to use each item of the high-dimensional image features as image feature data of the historical time series thermal image.
本发明实施例提供的装置中，所述第二处理单元604，可以配置为：In the apparatus provided by the embodiment of the present invention, the
输出子单元，用于将所述图像特征数据输入预先训练完成的分类神经网络中，使得所述分类神经网络输出所述历史时序热力图的图片分类数据，其中，所述图片分类数据中包含各项所述历史业务时间序列的时序流量的预测信息；The output subunit is used to input the image feature data into the pre-trained classification neural network, so that the classification neural network outputs the image classification data of the historical time series heat map, wherein the image classification data includes each Prediction information of the time series traffic of the historical business time series described in item;
调用子单元，用于调用预设的回归函数对所述图像分类数据进行回归处理，得到与所述图像分类数据对应的回归数据；a calling subunit for calling a preset regression function to perform regression processing on the image classification data to obtain regression data corresponding to the image classification data;
逆归一化处理子单元，用于对所述回归数据进行逆归一化处理，得每个所述历史业务时间序列的预测时序流量。The inverse normalization processing subunit is configured to perform inverse normalization processing on the regression data to obtain the predicted time series traffic of each of the historical business time series.
本发明实施例提供的装置中，还包括：In the device provided by the embodiment of the present invention, it also includes:
风险评估单元，用于基于每个所述历史业务时间序列的预测时序流量进行风险评估，以得到每个所述历史业务时间序列所对应的业务的风险评分。A risk assessment unit, configured to perform risk assessment based on the predicted time series traffic of each of the historical service time series, so as to obtain a risk score of the service corresponding to each of the historical service time series.
本发明实施例还提供了一种存储介质，所述存储介质包括存储的指令，其中，在所述指令运行时控制所述存储介质所在的设备执行以下操作：An embodiment of the present invention further provides a storage medium, where the storage medium includes stored instructions, wherein when the instructions are executed, the device where the storage medium is located is controlled to perform the following operations:
获取各个历史业务时间序列；Obtain each historical business time series;
对各个所述历史业务时间序列进行处理，得到历史时序热力图；Process each of the historical business time series to obtain a historical time series heat map;
提取所述历史时序热力图中的图像特征数据；extracting image feature data in the historical time series heatmap;
对所述图像特征数据进行处理，得到每个所述历史业务时间序列的预测时序流量。The image feature data is processed to obtain the predicted time series traffic of each of the historical business time series.
本发明实施例还提供了一种电子设备，其结构示意图如图7所示，具体包括存储器701，以及一个或者一个以上的指令702，其中一个或者一个以上指令702存储于存储器701中，且经配置以由一个或者一个以上处理器603执行所述一个或者一个以上指令702进行以下操作：An embodiment of the present invention further provides an electronic device, the schematic structural diagram of which is shown in FIG. 7 , and specifically includes a
获取各个历史业务时间序列；Obtain each historical business time series;
对各个所述历史业务时间序列进行处理，得到历史时序热力图；Process each of the historical business time series to obtain a historical time series heat map;
提取所述历史时序热力图中的图像特征数据；extracting image feature data in the historical time series heatmap;
对所述图像特征数据进行处理，得到每个所述历史业务时间序列的预测时序流量。The image feature data is processed to obtain the predicted time series traffic of each of the historical business time series.
上述各个实施例的具体实施过程及其衍生方式，均在本发明的保护范围之内。The specific implementation process of each of the above-mentioned embodiments and the derivatives thereof are all within the protection scope of the present invention.
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于系统或系统实施例而言，由于其基本相似于方法实施例，所以描述得比较简单，相关之处参见方法实施例的部分说明即可。以上所描述的系统及系统实施例仅仅是示意性的，其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性劳动的情况下，即可以理解并实施。The various embodiments in this specification are described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, as for the system or the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant part may refer to the partial description of the method embodiment. The systems and system embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, It can be located in one place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.
专业人员还可以进一步意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。Professionals may further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two, in order to clearly illustrate the possibilities of hardware and software. Interchangeability, the above description has generally described the components and steps of each example in terms of functionality. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of the present invention.
