No abstract

By using neural networks, correlations were established between chemical structure and boiling points of chlorofluorocarbons with 1, 1-2, or 1-4 carbon atoms (15, 62, and 276 compounds, respectively) as well as of halomethanes with up to four different halogens (48 compounds). The molecular descriptors included the number of carbon atoms and of each type of halogen atom as well as topological indices. Results were validated by the jackknifing procedure. The correlation coefficients were r = 0.985-0.995. Predictions were made for the boiling points of several haloethanes.

The language of computer science is laced with metaphor. We argue that computer science metaphors provide a conceptual framework in which to situate constantly emerging new ontologies in computational environments. But how computer science metaphors work does not fit neatly into prevailing general theories of metaphor. We borrow from these general theories while also considering the unique role of computer science metaphors in learning, design, and scientific analysis. We find that computer science metaphors trade on both preexisting and emerging similarities between computational and traditional domains, but owing to computer science's peculiar status as a discipline that creates its own subject matter, the role of similarity in metaphorical attribution is multifaceted.

No abstract

No abstract