Skip to content

codebook.md

Latest commit

 

History

History
45 lines (35 loc) · 1.22 KB

codebook.md

File metadata and controls

45 lines (35 loc) · 1.22 KB

CodeBook

Human Activity Recognition Using Smartphones

tidy data set for R

General information

  • Number of volunteers: 30
  • Ages: 19-48
  • Used Device: Samsung Galaxy SII (worn on the waist)
  • Measurements: 3-axial linear acceleration and 3-axial angular velocity at 50 Hz
  • Normalization: [-1 , 1]
  • Estimations: mean value and standard deviation
  • Agglomeration: each volunteer and each activity

Data description

  • Subject: number of volunteer (range: 1-30)
  • Activity: type of movement during measurement (WALKING, WALKING_UPSTAIRS, WALKING_DOWNSTAIRS, SITTING, STANDING, LAYING)

Measurement names contain 5-7 parts.

  • first part: prefix

  • t: linear acceleration and angular velocity in time context

  • f: Fast Fourier Transform (FFT) was applied

  • second part: force distinction

  • Body

  • Gravity

  • third part: measurement type

  • Acc: acceleration [g]

  • Gyro: angular velocity [rad/seconds]

  • fourth part: jerk signal (optional)

  • Jerk: jerk signal

  • fifth part: magnitude (optional)

  • Mag: magnitude (euclidean norm)

  • sixth part: estimated variable from signal

  • mean(): mean value

  • std(): standard deviation

  • seventh part: direction

  • X

  • Y

  • Z