WillRo: A Deep Learning App for Potential Phishing Threat Detection

Willy Sotelo; Alvaro Roque; Sandra Wong-Durand; Pedro Castaneda; Alejandra Onate-Andino

doi:10.48084/etasr.14161

WillRo: A Deep Learning App for Potential Phishing Threat Detection

Willy Sotelo
, Alvaro Roque
, Sandra Wong-Durand
, Pedro Castaneda
, Alejandra Onate-Andino

Universidad Peruana de Ciencias Aplicadas
Universidad Nacional Toribio Rodríguez de Mendoza de Amazonas
Escuela Superior Politécnica de Chimborazo

Research output: Contribution to journal › Article › peer-review

Abstract

This article presents an application called WillRo App, designed to detect potential phishing by analyzing website screenshots in real time. The system integrated Robotic Process Automation (RPA) to capture screenshots, and the YOLOv5 deep learning model in order to classify phishing and no-phishing content. The results demonstrated a precision of 85.80%, a recall of 93.00%, a [email protected] of 66.60%, and a [email protected] -0.95 of 32.70%. These values showed a reliable detection performance, making WillRo a possible model for phishing detection. Future work should focus on improving the model with additional features to increase its accuracy.

Original language	English
Pages (from-to)	30591-30598
Number of pages	8
Journal	Engineering, Technology and Applied Science Research
Volume	15
Issue number	6
DOIs	https://doi.org/10.48084/etasr.14161
State	Published - 8 Dec 2025
Externally published	Yes

Keywords

Robotic Process Automation (RPA)
deep learning
malicious content
phishing detection

Access to Document

10.48084/etasr.14161

Cite this

@article{fb1a18bfdb134573af9d35feb5ad43bf,

title = "WillRo: A Deep Learning App for Potential Phishing Threat Detection",

abstract = "This article presents an application called WillRo App, designed to detect potential phishing by analyzing website screenshots in real time. The system integrated Robotic Process Automation (RPA) to capture screenshots, and the YOLOv5 deep learning model in order to classify phishing and no-phishing content. The results demonstrated a precision of 85.80\%, a recall of 93.00\%, a [email protected] of 66.60\%, and a [email protected] -0.95 of 32.70\%. These values showed a reliable detection performance, making WillRo a possible model for phishing detection. Future work should focus on improving the model with additional features to increase its accuracy.",

keywords = "Robotic Process Automation (RPA), deep learning, malicious content, phishing detection",

author = "Willy Sotelo and Alvaro Roque and Sandra Wong-Durand and Pedro Castaneda and Alejandra Onate-Andino",

note = "Publisher Copyright: {\textcopyright} (c) by the authors",

year = "2025",

month = dec,

day = "8",

doi = "10.48084/etasr.14161",

language = "Ingl{\'e}s",

volume = "15",

pages = "30591--30598",

journal = "Engineering, Technology and Applied Science Research",

issn = "2241-4487",

publisher = "Dr D. Pylarinos",

number = "6",

}

TY - JOUR

T1 - WillRo

T2 - A Deep Learning App for Potential Phishing Threat Detection

AU - Sotelo, Willy

AU - Roque, Alvaro

AU - Wong-Durand, Sandra

AU - Castaneda, Pedro

AU - Onate-Andino, Alejandra

N1 - Publisher Copyright: © (c) by the authors

PY - 2025/12/8

Y1 - 2025/12/8

N2 - This article presents an application called WillRo App, designed to detect potential phishing by analyzing website screenshots in real time. The system integrated Robotic Process Automation (RPA) to capture screenshots, and the YOLOv5 deep learning model in order to classify phishing and no-phishing content. The results demonstrated a precision of 85.80%, a recall of 93.00%, a [email protected] of 66.60%, and a [email protected] -0.95 of 32.70%. These values showed a reliable detection performance, making WillRo a possible model for phishing detection. Future work should focus on improving the model with additional features to increase its accuracy.

AB - This article presents an application called WillRo App, designed to detect potential phishing by analyzing website screenshots in real time. The system integrated Robotic Process Automation (RPA) to capture screenshots, and the YOLOv5 deep learning model in order to classify phishing and no-phishing content. The results demonstrated a precision of 85.80%, a recall of 93.00%, a [email protected] of 66.60%, and a [email protected] -0.95 of 32.70%. These values showed a reliable detection performance, making WillRo a possible model for phishing detection. Future work should focus on improving the model with additional features to increase its accuracy.

KW - Robotic Process Automation (RPA)

KW - deep learning

KW - malicious content

KW - phishing detection

UR - https://www.scopus.com/pages/publications/105026864829

U2 - 10.48084/etasr.14161

DO - 10.48084/etasr.14161

M3 - Artículo

AN - SCOPUS:105026864829

SN - 2241-4487

VL - 15

SP - 30591

EP - 30598

JO - Engineering, Technology and Applied Science Research

JF - Engineering, Technology and Applied Science Research

IS - 6

ER -

WillRo: A Deep Learning App for Potential Phishing Threat Detection

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this