Imane Moustati

The Internet of Behaviors (IoB) emerged as the next frontier in data-driven innovation, leveraging the vast amount of behavioral data collected through Internet of Things (IoT) devices and online to understand, predict, and influence human behavior. It holds enormous advantages in shaping individuals' daily lives. The IoB has the potential to optimize processes, personalize experiences, and fundamentally reshape people's interactions with technology; however, the widespread adoption of the IoB also faces challenges. This paper offers a thorough state-of-the-art snapshot of the IoB research area, examines multiple potential use-case application scenarios of the IoB across various industries and disciplines, and presents critical ethical and technical challenges that should be addressed.

IoT systems have provided humans with incredible benefits. However, its features related to analyzing collected data and automatically executing established guidelines restrict them from comprehending the intent of these data, retrieving human intention, or understanding behaviors. The Internet of Behaviors (IoB) could be considered an ecosystem that combines technology, advanced data analysis, edge analytics, and behavioral science to aggregate, analyze, and comprehend human data gathered from IoT devices or online utilizing human psychology. Then, this understanding can be applied to change or affect human behavior. In this work, we present an in-depth review of the latest research in the IoB field and comprehensively describe the IoB, its fundamental features, and its relationship with IoT and the DKIW pyramid. Then, we examine key distinctions between IoB and IoT and delve into the IoB architecture and core enabling technologies. Furthermore, we comprehensively identify the major challenges to be addressed by the IoB framework, notably in terms of ethics, security, and privacy.

Effective investment decision-making in today’s volatile financial market demands the integration of advanced predictive analytics, alternative data sources, and behavioral insights. This paper introduces an innovative Internet of Behaviors (IoB) ecosystem that integrates real-time data acquisition, advanced feature engineering, predictive modeling, explainability, automated portfolio management, and an intelligent decision support engine to enhance financial decision-making. Our framework effectively captures complex temporal dependencies in financial data by combining robust technical indicators and sentiment-driven metrics—derived from BERT-based sentiment analysis—with a multi-layer LSTM forecasting model. To enhance the model’s performance and transparency and foster user trust, we apply XAI methods, namely, TimeSHAP and TIME. The IoB ecosystem also proposes a portfolio management engine that translates the predictions into actionable strategies and a continuous feedback loop, enabling the system to adapt and refine its strategy in real time. Empirical evaluations demonstrate the effectiveness of our approach: the LSTM forecasting model achieved an RMSE of 0.0312, an MAE of 0.0250, an MSE of 0.0010, and a directional accuracy of 95.24% on TSLA stock returns. Furthermore, the portfolio management algorithm successfully transformed an initial balance of USD 15,000 into a final portfolio value of USD 21,824.12, yielding a net profit of USD 6824.12. These results highlight the potential of IoB-driven methodologies to revolutionize financial services by enabling more personalized, transparent, and adaptive investment solutions.

JAIT is a scientific open access journal which focuses on empirical research results and critical analysis of technology development, use, management and impacts in information technology.