Hybrid models for future event prediction

Giuseppe Amodeo; Roi Blanco; Ulf Brefeld

doi:10.1145/2063576.2063870

Hybrid models for future event prediction

Giuseppe Amodeo^*
, Roi Blanco
, Ulf Brefeld

^*Corresponding author for this work

Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review

25 Citations (Scopus)

Abstract

We present a hybrid method to turn off-the-shelf information retrieval (IR) systems into future event predictors. Given a query, a time series model is trained on the publication dates of the retrieved documents to capture trends and periodicity of the associated events. The periodicity of historic data is used to estimate a probabilistic model to predict future bursts. Finally, a hybrid model is obtained by intertwining the probabilistic and the time-series model. Our empirical results on the New York Times corpus show that autocorrelation functions of time-series suffice to classify queries accurately and that our hybrid models lead to more accurate future event predictions than baseline competitors.

Original language	English
Title of host publication	Proceedings of the 20th ACM international conference on Information and knowledge management
Editors	Bettina Berendt, Arjen de Vries, Wenfei Fan, Craig Macdonald, Iadh Ounis, Ian Ruthven
Number of pages	4
Place of Publication	New York
Publisher	Association for Computing Machinery, Inc
Publication date	2011
Pages	1981-1984
ISBN (Electronic)	978-145030717-8
DOIs	https://doi.org/10.1145/2063576.2063870
Publication status	Published - 2011
Externally published	Yes
Event	20th ACM Conference on Information and Knowledge Management - CIKM '11 - Glasgow, United Kingdom Duration: 24.10.2011 → 28.10.2011 http://www.cikm2011.org/

Research areas and keywords

Informatics
Event prediction
Web searches
Forecasting
Information retrieval
Knowledge management
Regression analysis
Time series
World Wide Web
Business informatics

Access to Document

10.1145/2063576.2063870

Cite this

@inbook{038827f98b584417b80efd43a93e5235,

title = "Hybrid models for future event prediction",

abstract = "We present a hybrid method to turn off-the-shelf information retrieval (IR) systems into future event predictors. Given a query, a time series model is trained on the publication dates of the retrieved documents to capture trends and periodicity of the associated events. The periodicity of historic data is used to estimate a probabilistic model to predict future bursts. Finally, a hybrid model is obtained by intertwining the probabilistic and the time-series model. Our empirical results on the New York Times corpus show that autocorrelation functions of time-series suffice to classify queries accurately and that our hybrid models lead to more accurate future event predictions than baseline competitors.",

keywords = "Informatics, Event prediction, Web searches, Forecasting, Information retrieval, Knowledge management, Regression analysis, Time series, World Wide Web, Business informatics",

author = "Giuseppe Amodeo and Roi Blanco and Ulf Brefeld",

year = "2011",

doi = "10.1145/2063576.2063870",

language = "English",

pages = "1981--1984",

editor = "Berendt, \{Bettina \} and \{de Vries\}, Arjen and Wenfei Fan and Craig Macdonald and Iadh Ounis and Ian Ruthven",

booktitle = "Proceedings of the 20th ACM international conference on Information and knowledge management",

publisher = "Association for Computing Machinery, Inc",

address = "United States",

note = "20th ACM Conference on Information and Knowledge Management - CIKM '11, CIKM '11 ; Conference date: 24-10-2011 Through 28-10-2011",

url = "http://www.cikm2011.org/",

}

Amodeo, G, Blanco, R & Brefeld, U 2011, Hybrid models for future event prediction. in B Berendt, A de Vries, W Fan, C Macdonald, I Ounis & I Ruthven (eds), Proceedings of the 20th ACM international conference on Information and knowledge management. Association for Computing Machinery, Inc, New York, pp. 1981-1984, 20th ACM Conference on Information and Knowledge Management - CIKM '11, Glasgow, United Kingdom, 24.10.11. https://doi.org/10.1145/2063576.2063870

Hybrid models for future event prediction. / Amodeo, Giuseppe; Blanco, Roi; Brefeld, Ulf.
Proceedings of the 20th ACM international conference on Information and knowledge management. ed. / Bettina Berendt; Arjen de Vries; Wenfei Fan; Craig Macdonald; Iadh Ounis; Ian Ruthven. New York: Association for Computing Machinery, Inc, 2011. p. 1981-1984.

Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review

TY - CHAP

T1 - Hybrid models for future event prediction

AU - Amodeo, Giuseppe

AU - Blanco, Roi

AU - Brefeld, Ulf

PY - 2011

Y1 - 2011

N2 - We present a hybrid method to turn off-the-shelf information retrieval (IR) systems into future event predictors. Given a query, a time series model is trained on the publication dates of the retrieved documents to capture trends and periodicity of the associated events. The periodicity of historic data is used to estimate a probabilistic model to predict future bursts. Finally, a hybrid model is obtained by intertwining the probabilistic and the time-series model. Our empirical results on the New York Times corpus show that autocorrelation functions of time-series suffice to classify queries accurately and that our hybrid models lead to more accurate future event predictions than baseline competitors.

AB - We present a hybrid method to turn off-the-shelf information retrieval (IR) systems into future event predictors. Given a query, a time series model is trained on the publication dates of the retrieved documents to capture trends and periodicity of the associated events. The periodicity of historic data is used to estimate a probabilistic model to predict future bursts. Finally, a hybrid model is obtained by intertwining the probabilistic and the time-series model. Our empirical results on the New York Times corpus show that autocorrelation functions of time-series suffice to classify queries accurately and that our hybrid models lead to more accurate future event predictions than baseline competitors.

KW - Informatics

KW - Event prediction

KW - Web searches

KW - Forecasting

KW - Information retrieval

KW - Knowledge management

KW - Regression analysis

KW - Time series

KW - World Wide Web

KW - Business informatics

UR - https://www.scopus.com/pages/publications/83055161561

U2 - 10.1145/2063576.2063870

DO - 10.1145/2063576.2063870

M3 - Article in conference proceedings

AN - SCOPUS:83055161561

SP - 1981

EP - 1984

BT - Proceedings of the 20th ACM international conference on Information and knowledge management

A2 - Berendt, Bettina

A2 - de Vries, Arjen

A2 - Fan, Wenfei

A2 - Macdonald, Craig

A2 - Ounis, Iadh

A2 - Ruthven, Ian

PB - Association for Computing Machinery, Inc

CY - New York

T2 - 20th ACM Conference on Information and Knowledge Management - CIKM '11

Y2 - 24 October 2014 through 28 October 2011

ER -

Hybrid models for future event prediction

Abstract

Research areas and keywords

Access to Document

Fingerprint

Cite this