Machine translation apparatus having means for translating polysemous words using dominated codes

• US 5,285,386 A
• Filed: 07/02/1990
• Issued: 02/08/1994
• Est. Priority Date: 12/29/1989
• Status: Expired due to Term

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Kuo, June-Jei
• Primary Examiner(s): Envall, Jr., Roy N.
• Assistant Examiner(s): BODENDORF, ANDREW
• Application Number:
  US07/546,940
• Time in Patent Office: 1,317 Days
• Field of Search: 364/417, 364/DIG. 2
• US Class Current: 704/2
• CPC Class Codes: G06F 40/55 Rule-based translation

Abstract

A machine translation apparatus includes: a transfer dictionary for translation word searching for storing therein a semantic code representative of semantic categorization of a source language word, a dominated code which is a semantic code of a word relatable to the source language word, and a translation word of a target language corresponding to the dominated code; a dominated code and semantic code searching unit for executing a logical AND operation applicable to a word at each node of an intermediate structure obtained by parsing an inputted source language sentence, to be performed on the dominated code of the each node and the semantic code of a node related to the each node, which are respectively stored in the transfer dictionary for translation word searching, and determining proper ones of the dominated code and the semantic code, respectively and a translation word searching unit for searching a translation word for each node corresponding to the dominated code obtained by the dominated code and semantic code searching unit, from among target language translation words for the source language word stored in the transfer dictionary for translation word searching. With the aforesaid machine translation apparatus, a proper translation word for a polysemous word can be automatically selected.

Claims

1. A machine translation apparatus for translating a source language input sentence with a polysemous word into a target language sentence, the polysemous source language word in the input sentence having a plurality of possible translations in the target language, comprising:

a transfer dictionary for translation word searching, the transfer dictionary including means for storing a plurality of source language words, at least one semantic code for every source language word, and at least one target language word for every semantic code, wherein the source language words stored in the transfer dictionary include polysemous source language words, and wherein the transfer dictionary additionally includes means for storing semantic complex information for identifying the possible translations of the polysemous source language words, the semantic complex information including a plurality of dominated codes for each of the polysemous source language words;

a source language parsing and intermediate form generating unit means for parsing the input sentence to obtain an intermediate representation having a plurality of nodes which correspond to source language words in the input sentence, the nodes of the intermediate representation including a particular node which corresponds to the polysemous word in the input sentence and a neighboring node which corresponds to a word in the input sentence that is grammatically related to the polysemous word;

a dominated code and semantic code searching unit means for selecting a dominated code and semantic code for the polysemous word in the input sentence, the dominated code and semantic code searching unit means including means for comparing the dominated codes of the polysemous word which corresponds to the particular node of the intermediate representation with the at least one semantic code of the grammatically related word which corresponds to the neighboring node of the intermediate representation; and

a translation word searching unit means for determining appropriate target language words for the source language words corresponding to each node of the intermediate representation, the translation word searching unit means including means for using dominated codes selected by the dominated code and semantic code searching unit means as keys to search out appropriate target words from the transfer dictionary.

2. The translation apparatus of claim 1, wherein the transfer dictionary additionally stores a lexical category key for every source language word.

3. The translation apparatus of claim 1, wherein the dominated codes for a given polysemous source language word stored in the transfer dictionary are selected from the group consisting of a default code and semantic codes for other source language words.

4. A machine translation apparatus for translating a source language sentence into a target language sentence, comprising:

a transfer dictionary for translation word searching, the transfer dictionary including means for storing a plurality of source language words each having a single meaning or polysemous meanings, and means for storing a single set or multiple sets of semantic information corresponding to the single meaning or polysemous meanings of each of the words, each set of semantic information including a dominated code, a semantic code, and a translation word in the target language, the dominated codes and semantic codes having a hierarchical code structure to represent the semantic information of each meaning;

a source language parsing and intermediate form generating unit means for generating an intermediate representation of the source language sentence, the intermediate representation having a plurality of nodes which correspond to words in the source language sentence, nodes in the intermediate representation being neighboring nodes if they correspond to words that are grammatically related in the source language sentence;

a dominated code and semantic code searching unit means for selecting a dominated code and a semantic code for each word of the source language sentence, the dominated code and semantic code searching unit means including means for executing a logical operation applicable to the single set or multiple sets of semantic information corresponding to a word which corresponds to a particular node of the intermediate representation with the single set or multiple sets of semantic information corresponding to words which correspond to nodes of the intermediate representation that are neighboring nodes to the particular node, all the information being searched and retrieved from said transfer dictionary by using a source language word as a searching key; and

a translation word searching unit means for translation word searching for each word of the source language sentence, by using the dominated codes selected by said dominated code and semantic code searching unit means as searching keys to search out translation words from said transfer dictionary.

5. A machine translation apparatus as defined in claim 4, wherein said dominated code and semantic code searching unit means comprises means for executing a set inclusion operation applicable to the dominated code corresponding to the word which corresponds to the particular node with the semantic codes corresponding to words which correspond to the neighboring nodes in order to select the dominated code for the word corresponding to the particular node.

6. A machine translation apparatus as defined in claim 4, wherein the transfer dictionary additionally includes means for storing a lexical category key for every source language word, and wherein the dominated code and semantic code searching unit means comprises means for using the source language words and the lexical category keys as searching keys for transfer dictionary searching.