Student Research

Abstract

DeepSeek is a generative AI tool that can answer questions, solve problems, and formulate information’s necessary for the topic. This study determined the perceived usability and learning potential of DeepSeek among Grade 7 students at Marigondon National High School during the 2024 - 2025 academic year. It examined perceived usability (academic value, ease of use), feedback response (appearance, language structure, pro mpt response), and future potentials in (assignments, experiments, class reporting, and projects). Using a descriptive - correlational design employing a survey questionnaire, data revealed that DeepSeek is highly effective in Science learning with appealing visuals and concise responses. Also, it is perceived a great potential in the future. This also showed a significant relationship with a (p=0.00) between usability and feedback response, indicating that visual quality affects the perceived usefulness of D eepSeek as Robo Assistant in Science 7 learning. Thus, DeepSeek can enhance Science Education, recommending its integration as a Robo Assistant. Schools should explore its implementation to support student academic performance. However, proper ethical cons iderations are a must in using AI chatbots like DeepSeek in education.

Keywords

DeepSeek, Robo Assistant, Perceived Usability, and Future Potent

Research Information

Classification
Thesis
Research Type
Student Research
Year Completed
2025
DOI
Not specified
Publisher
Not specified
Pages
Not specified
ISSN / ISBN
Not specified

Sustainable Development Goals

SDG 4: Quality Education

Department / Program

Bachelor of Secondary Education Major in Science

Cordova Public College