Skip to content

Latest commit

 

History

History
74 lines (60 loc) · 3.12 KB

README.md

File metadata and controls

74 lines (60 loc) · 3.12 KB
Raw

Wearables-Driven Freeform Handwriting Dataset

The database (using PostgreSQL) is comprised of three tables and five experimental scenarios. This document will describe the dataset, database, and how to use it.

You can find the dataset here: https://drive.google.com/open?id=1trBm0BcdzZvYearEVcPB-UdQ797Tf_Pb

The Dataset

Our dataset includes the sensor data for the following experimental scenarios (please refer to the paper for further details):

  • Baseline Experiments
    • E1: The scenario in which users copied the same content.
    • E2: Users copied unique content from one another and across sessions.
    • E3: Emulating a freewriting scenario in which users experience a higher cognitive load, this experiment had users answer essay-style questions.
  • Attack Experiments
    • E4: Attackers are given copies of their target's handwriting to practice against.
    • E5: Each attacker viewed the video footage of their target's wrist and hand during the writing process.

The Database

This section describes the different tables included in our database and what the possible values mean.

Baseline Experiments Table

TABLE baseline_experiments (
    user_session_id TEXT PRIMARY KEY,
    user_id BIGINT NOT NULL,
    session BIGINT NOT NULL,
    scenario BIGINT NOT NULL
)

Definitions:

  • user_session_id: user_id + session + scenario.
  • user_id: the anonymized ID of the user.
  • session: 1 or 0, depending on if this was the training session.
  • scenario: 1 (E1), 2 (E2), or 3 (E3) depending on which scenario the data is from.

Impostor Experiments Table

TABLE impostor_experiments (
    user_session_id TEXT PRIMARY KEY,
    target_user BIGINT NOT NULL,
    attacking_user BIGINT NOT NULL,
    experiment BIGINT NOT NULL
    attack BIGINT NOT NULL
)

Definitions:

  • user_session_id: attacking_user + target_user + experiment + attack
  • target_user: the anonymized ID of the user.
  • attacking_user: the anonymized ID of the attacker.
  • experiment: 1 (E1), 2 (E2), or 3 (E3) depending on which scenario the training data is from.
  • attack: 4 (E4) or 5 (E5) depending on which attack method was used for this particular set of data.

Handwriting Data Table

TABLE handwriting_data (
    timestamp TEXT NOT NULL,
    x DOUBLE PRECISION NOT NULL,
    y DOUBLE PRECISION NOT NULL,
    z DOUBLE PRECISION NOT NULL,
    user_session_id TEXT NOT NULL,
    index BIGINT NOT NULL,
    PRIMARY KEY(user_session_id, index)
)

Definitions:

  • timestamp: The datetime difference between the start and end of the session.
  • x: the x component of the linear acceleration sensor data.
  • y: the y component of the linear acceleration sensor data.
  • z: the z component of the linear acceleration sensor data.
  • user_session_id: user_session_id from either the baseline or impostor experiment table.
  • index: the index of the data in a given csv file.

Usage Instructions

The dataset was exported with pgAdmin and can be imported with the psql command.

psql -U your_user_name your_db_name < dataset.sql