Publication Date

Fall 2025

Degree Type

Master's Project

Degree Name

Master of Science in Computer Science (MSCS)

Department

Computer Science

First Advisor

Sayma Akther

Second Advisor

William Andreopoulos

Third Advisor

Nithish Kumar Reddy Rajapuram

Keywords

App recommendation, behavioral modeling, context-aware computing, deep learning, intelligent assistants, mobile sensing, student life analytics.

Abstract

Wearable sensors and smartphones continuously generate lots of data that are reflective of various aspects of human behavior. Yet, these behavioral signals are rarely translated into meaningful proactive digital support. This study introduces an AI-powered context prediction framework that predicts the next activity that a user will perform and suggests mobile applications matching the predicted context. This system uses both classical machine learning models and deep learning models for analyzing multi-modal behavioral histories, which are collected from heterogeneous sensors such as accelerometers, GPS, and screen usage logs. Contrary to traditional recommendation systems that respond to past usage patterns, the proposed approach predicts upcoming human behavior for triggering timely and personalized app recommendations that enhance productivity, health, and overall well-being. The project uses four datasets: StudentLife, College Life Experience, custom smartwatch dataset, and MobileRec. The first three provide rich real-world sensor data for context modeling, whereas MobileRec provides app metadata for semantic retrieval and recommendation. Integrating those datasets allows a leap from passive sensing toward proactive behavioral guidance. Experimental results demonstrate that predictive context modeling together with semantic retrieval substantially improves accuracy and contextual relevance of recommendations compared to state-of-the-art baselines. The findings emphasize the potential of anticipatory, context-aware assistants that are able to adapt in an intelligent manner to users' goals and day-to-day life.

Download

Available for download on Saturday, December 19, 2026

Share

COinS