Skip to content

This dataset accompanies the ICTIR'20 paper "A Mixed-Method Analysis of Text and Audio Search Interfaceswith Varying Task Complexity"

Notifications You must be signed in to change notification settings

sashavtyurina/audio-serp-ictir-2020

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

This dataset accompanies the following paper:

"A Mixed-Method Analysis of Text and Audio Search Interfaceswith Varying Task Complexity"

Alxandra Vtyurina, Charles L.A. Clarke, Edith Law, Johanne R. Trippas, Horatiu Bota

ICTIR 2020

Structure of the folder:

  • AudioClips
  • TextSnippetsForAudio
  • Images
  • SearchTasks

AudioClips.

Audio clips that were played to the study participants. Each subdirectory corresponds to a task and contains 5 audio files that were presented to participants in Study 1 (AMT) and Study 2 (LAB).

TextSnippetsForAudio

This folder contains search snippets selected for each task using which audio results were generated. Each file corresponds to a search task. Each file has the folowing structure:

  • taskID -- id assigned to the task in [2].
  • snippets -- search snippets collected by issuing the topic of the task to Google search engine.
    • trueRank -- the rank of the result as displayed on Google search results page (SERP)
    • title -- the title of the search result as displayed on the SERP.
    • tlDomain -- the shortened URL of the website.
    • summary -- displayed summary of the result. File ``IN14.009_Phonograph_GOLDEN.txt'' contains only two search results, because this task was used as a quality check task in Study 1 (AMT).

SearchTasks

Contains a single file tasks.json, which lists all tasks used in this project and their corresponding backstories.

  • id -- id assigned to the task in [1];
  • code -- TREC code of the task;
  • complexity -- complexity level of the task according to taxonomy defined in [2]. A - Analyze, U - Understand, R - Remember.
  • search_query -- topic of the task as retrieved from TREC task collections. This search query was issued to collect search results for this task.
  • backstory -- a detailed backstory as defined by [2], explaining the supposed information need of the participant.
  • goldenHIT -- if True, this task was used as a quality check in the crowdsourced experiment (Study 2).
  • training -- if True, this task was used as a training task in the controlled experiment (Study 2).

References:

[1] Peter Bailey, Alistair Moffat, Falk Scholer, and Paul Thomas. User variability and IR system evaluation. SIGIR 2015.

[2] L.W. Anderson, D. R. Krathwohl, and B. S. Bloom. A Taxonomy for Learning, Teaching, and Assessing: A Revision of Bloom’s Taxonomy of Educational Objectives.

About

This dataset accompanies the ICTIR'20 paper "A Mixed-Method Analysis of Text and Audio Search Interfaceswith Varying Task Complexity"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published