A C4.5 Fuzzy Decision Tree Method for Multivariate Time Series Forecasting

Silva, Rafael R. C.; Caminhas, Walmir M.; Silva, Petrônio Cândido de Lima e; Guimarães, Frederico Gadelha

doi:10.1109/fuzz45933.2021.9494439

Cited by 5 publications

(2 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Type-2 Fuzzy logic [72,88,107,119], fuzzy inference system [120], rough set theory [42,[121][122][123], c-fuzzy-decision tree [27], fuzzy decision tree (FDT) based on C4.5 [124] and Fuzzy Cognitive Maps (FCMs) [125,126] are other types of intelligent techniques that have been used in conjunction with FTS.…”

Section: Matrix and Rule-based Modelsmentioning

confidence: 99%

A Tutorial on Fuzzy Time Series Forecasting Models: Recent Advances and Challenges

Lucas¹,

Orang²,

Silva³

et al. 2022

Learn. Nonlin. Mod.

View full text Add to dashboard Cite

Time series forecasting is a powerful tool in planning and decision making, from traditional statistical models to soft computing and artificial intelligence approaches several methods have been developed to generate increasingly accurate forecasts. Fuzzy Time Series (FTS) methods have been introduced in the early 1990’s to handle data uncertainty and to undercome the statistical assumptions of linearity. Many studies have been reporting their good accuracy, simplicity, potential for interpretability and reduced computational complexity. This paper presents a tutorial for FTS methods. First, a review of the relevant literature is made, offering a foundation on the main concepts and FTS-based models for different time series and different types of forecasts. Then, the current challenges and possible solutions, are discussed alongside a timeline of the research developed in this area by the authors that aims at filling some of these gaps. Finally, a tutorial on the pyFTS library is presented. PyFTS is an open and free library coded in Python programming language that was developed by the MINDS Lab (Laboratory of Machine Intelligence and Data Science) and, also provides a set of transformation functions for pre-processing time series and a set of metrics and databases for benchmarking, in addition to implementing several FTS models in the literature.

show abstract

Section: Matrix and Rule-based Modelsmentioning

confidence: 99%

A Tutorial on Fuzzy Time Series Forecasting Models: Recent Advances and Challenges

Lucas¹,

Orang²,

Silva³

et al. 2022

Learn. Nonlin. Mod.

View full text Add to dashboard Cite

show abstract

“…Neural network (NN) is an algorithm that can produce non-linear predictions, has the ability to accept errors, and is powerful in parallel processing (Deng et al, 2021;Syafar et al, 2014), however in this situation the NN approach has limitations such as over-fitting, convergence, and requires training with huge amounts of data (Buchori et (Yaman et al, 2021) can considerably facilitate the reduction of class imbalance in distribution. One of the boosting algorithms is adaboost, which is an ensemble learning method that can reduce variance (Silva et al, 2021), due to the bias impact of the ensemble average reducing variance from a series of classifications (Silhavy et al, 2019). Following are research that employ boosting strategies to improve accuracy, as was done Sonavane & Sonar in the categorization of brain tumor detection.…”

Section: Introductionmentioning

confidence: 99%

Untitled

2023

JOWUA

View full text Add to dashboard Cite

The goal of this research is to use the boosting technique with the C4.5 algorithm to lower the number of wrong classifications. One of the techniques for boosting is called Adaboost. It can balance the class by giving more weight to the level of classification error, which can change how the data is spread out. The Online Shoppers Purchasing Intention dataset from the UCI Machine Learning Repository is used to test this method. It has 12330 records and 18 attributes, with a label being one of those attributes. In this study, the results were looked at with the help of software called RapidMiner 9.10. Before the processing stage, the dataset is split into two parts: training and testing, with a 90:10 and 80:20 split, respectively. The results of the experiments with the C4.5 algorithm and the boosting technique at a ratio of 90:10 (accuracy of 95.07 percent and AUC of 0.966) and a ratio of 80:20 (accuracy of 95.07 percent and AUC of 0.966). While the experimental results of the C4.5 algorithm without the boosting technique at a ratio of 90:10 (accuracy of 89.02 percent and AUC of 0.845) and a ratio of 80:20 (accuracy of 88.1 percent and AUC of 0.845). Based on the results of the comparison, boosting on the C4.5 algorithm is much better than the standard C4.5 algorithm, with an average increase of 5.76 percent. We can say that using the boosting technique with the C4.5 algorithm can improve high accuracy and lower the amount of classification error.

show abstract

A Multi-Scale Residual Graph Convolution Network with hierarchical attention for predicting traffic flow in urban mobility

Ling,

Lan,

Huang

et al. 2024

Complex Intell. Syst.

View full text Add to dashboard Cite

Accurate prediction of traffic flow is essential for optimizing transportation resource allocation and enhancing urban mobility efficiency. However, traffic data generated daily are vast and complex, involving dynamic and intricate changes in the traffic road network and traffic flow. Therefore, real-time and accurate prediction of traffic flow is a challenging task that requires modeling the intricate spatial–temporal dynamics of traffic data. In this paper, we propose a novel approach for traffic flow prediction, based on a Multi-Scale Residual Graph Convolution Network with hierarchical attention. First, we design a novel encoder–decoder with multi-independent channels to capture traffic flow information from different time scales and diverse temporal dependencies. Second, we employ a coupled graph convolution network with residual graph attention to dynamically learn the varying spatial features among and within traffic stations. Third, we utilize channel attention to fuse the multi-scale spatial–temporal dependencies and accurately predict traffic flow. We evaluate the proposed approach on multiple benchmark datasets, and the experimental results demonstrate its superior performance compared to state-of-the-art approaches in terms of various metrics.

show abstract

A C4.5 Fuzzy Decision Tree Method for Multivariate Time Series Forecasting

Cited by 5 publications

References 35 publications

A Tutorial on Fuzzy Time Series Forecasting Models: Recent Advances and Challenges

A Tutorial on Fuzzy Time Series Forecasting Models: Recent Advances and Challenges

Untitled

A Multi-Scale Residual Graph Convolution Network with hierarchical attention for predicting traffic flow in urban mobility

Contact Info

Product

Resources

About