Abstract

The key theme of the analytics here encompasses data mining and knowledge discovery in data, and that comprises unsupervised classification, analytics of semantics, and what can be well considered as cross-disciplinarity and multi-disciplinarity. In analyzing data, there are requirements and also possible additional perspectives to have. This allows coverage of both quantitative and qualitative themes and aspects. A basis for much of the Correspondence Analysis, latent semantics, methodology here is the mapping of data into the Euclidean metric endowed factor space. The latter expresses and represents the information space, that can also be well displayed and visualized. New methods in this paper include: process convergence and its application; how analytical focus and contextualization are very important and how these are implemented, and further aspects of semantic analytics. Since semantics are underlying meanings, this indicates the importance here for decision support and for the well known saying that "correlation is not causation". The latter expression means that understanding causal actions and events cannot be purely reduced to the input data and starting point relative to the output data or finalization. Interesting and important results and outcomes, at issue here, include social media analytics; incorporating context in mental health analytics; and large-scale social media analytics, being Twitter text mining.