Predicting movie success with machine learning techniques: ways to improve accuracy

K. Lee, J. Park, I. Kim, Youngseok Choi

Research output: Contribution to journalArticle

7 Citations (Scopus)
49 Downloads (Pure)

Abstract

Previous studies on predicting the box-office performance of a movie using machine learning techniques have shown practical levels of predictive accuracy. Their works are technically- and methodologically-oriented, focusing mainly on what algorithms are better at predicting the movie performance. However, the accuracy of prediction model can also be elevated by taking other perspectives such as introducing unexplored features that might be related to the prediction of the outcomes. In this paper, we examine multiple approaches to improve the performance of the prediction model. First, we develop and add a new feature derived from the theory of transmedia storytelling. Such theory-driven feature selection not only increases the forecast accuracy, but also enhances the interpretability of a prediction model. Second, we use an ensemble approach, which has rarely been adopted in the research on predicting box-office performance. As a result, the proposed model, Cinema Ensemble Model (CEM), outperforms the prediction models from the past studies that use machine learning algorithms. We suggest that CEM can be extensively used for industrial experts as a powerful tool for improving decision-making process.
The final publication is available at Springer via http://dx.doi.org/[10.1007/s10796-016-9689-z
Original languageEnglish
Pages (from-to)577–588
JournalInformation Systems Frontiers
Volume20
Issue number3
Early online date19 Aug 2016
DOIs
Publication statusPublished - Jun 2018

Fingerprint

Prediction Model
Learning systems
Machine Learning
Ensemble
Storytelling
Interpretability
Feature Selection
Forecast
Learning Algorithm
Decision Making
Model
Prediction
Learning algorithms
Feature extraction
Decision making

Bibliographical note

The final publication is available at Springer via
http://dx.doi.org/[10.1007/s10796-016-9689-z

Keywords

  • Prediction model
  • Movie performance
  • Machine learning techniques
  • Cinema ensemble model
  • Transmedia storytelling
  • Feature selection

Cite this

Predicting movie success with machine learning techniques: ways to improve accuracy. / Lee, K.; Park, J.; Kim, I.; Choi, Youngseok.

In: Information Systems Frontiers, Vol. 20, No. 3, 06.2018, p. 577–588.

Research output: Contribution to journalArticle

Lee, K. ; Park, J. ; Kim, I. ; Choi, Youngseok. / Predicting movie success with machine learning techniques: ways to improve accuracy. In: Information Systems Frontiers. 2018 ; Vol. 20, No. 3. pp. 577–588.
@article{b6c13bbd81274b5b9a6332989444f097,
title = "Predicting movie success with machine learning techniques: ways to improve accuracy",
abstract = "Previous studies on predicting the box-office performance of a movie using machine learning techniques have shown practical levels of predictive accuracy. Their works are technically- and methodologically-oriented, focusing mainly on what algorithms are better at predicting the movie performance. However, the accuracy of prediction model can also be elevated by taking other perspectives such as introducing unexplored features that might be related to the prediction of the outcomes. In this paper, we examine multiple approaches to improve the performance of the prediction model. First, we develop and add a new feature derived from the theory of transmedia storytelling. Such theory-driven feature selection not only increases the forecast accuracy, but also enhances the interpretability of a prediction model. Second, we use an ensemble approach, which has rarely been adopted in the research on predicting box-office performance. As a result, the proposed model, Cinema Ensemble Model (CEM), outperforms the prediction models from the past studies that use machine learning algorithms. We suggest that CEM can be extensively used for industrial experts as a powerful tool for improving decision-making process.The final publication is available at Springer via http://dx.doi.org/[10.1007/s10796-016-9689-z",
keywords = "Prediction model, Movie performance, Machine learning techniques, Cinema ensemble model, Transmedia storytelling, Feature selection",
author = "K. Lee and J. Park and I. Kim and Youngseok Choi",
note = "The final publication is available at Springer via http://dx.doi.org/[10.1007/s10796-016-9689-z",
year = "2018",
month = "6",
doi = "10.1007/s10796-016-9689-z",
language = "English",
volume = "20",
pages = "577–588",
journal = "Information Systems Frontiers",
issn = "1387-3326",
publisher = "Springer Verlag",
number = "3",

}

TY - JOUR

T1 - Predicting movie success with machine learning techniques: ways to improve accuracy

AU - Lee, K.

AU - Park, J.

AU - Kim, I.

AU - Choi, Youngseok

N1 - The final publication is available at Springer via http://dx.doi.org/[10.1007/s10796-016-9689-z

PY - 2018/6

Y1 - 2018/6

N2 - Previous studies on predicting the box-office performance of a movie using machine learning techniques have shown practical levels of predictive accuracy. Their works are technically- and methodologically-oriented, focusing mainly on what algorithms are better at predicting the movie performance. However, the accuracy of prediction model can also be elevated by taking other perspectives such as introducing unexplored features that might be related to the prediction of the outcomes. In this paper, we examine multiple approaches to improve the performance of the prediction model. First, we develop and add a new feature derived from the theory of transmedia storytelling. Such theory-driven feature selection not only increases the forecast accuracy, but also enhances the interpretability of a prediction model. Second, we use an ensemble approach, which has rarely been adopted in the research on predicting box-office performance. As a result, the proposed model, Cinema Ensemble Model (CEM), outperforms the prediction models from the past studies that use machine learning algorithms. We suggest that CEM can be extensively used for industrial experts as a powerful tool for improving decision-making process.The final publication is available at Springer via http://dx.doi.org/[10.1007/s10796-016-9689-z

AB - Previous studies on predicting the box-office performance of a movie using machine learning techniques have shown practical levels of predictive accuracy. Their works are technically- and methodologically-oriented, focusing mainly on what algorithms are better at predicting the movie performance. However, the accuracy of prediction model can also be elevated by taking other perspectives such as introducing unexplored features that might be related to the prediction of the outcomes. In this paper, we examine multiple approaches to improve the performance of the prediction model. First, we develop and add a new feature derived from the theory of transmedia storytelling. Such theory-driven feature selection not only increases the forecast accuracy, but also enhances the interpretability of a prediction model. Second, we use an ensemble approach, which has rarely been adopted in the research on predicting box-office performance. As a result, the proposed model, Cinema Ensemble Model (CEM), outperforms the prediction models from the past studies that use machine learning algorithms. We suggest that CEM can be extensively used for industrial experts as a powerful tool for improving decision-making process.The final publication is available at Springer via http://dx.doi.org/[10.1007/s10796-016-9689-z

KW - Prediction model

KW - Movie performance

KW - Machine learning techniques

KW - Cinema ensemble model

KW - Transmedia storytelling

KW - Feature selection

U2 - 10.1007/s10796-016-9689-z

DO - 10.1007/s10796-016-9689-z

M3 - Article

VL - 20

SP - 577

EP - 588

JO - Information Systems Frontiers

JF - Information Systems Frontiers

SN - 1387-3326

IS - 3

ER -