List of Content
Back to Listing
| Title | Abstract | Action(s) |
|---|---|---|
| EVALUATING OBE STUDENTS LEARNING CAPACITY IN HIGHER EDUCATION: TOWARDS COMPREHENSIVE FRAMEWORK | Outcome-Based Education (OBE) is adopted in higher education to make sure that students acquire knowledge and skills that are useful for their future profession. Several strategies based on theoretical frameworks have been developed to improve education in higher education over the past two decades. Outcome-Based Education (OBE) aims to establish specific learning objectives that students are expected to achieve by the end of their educational journey and is also helpful for professional careers. Therefore, OBE approach develops the practical experience of students by integrating educational knowledge with practical skills. There is a lack of studies on the investigation of students' learning capacity and practical skills of OBE graduates in a real environment. To fill the research gap, this study aims to measure attainment of students' learning capacity and practical skills in OBE. Survey research is used to achieve research objectives. The survey respondents are employers of OBE graduates. A total of 140 employers participated in the survey. The result analysis of survey demonstrate that OBE graduates possess practical skills and learning capacity that enables them to effectively perform in real work environment. The research provides a systematic approach to assessing educational outcomes and guides the practitioners towards effectiveness of OBE. In the future, researchers can specifically measure student’s performance in OBE and assess other factors like organization objectives, teaching strategies, and assessment methods in OBE. |
|
| EFFORT ESTIMATION IN AGILE SOFTWARE DEVELOPMENT USING ENSEMBLE LEARNING MODEL | In the domain of software development, effort estimating is an essential procedure that entails projecting the size and schedule of a particular project. It becomes necessary to create an estimate before beginning any software project. Obtaining the required approvals and evaluating the project depends on this preliminary assessment. The importance of this procedure cannot be emphasized since a project's success or failure is solely dependent on how precisely and accurately effort is estimated. There are various cost and effort methods and techniques. These techniques have been utilized to construct several effort estimation models that are used in the software development process in the traditional model. This research explores the application of ensemble learning techniques, specifically stacking, to enhance the accuracy of effort estimation in Agile environments. Stacking involves combining multiple diverse base estimators to create a meta-estimator that outperforms individual models. This study includes a crucial step for gathering datasets because old dataset size is small and old. The proposed approach is assessed using real-world Agile project datasets, proving the advantages of the stacking model over agile software development estimation techniques. |
|
| Intelligent Feedback Controller Design For DC-DC Converter | In this research, a feedback control system for DC-DC converter is designed using artificial intelligence. PID control is widely used as the control method for DC-DC converters. The PID parameter tuning process parameters remains hard in nonlinear case scenarios. As in such scenarios system needs to handle disturbances during load changes and operate under dynamically changing conditions. The system control response with the inclusion of AI technology leads to an improvement in the operation of DC-DC converters through their feedback system and improved the overall ripple on output line. Also, it helps the system to become more stable across different loads at the same time and achieve maximum performance. Furthermore, the proposed AI-based controller has been successfully applied to a real-time Voltage-to-Grid (70V-380V) case scenario, replacing the PID controller in a Neutral Point Clamped (NPC) converter topology. This shows the versatility and effectiveness of this control technique in complex power conversions. Replacement of AI based control with PID helps system in continuous learning and adaptation and as a result real-time control, reduction in output voltage ripple, and enhanced system adaptability under dynamic load conditions and making it suitable for current applications of DC-DC converters. |
|
| MULTI OBJECT TRACKING USING NON-LINEAR FILTERING TECHNIQUES | In this research, the problem of redundant detections of tracks in distributed sensor networks for multi-object tracking is addressed. Measurement noise and network-induced errors often result in multiple detections of the same object, complicating the accurate estimation of object count and position. This makes it challenging to accurately determine the true number of objects and their locations. Track-to-track association algorithms help address this issue. Many such algorithms have been developed and can be broadly categorized into two types: statistical algorithms and clustering-based algorithms. A key clustering-based approach is the fuzzy track-to-track association algorithm, which is the focus of this research. A variation of this algorithm is tested on data generated from a model simulating a multi-sensor, multi-target environment. In real-world sensors, errors typically arise in azimuth, elevation, and range, so this thesis proposes an error model based on these parameters. The association algorithm’s resolutions are also grounded in this realistic error model. Additionally, time synchronization is critical before performing track association. This thesis employs a linear predictor to synchronize tracks before association, and the performance of the algorithm is analyzed under these conditions. |
|