The student dropout rate in universities is fascinating, especially among the students of Electrical Engineering. Even the most developed European countries face 40% to 50% dropout rate of engineering students during their first year, and the rate can be as high as 80% for some engineering disciplines. This problem calls attention of educators and university administration to take measures which can help in the reduction of the dropout rate and assist students in successfully completing their degree. Among many other solutions to control the student dropout rate, one is the adoption of a prediction mechanism whereby students can be warned about their potentially poor performance so that they can improve their performance resulting in better grades. Most of the existing prediction mechanisms apply various machine learning techniques on student cognitive features. In addition, non-cognitive features also have significant impact on students’ performance; however, they have been sparsely applied for prediction. This research aims at improving the existing prediction mechanism by exploiting both cognitive and non-cognitive features of students for predicting their results. It has been found in the result analysis that addition of cognitive features increases prediction accuracies of decision tree; however, the addition does not play a significant role in other techniques. The study also identified the individual cognitive features that should be considered by students and universities to cater for drop outs.

In the recent decade, the citation recommendation has emerged as an important research topic due to its need for the huge size of published scientific work. Among other citation recommendation techniques, the widely used content-based filtering (CBF) exploits research articles’ textual content to produce recommendations. However, CBF techniques are prone to the well-known cold-start problem. On the other hand, deep learning has shown its effectiveness in understanding the semantics of the text. The present paper proposes a citation recommendation system using deep learning models to classify rhetorical zones of the research articles and compute similarity using rhetorical zone embeddings that overcome the cold-start problem. Rhetorical zones are the predefined linguistic categories having some common characteristics about the text. A deep learning model is trained using ART and CORE datasets with an accuracy of 76 per cent. The final ranked lists of the recommendations have an average of 0.704 normalized discounted cumulative gain (nDCG) score involving ten domain experts. The proposed system is applicable for both local and global context-aware recommendations.

Requirements change management is a vital activity of the software development and even after maintenance. When we change one requirement it effects the other requirements of the system. A proper impact analysis maintains a repository of changing requirement, reusing previous experiences and reduces rework effort, is also essential for better requirement change management. There is also a need of a technique which improves requirement change management in a better way. In this paper, a model has been introduced which integrates Requirement Change Management and CBR (Case Based Reasoning) technique. Expert review method is used to evaluate our research. This frame work gives a better approach which manages requirement changes to the system. Our research outcome, i.e. the CBR technique used with the RCM, can provide a better approach to solve the issues of managing changes to the system. It maintains a repository by using CBR technique which is helpful to manage changes in future and we are able to solve the new change on the basis of previous experience.