Search results
1 – 10 of 299K.V. Sheelavathy and V. Udaya Rani
Internet of Things (IoT) is a network, which provides the connection with various physical objects such as smart machines, smart home appliance and so on. The physical objects are…
Abstract
Purpose
Internet of Things (IoT) is a network, which provides the connection with various physical objects such as smart machines, smart home appliance and so on. The physical objects are allocated with a unique internet address, namely, Internet Protocol, which is used to perform the data broadcasting with the external objects using the internet. The sudden increment in the number of attacks generated by intruders, causes security-related problems in IoT devices while performing the communication. The main purpose of this paper is to develop an effective attack detection to enhance the robustness against the attackers in IoT.
Design/methodology/approach
In this research, the lasso regression algorithm is proposed along with ensemble classifier for identifying the IoT attacks. The lasso algorithm is used for the process of feature selection that modeled fewer parameters for the sparse models. The type of regression is analyzed for showing higher levels when certain parts of model selection is needed for parameter elimination. The lasso regression obtains the subset for predictors to lower the prediction error with respect to the quantitative response variable. The lasso does not impose a constraint for modeling the parameters caused the coefficients with some variables shrink as zero. The selected features are classified by using an ensemble classifier, that is important for linear and nonlinear types of data in the dataset, and the models are combined for handling these data types.
Findings
The lasso regression with ensemble classifier–based attack classification comprises distributed denial-of-service and Mirai botnet attacks which achieved an improved accuracy of 99.981% than the conventional deep neural network (DNN) methods.
Originality/value
Here, an efficient lasso regression algorithm is developed for extracting the features to perform the network anomaly detection using ensemble classifier.
Details
Keywords
Ikhlaas Gurrib, Firuz Kamalov, Olga Starkova, Elgilani Eltahir Elshareif and Davide Contu
This paper aims to investigate the role of price-based information from major cryptocurrencies, foreign exchange, equity markets and key commodities in predicting the next-minute…
Abstract
Purpose
This paper aims to investigate the role of price-based information from major cryptocurrencies, foreign exchange, equity markets and key commodities in predicting the next-minute Bitcoin (BTC) price. This study answers the following research questions: What is the best sparse regression model to predict the next-minute price of BTC? What are the key drivers of the BTC price in high-frequency trading?
Design/methodology/approach
Least absolute shrinkage and selection operator and Ridge regressions are adopted using minute-based open-high-low-close prices, volume and trade count for eight major cryptos, global stock market indices, foreign currency pairs, crude oil and gold price information for February 2020–March 2021. This study also examines whether there was any significant break and how the accuracy of the selected models was impacted.
Findings
Findings suggest that Ridge regression is the most effective model for predicting next-minute BTC prices based on BTC-related covariates such as BTC-open, BTC-high and BTC-low, with a moderate amount of regularization. While BTC-based covariates BTC-open and BTC-low were most significant in predicting BTC closing prices during stable periods, BTC-open and BTC-high were most important during volatile periods. Overall findings suggest that BTC’s price information is the most helpful to predict its next-minute closing price after considering various other asset classes’ price information.
Originality/value
To the best of the authors’ knowledge, this is the first paper to identify the covariates of major cryptocurrencies and predict the next-minute BTC crypto price, with a focus on both crypto-asset and cross-market information.
Details
Keywords
Fengjun Tian, Yang Yang, Zhenxing Mao and Wenyue Tang
This paper aims to compare the forecasting performance of different models with and without big data predictors from search engines and social media.
Abstract
Purpose
This paper aims to compare the forecasting performance of different models with and without big data predictors from search engines and social media.
Design/methodology/approach
Using daily tourist arrival data to Mount Longhu, China in 2018 and 2019, the authors estimated ARMA, ARMAX, Markov-switching auto-regression (MSAR), lasso model, elastic net model and post-lasso and post-elastic net models to conduct one- to seven-days-ahead forecasting. Search engine data and social media data from WeChat, Douyin and Weibo were incorporated to improve forecasting accuracy.
Findings
Results show that search engine data can substantially reduce forecasting error, whereas social media data has very limited value. Compared to the ARMAX/MSAR model without big data predictors, the corresponding post-lasso model reduced forecasting error by 39.29% based on mean square percentage error, 33.95% based on root mean square percentage error, 46.96% based on root mean squared error and 45.67% based on mean absolute scaled error.
Practical implications
Results highlight the importance of incorporating big data predictors into daily demand forecasting for tourism attractions.
Originality/value
This study represents a pioneering attempt to apply the regularized regression (e.g. lasso model and elastic net) in tourism forecasting and to explore various daily big data indicators across platforms as predictors.
