Chaher Alzaman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Chaher Alzaman has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Amar Ramudhin
The Logistics Institute, University of Hull, Hull, UK
2 shared publications Akif Asil Bulgak
Department of Mechanical and Industrial Engineering, Concordia University, 1455 de Maisonneuve Blvd., West, Montreal, Quebec H3G 1M8 Canada
1 shared publication- Akif Asil Bulgak
Department of Mechanical, Industrial, and Aerospace Engineering, Gina Cody School of Engineering and Computer Science, Concordia University, Montreal, Canada
1 shared publication Akif Bulgak
Concordia University
1 shared publicationAli Diabat
New York University Abu Dhabi
1 shared publicationJiangtao Hong
Shanghai University of International, Business and Economics
1 shared publication
Latest Publications
- Open Access
- 15 August 2026
- Published by Elsevier in SSRN Electronic Journal
Reinforcement Learning with Large Language Model–Generated Event-Based Sentiment Index for Adaptive Model Selection in Financial Markets
Traditional financial models and deep learning struggle to capture qualitative market nuances and conflicting signals. This research introduces a dynamic reinforcement learning agent acting as a meta-controller to adaptively allocate capital across diverse base models.Instead of choosing stocks directly, the agent combines recent model performance with a novel Event-Based Synthetic Stock Sentiment Index, which is generated by a large language model and models firm-level sentiment using a baseline, event shocks, sector trends, and random variation. To prevent over-trading, a switching-penalty mechanism enforces stability. Empirical results show that the agent successfully identifies when different base models perform best, increasing exploration during market turbulence and maintaining consistency in stable periods. Ultimately, this framework improves cumulative returns and risk-adjusted efficiency while reducing return variance, tail risk, and downside exposure compared to individual baselines and the market benchmark.
- 15 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessReinforcement Learning with Large Language Model–Generated Event-Based Sentiment Index for Adaptive Model Selection in Financial Markets
Traditional financial models and deep learning struggle to capture qualitative market nuances and conflicting signals. This research introduces a dynamic reinforcement learning agent acting as a meta-controller to adaptively allocate capital across diverse base models.Instead of choosing stocks directly, the agent combines recent model performance with a novel Event-Based Synthetic Stock Sentiment Index, which is generated by a large language model and models firm-level sentiment using a baseline, event shocks, sector trends, and random variation. To prevent over-trading, a switching-penalty mechanism enforces stability. Empirical results show that the agent successfully identifies when different base models perform best, increasing exploration during market turbulence and maintaining consistency in stable periods. Ultimately, this framework improves cumulative returns and risk-adjusted efficiency while reducing return variance, tail risk, and downside exposure compared to individual baselines and the market benchmark.
- 5 January 2026
- Published by Elsevier in Applied Soft Computing
- Vol. 190
Product selection in drop-shipping: Reinforcement learning stacked on multiple machine learning algorithms
- 5 January 2026
- Published by Elsevier in Applied Soft Computing
- Vol. 190
Product selection in drop-shipping: Reinforcement learning stacked on multiple machine learning algorithms
- Top Cited Papers
- 8 January 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 269, 126430
Optimizing portfolio selection through stock ranking and matching: A reinforcement learning approach
- 8 January 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 269, 126430
Top Cited PapersOptimizing portfolio selection through stock ranking and matching: A reinforcement learning approach
- Open Access
- 26 June 2024
- Published by Universitat Politecnica de Valencia
- p. 220-234
Unlocking the Potential of Machine Learning in Portfolio Selection: A Hybrid Approach with Genetic Optimization
In the field of financial market predictions, machine learning has been widely used to identify patterns and gain valuable insights. However, for success in portfolio selection, it is crucial to optimize factors that impact accuracy. This study focuses on combining machine learning and optimization to enhance stock selection and prediction capabilities. The work starts with hyperparameter optimization and utilizes three different machine learning algorithms: XGBoost, LSTM, and Deep RankNet. Our findings show a 40% improvement in results through the use of a genetic-based optimization technique, as well as a promising daily average return of 0.47% through a novel feature engineering approach. The study provides a framework for optimizing and learning in financial portfolio selection, with promising results for medium and small-sized traders.
- 26 June 2024
- Published by Universitat Politecnica de Valencia
- p. 220-234
Open AccessUnlocking the Potential of Machine Learning in Portfolio Selection: A Hybrid Approach with Genetic Optimization
In the field of financial market predictions, machine learning has been widely used to identify patterns and gain valuable insights. However, for success in portfolio selection, it is crucial to optimize factors that impact accuracy. This study focuses on combining machine learning and optimization to enhance stock selection and prediction capabilities. The work starts with hyperparameter optimization and utilizes three different machine learning algorithms: XGBoost, LSTM, and Deep RankNet. Our findings show a 40% improvement in results through the use of a genetic-based optimization technique, as well as a promising daily average return of 0.47% through a novel feature engineering approach. The study provides a framework for optimizing and learning in financial portfolio selection, with promising results for medium and small-sized traders.
