I need to define a rule but I'm struggling to get through the UIMA Ruta syntax. I have a dictionary, which, if it finds a matching word in the text, looks two words ahead and two words behind this word. If the word has uppercase letters, the condition is fulfilled.

Example: "Prime Minister" Margaret Thatcher

The phrase "Prime Minister" is in the dictionary, and the condition of two words after it starting with an uppercase letter is satisfied. The result is Margaret Thatcher.

This is how my very rough rule is written, but the syntax is incorrect, and I cannot figure out how to improve it:

WORDLIST MyWordList = 'test_words.txt';

// Declaration of annotations for marked words
DECLARE MyAnnotation;
DECLARE CapitalizedAnnotation;

Document {-> MyWordList 
            -2, -1 {
            (REGEXP("[A-ZÁ-Ž]") && NOT(REGEXP("[a-zá-ž]"))){ 
                -> MARK(CapitalizedAnnotation); 
            }
        }
        
        +1, +2 {

            (REGEXP("[A-ZÁ-Ž]") && NOT(REGEXP("[a-zá-ž]"))){ 
                -> MARK(CapitalizedAnnotation); 
            }
        }
    }
};

I need to define a rule but I'm struggling to get through the UIMA Ruta syntax. I have a dictionary, which, if it finds a matching word in the text, looks two words ahead and two words behind this word. If the word has uppercase letters, the condition is fulfilled.

Example: "Prime Minister" Margaret Thatcher

The phrase "Prime Minister" is in the dictionary, and the condition of two words after it starting with an uppercase letter is satisfied. The result is Margaret Thatcher.

This is how my very rough rule is written, but the syntax is incorrect, and I cannot figure out how to improve it:

WORDLIST MyWordList = 'test_words.txt';

// Declaration of annotations for marked words
DECLARE MyAnnotation;
DECLARE CapitalizedAnnotation;

Document {-> MyWordList 
            -2, -1 {
            (REGEXP("[A-ZÁ-Ž]") && NOT(REGEXP("[a-zá-ž]"))){ 
                -> MARK(CapitalizedAnnotation); 
            }
        }
        
        +1, +2 {

            (REGEXP("[A-ZÁ-Ž]") && NOT(REGEXP("[a-zá-ž]"))){ 
                -> MARK(CapitalizedAnnotation); 
            }
        }
    }
};

• uima
• ruta

Share Improve this question Follow asked Mar 25 at 11:21 Pavlína Z.Pavlína Z. 1

Add a comment  | 

1 Answer 1

Sorted by: Reset to default 0

Maybe something like that?

WORDLIST MyWordList = 'test_words.txt';
// Declaration of annotations for marked words
DECLARE MyAnnotation;
DECLARE CapitalizedAnnotation;
MARKFAST(MyAnnotation, MyWordList);

MyAnnotation CW[2,2]{-> CapitalizedAnnotation};
CW[2,2]{-> CapitalizedAnnotation} @MyAnnotation;