Details
Keywords
Ajit Kumar and A.K. Ghosh
The purpose of this study is to estimate aerodynamic parameters using regularized regression-based methods.
Abstract
Purpose
The purpose of this study is to estimate aerodynamic parameters using regularized regression-based methods.
Design/methodology/approach
Regularized regression methods used are LASSO, ridge and elastic net.
Findings
A viable option of aerodynamic parameter estimation from regularized regression-based methods is found.
Practical implications
Efficacy of the methods is examined on flight test data.
Originality/value
This study provides regularized regression-based methods for aerodynamic parameter estimation from the flight test data.
Details
Keywords
James Mitchell, Aubrey Poon and Gian Luigi Mazzi
This chapter uses an application to explore the utility of Bayesian quantile regression (BQR) methods in producing density nowcasts. Our quantile regression modeling strategy is…
Abstract
This chapter uses an application to explore the utility of Bayesian quantile regression (BQR) methods in producing density nowcasts. Our quantile regression modeling strategy is designed to reflect important nowcasting features, namely the use of mixed-frequency data, the ragged-edge, and large numbers of indicators (big data). An unrestricted mixed data sampling strategy within a BQR is used to accommodate a large mixed-frequency data set when nowcasting; the authors consider various shrinkage priors to avoid parameter proliferation. In an application to euro area GDP growth, using over 100 mixed-frequency indicators, the authors find that the quantile regression approach produces accurate density nowcasts including over recessionary periods when global-local shrinkage priors are used.
Details
Keywords
Philip Kostov, Thankom Arun and Samuel Annim
This paper aims to understand household’s latent behaviour decision-making in accessing financial services. In this analysis, the determinants of the choice of the pre-entry…
Abstract
Purpose
This paper aims to understand household’s latent behaviour decision-making in accessing financial services. In this analysis, the determinants of the choice of the pre-entry Mzansi account by consumers in South Africa is looked at.
Design/methodology/approach
In this study, 102 variables, grouped in the following categories: basic literacy, understanding financial terms, targets for financial advice, desired financial education and financial perception. Using a computationally efficient variable selection algorithm, variables that can satisfactorily explain the choice of a Mzansi account were studied.
Findings
The Mzansi intervention is appealing to individuals with basic but insufficient financial education. Aspirations seem to be very influential in revealing the choice of financial services, and, to this end, Mzansi is perceived as a pre-entry account not meeting the aspirations of individuals aiming to climb up the financial services ladder. It was found that Mzansi holders view the account mainly as a vehicle for receiving payments, but, on the other hand, are debt-averse and inclined to save. Hence, although there is at present no concrete evidence that the Mzansi intervention increases access to finance via diversification (i.e. by recruiting customers into higher-level accounts and services), this analysis shows that this is very likely to be the case.
Originality/value
The issue of demand-side constraints on access to finance have been largely been ignored in the theoretical and empirical literature. This paper undertakes some preliminary steps in addressing this gap.
Details
Keywords
Suman Chhabri, Krishnendu Hazra, Amitava Choudhury, Arijit Sinha and Manojit Ghosh
Because of the mechanical properties of aluminium (Al), an accurate prediction of its properties has been challenging. Researchers are seeking reliable models for predicting the…
Abstract
Purpose
Because of the mechanical properties of aluminium (Al), an accurate prediction of its properties has been challenging. Researchers are seeking reliable models for predicting the mechanical strength of Al alloys owing to the continuous emergence of new Al alloys and their applications. There has been widespread use of empirical and statistical models for the prediction of different mechanical properties of Al and Al alloy, such as linear and nonlinear regression. Nevertheless, the development of these models requires laborious experimental work, and they may not produce accurate results depending on the relationship between the Al properties, mix of other compositions and curing conditions.
Design/methodology/approach
Numerous machine learning (ML) models have been proposed as alternative approaches for predicting the strengths of Al and its alloys. The hardness of Al alloys has been predicted by implementing various ML algorithms, such as linear regression, ridge regression, lasso regression and artificial neural network (ANN). This investigation critically analysed and discussed the application and performance of models generated by linear regression, ridge regression, lasso regression and ANN algorithms using different mechanical properties as training parameters.
Findings
Considering the definition of the problem, linear regression has been found to be the most suitable algorithm in predicting the hardness values of AA7XXX alloys as the model generated by it best fits the data set.
Originality/value
The work presented in this paper is original and not submitted anywhere else.
Details
Keywords
The purpose of the present study was to improve the fit of women’s bifurcated garments by developing an equation that can predict the crotch length accurately by using a few basic…
Abstract
Purpose
The purpose of the present study was to improve the fit of women’s bifurcated garments by developing an equation that can predict the crotch length accurately by using a few basic body measurements. This equation could provide a simple mass-customization approach to the design of bifurcated garments.
