C. Thompson and R. Mooney. Automatic Construction of Semantic Lexicons for Learning Natural Language Interfaces. Proceedings of the Sixteenth National Conference on Artificial Intelligence, 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....[15] terrorist attacks, management transitions in industries, and the launching of space vehicles. In order to address the problem of creating such systems economically, some researchers have looked at techniques for automatically or semi automatically constructing lexicons ( 23] 24] 26] [30]) or extraction rules for the domain ( 7] 28] 20] 16] 17] 25] Most of these techniques have applied machine learning approaches to learn rules based on texts that have been semantically annotated. Either pre annotation of text is done by a human expert, or the rules are post processed ....

C. Thompson and R. Mooney. Automatic Construction of Semantic Lexicons for Learning Natural Language Interfaces. Proceedings of the Sixteenth National Conference on Artificial Intelligence, 1999.

....in exactly that order; the set of parsing actions resemble a decision list in which the first applicable choice is taken. The parser also requires a lexicon to interpret meaning of phrases into specific logical forms. The lexicon can be learned from a given set of sample sentence and query pairs (Thompson Mooney, 1999). We will briefly illustrate what action each template does here by showing a trace of parsing a simple example: Sentence: What is the capital of Texas Logical Query: answer(C, capital(C,S) const(S, stateid(texas) The first thing we need is a lexicon. A very simple lexicon that maps ....

Thompson, C. A., & Mooney, R. J. (1999). Automatic construction of semantic lexicons for learning natural language interfaces. In Proceedings of the Sixteenth National Conference on Artificial Intelligence, pp. 487--493 Orlando, FL.

....on 225 examples, the system was able to answer an average of 70 of novel queries correctly compared to 57 for an interface developed by a human programmer. Similar results were obtained for semantic parsing of other languages after translating the corpus into Spanish, Turkish, and Japanese (Thompson Mooney, 1999). 20.6.6 Information Extraction Information extraction is a form of shallow text processing that locates a speci ed set of relevant items in a natural language document (see Chapter 30) Figure 8 shows an example of extracting values for a set of labeled slots from a job announcement posted to ....

Thompson, C. A., & Mooney, R. J. (1999). Automatic construction of semantic lexicons for learning natural language interfaces. In Proceedings of the Sixteenth National Conference on Articial Intelligence (AAAI-99), pp. 487-493 Orlando, FL.

....on the stack. SHIFT simply pushes a word from the input bu er onto the stack. The parsing actions are tried in exactly this order. The parser also requires a lexicon to map phrases in the input into speci c predicates, this lexicon can also be learned automatically from the training data (Thompson and Mooney, 1999). Let s go through a simple trace of parsing the request What is the capital of Texas A lexicon that maps capital to capital( and Texas to const( stateid(texas) suces here. Interrogatives like what may be mapped to predicates in the lexicon if necessary. The parser begins with ....

C. A. Thompson and R. J. Mooney. 1999. Automatic construction of semantic lexicons for learning natural language interfaces. In Proceedings of the Sixteenth National Conference on Articial Intelligence, pages 487-493, Orlando, FL, July.

....on the stack. SHIFT simply pushes a word from the input bu er onto the stack. The parsing actions are tried in exactly this order. The parser also requires a lexicon to map phrases in the input into speci c predicates, this lexicon can also be learned automatically from the training data (Thompson and Mooney, 1999). Let s go through a simple trace of parsing the request What is the capital of Texas A very simple lexicon that maps capital to capital( and Texas to const( stateid(texas) suces here. The parser begins with an initial stack and a bu er holding the input sentence. Each predicate ....

C. A. Thompson and R. J. Mooney. 1999. Automatic construction of semantic lexicons for learning natural language interfaces. In Proceedings of the Sixteenth National Conference on Articial Intelligence, pages 487-493, Orlando, FL, July.

....methods for these tasks. In contrast, most of our own recent research on applying ILP to NLP has focused on learning to parse natural language database queries into a semantic logical form that produces an answer when executed in Prolog (Zelle Mooney, 1993, 1994, 1996; Zelle, 1995; Mooney, 1997; Thompson Mooney, 1999; Thompson, 1998; Thompson, Califf, Mooney, 1999) There is a long tradition of representing the meaning of natural language statements and queries in first order logic (Allen, 1995; Dowty, Wall, Peters, 1981; Woods, 1978) However, we know of no other recent research specifically on learning ....

....as background knowledge that provides the possible logical representations of specific words and phrases. Chill initially required the user to provide this lexicon; however, we have recently developed a system called Wolfie that learns this lexicon automatically from the same training corpus (Thompson Mooney, 1999; Thompson, 1998) Chill has been used successfully to learn natural language interfaces for three separate databases: 1) a small database on U.S. geography, 2) a database of thousands of restaurants in northern California, and 3) a database of computer jobs automatically extracted from the ....

Thompson, C. A., & Mooney, R. J. (1999). Automatic construction of semantic lexicons for learning natural language interfaces. In Proceedings of the Sixteenth National Conference on Artificial Intelligence Orlando, FL.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC