Information System (IS) becomes a big priority of organization for the majority of firms and government institutions. Among the main reasons of using information system, we can mention information availability or reliability, better data circulation or communication, and finally insure information visibility and ease of access. For the reasons above, all information system's components, namely human resources, hardware, software, procedures and data, must have a definite level of quality. The review of literature reveals that all existing models are limited to the software quality as a substitution of information system quality, in addition, almost all the surveys and studies done to measure the information system quality on different organizations are considering only the developers or the technical staff opinion and neglecting the managers, the users and the operating staff opinion. In this article, we will highlight the limits of existing models and propose a hybrid model integrating quality indicators measurements for all information system components; we will also give adapted surveys to each kind of information system player.
Using machine learning to predict students’ dropout in higher education institutions and programs has proven to be effective in many use cases. In an approach based on machine learning algorithms to detect students at risk of dropout, there are three main factors: the choice of features likely to influence a partial or total stop of the student, the choice of the algorithm to implement a prediction model, and the choice of the evaluation metrics to monitor and assess the credibility of the results. This paper aims to provide a diagnosis of machine learning techniques used to detect students’ dropout in higher education programs, a critical analysis of the limitations of the models proposed in the literature, as well as the major contribution of this arti-cle is to present recommendations that may resolve the lack of global model that can be generalized in all the higher education institutions at least in the same country or in the same university.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.