Predicting socio-economic indicators using news events

Sunandan Chakraborty, Ashwin Venkataraman, Srikanth Jagabathula, Lakshminarayanan Subramanian

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Many socio-economic indicators are sensitive to real-world events. Proper characterization of the events can help to identify the relevant events that drive fluctuations in these indicators. In this paper, we propose a novel generative model of real-world events and employ it to extract events from a large corpus of news articles. We introduce the notion of an event class, which is an abstract grouping of similarly themed events. These event classes are manifested in news articles in the form of event triggers which are specific words that describe the actions or incidents reported in any article. We use the extracted events to predict fluctuations in different socioeconomic indicators. Specifically, we focus on food prices and predict the price of 12 different crops based on real-world events that potentially influence food price volatility, such as transport strikes, festivals etc. Our experiments demonstrate that incorporating event information in the prediction tasks reduces the root mean square error (RMSE) of prediction by 22% compared to the standard ARIMA model. We also predict sudden increases in the food prices (i.e. spikes) using events as features, and achieve an average 5-10% increase in accuracy compared to baseline models, including an LDA topic-model based predictive model.

Original languageEnglish (US)
Title of host publicationKDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Pages1455-1464
Number of pages10
Volume13-17-August-2016
ISBN (Electronic)9781450342322
DOIs
StatePublished - Aug 13 2016
Event22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016 - San Francisco, United States
Duration: Aug 13 2016Aug 17 2016

Other

Other22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016
CountryUnited States
CitySan Francisco
Period8/13/168/17/16

Fingerprint

Economics
Mean square error
Crops
Experiments

ASJC Scopus subject areas

  • Software
  • Information Systems

Cite this

Chakraborty, S., Venkataraman, A., Jagabathula, S., & Subramanian, L. (2016). Predicting socio-economic indicators using news events. In KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (Vol. 13-17-August-2016, pp. 1455-1464). Association for Computing Machinery. https://doi.org/10.1145/2939672.2939817

Predicting socio-economic indicators using news events. / Chakraborty, Sunandan; Venkataraman, Ashwin; Jagabathula, Srikanth; Subramanian, Lakshminarayanan.

KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. 13-17-August-2016 Association for Computing Machinery, 2016. p. 1455-1464.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chakraborty, S, Venkataraman, A, Jagabathula, S & Subramanian, L 2016, Predicting socio-economic indicators using news events. in KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. vol. 13-17-August-2016, Association for Computing Machinery, pp. 1455-1464, 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016, San Francisco, United States, 8/13/16. https://doi.org/10.1145/2939672.2939817
Chakraborty S, Venkataraman A, Jagabathula S, Subramanian L. Predicting socio-economic indicators using news events. In KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. 13-17-August-2016. Association for Computing Machinery. 2016. p. 1455-1464 https://doi.org/10.1145/2939672.2939817
Chakraborty, Sunandan ; Venkataraman, Ashwin ; Jagabathula, Srikanth ; Subramanian, Lakshminarayanan. / Predicting socio-economic indicators using news events. KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. 13-17-August-2016 Association for Computing Machinery, 2016. pp. 1455-1464
@inproceedings{de09e073dc8d4f22af744ae5a28e2aa0,
title = "Predicting socio-economic indicators using news events",
abstract = "Many socio-economic indicators are sensitive to real-world events. Proper characterization of the events can help to identify the relevant events that drive fluctuations in these indicators. In this paper, we propose a novel generative model of real-world events and employ it to extract events from a large corpus of news articles. We introduce the notion of an event class, which is an abstract grouping of similarly themed events. These event classes are manifested in news articles in the form of event triggers which are specific words that describe the actions or incidents reported in any article. We use the extracted events to predict fluctuations in different socioeconomic indicators. Specifically, we focus on food prices and predict the price of 12 different crops based on real-world events that potentially influence food price volatility, such as transport strikes, festivals etc. Our experiments demonstrate that incorporating event information in the prediction tasks reduces the root mean square error (RMSE) of prediction by 22{\%} compared to the standard ARIMA model. We also predict sudden increases in the food prices (i.e. spikes) using events as features, and achieve an average 5-10{\%} increase in accuracy compared to baseline models, including an LDA topic-model based predictive model.",
author = "Sunandan Chakraborty and Ashwin Venkataraman and Srikanth Jagabathula and Lakshminarayanan Subramanian",
year = "2016",
month = "8",
day = "13",
doi = "10.1145/2939672.2939817",
language = "English (US)",
volume = "13-17-August-2016",
pages = "1455--1464",
booktitle = "KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - Predicting socio-economic indicators using news events

AU - Chakraborty, Sunandan

AU - Venkataraman, Ashwin

AU - Jagabathula, Srikanth

AU - Subramanian, Lakshminarayanan

PY - 2016/8/13

Y1 - 2016/8/13

N2 - Many socio-economic indicators are sensitive to real-world events. Proper characterization of the events can help to identify the relevant events that drive fluctuations in these indicators. In this paper, we propose a novel generative model of real-world events and employ it to extract events from a large corpus of news articles. We introduce the notion of an event class, which is an abstract grouping of similarly themed events. These event classes are manifested in news articles in the form of event triggers which are specific words that describe the actions or incidents reported in any article. We use the extracted events to predict fluctuations in different socioeconomic indicators. Specifically, we focus on food prices and predict the price of 12 different crops based on real-world events that potentially influence food price volatility, such as transport strikes, festivals etc. Our experiments demonstrate that incorporating event information in the prediction tasks reduces the root mean square error (RMSE) of prediction by 22% compared to the standard ARIMA model. We also predict sudden increases in the food prices (i.e. spikes) using events as features, and achieve an average 5-10% increase in accuracy compared to baseline models, including an LDA topic-model based predictive model.

AB - Many socio-economic indicators are sensitive to real-world events. Proper characterization of the events can help to identify the relevant events that drive fluctuations in these indicators. In this paper, we propose a novel generative model of real-world events and employ it to extract events from a large corpus of news articles. We introduce the notion of an event class, which is an abstract grouping of similarly themed events. These event classes are manifested in news articles in the form of event triggers which are specific words that describe the actions or incidents reported in any article. We use the extracted events to predict fluctuations in different socioeconomic indicators. Specifically, we focus on food prices and predict the price of 12 different crops based on real-world events that potentially influence food price volatility, such as transport strikes, festivals etc. Our experiments demonstrate that incorporating event information in the prediction tasks reduces the root mean square error (RMSE) of prediction by 22% compared to the standard ARIMA model. We also predict sudden increases in the food prices (i.e. spikes) using events as features, and achieve an average 5-10% increase in accuracy compared to baseline models, including an LDA topic-model based predictive model.

UR - http://www.scopus.com/inward/record.url?scp=84984985105&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84984985105&partnerID=8YFLogxK

U2 - 10.1145/2939672.2939817

DO - 10.1145/2939672.2939817

M3 - Conference contribution

AN - SCOPUS:84984985105

VL - 13-17-August-2016

SP - 1455

EP - 1464

BT - KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -