tic knowledge programmed in. In the latter approach, one may use knowledge from statistical corpus analysis, one may use electronic resources like WordNet and FrameNet, and one may use sophisticated, specialized tools like natural language parsers with hand-coded grammars. Rather than trying to emulate

ords; a graph between word senses is drawn, and then solved as a Markov chain. In the original formulation, word senses are defined by appealing to WordNet, and affinity between word-senses is obtained via one of several similarity measures. Neither of these can be applied in learning a language de novo

nature of human linguistic concepts is well known, and is illustrated in Figure 15.2 for the commonsense knowledge domain (using a graph drawn from WordNet, a huge con- cept hierarchy covering 50K+ English-language concepts), and in Figure 15.4 for a specialized knowledge subdomain, genetics. Due to thi

l nature of human linguistic concepts is well known, and is illustrated in Figure 15.2 for the commonsense knowledge domain (using a graph drawn from WordNet, a huge con- cept hierarchy covering 50K+ English-language concepts), and in Figure 15.4 for a specialized knowledge subdomain, genetics. Due to this