Knowledge Extraction and Discovery about Web System Based on the Benchmark Application of Online Stock Trading System

Borowiec, Marcin; Piszko, Rafał; Rak, Tomasz

doi:10.3390/s23042274

Cited by 1 publication

(6 citation statements)

References 32 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using data mining techniques, they identify user behavior and patterns from weblogs, albeit without employing ML techniques. Meanwhile, Piszko et al [2] perform an in-depth exploratory data analysis of a benchmark application log dataset, focusing on a stock market web application using statistical methods and data visualization techniques. In the same way, Meng et al [15] evaluate the efficiency of different ML classifiers in predicting user behavior based on weblogs, utilizing datasets similar to our study.…”

Section: Related Workmentioning

confidence: 99%

“…The experimental environment of the OSTS stock exchange system, discussed in [2], was used to obtain the system logs. It consists of two applications: a scalable stock exchange application (APP1) and an application generating traffic on the stock exchangean automatic client (APP2).…”

Section: Experimental Environmentmentioning

confidence: 99%

“…All these events were recorded in databases (of both applications) and, at the end of the simulation, saved in the form of logs (converted to CSV format). The types of queries generated by APP2 are discussed in detail in [2]:…”

Section: Ostsmentioning

confidence: 99%

“…These logs are generated during the operation of a benchmark that tests the hardware architecture on which the application has been deployed. The exploratory data analysis (EDA), which allows understanding the characteristics of the data through their visualization and statistical analysis, has been presented in [2].…”

Section: Introductionmentioning

confidence: 99%

“…Diving into data analysis [2]. • Contrasting the employed classification techniques and studying their efficacy.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Advanced Examination of User Behavior Recognition via Log Dataset Analysis of Web Applications Using Data Mining Techniques

Borowiec,

Rak

2023

Electronics

Self Cite

View full text Add to dashboard Cite

As web systems based on containerization increasingly attract research interest, the need for effective analytical methods has heightened, with an emphasis on efficiency and cost reduction. Web client simulation tools have been utilized to further this aim. While applying machine learning (ML) methods for anomaly detection in requests is prevalent, predicting patterns in web datasets is still a complex task. Prior approaches incorporating elements such as URLs, content from web pages, and auxiliary features have not provided any satisfying results. Moreover, such methods have not significantly improved the understanding of client behavior and the variety of request types. To overcome these shortcomings, this study introduces an incremental approach to request categorization. This research involves an in-depth examination of various established classification techniques, assessing their performance on a selected dataset to determine the most effective model for classification tasks. The utilized dataset comprises 8 million distinct records, each defined by performance metrics. Upon conducting meticulous training and testing of multiple algorithms from the CART family, Extreme Gradient Boosting was deemed to be the best-performing model for classification tasks. This model outperforms prediction accuracy, even for unrecognized requests, reaching a remarkable accuracy of 97% across diverse datasets. These results underline the exceptional performance of Extreme Gradient Boosting against other ML techniques, providing substantial insights for efficient request categorization in web-based systems.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Experimental Environmentmentioning

confidence: 99%

Section: Ostsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…Diving into data analysis [2]. • Contrasting the employed classification techniques and studying their efficacy.…”

mentioning

confidence: 99%