Application of multi-dimensional scaling and artificial neural networks for biologically inspired opinion mining
The way people express their opinions has radically changed in the past few years thanks to the advent of online collaborative media. The distillation of knowledge from this huge amount of unstructured information can be a key factor for marketers who want to create an identity for their product or brand in the minds of their customers. These online social data, however, remain hardly accessible to computers, as they are specifically meant for human consumption. Existing approaches to opinion mining, in fact, are still far from being able to infer the cognitive and affective information associated with natural language as they mainly rely on knowledge bases that are too limited to efficiently process text at concept-level. In this context, standard clustering techniques have been previously employed on an affective common-sense knowledge base in attempt to discover how different natural language concepts are semantically and affectively related to each other and, hence, to accordingly mine on-line opinions. In this work, a novel cognitive model based on the combined use of multi-dimensional scaling and artificial neural networks is exploited for better modelling the way multi-word expressions are organised in a brain-like universe of natural language concepts. The integration of a biologically inspired paradigm with standard principal component analysis helps to better grasp the non-linearities of the resulting vector space and, hence, improve the affective common-sense reasoning capabilities of the system.