Various language analysis methods form the basis for the Connexor technology implementations.
Connexor's language analysers can identify the following linguistic features:
Sentence and word boundaries
Often it is desirable to know where a sentence or a word begins and ends.
Finding which words in the text are derived from the same root.
Lexical categories and parts-of-speech
Knowing which words are adjectives, nouns, verbs etc. can be useful, as often we are looking for just certain types of words.
Details on which inflection is used in each word.
Details on what roles each word has in a sentence and how words in a sentence are related to each other.
Details on semantics of words, eg. whether a word refers to a living being or not. This kind of information is helpful especially in machine translation.
Noun phrase detection
Details on which words form noun phrases, which are often the most interesting parts of text.
Details on which words are proper nouns.
As providing more detailed information on text structure requires more processing power, Connexor analysers come in different varieties to fill different tasks' needs in speed vs. analysis detail ratio.