Articles by Lee CP's Profile | PeerJ Journalist

Cross-domain sentiment classification in Slovak: a comparative analysis of transformer-based models

Mar 17, 2026 |

| PeerJ

The sentiment classification task has gained increasing importance across various domains (for instance finance, social media, movies, politics, and product reviews) due to the abundance of textual data and its core objective is to determine the emotional tone of a text. This emotional tone can be analyzed at different levels of granularity. The simplest taxonomies divide sentiment into positive sentiment and negative sentiment.

Open in Who Shared

Two-hand static and dynamic Arabic sign language recognition using keypoints and shape descriptors with attention-driven feature fusion

Nov 26, 2025 |

By Lee CP, Al Abdullah BA, Zarnab Kausar, Shaheryar Najam

| PeerJ

Hearing loss is a growing global issue, with the World Health Organization projecting that 2.5 billion people will have some level of hearing impairment by 2050, increasing the need for sign language communication (World Health Organization, 2021). Sign languages like American Sign Language (ASL), Chinese Sign Language (CSL), and Arabic Sign Language (ArSL) have unique linguistic structures and visual vocabularies.

Open in Who Shared

GaitTriViT and GaitVViT: Transformer-based methods emphasizing spatial or temporal aspects in gait recognition

Aug 20, 2025 |

By Lee CP, Hongyun Sheng, Wei S-E, Breckon TP

| PeerJ

Gait is defined as the physical and behavioral biological characteristics exhibited by a human when walking upright, and it can be used to describe an individual’s walking pattern. Gait recognition is the technology that identifies individuals based on their distinct walking patterns.

Open in Who Shared

Topic adversarial neural network for cross-topic cyberbullying detection

Jun 19, 2025 |

By Lee CP, Shufeng Xiong, Henan Agricultural, Wenzhuo Liu

| PeerJ

With the rapid development of social media, online platforms have reshaped not only how people access information and express opinions, but also become fertile ground for the emergence of harmful online behavior (Oksanen et al., 2022). Cyberbullying is one of the most prominent and pervasive issues. It refers to the intentional and repeated use of language in digital spaces to insult, demean, threaten, or humiliate others.

Open in Who Shared

A majority voting framework for reliable sentiment analysis of product reviews

Feb 28, 2025 |

By Noah SAM, Kim RY, Lee CP, Wai Chung Y

| PeerJ

Businesses increasingly rely on sentiment analysis to understand customer preferences, perform market analysis, monitor brand reputation, and gauge customer experiences, making it a powerful tool for decision-making in competitive markets (Geetha & Karthika Renuka, 2021). While originally intended as a feedback tool to help companies improve, product reviews now also influence consumer perception and drive sales rankings (Catelli et al., 2022; Kim, 2024).

Open in Who Shared