ASK

A taxonomy of accuracy, social, and knowledge information seeking posts in social question and answering

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Many people turn to their social networks to find information through the practice of question and answering. We believe it is necessary to use different answering strategies based on the type of questions to accommodate the different information needs. In this research, we propose the ASK taxonomy that categorizes questions posted on social networking sites into three types according to the nature of the questioner's inquiry of accuracy, social, or knowledge. To automatically decide which answering strategy to use, we develop a predictive model based on ASK question types using question features from the perspectives of lexical, topical, contextual, and syntactic as well as answer features. By applying the classifier on an annotated data set, we present a comprehensive analysis to compare questions in terms of their word usage, topical interests, temporal and spatial restrictions, syntactic structure, and response characteristics. Our research results show that the three types of questions exhibited different characteristics in the way they are asked. Our automatic classification algorithm achieves an 83% correct labeling result, showing the value of the ASK taxonomy for the design of social question and answering systems.

Original languageEnglish
JournalJournal of the Association for Information Science and Technology
DOIs
Publication statusAccepted/In press - 2016

Fingerprint

Taxonomies
taxonomy
Syntactics
predictive model
research results
networking
social network
Labeling
Classifiers
Values
Amplitude shift keying
Taxonomy
Information seeking
Social networks
Classifier
Social networking sites
Information needs

ASJC Scopus subject areas

  • Information Systems and Management
  • Library and Information Sciences
  • Computer Networks and Communications
  • Information Systems

Cite this

@article{de74503a35b543fa899f4f4c1308fd3d,
title = "ASK: A taxonomy of accuracy, social, and knowledge information seeking posts in social question and answering",
abstract = "Many people turn to their social networks to find information through the practice of question and answering. We believe it is necessary to use different answering strategies based on the type of questions to accommodate the different information needs. In this research, we propose the ASK taxonomy that categorizes questions posted on social networking sites into three types according to the nature of the questioner's inquiry of accuracy, social, or knowledge. To automatically decide which answering strategy to use, we develop a predictive model based on ASK question types using question features from the perspectives of lexical, topical, contextual, and syntactic as well as answer features. By applying the classifier on an annotated data set, we present a comprehensive analysis to compare questions in terms of their word usage, topical interests, temporal and spatial restrictions, syntactic structure, and response characteristics. Our research results show that the three types of questions exhibited different characteristics in the way they are asked. Our automatic classification algorithm achieves an 83{\%} correct labeling result, showing the value of the ASK taxonomy for the design of social question and answering systems.",
author = "Zhe Liu and Bernard Jansen",
year = "2016",
doi = "10.1002/asi.23655",
language = "English",
journal = "Journal of the Association for Information Science and Technology",
issn = "2330-1635",
publisher = "John Wiley and Sons Ltd",

}

TY - JOUR

T1 - ASK

T2 - A taxonomy of accuracy, social, and knowledge information seeking posts in social question and answering

AU - Liu, Zhe

AU - Jansen, Bernard

PY - 2016

Y1 - 2016

N2 - Many people turn to their social networks to find information through the practice of question and answering. We believe it is necessary to use different answering strategies based on the type of questions to accommodate the different information needs. In this research, we propose the ASK taxonomy that categorizes questions posted on social networking sites into three types according to the nature of the questioner's inquiry of accuracy, social, or knowledge. To automatically decide which answering strategy to use, we develop a predictive model based on ASK question types using question features from the perspectives of lexical, topical, contextual, and syntactic as well as answer features. By applying the classifier on an annotated data set, we present a comprehensive analysis to compare questions in terms of their word usage, topical interests, temporal and spatial restrictions, syntactic structure, and response characteristics. Our research results show that the three types of questions exhibited different characteristics in the way they are asked. Our automatic classification algorithm achieves an 83% correct labeling result, showing the value of the ASK taxonomy for the design of social question and answering systems.

AB - Many people turn to their social networks to find information through the practice of question and answering. We believe it is necessary to use different answering strategies based on the type of questions to accommodate the different information needs. In this research, we propose the ASK taxonomy that categorizes questions posted on social networking sites into three types according to the nature of the questioner's inquiry of accuracy, social, or knowledge. To automatically decide which answering strategy to use, we develop a predictive model based on ASK question types using question features from the perspectives of lexical, topical, contextual, and syntactic as well as answer features. By applying the classifier on an annotated data set, we present a comprehensive analysis to compare questions in terms of their word usage, topical interests, temporal and spatial restrictions, syntactic structure, and response characteristics. Our research results show that the three types of questions exhibited different characteristics in the way they are asked. Our automatic classification algorithm achieves an 83% correct labeling result, showing the value of the ASK taxonomy for the design of social question and answering systems.

UR - http://www.scopus.com/inward/record.url?scp=84959576682&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959576682&partnerID=8YFLogxK

U2 - 10.1002/asi.23655

DO - 10.1002/asi.23655

M3 - Article

JO - Journal of the Association for Information Science and Technology

JF - Journal of the Association for Information Science and Technology

SN - 2330-1635

ER -