Design/methodology/approach
Demographic characteristics and easy-to-record body measurements available in the size USA database were used to predict the crotch length. Different methodologies including best subset regression, lasso regression and principal components regression were experimented with to identify the most important predictor variables and establish a relationship between the significant predictors and crotch length.
Findings
The lasso regression model provided the highest accuracy, required only five body dimensions and dealt with multicollinearity. The preliminary pattern preparation and garment fit tests indicated that by utilizing the proposed equation, patterns of customized garments could be successfully altered to match the crotch length of the customer, thereby, improving the precision and efficiency of the pattern making process.
Originality/value
Crotch length is a crucial measurement as it determines bifurcated garment comfort as well as aesthetic fit. The crotch length is usually estimated arbitrarily based on non-scientific methods while drafting patterns, and this increases the likelihood of dissatisfaction with the fit of the lower-body garments. The present study suggested an algorithm that could predict crotch length with 90.53% accuracy using the body dimensions height, hips, waist height, knee height and arm length.
Details
Keywords
Sasanka Choudhury, Dhirendra Nath Thatoi, Jhalak Hota, Suman Sau and Mohan D. Rao
The purpose of this paper is to identify the crack in beam-like structures before the complete failure or damage occurs to the structure. The beam-like structure plays an…
Abstract
Purpose
The purpose of this paper is to identify the crack in beam-like structures before the complete failure or damage occurs to the structure. The beam-like structure plays an important role in modern architecture; hence, the safety of this structure is much dependent on the safety of the beam. Hence, predicting the cracks is much more important for the safety of the overall structure.
Design/methodology/approach
In the present work, the regression analysis has been carried out through LASSO and Ridge regression models. Both the statistical models have been well implemented in the detection of crack depth and crack location. A cantilever beam-like structure has been taken for the analysis in which the first three natural frequencies have been considered as the independent variable and crack location and depth is used as the dependent variable. The first three natural frequencies, f1, f2 and f3 are used as an independent variable. The crack location and crack depth are estimated though the regressor models and the accuracy are compared, to verify the correctness of the estimation.
Findings
As stated in the purpose of work, the main aim of the present work is to identify the crack parameters using an inverse technique, which will be more effective and will provide the results with less time. The data used for regression analysis are obtained from theoretical analysis and later the theoretical results are also verified through experimental analysis. The regression model developed is tested for its Bias Variance Trade-off (“Bias” – Overfitting, “variance” – generalization). The regression results have been compared with the theoretical results to check the robustness in the subsequent result section.
Originality/value
The idea is an amalgamation of existing and well-established technologies, that is aimed to achieve better performance for the given task. A regressor is trained from the data obtained through numerical simulation. The model is developed taking bias variance trade-off into consideration. This generalized model gives us very much acceptable performance.
Details
Keywords
Jiaming Liu, Liuan Wang, Linan Zhang, Zeming Zhang and Sicheng Zhang
The primary objective of this study was to recognize critical indicators in predicting blood glucose (BG) through data-driven methods and to compare the prediction performance of…
Abstract
Purpose
The primary objective of this study was to recognize critical indicators in predicting blood glucose (BG) through data-driven methods and to compare the prediction performance of four tree-based ensemble models, i.e. bagging with tree regressors (bagging-decision tree [Bagging-DT]), AdaBoost with tree regressors (Adaboost-DT), random forest (RF) and gradient boosting decision tree (GBDT).
Design/methodology/approach
This study proposed a majority voting feature selection method by combining lasso regression with the Akaike information criterion (AIC) (LR-AIC), lasso regression with the Bayesian information criterion (BIC) (LR-BIC) and RF to select indicators with excellent predictive performance from initial 38 indicators in 5,642 samples. The selected features were deployed to build the tree-based ensemble models. The 10-fold cross-validation (CV) method was used to evaluate the performance of each ensemble model.
Findings
The results of feature selection indicated that age, corpuscular hemoglobin concentration (CHC), red blood cell volume distribution width (RBCVDW), red blood cell volume and leucocyte count are five most important clinical/physical indicators in BG prediction. Furthermore, this study also found that the GBDT ensemble model combined with the proposed majority voting feature selection method is better than other three models with respect to prediction performance and stability.
Practical implications
This study proposed a novel BG prediction framework for better predictive analytics in health care.
Social implications
This study incorporated medical background and machine learning technology to reduce diabetes morbidity and formulate precise medical schemes.
Originality/value
The majority voting feature selection method combined with the GBDT ensemble model provides an effective decision-making tool for predicting BG and detecting diabetes risk in advance.
Details