Ontology Learning as a Use-Case for Neural-Symbolic Integration (position paper)

We argue that the field of neural-symbolic integra-tion is in need of identifying application scenarios for guiding further research. We furthermore argue that ontology learning — as occuring in the con-text of semantic technologies — provides such an application scenario with potential for success and high impact on neural-symbolic integration. 1 Neural-Symbolic Integration Intelligent systems based on symbolic knowledge process-ing, on the one hand, and on artificial neural networks (also called connectionist systems), on the other, differ substan-tially. Nevertheless, these are both standard approaches to artificial intelligence and it would be very desirable to com-bine the robustness of neural networks with the expressiv-ity of symbolic knowledge representation. This is the reason why the importance of the efforts to bridge the gap between the connectionist and symbolic paradigms of Artificial Intel-ligence has been widely recognised. As the amount of hybrid data containing symbolic and statistical elements as well as noise increases in diverse areas such as bioinformatics or text and web mining, neural-symbolic learning and reasoning be-comes of particular practical importance. Notwithstanding, this is not an easy task, as illustrated in the sequel. The merging of theory (background knowledge) and data learning (learning from examples) in neural networks has been indicated to provide learning systems that are more ef-fective than purely symbolic and purely connectionist sys-tems, especially when data are noisy [16]. This has con-tributed decisively to the growing interest in developing neural-symbolic systems, i.e. hybrid systems based on neu-ral networks that are capable of learning from examples and background knowledge, and of performing reasoning tasks in a massively parallel fashion. Typically, translation algo-rithms from a symbolic to a connectionist representation and vice-versa are employed to provide either (i) a neural imple-mentation of a logic, (ii) a logical characterization of a neura