1

Three columns/values in the dataset may be unclear without documentation:

  1. S5_CO2_Slope: This column represents the change rate of carbon dioxide (CO2) levels from sensor 5. Without documentation, it might not be obvious that this column indicates the slope of CO2 level changes over time, potentially leading to misinterpretation.
  2. S6_PIR and S7_PIR: These columns contain data from Passive Infrared (PIR) sensors. Without documentation, it might not be immediately clear what these abbreviations signify or how the readings are relevant. Understanding their association with PIR sensors aids in interpreting their role in the dataset.
  3. Room_Occupancy_Count: While the name implies it denotes the number of occupants in the room, lacking documentation might obscure how this count is calculated or its significance. Additionally, it might not be apparent whether this count reflects current occupancy or accumulative occupancy over time.

The encoding choices likely stem from factors like data collection requirements, sensor capabilities, and specific application needs. For example, including the S5_CO2_Slope column could provide insights into CO2 concentration trends for assessing air quality.

Not consulting the documentation could lead to misinterpretations. For instance:

  • Interpreting S5_CO2_Slope as a static CO2 level might yield inaccurate conclusions about air quality trends.
  • Misunderstanding the purpose of PIR sensors (S6_PIR and S7_PIR) could result in overlooking their importance in detecting motion or occupancy.
  • Incorrectly assuming the nature of Room_Occupancy_Count could lead to erroneous assumptions about room usage or occupancy patterns.

2

Even after reading the documentation, one thing that’s still unclear about the data is what exactly the numbers in the columns for light intensity (S1_Light to S4_Light) mean. We know they come from different sensors and measure how bright it is, but we’re not sure if they’re using units like lux or lumens, or some other measurement. Knowing this would help us understand how bright the environment is. Without this information, it’s hard to make accurate comparisons or understand what the numbers really tell us about the level of light.

3

data <- read.csv("/Users/manojdannana/Desktop/Manoj/IUPUI/class work/intro to statistics/datasets/Occupancy_Estimation.csv")
# Sample data for illustration
time <- c('10:49:41', '10:50:12', '10:50:42', '10:51:13', '10:51:44')
S1_Light <- c(121, 121, 121, 121, 120)  # Sample light intensity readings from sensor 1

# Convert time to numeric values for plotting
time_numeric <- as.numeric(as.POSIXct(time, format="%H:%M:%S"))

# Plotting the data
plot(time_numeric, S1_Light, type = "b", col = "blue", xlab = "Time", ylab = "Light Intensity", main = "Light Intensity Readings from Sensor 1")
grid()

# Annotation to highlight the issue
text(x = c(time_numeric[3], time_numeric[5]), y = c(121, 120), labels = "Unclear unit of measurement", pos = 3, col = "red")

Explanation:

In this picture, the horizontal line shows time, and the vertical line shows how bright the light is from Sensor 1 (S1_Light column). Each dot represents a specific time when the light intensity was measured.

The problem we’re highlighting is that we don’t know what exactly the numbers on the vertical line mean. We see a red label that says “Unclear unit of measurement” pointing to one of the dots. This means we’re not sure if the number (like 121) represents brightness in units like “lux” or “lumens.” Because of this uncertainty, it’s hard to know how bright the light really is.

Risks and Mitigation Methods:

The risk of not knowing the unit is that we might misunderstand how bright the light is, leading to incorrect conclusions about the lighting conditions. To fix this, we could try to find more information about the unit, test different possibilities, or make sure to tell others about our uncertainty when discussing the data.