| Project/Area Number |
63460225
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Informatics
|
|---|
| Research Institution | Fukuoka Institute of Technology |
|---|
Principal Investigator |
TANAKA Takushi Fukuoka Institute of Technology, Department of Communication and Computer Engineering, Professor, 工学部, 教授 (50038048)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NISHIHARA Suzuko The National Language Research Institute, Department of External Services, Direc, 日本語教育センター, 部長 (60189298)
TAMACHI Tsuneo Fukuoka Institute of Technology, Department of Communication and Computer Engine, 工学部, 教授 (90037656)
|
|---|
| Project Period (FY) |
1988 – 1990
|
| Project Status |
Completed (Fiscal Year 1990)
|
| Budget Amount *help |
¥7,300,000 (Direct Cost: ¥7,300,000)
Fiscal Year 1990: ¥1,000,000 (Direct Cost: ¥1,000,000)
Fiscal Year 1989: ¥1,800,000 (Direct Cost: ¥1,800,000)
Fiscal Year 1988: ¥4,500,000 (Direct Cost: ¥4,500,000)
|
|---|
| Keywords | Language Understanding / Knowledge Representation / Logic Grammar / Conceptual Structure / Inference / 述語論理 / 知織表現 / 言語理解 / 確定節文法 / DCSG |
|---|
| Research Abstract |
The purpose of this study is to investigate human mechanisms of language understanding and to make computer models of the language understanding. We first developed a logic grammar called DCSG (Definite Clause Set Grammars) for free word-order language. The DCSG is a method for parsing free word-order sentences such as Japanese in logic programming Then we extended the DCSG as a mechanism for deductive inference. Using the extended DCSG, we realized two computer models for understanding language. The one is a model for understanding story. The other is a model for understanding explanatory sentences.
(1) Understanding Story
Knowledges of the story world are coded into deductive rules in the extended DCSG. From input sentences, many theorems are derived using the deductive rules. A set of the theorems can be viewed as a computer state of understanding the story. Limitations of this model which consists of only deductive inferences were also clarified by computer experiments. As a next step toward story understanding, we have to investigate hypothetical inferences and inductive inferences in language understanding.
(2) Understanding Explanations
We have made a model in a field of cooking as an example of explanations. Most of the sentences in cooking describe operations for physical world. Knowledges of the operations are defined as changing states of the physical world. Using the knowledge, each input sentence changes a conceptual image of the physical world. Computer experiments of this model clarified the significance of situations assumed in each sentence.
|
