Software Defect Prediction (SDP) aims to detect bugs at an early stage of the software development process, helping to improve quality while reducing costs and development time. Machine learning (ML) models, such as LightGBM, have shown strong performance, their effectiveness depends heavily on proper hyperparameter optimization. This paper introduces MS-FSOA, a multi-strategy enhancement of the Starfsh Optimization Algorithm, to optimize LightGBM for SDP. MS-FSOA integrates enhanced population initialization, Lévy flight, and adaptable cooperative hunting to improve search quality and maintain diversity throughout the optimization process. Experiments on three PROMISE datasets show that MS-FSOA significantly improves prediction performance. Compared to traditional algorithms, it boosts LightGBM’s accuracy and robustness in software defect classification.

This research paper deals with the modeling and code generation of Reconfigurable Control Systems (RCS) following UML and B methods. Reconfiguration means dynamic changes of the system behavior at run-time according to well-defined conditions to adapt it to its environment. A reconfiguration scenario is applied as a response to user requirements or any possible evolution in its environment. We affect a Reconfiguration Agent (RA) to RCS to apply an automatic reconfiguration. A new approach called (R-UML-B) is proposed. It consists of three complementary phases: UML specification, B specification and the simulation phase. The first phase models the RCS following UML class and state diagrams. The second phase translates UML specification into B specification according to the well-defined rules and R-UML-B formalism to define the Behavior, Control, Listener, Database and Executive modules of the RCS. Then, we determine the refinement model and the code generation of the B abstract model in C code. We verify the RCS by following the B method in order to guarantee the consistency and the correctness of the specification, refinement and code generation levels. The third phase imports the generated C code to implement a simulator, named B Simulator in order to test and validate the proposed approach. All the contributions of this work are applied to the benchmark production system EnAS.

The paper deals with the modeling and verification of B method-based reconfigurable control systems. Reconfiguration means the dynamic changes of the system behavior at run-time according to well-defined conditions to adapt it to its environment. A reconfiguration scenario is applied as a response to improve the system's performance, or also to recover and prevent hardware/software errors, or also to adapt its behavior to new requirements according to the environment evolution. A new extension called Reconfigurable B “R-B” is proposed to specify reconfigurable control systems. It consists of two modules: Behavior and Control. The first defines all possible behaviors of the system, and whereas the second is a set of reconfiguration functions applied to change the system from a behavioral configuration to another one at run-time. We verify a reconfigurable control system by using the B method. The goal is to guarantee the consistency and the correctness of the abstract specification level. The second contribution of this paper deals with the verification of the reconfigurable system by avoiding redundant checking of different behaviors sharing similar operations. In order to control the complexity of verification, an optimal algorithm is developed and a prototyped tool called “Check R-B” is implemented. The paper's contribution is applied to a benchmark production system FESTO.