Browsing by Author "Nazir, Amril"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item A Comparative Study of Autoregressive and Neural Network Models: Forecasting the GARCH Process(Springer Science and Business Media Deutschland GmbH, 2022) Kamalov, Firuz; Gurrib, Ikhlaas; Moussa, Sherif; Nazir, AmrilThe Covid-19 pandemic has highlighted the importance of forecasting in managing public health. The two of the most commonly used approaches for time series forecasting methods are autoregressive (AR) and deep learning models (DL). While there exist a number of studies comparing the performance of AR and DL models in specific domains, there is no work that analyzes the two approaches in the general context of theoretically simulated time series. To fill the gap in the literature, we conduct an empirical study using different configurations of generalized autoregressive conditionally heteroskedastic (GARCH) time series. The results show that DL models can achieve a significant degree of accuracy in fitting and forecasting AR-GARCH time series. In particular, DL models outperform the AR-based models over a range of parameter values. However, the results are not consistent and depend on a number of factors including the DL architecture, AR-GARCH configuration, and parameter values. The study demonstrates that DL models can be an effective alternative to AR-based models in time series forecasting. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.Item Comparative analysis of activation functions in neural networks(Institute of Electrical and Electronics Engineers Inc., 2021) Kamalov, Firuz; Nazir, Amril; Safaraliev, Murodbek; Cherukuri, Aswani Kumar; Zgheib, RitaAlthough the impact of activations on the accuracy of neural networks has been covered in the literature, there is little discussion about the relationship between the activations and the geometry of neural network model. In this paper, we examine the effects of various activation functions on the geometry of the model within the feature space. In particular, we investigate the relationship between the activations in the hidden and output layers, the geometry of the trained neural network model, and the model performance. We present visualizations of the trained neural network models to help researchers better understand and intuit the effects of activation functions on the models. © 2021 IEEE.Item Suspicious Behavior Detection with Temporal Feature Extraction and Time-Series Classification for Shoplifting Crime Prevention(Multidisciplinary Digital Publishing Institute (MDPI), 2023-07) Nazir, Amril; Mitra, Rohan; Sulieman, Hana; Kamalov, FiruzThe rise in crime rates in many parts of the world, coupled with advancements in computer vision, has increased the need for automated crime detection services. To address this issue, we propose a new approach for detecting suspicious behavior as a means of preventing shoplifting. Existing methods are based on the use of convolutional neural networks that rely on extracting spatial features from pixel values. In contrast, our proposed method employs object detection based on YOLOv5 with Deep Sort to track people through a video, using the resulting bounding box coordinates as temporal features. The extracted temporal features are then modeled as a time-series classification problem. The proposed method was tested on the popular UCF Crime dataset, and benchmarked against the current state-of-the-art robust temporal feature magnitude (RTFM) method, which relies on the Inflated 3D ConvNet (I3D) preprocessing method. Our results demonstrate an impressive 8.45-fold increase in detection inference speed compared to the state-of-the-art RTFM, along with an F1 score of 92%,outperforming RTFM by 3%. Furthermore, our method achieved these results without requiring expensive data augmentation or image feature extraction. © 2023 by the authors.