Abstract
Learning by reading requires integrating several strands of AI research. We describe a prototype system, Learning Reader, which combines natural language processing, a large-scale knowledge base, and analogical processing to learn by reading simplified language texts. We outline the architecture of Learning Reader and some of system-level results, then explain how these results arise from the components. Specifically, we describe the design, implementation, and performance characteristics of a natural language understanding model (DMAP) that is tightly coupled to a knowledge base three orders of magnitude larger than previous attempts. We show that knowing the kinds of questions being asked and what might be learned can help provide more relevant, efficient reasoning. Finally, we show that analogical processing provides a means of generating useful new questions and conjectures when the system ruminates off-line about what it has read.
| Original language | English (US) |
|---|---|
| Title of host publication | AAAI-07/IAAI-07 Proceedings |
| Subtitle of host publication | 22nd AAAI Conference on Artificial Intelligence and the 19th Innovative Applications of Artificial Intelligence Conference |
| Pages | 1542-1547 |
| Number of pages | 6 |
| Volume | 2 |
| State | Published - Nov 28 2007 |
| Event | AAAI-07/IAAI-07 Proceedings: 22nd AAAI Conference on Artificial Intelligence and the 19th Innovative Applications of Artificial Intelligence Conference - Vancouver, BC, Canada Duration: Jul 22 2007 → Jul 26 2007 |
Other
| Other | AAAI-07/IAAI-07 Proceedings: 22nd AAAI Conference on Artificial Intelligence and the 19th Innovative Applications of Artificial Intelligence Conference |
|---|---|
| Country | Canada |
| City | Vancouver, BC |
| Period | 7/22/07 → 7/26/07 |
Fingerprint
ASJC Scopus subject areas
- Software
- Artificial Intelligence
Cite this
}
Integrating natural language, knowledge representation and reasoning, and analogical processing to learn by reading. / Forbus, Kenneth D; Riesbeck, Christopher K; Birnbaum, Lawrence A; Livingston, Kevin; Sharma, Abhishek; Ureel, Leo.
AAAI-07/IAAI-07 Proceedings: 22nd AAAI Conference on Artificial Intelligence and the 19th Innovative Applications of Artificial Intelligence Conference. Vol. 2 2007. p. 1542-1547.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Integrating natural language, knowledge representation and reasoning, and analogical processing to learn by reading
AU - Forbus, Kenneth D
AU - Riesbeck, Christopher K
AU - Birnbaum, Lawrence A
AU - Livingston, Kevin
AU - Sharma, Abhishek
AU - Ureel, Leo
PY - 2007/11/28
Y1 - 2007/11/28
N2 - Learning by reading requires integrating several strands of AI research. We describe a prototype system, Learning Reader, which combines natural language processing, a large-scale knowledge base, and analogical processing to learn by reading simplified language texts. We outline the architecture of Learning Reader and some of system-level results, then explain how these results arise from the components. Specifically, we describe the design, implementation, and performance characteristics of a natural language understanding model (DMAP) that is tightly coupled to a knowledge base three orders of magnitude larger than previous attempts. We show that knowing the kinds of questions being asked and what might be learned can help provide more relevant, efficient reasoning. Finally, we show that analogical processing provides a means of generating useful new questions and conjectures when the system ruminates off-line about what it has read.
AB - Learning by reading requires integrating several strands of AI research. We describe a prototype system, Learning Reader, which combines natural language processing, a large-scale knowledge base, and analogical processing to learn by reading simplified language texts. We outline the architecture of Learning Reader and some of system-level results, then explain how these results arise from the components. Specifically, we describe the design, implementation, and performance characteristics of a natural language understanding model (DMAP) that is tightly coupled to a knowledge base three orders of magnitude larger than previous attempts. We show that knowing the kinds of questions being asked and what might be learned can help provide more relevant, efficient reasoning. Finally, we show that analogical processing provides a means of generating useful new questions and conjectures when the system ruminates off-line about what it has read.
UR - http://www.scopus.com/inward/record.url?scp=36348957836&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=36348957836&partnerID=8YFLogxK
M3 - Conference contribution
SN - 1577353234
SN - 9781577353232
VL - 2
SP - 1542
EP - 1547
BT - AAAI-07/IAAI-07 Proceedings
ER -