Mohammad Reza Chalak Qazani

Dr. Mohammad Reza Chalak Qazani

Faculty of Computing and Information Technology, Sohar University, Sohar, Oman
orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 May 2024 in IEEE Access
Verified
Journal Article
Open access
APAli Peivandizadeh
Ali Peivandizadeh
SHSima Hatami
Sima Hatami
ANAmirhossein Nakhjavani
Amirhossein Nakhjavani
LKLida Khoshsima
Lida Khoshsima
Mohammad Reza Chalak QazaniMohammad Reza Chalak Qazani
Mohammad Reza Chalak Qazani
RARoohallah Alizadehsani
Roohallah Alizadehsani
In an era dominated by digital communication, the vast amounts of data generated from social media and financial markets present unique opportunities and challenges for forecasting stock market prices. This paper proposes an innovative approach that harnesses the power of social media sentiment analysis combined with stock market data to predict stock prices, directly addressing the critical challenges in this domain. A major challenge in sentiment analysis is the uneven distribution of data across different sentiment categories. Traditional models struggle to accurately identify fewer common sentiments (minority class) due to the overwhelming presence of more common sentiments (majority class). To tackle this, we introduce the Off-policy Proximal Policy Optimization (PPO) algorithm, specifically designed to handle class imbalance by adjusting the reward mechanism in the training phase, thus favoring the correct classification of minority class instances. Another challenge is effectively integrating the temporal dynamics of stock prices with sentiment analysis results. Our solution is implementing a Transductive Long Short-Term Memory (TLSTM) model that incorporates sentiment analysis findings with historical stock data. This model excels at recognizing temporal patterns and gives precedence to data points that are temporally closer to the prediction point, enhancing the prediction accuracy. Ablation studies confirm the effectiveness of the Off-policy PPO and TLSTM components on the overall model performance. The proposed approach advances the field of financial analytics by providing a more nuanced understanding of market dynamics but also offers actionable insights for investors and policymakers seeking to navigate the complexities of the stock market with greater precision and confidence.

24 December 2019 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
Verified
Journal Article
Closed access
Mohammad Reza Chalak QazaniMohammad Reza Chalak Qazani
Mohammad Reza Chalak Qazani
Houshyar AsadiHoushyar Asadi
Houshyar Asadi
SKSuiyang Khoo
Suiyang Khoo
SNSaeid Nahavandi
Saeid Nahavandi
The hexapod manipulator is the most common motion platform, which is widely used as a simulation-based motion platform (SBMP). As the hexapod manipulator has a limited workspace, it is not physically possible to regenerate the real vehicle motion signals using the SBMP. The motion cueing algorithm (MCA) is responsible for regenerating a realistic vehicle motion sensation for the user when the SBMP operates within its physical and dynamical limitations. Recently, model predictive control (MPC) has been introduced to extract the optimal input motion signals while considering the SBMP limitations in the Cartesian coordinate space, which leads to a linear time-invariant (LTI) MPC-based MCA methods. Unfortunately, the existing LTI MPC-based MCA methods are still not able to consider the parameters of the SBMP's hexapod mechanisms inside their models. In general, the current studies only consider the constraints in the Cartesian coordinate system of the hexapod mechanism, instead of its design parameters. This consideration results in a poor usage of the hexapod workspace due to the conservative assumptions; consequently, the SBMP users do not experience realistic motions. The main contribution of this article is to take the SBMP's physical limitations into account in the MPC model such that more precise motion cues can be extracted for the users. A linear time-varying (LTV) MPC-based MCA method is designed for the first time in this article to consider the parameters of the hexapod mechanism in the MPC model. The proposed model (LTV MPC-based MCA) is validated using the MATLAB software, and the results depict better motion sensation with more accurate motion signals as compared with those from the existing LTI MPC-based MCA methods.

14 September 2020 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
Mohammad Reza Chalak QazaniMohammad Reza Chalak Qazani
Mohammad Reza Chalak Qazani
Houshyar AsadiHoushyar Asadi
Houshyar Asadi
TBTobias Bellmann
Tobias Bellmann
SMShady Mohamed
Shady Mohamed
Chee Peng LimChee Peng Lim
Chee Peng Lim
SNSaeid Nahavandi
Saeid Nahavandi
Motion simulation platforms (MSPs) are widely used to generate driving/flying motion sensations for the users. The MSPs have a restricted workspace area due to the dynamical and physical restrictions of the Motion Platforms active joints as well as the physical limitations of its passive joints. The motion cueing algorithm (MCA) is the reproduction of the motion signal including linear accelerations and angular velocities. It aims to simultaneously respect the MSP's workspace limitations and make the same motion feeling for the user as a real vehicle. The Classical washout filter (WF) is a well-known type of MCA. The classical WF is easy to set-up, offers a low computational burden and high functionality but has some major drawbacks such as fixed WF parameters tuned according to worst-case scenarios and no consideration of the human vestibular system. As a result, adaptive WFs were developed to consider the human vestibular system and enhance the efficiency of the method using time-varying filters. The existing adaptive WFs only cogitate the boundaries of the end-effector in the Cartesian coordinate space as a substitute for the active and passive joints limitations, which is MSP's main limiting factor. This conservative assumption reduces the available workspace area of the MSP and increases the motion sensation error for the MSPs user. In this study, a fuzzy logic-based WF is developed, to consider the dynamical and physical boundaries of the active joints as well as the physical boundaries of the passive joints. A genetic algorithm is used to select the membership functions' values of the active and passive joints' boundaries. The model is designed using MATLAB /Simulink and the outcomes demonstrate the efficiency of the proposed method versus existing adaptive WFs.

1 July 2020 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
Mohammad Reza Chalak QazaniMohammad Reza Chalak Qazani
Mohammad Reza Chalak Qazani
Houshyar AsadiHoushyar Asadi
Houshyar Asadi
SMShady Mohamed
Shady Mohamed
SNSaeid Nahavandi
Saeid Nahavandi
The motion cueing algorithms (MCAs) is the method that reproduces the motion sensation of the real vehicle for the users of simulation-based motion platforms (SBMPs). Classical MCA is the most common type of MCA due to its simplicity, easy tunning procedure, and higher speed. The fixed neutral position of the SBMP restricts the available workspace and makes the motions of the platform conservative. Then, the high frequency motion inputs cannot be reproduced precisely using SBMP and it increases the motion sensation error between the SBMP and real vehicle driver. The main objective of this study is to enlarge the reachable workspace of the SBMP adaptively using the new prepositioning technique based on fuzzy logic and artificial neural network (ANN). Prepositioning of the SBMPs to an off-centre point can increase the available linear space of the SBMPs based on different driving scenarios to decrease the motion sensation error. The fuzzy logic-based units are working based on the current position of the SBMP and the predicted motion signals. Therefore, the ANN is employed to forecast the motion signal of the SBMP on the finite history of input. The prepositioning of the SBMP using fuzzy logic and ANN is established for the first time in this research with consideration of the road information and vehicle motion information. The proposed method is performed and validated using MATLAB/Simulink and the outcomes show that using the new prepositioning method will lead to better driving sensation with less wrong motion sensation errors compared with the non and previous prepositioning methods.

31 October 2019 in Robotica
Verified
Journal Article
Closed access
PDF file
Mohammad Reza Chalak QazaniMohammad Reza Chalak Qazani
Mohammad Reza Chalak Qazani
Houshyar AsadiHoushyar Asadi
Houshyar Asadi
SNSaeid Nahavandi
Saeid Nahavandi
SUMMARY: The 3 degree-of-freedom Gantry-Tau manipulator with the addition of the spherical wrist mechanism which is called Gantry-Tau-3R is designed as a high-G simulation-based motion platform (SBMP) with the capability of generating the large linear and angular displacement. The combination of both parallel and serial manipulator in newly designed Gantry-Tau-3R mechanism improves the ability of the mechanism to regenerate larger motion signals with higher linear acceleration and angular velocity. The high-frequency signals are reproduced using the parallel part of the mechanism, and sustainable low-frequency accelerations are regenerated via the serial part due to the larger rotational motion capability, which will be used through motion cueing algorithm tilt coordination channel. The proportional integral derivative (PID) and fuzzy incremental controller (FIC) are developed for the proposed mechanism to show the high path tracking performance as a motion platform. FIC reduces the motion tracking error of the newly designed Gantry-Tau-3R and increases the motion fidelity for the users of the proposed SBMP. The proposed method is implemented using Matlab/Simulink software. Finally, the results demonstrate the accurate motion signal generation using linear model predictive motion cues with a fuzzy controller, which is not possible using the common parallel and serial manipulators.

11 April 2018 in International Journal of Precision Engineering and Manufacturing
Verified
Journal Article
Closed access
PDF file
Mohammad Reza Chalak QazaniMohammad Reza Chalak Qazani
Mohammad Reza Chalak Qazani
Siamak PedrammehrSiamak Pedrammehr
Siamak Pedrammehr
MNMohammad Javad Nategh
Mohammad Javad Nategh