Choosing the content of textual summaries of large time-series data sets

Jin Yu; Ehud Baruch Reiter; James Ritchie Wallace Hunter; Christopher Stuart Mellish

doi:10.1017/S1351324905004031

Choosing the content of textual summaries of large time-series data sets

Jin Yu, Ehud Baruch Reiter, James Ritchie Wallace Hunter, Christopher Stuart Mellish

Computing Science

Research output: Contribution to journal › Article › peer-review

74 Citations (Scopus)

Abstract

Natural Language Generation (NLG) can be used to generate textual summaries of numeric data sets. In this paper we develop an architecture for generating short (a few sentences) summaries of large (100KB or more) time-series data sets. The architecture integrates pattern recognition, pattern abstraction, selection of the most significant patterns, microplanning (especially aggregation), and realisation. We also describe and evaluate SumTime-Turbine, a prototype system which uses this architecture to generate textualsummaries of sensor data from gas turbines.

Original language	English
Pages (from-to)	25-49
Number of pages	24
Journal	Natural Language Engineering
Volume	13
Issue number	1
DOIs	https://doi.org/10.1017/S1351324905004031
Publication status	Published - Mar 2007

Access to Document

10.1017/S1351324905004031

Mainstream communication of big data using natural language generation (NLG)
Ehud Reiter (Coordinator) & Gowri Sripada (Coordinator)
Impact: Economic Impacts

Cite this

@article{ca543b76c9a445d1a0676f85cd031323,

title = "Choosing the content of textual summaries of large time-series data sets",

abstract = "Natural Language Generation (NLG) can be used to generate textual summaries of numeric data sets. In this paper we develop an architecture for generating short (a few sentences) summaries of large (100KB or more) time-series data sets. The architecture integrates pattern recognition, pattern abstraction, selection of the most significant patterns, microplanning (especially aggregation), and realisation. We also describe and evaluate SumTime-Turbine, a prototype system which uses this architecture to generate textualsummaries of sensor data from gas turbines.",

author = "Jin Yu and Reiter, {Ehud Baruch} and Hunter, {James Ritchie Wallace} and Mellish, {Christopher Stuart}",

year = "2007",

month = mar,

doi = "10.1017/S1351324905004031",

language = "English",

volume = "13",

pages = "25--49",

journal = "Natural Language Engineering",

issn = "1351-3249",

publisher = "Cambridge University Press",

number = "1",

}

TY - JOUR

T1 - Choosing the content of textual summaries of large time-series data sets

AU - Yu, Jin

AU - Reiter, Ehud Baruch

AU - Hunter, James Ritchie Wallace

AU - Mellish, Christopher Stuart

PY - 2007/3

Y1 - 2007/3

N2 - Natural Language Generation (NLG) can be used to generate textual summaries of numeric data sets. In this paper we develop an architecture for generating short (a few sentences) summaries of large (100KB or more) time-series data sets. The architecture integrates pattern recognition, pattern abstraction, selection of the most significant patterns, microplanning (especially aggregation), and realisation. We also describe and evaluate SumTime-Turbine, a prototype system which uses this architecture to generate textualsummaries of sensor data from gas turbines.

AB - Natural Language Generation (NLG) can be used to generate textual summaries of numeric data sets. In this paper we develop an architecture for generating short (a few sentences) summaries of large (100KB or more) time-series data sets. The architecture integrates pattern recognition, pattern abstraction, selection of the most significant patterns, microplanning (especially aggregation), and realisation. We also describe and evaluate SumTime-Turbine, a prototype system which uses this architecture to generate textualsummaries of sensor data from gas turbines.

U2 - 10.1017/S1351324905004031

DO - 10.1017/S1351324905004031

M3 - Article

VL - 13

SP - 25

EP - 49

JO - Natural Language Engineering

JF - Natural Language Engineering

SN - 1351-3249

IS - 1

ER -

Choosing the content of textual summaries of large time-series data sets

Abstract

Access to Document

Fingerprint

Impacts

Mainstream communication of big data using natural language generation (NLG)

Cite this