The authors acknowledge some of the weaknesses in their tagger, including its speed, its ability to analyze substantive texts, and its inability to process less common constructions in which word order is loose.
Commentary
Oflazer and Kuruz suggest that their tagger would be greatly improved if tagging were not done incrementally but rather, globally. Natural language analysis must take into account real-world language uses, which are frequently loose. However, preliminary results of the Oflazer and Kuruz system seem promising. Combining a rule-based system with a statistics-based one minimizes general tagging errors as well as specifically targeting the morphological ambiguities in specific linguistic subsets. Potentially, the Oflazer and Kuruz could be designed as smart technology that logs correct and incorrect hits to generate new internal rules for tagging.
The authors carried out their research on a...
Their test was small but yielded results promising enough to warrant attention and extended applications. For example, Oflazer and Kuruz minimized errors by ensuring that tagging depends on variable unification and consistency. Expanding the set and type of variables included in the tagging and analytical process would help reduce errors and increase the possibility of employing the tagger for use on more substantive textual bodies. The current tagger is, however, based on natural language use and takes into account variables such as word order in a current sentence. Parsing accuracy is also ensured by a dual reliance on statistical probability and on a set of rules and constraints.
Works Cited
Oflazer, Kemal and Kuruz, ?lker. "Tagging and Morphological Disambiguation of Turkish Text."
