An online reversed French Sign Language dictionary based on a learning approach for signs classification

• Searching visual information by using uncompleted parameters.
• Development of a self-adaptive system based on cognitive experiments.
• Indexing a French Sign Language dictionary by non-textual inputs.

Sign Language contains very rich vocabulary units which can vary by a simple change in shape, position or movement. The characteristic of this language, to be practiced in space, has encouraged us to adopt a pragmatic approach, which facilitates access to its vocabulary. In this paper, we take advantage of digital techniques to build a human machine interface to create a dictionary of French Sign Language (FSL) to French as a web application. We notice that the FSL has not yet any fixed set of rules to learn this language. We hope that this application could contribute both in spreading and stabilizing the language. To this end, we develop a scalable research system, able to find the meaning of a FSL sign from some features of the sign itself. In this work, we set up a web platform for signs and we analyzed user requests that are connected to this platform. This analysis based on some experiments; aim to identify the parameters necessary to develop an application able to easily find the meaning of the sign in French. Therefore, by studying user behavior, we build a cognitive system that meets to needs.