Natural Language interface with household devices

Author

Ashadeepthi Kobbari, Southern Illinois University CarbondaleFollow

Date of Award

5-1-2025

Degree Name

Master of Science

Department

Computer Science

First Advisor

Henry H, Hexmoor

Abstract

The advent of smart homes has transformed everyday living by integrating household deviceswith advanced automation systems. A significant challenge lies in enabling seamless, natural interactions between users and these devices. This research investigates the design and implementation of natural language interfaces (NLI) for household device control, focusing on two distinct approaches: persona-based and command-based interactions.Through a comparative study, this thesis explores the trade-offs between user engagement, efficiency, and system performance in natural language interactions. The persona-based model provides an immersive, conversational interface, offering users a more human-like interaction with smart devices, while the command-based model prioritizes straightforward, task-oriented commands, optimizing for speed and simplicity. The study evaluates both models based on user satisfaction, task completion time, and overall system effectiveness. Experimental data from user testing and surveys will be analyzed to determine which model provides a superior user experience and better meets the needs of a smart home environment. The results aim to provide insights into the ideal design for natural language interfaces in household device control, with implications for future smart home systems and human-computer interaction.This research contributes to the field of human-computer interaction and smart home technology by providing a comprehensive comparison of interaction models, offering practical recommendations for the development of next-generation NLI systems in household automation.