Use of artificial intelligence in the early detection of school dropout: Theoretical elaboration of a major problem in the college cycle in Morocco
Bibliographic record
Abstract
In developing countries, the prospects for reducing dropout in the education system are still slim, given the magnitude of the socio-economic challenges that are considered essential to keeping students in school (Mduma et al., 2019). Dropping out of school is therefore one of the challenges faced by most schools in these countries. The development of solution approaches for the control of dropout requires a thorough understanding of the underlying factors. Several researchers have identified and proposed causes, methods and strategies that will help reduce or suppress the problem. However, most of the proposed solutions have not shown promising results and the dropout trend seems to continue in the education systems of several developing countries: in Morocco, the dropout rate increased from 10.8% in 2010-2011 to 10.4% in 2019-2020 in the college cycle according to data from the Ministry of National Education. Furthermore, to prevent dropout researchers have used supervised and unsupervised learning techniques, survival analysis methods, matrix factorization and neural networks (Hung et al., 2017; Elbadrawy et al., 2016). In addition, machine learning has attracted a lot of attention when it comes to solving societal problems in different sectors, including the education sector (Elbadrawy et al., 2016; Xu et al., 2017). In order to contribute to the analysis and reduction of the phenomenon, this research uses recent advances in data science and educational technologies to understand and model the dropout phenomenon in order to lay the foundation for an early dropout detection system in Moroccan junior high schools. From a methodological point of view, we use a four-step approach. We propose to conduct a systematic review of the determinants of school dropout in Africa on the one hand and the various options for combating school dropout, including the use of artificial intelligence, on the other. We then adapt the questionnaire developed by the Quebec team for dropout screening (Fortin et al., 2007) to Moroccan conditions, followed by the training of a predictive model for early detection of dropout in the college cycle in Morocco. Finally, we propose a model of argumentation applied to the case of school dropout by providing justifications for the steps leading to a result and making explicit the arguments that support the decisions.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".