Adaptive e-learning using Genetic Algorithm and Sentiments Analysis in a Big Data System

In this article we describe our adaptive e-learning system, which allows the learner to take courses adapted to his profile and to the pedagogical objectives set by the teacher, we use for adaptation the genetic algorithms to give the learner the concepts that must learn in an optimal way by seeking the objectives most adapted to his profile. And after a second level of adaptation using one of the social networks of the learner (twitter, facebook, Google + ...), based on his post on one of these social networks we propose two levels of analysis. The first one is to look for the period of activity which gives us an idea about the period when the learner is active and the second consists of making an analysis of the feelings on the publications that are published during the period of activity and related to education. Our work therefore is to adapt the profile of the learner with the pedagogical objectives by using the genetic algorithm and the notions of the research of information by doing this work in a Big Data system, that is to say we parallelize the search problem using Hadoop with Hadoop distributed file system (HDFS) and the MapReduce programming model,and after using information from a social network of the learner, we look for the period of activity of the learner and the feeling (sentiment analysis) related to the publications of the period of activity.

@article{MADANI2017,
  title     = {Adaptive e-learning using Genetic Algorithm and Sentiments Analysis in a Big Data System},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {8},
  number    = {8},
  year      = {2017},
  publisher = {The Science and Information Organization},
  author    = {Youness MADANI and Jamaa BENGOURRAM and Mohammed ERRITALI and Badr HSSINA and Marouane Birjali},
  doi       = {10.14569/IJACSA.2017.080851},
  url       = {https://doi.org/10.14569/IJACSA.2017.080851}
}