Evaluating classification schemes for second screen interactions

Partha Mukherjee, Bernard Jansen

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

3 Citations (Scopus)

Abstract

We analyze the performance of classification schemes on information collected from social conversation posted in Twitter among audiences of a popular US based TV show. In this research, we consider entropy as a measure of information exchange in a group conversation that is related to social, temporal, and second screen device features. The group conversations are identified by hashtags present in tweets where the number of members in the group interacting is at least two. We apply different classification schemes to more than 4,900 groups identified from 318,000 tweets. The result shows that 5-nn algorithm outperformed the other procedures in terms of misclassification error to identify the informed groups.

Original languageEnglish
Title of host publication2015 International Conference on Computing, Networking and Communications, ICNC 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages879-883
Number of pages5
ISBN (Print)9781479969593
DOIs
Publication statusPublished - 26 Mar 2015
Externally publishedYes
Event2015 International Conference on Computing, Networking and Communications, ICNC 2015 - Garden Grove, United States
Duration: 16 Feb 201519 Feb 2015

Other

Other2015 International Conference on Computing, Networking and Communications, ICNC 2015
CountryUnited States
CityGarden Grove
Period16/2/1519/2/15

Fingerprint

Entropy

Keywords

  • entropy
  • informed groups
  • misclassification
  • PCA
  • second screen
  • social TV

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Mukherjee, P., & Jansen, B. (2015). Evaluating classification schemes for second screen interactions. In 2015 International Conference on Computing, Networking and Communications, ICNC 2015 (pp. 879-883). [7069462] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCNC.2015.7069462

Evaluating classification schemes for second screen interactions. / Mukherjee, Partha; Jansen, Bernard.

2015 International Conference on Computing, Networking and Communications, ICNC 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 879-883 7069462.

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

Mukherjee, P & Jansen, B 2015, Evaluating classification schemes for second screen interactions. in 2015 International Conference on Computing, Networking and Communications, ICNC 2015., 7069462, Institute of Electrical and Electronics Engineers Inc., pp. 879-883, 2015 International Conference on Computing, Networking and Communications, ICNC 2015, Garden Grove, United States, 16/2/15. https://doi.org/10.1109/ICCNC.2015.7069462
Mukherjee P, Jansen B. Evaluating classification schemes for second screen interactions. In 2015 International Conference on Computing, Networking and Communications, ICNC 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 879-883. 7069462 https://doi.org/10.1109/ICCNC.2015.7069462
Mukherjee, Partha ; Jansen, Bernard. / Evaluating classification schemes for second screen interactions. 2015 International Conference on Computing, Networking and Communications, ICNC 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 879-883
@inproceedings{fdde0ea209d448cabefd3512b0d2a270,
title = "Evaluating classification schemes for second screen interactions",
abstract = "We analyze the performance of classification schemes on information collected from social conversation posted in Twitter among audiences of a popular US based TV show. In this research, we consider entropy as a measure of information exchange in a group conversation that is related to social, temporal, and second screen device features. The group conversations are identified by hashtags present in tweets where the number of members in the group interacting is at least two. We apply different classification schemes to more than 4,900 groups identified from 318,000 tweets. The result shows that 5-nn algorithm outperformed the other procedures in terms of misclassification error to identify the informed groups.",
keywords = "entropy, informed groups, misclassification, PCA, second screen, social TV",
author = "Partha Mukherjee and Bernard Jansen",
year = "2015",
month = "3",
day = "26",
doi = "10.1109/ICCNC.2015.7069462",
language = "English",
isbn = "9781479969593",
pages = "879--883",
booktitle = "2015 International Conference on Computing, Networking and Communications, ICNC 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Evaluating classification schemes for second screen interactions

AU - Mukherjee, Partha

AU - Jansen, Bernard

PY - 2015/3/26

Y1 - 2015/3/26

N2 - We analyze the performance of classification schemes on information collected from social conversation posted in Twitter among audiences of a popular US based TV show. In this research, we consider entropy as a measure of information exchange in a group conversation that is related to social, temporal, and second screen device features. The group conversations are identified by hashtags present in tweets where the number of members in the group interacting is at least two. We apply different classification schemes to more than 4,900 groups identified from 318,000 tweets. The result shows that 5-nn algorithm outperformed the other procedures in terms of misclassification error to identify the informed groups.

AB - We analyze the performance of classification schemes on information collected from social conversation posted in Twitter among audiences of a popular US based TV show. In this research, we consider entropy as a measure of information exchange in a group conversation that is related to social, temporal, and second screen device features. The group conversations are identified by hashtags present in tweets where the number of members in the group interacting is at least two. We apply different classification schemes to more than 4,900 groups identified from 318,000 tweets. The result shows that 5-nn algorithm outperformed the other procedures in terms of misclassification error to identify the informed groups.

KW - entropy

KW - informed groups

KW - misclassification

KW - PCA

KW - second screen

KW - social TV

UR - http://www.scopus.com/inward/record.url?scp=84928036826&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84928036826&partnerID=8YFLogxK

U2 - 10.1109/ICCNC.2015.7069462

DO - 10.1109/ICCNC.2015.7069462

M3 - Conference contribution

SN - 9781479969593

SP - 879

EP - 883

BT - 2015 International Conference on Computing, Networking and Communications, ICNC 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -