Incremental Missing-Data Imputation for Evolving Fuzzy Granular Prediction

Garcia, Cristiano Mesquita; Leite, Daniel; Škrjanc, Igor

doi:10.1109/tfuzz.2019.2935688

Cited by 58 publications

(16 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…An incremental model (i.e. model that learns onthe-fly, without retraining [50], [51]) would be the ideal for this scenario and that sort of model could be more deeply studied.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Data Mining applied on Web Robots Detection: A Systematic Mapping

Abílio¹,

Garcia²,

Fernandes³

2021

Anais Do 15. Congresso Brasileiro De Inteligência Computacional

Self Cite

View full text Add to dashboard Cite

Browsing on Internet is part of the world population’s daily routine. The number of web pages is increasing and so is the amount of published content (news, tutorials, images, videos) provided by them. Search engines use web robots to index web contents and to offer better results to their users. However, web robots have also been used for exploiting vulnerabilities in web pages. Thus, monitoring and detecting web robots’ accesses is important in order to keep the web server as safe as possible. Data Mining methods have been applied to web server logs (used as data source) in order to detect web robots. Then, the main objective of this work was to observe evidences of definition or use of web robots detection by analyzing web server-side logs using Data Mining methods. Thus, we conducted a systematic Literature mapping, analyzing papers published between 2013 and 2020. In the systematic mapping, we analyzed 34 studies and they allowed us to better understand the area of web robots detection, mapping what is being done, the data used to perform web robots detection, the tools, and algorithms used in the Literature. From those studies, we extracted 33 machine learning algorithms, 64 features, and 13 tools. This study is helpful for researchers to find machine learning algorithms, features, and tools to detect web robots by analyzing web server logs.

show abstract

“…An incremental model (i.e. model that learns onthe-fly, without retraining [50], [51]) would be the ideal for this scenario and that sort of model could be more deeply studied.…”

Section: Discussionmentioning

confidence: 99%

“…As future work, we suggest: the study of incremental classification methods [50], [51]; the definition of sets of features that could be selected according to web server logs; and the development of a (visual) tool to help the analysis of web robots requests.…”

Section: Discussionmentioning

confidence: 99%

Data Mining applied on Web Robots Detection: A Systematic Mapping

Abílio¹,

Garcia²,

Fernandes³

2021

Anais Do 15. Congresso Brasileiro De Inteligência Computacional

Self Cite

View full text Add to dashboard Cite

show abstract

“…Last but not least, the final category consists of fuzzy imputation techniques such as the evolving granular fuzzy-rule-based model [31], [32], fuzzy c-means clustering [16], fuzzy clustering-based EM imputation [6], decision trees and fuzzy clustering with iterative learning (DIFC) [33], and imputation using fuzzy neighborhood density-based clustering [19].…”

Section: Related Workmentioning

confidence: 99%

EPEM: Efficient Parameter Estimation for Multiple Class Monotone Missing Data

Nguyen

et al. 2021

Information Sciences

View full text Add to dashboard Cite

The missing data problem has been broadly studied in the last few decades and has various applications in different areas such as statistics or bioinformatics.Even though many methods have been developed to tackle this challenge, most of those are imputation techniques that require multiple iterations through the data before yielding convergence. In addition, such approaches may introduce extra biases and noises to the estimated parameters. In this work, we propose novel algorithms to find the maximum likelihood estimates (MLEs) for a oneclass/multiple-class randomly missing data set under some mild assumptions.As the computation is direct without any imputation, our algorithms do not require multiple iterations through the data, thus promising to be less timeconsuming than other methods while maintaining superior estimation performance. We validate these claims by empirical results on various data sets of different sizes and release all codes in a GitHub repository to contribute to the research community related to this problem.

show abstract

“…FBeMé baseado no conceito de cobertura (granulação fuzzy) do espaço dos dados. Suas regras são interpretáveis linguisticamente e, portanto,úteis para auxílioà tomada de decisão (Garcia et al, 2019).…”

Section: Introductionunclassified

“…O método não requer dados a priori para aprender. Regras e grânulos de informação fuzzy são criados dinamicamente e adaptados ao longo do tempo (Garcia et al, 2019). Para cada grânulo existe uma regra correspondente.…”

unclassified

Classificadores Não-Supervisionados baseados em Conjuntos Fuzzy e Nuvens de Dados para Detecção de Spam

Soares

Garcia

Poucas

et al. 2019

Anais Do 14º Simpósio Brasileiro De Automação Inteligente

Self Cite

View full text Add to dashboard Cite

Technological advancements have made individuals and organizations more dependent on e-mails to communicate and share information. The increasing use of e-mails has led to an increased production of unsolicited commercial messages, known as spam. Spam classification systems able to self-adapt over time, with no human intervention, are rare. Adaptation is interesting as spams vary over time due to the use of different message-masking techniques. Moreover, classification models that handle large volumes of data are essential. Evolving intelligent systems are able to adapt their parameters and structure according to the data stream. This study applies the evolving methods TEDA (Typicality and Eccentricity based Data Analytics) and FBeM (Fuzzy Set-Based Evolving Modeling) for online unsupervised classification of spams. TEDA and FBeM are compared in terms of accuracy, model compactness, and processing time. For dimensionality reduction, a non-parametric Spearman-correlation-based feature selection method is employed. A dataset containing 25,745 samples, being 7,830 spams and 17,915 legitimate e-mails, is considered. 711 features extracted from an e-mail server describe each sample. Resumo: O avanço de tecnologias tem tornado indivíduos e organizações mais dependentes de e-mails para comunicação e compartilhamento de informação. O uso crescente de e-mail tem levadoà produção de mensagens comerciais não-solicitadas, conhecidas como spam. Sistemas de classificação de spams capazes de se adaptar ao longo do tempo, sem intervenção humana, são raros. A adaptaçãoé interessante já que spams variam no tempo devido ao uso de diferentes técnicas de mascaramento de mensagens. Além disso, modelos classificadores que lidam com grandes volumes de dados são essenciais. Sistemas inteligentes evolutivos são capazes de adaptar seus parâmetros e estrutura de acordo com o fluxo de dados. Este estudo aplica os métodos evolutivos TEDA (Typicality and Eccentricity based Data Analytics) e FBeM (Fuzzy Set-Based Evolving Modeling) para classificação não-supervisionada de spam. TEDA e FBeM são comparados em termos de acurácia, compactação do modelo, e tempo de processamento. Para redução da dimensionalidade, um método não-paramétrico baseado em correlação de Spearmań e empregado. Uma base de dados contendo 25745 amostras, sendo 7830 spams e 17915 e-mails legítimos, foi elaborada. 711 atributos extraídos de um servidor descrevem as amostras.

show abstract

Incremental Missing-Data Imputation for Evolving Fuzzy Granular Prediction

Cited by 58 publications

References 43 publications

Data Mining applied on Web Robots Detection: A Systematic Mapping

Data Mining applied on Web Robots Detection: A Systematic Mapping

EPEM: Efficient Parameter Estimation for Multiple Class Monotone Missing Data

Classificadores Não-Supervisionados baseados em Conjuntos Fuzzy e Nuvens de Dados para Detecção de Spam

Contact Info

Product

Resources

About