Hybrid Systems. Recently, combinations of the two approaches are being developed in order to improve the translation output, in the so called hybrid systems. These systems combine advantages from both the statistics based and the rule-based systems. The development of rule-based machine translation systems is time consuming and very expensive. The statistical machine translation model has high computational costs, but is robust in processing all kinds of texts. The linguistic quality depends on how much a new text corresponds to a narrowly defined domain. Example-based machine translation systems rely on bilingual texts which must be of high quality in order to result in good translation output. Having collected a good set of relevant translation memories, the translation output is of high quality, if a new text is similar to the information stored in terms of subject area and content. These systems become more useful in concrete domains like technical and medical translations, where the vocabulary and construction of sentences is more restricted. Therefore, there is a trend towards the creation of hybrid systems which combine different approaches, making the most out of each individual system.
Hybrid Systems. Some hybrid systems have been investigated. For example, it is possible to adsorb molecular dyes onto the surface of a semiconductor and thus sensitize the action of the system to wavelengths longer than those that would be absorbed directly by the semiconductor itself. Also there have been reports of photoelectrochemical cells based on chloroplast preparations in the electrolyte or adsorbed onto an electrode. However, none of these systems have been successful in efficient hydrogen production.
Hybrid Systems. Since no hybrid systems have yet been developed that give significant amounts of water splitting, this option is not viable at present.
Hybrid Systems. In the above subsections, the simple discovery mechanisms, the mechanisms based on overlay networks and DHTs, and the ontology-driven semantic discovery have been discussed. In this section, the hybrid systems that are ontology-driven on top of DHTs are discussed.
