X-STREAMS: Cross- Stream Textual Realtime Multi-document Summarizer
Navy SBIR FY2012.1
Sol No.: |
Navy SBIR FY2012.1 |
Topic No.: |
N121-078 |
Topic Title: |
X-STREAMS: Cross- Stream Textual Realtime Multi-document Summarizer |
Proposal No.: |
N121-078-1104 |
Firm: |
Manifest Labs, Inc. 2900 W. Anderson Lane
C-200-301
Austin, Texas 78757-1159 |
Contact: |
Stephen Hilderbrand |
Phone: |
(512) 461-1978 |
Web Site: |
www.manifestlabs.com |
Abstract: |
In response to the Navy's N121-078 solicitation, Manifest Labs, Inc.,
proposes X-STREAMS, a real-time summarization system that improves
upon the current state-of-the-art results on the novel information
reporting of entities and events found in textual data sources. Using
a novel combination of mature techniques, and a new semantic layering
methodology, X-STREAMS will increase the value of streaming document
summarization capabilities, by merging information across streams and
improving the timeliness and accuracy of automated knowledge
discovery. The ultimate goal of the X-STREAMS research is to automate
much of the summarization of documents and other forms of
communication which may be represented as text, such as IM chat, voice
and image transcriptions.
The strength of X-STREAMS is that it uses a data-driven, unsupervised
learning approach to train adaptable summarization models. These
models can be trained in any language, and do not require special
rules or linguists to develop or maintain the system. To minimize
redundant information in reports, X-STREAMS employs a parallelized
implementation of the leading methodology for determining the maximum
marginal relevance in automated document summarization. |
Benefits: |
The X-STREAMS SBIR effort will produce an automated multi-document
summarization and presentation system that will serve as the primary
automated information awareness component in operational scenarios. In
this manner, X-STREAMS represents a force multiplier for analysts
responsible for reporting on activities occurring within their areas
of responsibility, streamlining the commander's information collection
needs. X-STREAMS will replace much of the manual reading activities
that analysts currently perform, as well as legacy semi-automated
systems, minimizing the underreporting that occurs during periods of
intense operational activity. X-STREAMS will significantly improve
current, state of the art capacity for persistent information
awareness with reduced manpower, allowing analysts and soldiers to
focus on deeper analysis tasks that only humans perform well, such as
higher-level cognitive functions and asking the right questions to
drive further analysis.
The strength of X-STREAMS is that it uses a data-driven, unsupervised
learning approach to train adaptable summarization models. These
models can be rapidly trained in any language, and do not require
special rules or experts to develop or maintain the system. The UI
will enable analysts to trace information back to the source documents
to assist in corroboration or conflicting intelligence resolution, as
well as to provide iterative feedback on the system to drive
improvements. |
Return
|