An Intent Prediction System for Wireless Fidelity (Wi-Fi) Sensor Data Using an Emerging Neural Machine Intelligence

Abstract

This paper presents a User Intent (UI) mining scheme based on an emerging neural machine intelligence technique called the Neuronal Auditory Machine Intelligence (NeuroAMI) and considering a Wi-Fi sessions dataset containing about 8000 data points for intent prediction. The results of simulations considering graded increase in samples from 50samples to 999samples showed that substantial increases in accuracies are achievable. When compared to the Long Short-Term Memory (LSTM) method, the results showed that the NeuroAMI will outperform the LSTM by a factor of about 2 considering percentage accuracies.

Introduction

Machine learning (ML), inspired by neural and animal behaviors, is widely used for classification, regression, and forecasting problems. Deep Learning (DL), an advanced neural network approach, is prominent in industry and academia. Intent prediction—predicting likely actions of humans or devices—is a critical ML problem, with User Intention (UI) mining emerging as a key research area offering applications in web opinion analysis, product offers, search services, and smart retail navigation.

Recent UI mining research employs neural machine intelligence and online learning with adaptable classifiers, validated through simulations, particularly for wireless sensor data in smart retail.

Several studies apply various ML techniques like Deep LSTM, CNN, Capsule Networks, and hybrid models for intent detection in contexts including driver behavior, pedestrian path prediction, prosthesis control, and chatbot interaction. Methods often combine classifiers, use attention mechanisms, or integrate sensor data (e.g., EMG, kinematic) to improve accuracy. Genetic algorithms, fuzzy classifiers, and Bayesian models have also been explored.

Intent prediction also extends to retail, with association rule mining and frequent pattern mining used to predict customer purchase and revisit behavior based on sensor and Wi-Fi data.

One notable dataset used in experiments involves real-world Wi-Fi sensor data from Korean supermarkets to predict customer revisit intention, leveraging location and session data from customers’ Wi-Fi-enabled devices.

Conclusion

In this research, an emerging machine intelligence approach is proposed for the online classification of wireless (Wi-Fi) sensor device data. The approach is based on an emerging neural approach (NeuroAMI) inspired by auditory sensations in the mammalian brain. The approach has shown promising results for both low Wi-Fi sensor training sample sizes and larger samples. The approach has also been compared to a state-of-the-art deep learning technique – the LSTM and showed superior performances. With the proposed solution, it is possible to investigate other wireless sensor data domains particularly in the area of security and anomaly detection in case of privacy breaches. This remains an area for future studies.

References

[1] Casas, S., Luo, W., & Urtasun, R. (2018, October). Intentnet: Learning to predict intention from raw sensor data. In Conference on Robot Learning (pp. 947-956). [2] Ertler, C., Possegger, H., Opitz, M., & Bischof, H. (2018, November). An Intent-Based Automated Traffic Light for Pedestrians. In 2018 15th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) (pp. 1-6). IEEE. [3] Hussein, S. E., & Granat, M. H. (2002). Intention detection using a neuro-fuzzy EMG classifier. IEEE Engineering in Medicine and Biology Magazine, 21(6), 123-129. [4] Im, I., Dunn, B. K., Lee, D. I., Galletta, D. F., & Jeong, S. O. (2019). Predicting the intent of sponsored search users: An exploratory user session-level analysis. Decision Support Systems, 121, 25-36. [5] Jagadish, K., & Sahini, S. (2013). Predicting Missing Items In Shopping Cart Using Associative Classification Mining. International Journal Of Computer Science And Mobile Computing, 2(11). [6] Kim, S., & Lee, J. G. (2018, November). Utilizing in-store sensors for revisit prediction. In 2018 IEEE International Conference on Data Mining (ICDM) (pp. 217-226). IEEE. [7] Kyeong, S., Shin, W., & Kim, J. (2018, July). Predicting Walking Intentions using sEMG and Mechanical sensors for various environment. In 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (pp. 4414-4417). IEEE. [8] Lin, T. Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017). Focal loss for dense object detection. In Proceedings of the IEEE international conference on computer vision (pp. 2980-2988). [9] Lin, T. E., & Xu, H. (2019). A post-processing method for detecting unknown intent of dialogue system via pre-trained deep neural network classifier. Knowledge-Based Systems, 186, 104979. [10] Liu, M., Luo, S., Han, K., DeMara, R. F., & Bai, Y. (2022). Autonomous binarized focal loss enhanced model compression design using tensor train decomposition. Micromachines, 13(10), 1738. [11] Lin, T. E., & Xu, H. (2019). A post-processing method for detecting unknown intent of dialogue system via pre-trained deep neural network classifier. Knowledge-Based Systems, 186, 104979. [12] Osegi, E. N., & Anireh, V. (2020). AMI: an auditory machine intelligence algorithm for predicting sensory-like data. Computer Science, 5(2), 71-89. [13] Osegi, E. N. (2023). Neuronal Auditory Machine Intelligence (Neuro-AMI) In Perspective. http://dx.doi.org/10.13140/RG.2.2.10131.45605. [14] Ravichandar, H. C., & Dani, A. (2017). Intention inference for human-robot collaboration in assistive robotics. In Human Modelling for Bio-Inspired Robotics (pp. 217-249). Academic Press. [15] Saleh, K., Hossny, M., & Nahavandi, S. (2018). Intent prediction of pedestrians via motion trajectories using stacked recurrent neural networks. IEEE Transactions on Intelligent Vehicles, 3(4), 414-424. [16] Srivatsan, M., Kumar, S., & Sri Venkateswara, V. V. (2011). Predicting missing items in shopping carts using fast algorithm. [17] Stojanov, P., Awadallah, A. H., Bennett, P., & Hosseini, S. (2019, September). On Domain Transfer When Predicting Intent in Text. In Workshop on Document Intelligence at NeurIPS 2019. [18] Tang, J., Yu, S., Liu, F., Chen, X., & Huang, H. (2019). A hierarchical prediction model for lane-changes based on combination of fuzzy C-means and adaptive neural network. Expert systems with applications, 130, 265-275. [19] Tran, O. T., & Luong, T. C. (2020). Understanding what the users say in chatbots: A case study for the Vietnamese language. Engineering Applications of Artificial Intelligence, 87, 103322. [20] Varol, H. A., Sup, F., & Goldfarb, M. (2009). Multiclass real-time intent recognition of a powered lower limb prosthesis. IEEE Transactions on Biomedical Engineering, 57(3), 542-551. [21] Wu, S., Zhang, Q., Chen, W., Liu, J., & Liiu, L. (2019). Research on trend prediction of internet user intention understanding and public intelligence mining based on fractional differential method. Chaos, Solitons & Fractals, 128, 331-338. [22] Xu, P., & Sarikaya, R. (2013, December). Convolutional neural network based triangular crf for joint intent detection and slot filling. In 2013 ieee workshop on automatic speech recognition and understanding (pp. 78-83). IEEE. [23] Young, A. J., Simon, A. M., Fey, N. P., & Hargrove, L. J. (2014a). Intent recognition in a powered lower limb prosthesis using time history information. Annals of biomedical engineering, 42(3), 631-641. [24] Young, A. J., Kuiken, T. A., & Hargrove, L. J. (2014b). Analysis of using EMG and mechanical sensors to enhance intent recognition in powered lower limb prostheses. Journal of neural engineering, 11(5), 056021.

Copyright

Copyright © 2025 Collins Iyaminapu Iyoloma, Solomon Malcolm Ekolama, Tamunotonye Sotonye Ibanibo. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.