Ids Prototype for Intrusion Detection With Machine Learning Models in Iot Systems of the Industry 4.0

Aveleira‐Mata, José; Castañeda, Ángel Luis Muñoz; Ordás, María Teresa García; Benavides, Carmen; Bení­tez-Andrades, José Alberto; Aláiz‐Moretón, Héctor

doi:10.6036/10011

Cited by 8 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The specific functions and powers of smart government information management department, supervision department, and accountability department should be clarified. Various measures can also speed up the establishment of legal systems and standards for the government to use the IoT, 5G, artificial intelligence, and other ICT [34].…”

Section: Discussionmentioning

confidence: 99%

Influencing Factors of Smart Government Information Security: Experience from China

Wang

Zhang

2022

Teh. vjesn.

View full text Add to dashboard Cite

Based on information technology, smart government processes information data to help improve the efficiency of government operations. Information security has become the key to the transformation of government wisdom and improvement of government service efficiency and transparency. To explore the influencing factors of government information security, 27 influencing factors from 6 dimensions of personnel, facilities, information, personnel management, system, and environment were extracted. The decision-making trial and evaluation laboratory (DEMATEL)-interpretative structural modeling (ISM) composite model was used to determine the importance of each factor, the mechanism between the influencing factors was analyzed, and a smart government information security interpretation structure model was constructed. Results show that policies and regulations are the root factors affecting the information security of the smart government. The international environment is a deep-layer influencing factor on government information security. Infrastructure, moral training, and other factors are middle-layer influencing factors that are influenced by the superior factors. Psychological factors, platform construction, and others are surface-layer influencing factors that play a connecting role. Security awareness, behavior security, moral level, and others are directly influencing factors and determine the level of smart government information security. Results of this study also show the feasibility of DEMATEL-ISM model in identifying the relevant factors and analyzing their influencing mechanism on smart government information security. A novel modeling method and its analysis approach are provided for the government to improve the level of information security.

show abstract

Section: Discussionmentioning

confidence: 99%

Influencing Factors of Smart Government Information Security: Experience from China

Wang

Zhang

2022

Teh. vjesn.

View full text Add to dashboard Cite

show abstract

“…The PCA revealed the best Area Under the Curve (AUC) of over 89% and the shortest training time. Also, a research [16] developed an IDS prototype, which detects anomalies in MQTT-based IoT environments. After training and testing the DT classifier and k-nearest neighbors algorithm (KNN), the results showed that DT's performance outperformed the other with an average Fmeasure of 93.81%.…”

Section: Related Workmentioning

confidence: 99%

“…After training and testing the DT classifier and k-nearest neighbors algorithm (KNN), the results showed that DT's performance outperformed the other with an average Fmeasure of 93.81%. However, the imbalanced dataset issue was not addressed or resolved in either of the three previous papers [13], [15], [16].…”

Section: Related Workmentioning

confidence: 99%

Protecting Smart-Home IoT Devices From MQTT Attacks: An Empirical Study of ML-Based IDS

Alasmari,

Alhogail

2024

IEEE Access

View full text Add to dashboard Cite

Smart homes are becoming more and more popular in the world, and they are mainly on Internet of Things (IoT) technologies to enable their functionality. However, since IoT devices have limited computing power and resources, implementing strong security measures is difficult, which makes the use of intrusion detection systems (IDS) an appropriate option. In this paper, we propose an optimized model with high performance for intrusion detection in Message Queue Telemetry Transport protocol (MQTT)based IoT networks for smart homes. This is done by studying a total of 22 Machine Learning (ML) algorithms based on an extended two-stage evaluation approach that includes several aspects for optimizing and validating performance for finding the ideal model. Based on the empirical evaluation, the GLM classifier with the random over-sampling technique produces the best detection performance with 100% accuracy and an f-score of 100%, outperforming the previous studies. This paper also investigates the influence of automatic feature engineering technique on the performance of algorithms. With the automatic feature engineering technique, the performance increased by up to 38.9%, and the time consumed to classify attacks decreased by up to 67.7%. This shows that automatic feature engineering can improve performance and reduce detection time.

show abstract

“…For perishable food, the key dimensions in cold supply chain management, such as information technology, equipment, infrastructure and supervision, must be interconnected (Bremer, 2018). This, combined with remote access functions and the cloud processing enabled by the Internet of Things (IoT), provides information that optimizes processes and supports decision‐making (Aveleira‐Mata et al, 2021). Artificial intelligence, robotics, data analytics, digital platforms and automation, among other technologies, are prompting a swift and profound transformation of work (Rodriguez‐Lluesma et al, 2021).…”

Section: The State Of the Artmentioning

confidence: 99%

Expert system and entropy‐ordered weighted average method in cold supply chain risk evaluation

Shen

Yan

2022

Expert Systems

View full text Add to dashboard Cite

If not well identified and controlled, risks in systematically engineered cold supply chains can lead directly to food safety incidents. In current approaches to cold supply chain risk evaluation, there is a lack of systematic classification and quantitative analysis of the influences of risk factors in the chain. To accurately evaluate unknown risks that can exert a fluctuating influence and result in great losses, this study builds a knowledge base of expert systems based on expertise and extensive experience and modifies the expert weights in conjunction with entropy weights to reduce subjective error. An ordered weighted average (OWA) operator is used to evaluate risk in a cold supply chain effectively. First, in combination with a case study of a typical fresh food e‐commerce enterprise, a typical fresh food e‐commerce enterprise, this paper identifies the risks in a food refrigeration supply chain. Second, relying on an expert group with professional knowledge and rich experience, an expert system knowledge base is constructed, risks are assigned, and the expert weight is modified in combination with the entropy weight method. Finally, based on the OWA operator, the modified risk assignment is effectively evaluated, and the risk value is obtained and sorted. Specific risk control measures are put forward according to the results of the calculation. The results show that an evaluation method combining an expert system knowledge base and the entropy‐OWA method can effectively depict and process the evaluation information needed for risk rankings and solve the problems of supply chain risks more rapidly and effectively.

show abstract

Ids Prototype for Intrusion Detection With Machine Learning Models in Iot Systems of the Industry 4.0

Cited by 8 publications

References 0 publications

Influencing Factors of Smart Government Information Security: Experience from China

Influencing Factors of Smart Government Information Security: Experience from China

Protecting Smart-Home IoT Devices From MQTT Attacks: An Empirical Study of ML-Based IDS

Expert system and entropy‐ordered weighted average method in cold supply chain risk evaluation

Contact Info

Product

Resources

About