Large-Scale Paraphrasing for Natural Language Understanding

reportActive / Technical Report | Accession Number: AD1050977 | Open PDF

Abstract:

In this project, we researched and developed technologies to automatically extract large-volumes of paraphrases to aid in natural language understanding NLU tasks. We developed three core algorithms to 1 generate extremely large paraphrase databases, and 2 adapt paraphrase databases to new domains, and 3 augment paraphrase rules with fine-grained semantic entailment relations. Our work introduced the paraphrase database PPDB, the largest paraphrase resource developed to date. The resource contains over 100 million paraphrases for English. We generated paraphrase databases for 23 foreign languages.

Author(s):
Callison-burch, Chris ; Van Durme, Benjamin
Author Organization(s):
Johns Hopkins University Baltimore United States
Descriptive Note:
Technical Report
Pagination:
0071

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution:
Approved For Public Release
Distribution Statement:
Approved For Public Release;

RECORD

Collection: TR
Identifying Numbers
Report Number(s):
AFRL/RI-AFRL-RI-RS-TR-2018-098
Task Number(s):
12
Project Number(s):
DEFT
Monitor Series:
AFRL-RI-RS-TR-2018-098
 Subject Terms
Joint Capability Areas:
JCA_5_Command and Control; JCA_5.2_Understand; JCA_5.2.3_Share Knowledge and Situational Awareness; JCA_5.6.1_Assess Compliance with Guidance; JCA_5.6_Monitor; JCA_5.5.2_Task; JCA_5.5_Direct; JCA_1.2.1_Training; JCA_5.3_Planning; JCA_1.2.7_Experimentation; JCA_5.2.2_Develop Knowledge and Situational Awareness; JCA_1.3.2_Personnel Management; JCA_1.2.3_Educating; JCA_1.2.5_Lessons Learned; JCA_1.2.6_Concepts; JCA_5.4_Decide; JCA_5.4.5_Intuit
Modernization Areas:
AI and Machine Learning
Communities of Interest:
Energy and Power Technologies
Descriptor(s):
natural language computing, computational linguistics, natural language understanding, supervised machine learning, ontologies, artificial neural networks, artificial intelligence software, automated text summarization, algorithms, data set, machine translation, GRAPHS, CLUSTERING
Field(s)/Group(s):
Linguistics, Cybernetics
Keyword(s):
Paraphrase, knowledge base, FrameNet, semantic entailment, graph clustering
Report Date:
2018 Apr 01
Creation Date:
2018 May 01