StanfordHCI/dreamteam-task-set

Name: dreamteam-task-set

Owner: Stanford HCI

Description: DreamTeam (CHI 2018) Codenames task set

Created: 2018-05-23 04:00:48.0

Updated: 2018-05-23 04:34:50.0

Pushed: 2018-05-23 04:34:36.0

Homepage: http://hci.stanford.edu/publications/2018/dreamteam/zhou_sharon_dreamteam.pdf

Size: 6

Language: null

GitHub Committers

UserMost Recent Commit# Commits

Other Committers

UserEmailMost Recent Commit# Commits

README

DreamTeam (CHI 2018) Codenames task set

This is the Codenames task set used for DreamTeam, a system introduced in the CHI 2018 paper In Search of the Dream Team: Temporally Constrained Multi-Armed Bandits for Identifying Effective Team Structures (Sharon Zhou, Melissa Valentine, Michael Bernstein).

Using this task set

There are 11 game boards in a list of dictionary objects. Each board is enclosed in a dictionary object, e.g.


"dice":
    [
        "processor",
        "dots",
        "chance"
    ],
"invalid":
    [
        "null",
        "patient"
    ],
"compress":
    [
        "swelling",
        "encode"
    ]

Here, dice, invalid, and compress are clues used for the board. The list of words after each clue contains the corresponding answers.

The board itself does not include the clues, and is comprised of the words processor, dots, chance, null, patient, swelling, and encode. Ideally, these are scrambled in random order.

The common way of presenting clues is dice, 3 which indicates that there are 3 words on the board (processor, dots, chance) corresponding to the clue dice.

About Codenames and this adaptation

This task is adapted from the popular board game Codenames.

The premise of the game is that there is a board of words, e.g. cat, tree, dog. There are then clues that correspond to a subset of words on the board, e.g. pet, 2 where pet refers to 2 words on the board and plant, 1 where plant refers to 1 word on the board.

The players do not know which words on the board correspond to each clue and have to decide on which words to guess. In this adaptation, each word has only one clue. In traditional Codenames, words can have many clues. The game can be collaborative between all players, or be competitive between teams of players who get different clues. There are variations online, and we leave it open to your discretion on how to further employ this task set.


This work is supported by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number U24TR002306. This work is solely the responsibility of the creators and does not necessarily represent the official views of the National Institutes of Health.