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本发明。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本发明的精神或范围的情况下，在其它实施例中实现。因此，本发明将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。The above description of the disclosed embodiments enables any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
Translated fromChinesePriority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210203124.4ACN114627330A (en) | 2022-03-02 | 2022-03-02 | Time sequence flow prediction method and device, storage medium and electronic equipment |
| PCT/CN2022/127217WO2023165145A1 (en) | 2022-03-02 | 2022-10-25 | Time sequence traffic prediction method and apparatus, storage medium, and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210203124.4ACN114627330A (en) | 2022-03-02 | 2022-03-02 | Time sequence flow prediction method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114627330Atrue CN114627330A (en) | 2022-06-14 |
Family
ID=81899895
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210203124.4APendingCN114627330A (en) | 2022-03-02 | 2022-03-02 | Time sequence flow prediction method and device, storage medium and electronic equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114627330A (en) |
| WO (1) | WO2023165145A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023165145A1 (en)* | 2022-03-02 | 2023-09-07 | 北京沃东天骏信息技术有限公司 | Time sequence traffic prediction method and apparatus, storage medium, and electronic device |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118017478B (en)* | 2024-01-23 | 2024-10-29 | 中国铁塔股份有限公司 | Intelligent power saving system based on business scene and intelligent space-time prediction |
| CN118071196B (en)* | 2024-02-21 | 2024-11-05 | 华能国际电力股份有限公司大连电厂 | Power generation set level index management system based on table format visualization |
| CN118212535B (en)* | 2024-05-17 | 2024-08-27 | 江西省林业科学院 | Wetland aquatic plant cultivation management system and method |
| CN118856363B (en)* | 2024-09-18 | 2025-03-11 | 新疆正通路桥工程有限公司 | Intelligent combustion control system and method for quartz sand drying |
| CN118936113B (en)* | 2024-09-19 | 2025-04-04 | 新疆宝信智能技术有限公司 | Intelligent combustion control system and method for atmosphere furnace |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109657582A (en)* | 2018-12-10 | 2019-04-19 | 平安科技（深圳）有限公司 | Recognition methods, device, computer equipment and the storage medium of face mood |
| CN110798365A (en)* | 2020-01-06 | 2020-02-14 | 支付宝(杭州)信息技术有限公司 | Neural network-based traffic prediction method and device |
| CN110970090A (en)* | 2019-11-18 | 2020-04-07 | 华中科技大学 | Method for judging similarity between polypeptide to be processed and positive data set peptide fragment |
| CN112468326A (en)* | 2020-11-11 | 2021-03-09 | 北京工业大学 | Access flow prediction method based on time convolution neural network |
| CN112905839A (en)* | 2021-02-10 | 2021-06-04 | 北京有竹居网络技术有限公司 | Model training method, model using device, storage medium and equipment |
| CN113469399A (en)* | 2020-03-31 | 2021-10-01 | 阿里巴巴集团控股有限公司 | Service prediction method and device |
| CN113538903A (en)* | 2021-06-21 | 2021-10-22 | 东南大学 | A Traffic Congestion Prediction Method Based on Traffic Flow Feature Extraction and Classification |
| CN113660176A (en)* | 2021-08-16 | 2021-11-16 | 中国电信股份有限公司 | Traffic prediction method and device for communication network, electronic device and storage medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109995562B (en)* | 2017-12-30 | 2022-11-08 | 中国移动通信集团河北有限公司 | Network traffic forecasting method, device, equipment and medium |
| US20200402665A1 (en)* | 2019-06-19 | 2020-12-24 | GE Precision Healthcare LLC | Unplanned readmission prediction using an interactive augmented intelligent (iai) system |
| CN110674019B (en)* | 2019-08-30 | 2023-02-07 | 中国人民财产保险股份有限公司 | Method and device for predicting system fault and electronic equipment |
| CN113259950B (en)* | 2021-05-13 | 2022-03-11 | 重庆邮电大学 | Low-orbit satellite spot beam closing method based on service prediction |
| CN114118622A (en)* | 2021-12-08 | 2022-03-01 | 北京北大软件工程股份有限公司 | Data trend prediction method and system based on time series |
| CN114627330A (en)* | 2022-03-02 | 2022-06-14 | 北京沃东天骏信息技术有限公司 | Time sequence flow prediction method and device, storage medium and electronic equipment |
