Posts Tagged ‘Interlingua’

PostHeaderIcon Universal Networking Language (UNL)

Universal Networking Language (UNL) is an Interlingua developed by UNDL foundation. UNL is in the form of semantic network to represent and exchange information. Concepts and relations enable encapsulation of the meaning of sentences. In UNL, a sentence can be considered as a hypergraph where each node is the concept and the links or arcs represent the relations between the concepts. UNL knowledge base provides concepts for the words in the natural language sentences.

The UNL consists of Universal Words (UWs), Relations and Attributes and knowledge base.

Universal Words (UWs)

Universal words are UNL words that carry knowledge or concepts. UWs are simply nodes in the UNL graph. There are two type of UWs: permanent and temporary. Permanent UWs represent concepts of common use and are included in the UW dictionary. Temporary UWs may represent new concepts, too specific or not translatable so that they are not included in the dictionary.

Examples:

bucket(icl>container)
water(icl>liquid)

Relations

Relations are labelled arcs that connect nodes (Uws) in the UNL graph. The relations are binary and usually represent semantic cases and semantic roles.

Examples:

agt ( break(agt>thing,obj>thing), John(iof>person) )

Attributes

Attributes are annotations used to represent grammatical categories, mood, aspect, etc. Every attribute starts with “@” symbol.

Example:

work(agt>human).@past
– means that the tense of the verb work is past tense, that is , it is actually worked in the sentence.

Knowledge Base

The UNL Knowledge Base contains entries that define possible binary relations between UWs.

PostHeaderIcon Interlingual Machine Translation

Interlingua

According to Collins dictionary Interlingua is an artificial language based on words common to English and the Romance languages. Interlingua is any artificial language used to represent the meaning of natural languages, as for purposes of machine translation. Interlingua is an intermediate form between two or more languages.

Interlingual Machine Translation

Interlingual Machine Translation is a methodology that employs interlingua for translation. Ideally the interlingual representation of the text should be sufficient to generate sentences in any language. Languages can have different parts of speech. In some cases two or more words in one language have a equivalent single word in another language. Interlingua approach addresses these structural differences between languages. The disadvantage is that the design of interlingua is too complex. This is due to the fact that there is no clear methodology developed so far to build a perfect interlingual representation.

An interlingual lexicon is necessary to store information about the nature and behavior of each word in the language. The information include events and actions.

Figure: Interlingual Machine Translation

Figure: Interlingual Machine Translation

A typical interlingual MT system has analyzer and synthesizer for each language. The analyzer produces interlingual represention of the meaning of the given text. The synthesizer produces one or more sentences with the meaning given by the analyzer. KANT is a interlingual machine translation system openational in a commercial setting.

Related Articles

Machine Translation Overview
Rule-Based Machine Translation
Example-Based Machine Translation