Automatic Morphological Analysis of English Noun
3
The main idea
Implementation Area
To create a basis for computerized English grammar starting with the Morphology of a Noun.
So, the object of our research is the Morphology of English, the subject - paradigmatic Noun classes.
- Syntactical and Semantical Processors;
- Machine Translation;
- Natural Language Processing;
- Error Search Systems;
- Information Retrieval Systems;
- Automatic Text Editing and Summarising.
Automatic Morphological Analysis of English Noun
Research Objectives
1. Creating paradigmatic Noun classes;
2. Working with alphabetic frequency dictionary;
3. Tagging words` stems by belonging to certain Parts of Speech;
4. Tagging Nouns by belonging to certain Paradigmatic Class.
English Morphological Analyzer made by the Department of Computer and Information Science in the University of Pennsylvania
Noun processing:
mice N_Root1 N (mouse) PL
mouse N_Root1 N (mouse) SG
ambassador N_Root2 N (ambassador)
Phase 1
In the first phase we have completed the following tasks:
1. create the dictionary of stems ;
(for example, if you have ability and abilities, the stem abilit remains, which later will be attributed to the paradigmatic class with the inflection y -> #, + ies);
2. to cope with morphologically homonymous word forms ;
(for example, the word absolute can be used as a Noun with the meaning "absolute value" and as an Adjective with the meaning "total", so we added to the base the same stem with 2 different codes).
• Noun – N
• Verb – V
• Adjective – Adj
• Adverb – Adv
• Pronoun – Pron
• Preposition - Prep
• Numeral - Num
• Conjunction – Conj
• Interjection – Interj
• Article – Art
Results and Statistics
28 paradigmatic classes of the Noun,
12 000 word forms processed.
40% — class «Noun»,
15% — Class «Verb»,
20% — «Adjective»,
25% — «Adverb», «Preposition», «Pronoun», «Conjunction», «Interjection», «Article».
The most frequent Noun classes:
1-4 (the most common nouns with the traditional way of inflection).
Phase 2
1) develop a classification by the Noun paradigmatic classes;
2) ascribe paradigmatic class code to the Noun stems.
Noun is N; S - singular; P - plural; C - singular, possessive case; L - plural, possessive case