Extracting knowledge from huge data in a reasonable time is still a challenging problem. Most real data (structured or not) can be mapped to an equivalent binary context, with or without using a scaling method, as for extracting associations between words in a text, or in machine learning...