- 9 September 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 237
Deep learning in stock portfolio selection and predictions
- 9 September 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 237
Deep learning in stock portfolio selection and predictions
- Open Access
- 9 May 2023
- Published by Elsevier in Systems and Soft Computing
- Vol. 5
Forecasting and optimization stock predictions: Varying asset profile, time window, and hyperparameter factors
- 9 May 2023
- Published by Elsevier in Systems and Soft Computing
- Vol. 5
Open AccessForecasting and optimization stock predictions: Varying asset profile, time window, and hyperparameter factors
- 16 October 2018
- Published by Springer Nature in Annals of Operations Research
- Vol. 275 (2) , 339-366
Sustainability dimensions and PM2.5 in supply chain logistics
- 16 October 2018
- Published by Springer Nature in Annals of Operations Research
- Vol. 275 (2) , 339-366
Sustainability dimensions and PM2.5 in supply chain logistics
- 1 January 2010
- Published by Taylor & Francis in International Journal of Management Science and Engineering Management
- Vol. 5 (6) , 411-421
Gradient method in solving nonlinear cost of quality functions in supply chain network design
The design of a supply chain network is of great importance as it aims at optimizing costs throughout the supply chain. Traditionally, this optimization has considered operations related costs such as sourcing, production, transportation, inventory at each site in the network. Quality related costs were either ignored or considered indirectly. In a previous work, we showed how Cost of Quality (COQ) functions can be used as a viable indicator of the level of quality at each site and integrated in supply chain network design (SCND) models. This results in a non-linear MIP model, which poses challenges to solve. In this work, we develop a gradient-search based solution procedure to solve the model effectively for practical implications. The integration of COQ functions allows managers to explicitly consider the maturity levels of both internal sites (plants) and external partners such as suppliers or subcontractors. More importantly, it can be used as an additional metric by which the partners are measured during the annual evaluation process to see if they have progressed in their maturity levels, i.e. in mastering the costs of quality. COQ also opens a new avenue of research in SCND as the model and hence the solution procedure will vary based on the different ways COQ functions are incorporated in the model.
- 1 January 2010
- Published by Taylor & Francis in International Journal of Management Science and Engineering Management
- Vol. 5 (6) , 411-421
Gradient method in solving nonlinear cost of quality functions in supply chain network design
The design of a supply chain network is of great importance as it aims at optimizing costs throughout the supply chain. Traditionally, this optimization has considered operations related costs such as sourcing, production, transportation, inventory at each site in the network. Quality related costs were either ignored or considered indirectly. In a previous work, we showed how Cost of Quality (COQ) functions can be used as a viable indicator of the level of quality at each site and integrated in supply chain network design (SCND) models. This results in a non-linear MIP model, which poses challenges to solve. In this work, we develop a gradient-search based solution procedure to solve the model effectively for practical implications. The integration of COQ functions allows managers to explicitly consider the maturity levels of both internal sites (plants) and external partners such as suppliers or subcontractors. More importantly, it can be used as an additional metric by which the partners are measured during the annual evaluation process to see if they have progressed in their maturity levels, i.e. in mastering the costs of quality. COQ also opens a new avenue of research in SCND as the model and hence the solution procedure will vary based on the different ways COQ functions are incorporated in the model.
- 1 July 2009
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 985-990
Heuristic procedures to solve a binary nonlinear supply chain model: A case study from the Aerospace industry
In the aerospace industry quality is an important factor and suppliers are chosen on the basis of quality as well as cost. By integrating cost of quality (COQ) into supply chain network design (SCND), managers could make decisions based on operational costs and also based on quality non-conformance costs. This paper presents a case study for a leading company in the Aerospace industry with COQ integrated into SCND. Further in this paper, we present a model that tailors to a real life supply chain with cost of quality (COQ), binary variables for assignable cost representation of supply chain nodes, and an n level bill of material (BOM) integrated; the model resulting is a nonlinear model and is challenging to solve. Six known heuristics and a new hybrid heuristic are tested and the performance of each heuristic is compared to a lower bound on the optimal solution.
- 1 July 2009
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 985-990
Heuristic procedures to solve a binary nonlinear supply chain model: A case study from the Aerospace industry
In the aerospace industry quality is an important factor and suppliers are chosen on the basis of quality as well as cost. By integrating cost of quality (COQ) into supply chain network design (SCND), managers could make decisions based on operational costs and also based on quality non-conformance costs. This paper presents a case study for a leading company in the Aerospace industry with COQ integrated into SCND. Further in this paper, we present a model that tailors to a real life supply chain with cost of quality (COQ), binary variables for assignable cost representation of supply chain nodes, and an n level bill of material (BOM) integrated; the model resulting is a nonlinear model and is challenging to solve. Six known heuristics and a new hybrid heuristic are tested and the performance of each heuristic is compared to a lower bound on the optimal solution.