An Automated Forecasting Framework based on Method Recommendation for Seasonal Time Series

Bauer, André; Züfle, Marwin; Grohmann, Johannes; Schmitt, Norbert; Herbst, Nikolas; Kounev, Samuel

doi:10.1145/3358960.3379123

Cited by 11 publications

(16 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Statistics is a pillar field of study for both machine learning and deep learning. Many Statistical methods and Machine Learning approaches have been formulated to support accurate forecasting results [3], including the analysis of seasonal time series [3,12,13,32,37]. Typical examples of Statistical models include seasonal autoregressive models of integrated moving averages (SARIMA), Holt-Winters models, periodic autoregressive moving average models (PARMA), etc [14, 15, 19-23, 26, 29-31, 40, 45].…”

Section: Conventional Statistical Methods Versus Machine Learning Methodsmentioning

confidence: 99%

“…Machine learning exists at the intersection of traditional mathematics and statistics with software engineering and computer science [9]. Machine Learning techniques that allow to find patterns in time series include the soft computing technique called Recurrent Neural Networks, Support Vector Machines, Least-Square Support Vector Machine models, and ensemble methods such as Random Forests and Gradient Boosting, [3,8,17,32,37,42]. Given this variety of methods, choosing and configuring the best performing method for a given seasonal time series becomes a challenging task for a data science expert.…”

Section: Conventional Statistical Methods Versus Machine Learning Methodsmentioning

confidence: 99%

“…All of these steps help the machine to "learn", but, because of the large number of variables in data sets, the model developed can be extremely complex and difficult to interpret, such is the case of computing to find the least of the function of cost of neural networks, which is quite complex and time-consuming. Furthermore, Machine Mearning-based methods struggle with non-stationary data [3,35].…”

Section: Conventional Statistical Methods Versus Machine Learning Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Structural Models: A Computational Look for Signal Extraction and Forecasting Seasonal Time Series

Puindi

2021

SN COMPUT. SCI.

View full text Add to dashboard Cite

Data analysis requires doing statistical programming which can be done at a simpler or more complex level. This article presents the main computational aspects for signal extraction, estimation and forecasting seasonal time series. From a structural model with covariates and different errors affecting the observations and the states, intelligent computational procedures are designed. First, the intelligent computational algorithms to obtain the main matrices of the system are derived; second, a general intelligent computational procedure and an algorithm that performs the Kalman filter and model estimation are also derived. Furthermore, the intelligent computational procedures and the structural model are evaluated using real seasonal time series, and the results demonstrate that the proposed method and model are very attractive and promising for model estimation, signal extraction and forecasting tasks.

show abstract

Section: Conventional Statistical Methods Versus Machine Learning Methodsmentioning

confidence: 99%

Section: Conventional Statistical Methods Versus Machine Learning Methodsmentioning

confidence: 99%

Section: Conventional Statistical Methods Versus Machine Learning Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Structural Models: A Computational Look for Signal Extraction and Forecasting Seasonal Time Series

Puindi

2021

SN COMPUT. SCI.

View full text Add to dashboard Cite

show abstract

“…Hence, there exist different methods for forecasting that are of use for the aforementioned techniques. Wold [ [109], and Bauer et al [110] aim at creating or improving general forecasting techniques. Note that those works do not specifically consider anomaly detection or performance prediction but propose generally applicable techniques for forecasting.…”

Section: Detection Based On Time Series Forecastingmentioning

confidence: 99%

A Taxonomy of Techniques for SLO Failure Prediction in Software Systems

Grohmann

Herbst

Chalbani³

et al. 2020

Computers

Self Cite

View full text Add to dashboard Cite

Failure prediction is an important aspect of self-aware computing systems. Therefore, a multitude of different approaches has been proposed in the literature over the past few years. In this work, we propose a taxonomy for organizing works focusing on the prediction of Service Level Objective (SLO) failures. Our taxonomy classifies related work along the dimensions of the prediction target (e.g., anomaly detection, performance prediction, or failure prediction), the time horizon (e.g., detection or prediction, online or offline application), and the applied modeling type (e.g., time series forecasting, machine learning, or queueing theory). The classification is derived based on a systematic mapping of relevant papers in the area. Additionally, we give an overview of different techniques in each sub-group and address remaining challenges in order to guide future research.

show abstract

“…There exists no forecasting method that performs best on all types of time-series [18]. However, efforts to create more universal approaches seek to adjust ensemble size and choose potential members or weights adaptively based on dynamics we try to extrapolate into the future.…”

Section: Introductionmentioning

confidence: 99%