Today, symbolic and connectionist artificial intelligence have proved their complementary efficiency for various aspects of cognitive processing. Both approaches can be seen as complementarily acting on specific parts of information, namely data and knowledge. Such a dichotomy can also be observed as one consider real world applications. A general theory is rarely available to build a complete knowledge based system. Conversely, data can generally be extracted from the problem but never cover the whole problem. Accordingly, the idea has emerged that the combination of symbolic and connectionist tools could be a way to benefit from the advantages of both approaches. Neurosymbolic integration that will be presented here is the domain whose goal is to define strategies and propose tools for the cooperation of symbolic and connectionist artificial intelligence.