Review of ML and AutoML Solutions to Forecast Time-Series Data

Alsahref, Ahmad; Aggarwal, Karan; Sonia, .; Kumar, M V Suresh; Mishra, Ashutosh

doi:10.1007/s11831-022-09765-0

Cited by 56 publications

(34 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The Internet of Things, as shown in Figure 2, is made up of devices that generate, process, and share massive volumes of security and safety-critical data as well as privacy-sensitive data, making them attractive targets for cyberattacks [43]. Many of the new networkable devices that make up the Internet of Things [11] are low-power and lightweight.…”

Section: The Internet Of Thingsmentioning

confidence: 99%

See 1 more Smart Citation

Next Generation IoT and Blockchain Integration

et al. 2022

View full text Add to dashboard Cite

The Internet of Things (IoT) refers to the interconnection of smart devices to collect data and make intelligent decisions. However, a lack of intrinsic security measures makes next generation IoT more vulnerable to privacy and security threats. With its “security by design,” Blockchain (BC) can help in addressing major security requirements in IoT. Blockchain is an ever-growing list of records that are linked and protected using cryptographic methods. It offers its users the flexibility to conduct transactions with lower costs and faster speeds. Blockchain ledgers are also decentralized and a ledger is maintained at each node in the network. Blockchain’s security and adaptability help in making even entire systems on it a much easily task with the benefit of decentralization. BC capabilities like immutability, transparency, auditability, data encryption, and operational resilience can help solve most architectural shortcomings of IoT. In the vision of the Internet of Things, traditional devices are becoming smarter and more autonomous. This vision is becoming reality as technology advances but there are still challenges to be resolved. This is especially true in a security domain like data trust, and with the expected evolution of the IoT in the coming years, it is important to ensure that this great source of data arrives. This paper began with an overview of blockchain and IoT, as well as explore the IoT blockchain application challenges. This article also focuses to review the most relevant tasks to analyse how IoT blockchain can improve and examine current research concerns and developments in the use of blockchain-related techniques and technologies in the context of IoT security in depth. One of the best parts of working or learning about blockchain and its application is the curiosity about how it can impact the things that we have been accustomed to without trying to improve and make things more efficient and productive.

show abstract

Section: The Internet Of Thingsmentioning

confidence: 99%

“…Brilliant agreements-although brilliant agreements have been identified as the ideal application of blockchain innovation, there are still a few issues to be resolved, as previously said. The use of clever contracts in IoT [42][43][44] might be beneficial, but the way they integrate into IoT applications is different [45,46].…”

Section: Challengesmentioning

confidence: 99%

Next Generation IoT and Blockchain Integration

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Time-Series Forecasting research dates back to 1985 [11], and since then, it has been a constantly expanding research area, especially in the past decade [2], due to the expansion of data volumes arising from users, industries, and markets, as well as the centrality of forecasting in various applications, such as economic, weather, stock price, business development, and health. As a result, numerous forecasting models have been developed, including ARIMA [20], SARIMA [20], ARIMAX [8], SARIMAX [8], N-BEATS [19], DeepAR [24], Long Short-Term Memory Neural Network (LSTM) [16], Gated Recurrent Unit Neural Networks (GRU) [29], and Temporal Fusion Transformer (TFT) [15].…”

Section: Related Workmentioning

confidence: 99%

“…Our primary contributions are as follows: (1) Conducting a comprehensive exploratory data analysis on this relatively new dataset in order to identify underlying patterns. (2) Examining the relationship between chickenpox cases and other variables such as the population. (3) Conducting comprehensive experiments on multiple time-series models and selecting the model that produces the best results for each county and at the national level.…”

Section: Introductionmentioning

confidence: 99%

Towards Automatic Forecasting: Evaluation of Time-Series Forecasting Models for Chickenpox Cases Estimation in Hungary

Skaf¹,

Arzu²,

Várkonyi³

2022

Preprint

View full text Add to dashboard Cite

Time-Series Forecasting is a powerful data modeling discipline that analyzes historical observations to predict future values of a time-series. It has been utilized in numerous applications, including but not limited to economics, meteorology, and health. In this paper, we use time-series forecasting techniques to model and predict the future incidence of chickenpox. To achieve this, we implement and simulate multiple models and data preprocessing techniques on a Hungary-collected dataset. We demonstrate that the LSTM model outperforms all other models in the vast majority of the experiments in terms of county-level forecasting, whereas the SARIMAX model performs best at the national level. We also demonstrate that the performance of the traditional data preprocessing method is inferior to that of the data preprocessing method that we have proposed.

show abstract

“…Closing the gap between domain experts/practitioners and ML specialists has recently been a topic of interest as ML processes are increasingly automated, and applied to solve practical problems in the industry [68]. Much of AutoML research has been focused on supervised learning, but recent research has diverged to tackle a larger range of ML problems such as unsupervised learning, time-series forecasting, and anamoly detection [69,70].…”

Section: Automated Machine Learning (Automl)mentioning

confidence: 99%

Spatial Decision Support Systems with Automated Machine Learning: A Review

Wen¹,

Li²

2022

Preprint

View full text Add to dashboard Cite

Many spatial decision support systems suffer from user adoption issues in practice due to lack of trust, technical expertise, and resources. Automated machine learning has recently allowed non-experts to explore and apply machine learning models in the industry without requiring abundant expert knowledge and resources. This paper reviews recent literature from 136 papers, and proposes a general framework for integrating spatial decision support systems with automated machine learning to lower major user adoption barriers. Challenges of data quality, model interpretability, and practical usefulness were discussed as general considerations for system implementation. Research opportunities related to spatially explicit models in AutoML, and resource-aware, collaborative/connected, and human-centered systems were also discussed to address these challenges. This paper argues that integrating spatial decision support systems with automated machine learning can not only encourage user adoption, but also mutually benefit research in both fields — bridging human-related and technical advancements for fostering future developments in spatial decision support systems and automated machine learning.

show abstract

Review of ML and AutoML Solutions to Forecast Time-Series Data

Cited by 56 publications

References 78 publications

Next Generation IoT and Blockchain Integration

Next Generation IoT and Blockchain Integration

Towards Automatic Forecasting: Evaluation of Time-Series Forecasting Models for Chickenpox Cases Estimation in Hungary

Spatial Decision Support Systems with Automated Machine Learning: A Review

Contact Info

Product

Resources

About