- 2022
- 2022-03-02CNCN202210203124.4Apatent/CN114627330A/enactivePending
- 2022-10-25WOPCT/CN2022/127217patent/WO2023165145A1/ennot_activeCeased
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109657582A (en)* | 2018-12-10 | 2019-04-19 | 平安科技（深圳）有限公司 | Recognition methods, device, computer equipment and the storage medium of face mood |
| CN110970090A (en)* | 2019-11-18 | 2020-04-07 | 华中科技大学 | Method for judging similarity between polypeptide to be processed and positive data set peptide fragment |
| CN110798365A (en)* | 2020-01-06 | 2020-02-14 | 支付宝(杭州)信息技术有限公司 | Neural network-based traffic prediction method and device |
| CN113469399A (en)* | 2020-03-31 | 2021-10-01 | 阿里巴巴集团控股有限公司 | Service prediction method and device |
| CN112468326A (en)* | 2020-11-11 | 2021-03-09 | 北京工业大学 | Access flow prediction method based on time convolution neural network |
| CN112905839A (en)* | 2021-02-10 | 2021-06-04 | 北京有竹居网络技术有限公司 | Model training method, model using device, storage medium and equipment |
| CN113538903A (en)* | 2021-06-21 | 2021-10-22 | 东南大学 | A Traffic Congestion Prediction Method Based on Traffic Flow Feature Extraction and Classification |
| CN113660176A (en)* | 2021-08-16 | 2021-11-16 | 中国电信股份有限公司 | Traffic prediction method and device for communication network, electronic device and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023165145A1 (en)* | 2022-03-02 | 2023-09-07 | 北京沃东天骏信息技术有限公司 | Time sequence traffic prediction method and apparatus, storage medium, and electronic device |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023165145A1 (en) | 2023-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114627330A (en) | Time sequence flow prediction method and device, storage medium and electronic equipment | |
| US11604980B2 (en) | Targeted crowd sourcing for metadata management across data sets | |
| US20130191395A1 (en) | Social media data analysis system and method | |
| JP6911603B2 (en) | How to generate predictive models for the categories of facilities visited by users, programs, server equipment, and processing equipment | |
| US11115520B2 (en) | Signal discovery using artificial intelligence models | |
| CN107944032B (en) | Method and apparatus for generating information | |
| CN109978020B (en) | Social network account number vest identity identification method based on multi-dimensional features | |
| CN106354856A (en) | Deep neural network enhanced search method and device based on artificial intelligence | |
| US10678821B2 (en) | Evaluating theses using tree structures | |
| US11521601B2 (en) | Detecting extraneous topic information using artificial intelligence models | |
| CN111061837A (en) | Topic identification method, device, equipment and medium | |
| CN103617146B (en) | A kind of machine learning method and device based on hardware resource consumption | |
| WO2017071474A1 (en) | Method and device for processing language data items and method and device for analyzing language data items | |
| CN111723180A (en) | Interviewing method and device | |
| CN111582757A (en) | Fraud risk analysis method, device, equipment and computer-readable storage medium | |
| CN113886697A (en) | Activity recommendation method, device, device and storage medium based on clustering algorithm | |
| CN113919921B (en) | A product recommendation method based on multi-task learning model and related equipment | |
| WO2023221359A1 (en) | User security level identification method and apparatus based on multi-stage time sequence and multiple tasks | |
| CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
| CN112200602B (en) | Neural network model training method and device for advertisement recommendation | |
| CN114638299A (en) | Intelligent zombie license identification method, device, equipment and storage medium | |
| CN114240250A (en) | An intelligent management method and system for occupational evaluation | |
| CN113283979A (en) | Loan credit evaluation method and device for loan applicant and storage medium | |
| Guan et al. | Fila: Online auditing of machine learning model accuracy under finite labelling budget | |
| WO2019000051A1 (en) | Data analysis method and learning system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |