Google Ngram Viewer

On the page above, there is such a sentence:

Graph these case-sensitive comma-separated phrases between 1800 and 2000 from the corpus English with a smoothing of 3.

What does smoothing mean in that context?

closed as off-topic by Kris, Rory Alsop, tchrist, Matt E. Эллен, choster Sep 4 '13 at 13:38

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "Questions that can be answered using commonly-available references are off-topic. A list of these references can be found here: List of general references" – Kris, Rory Alsop, tchrist, Matt E. Эллен, choster
If this question can be reworded to fit the rules in the help center, please edit the question.

  • A "peak" of the graph of only height 3 will not appear visibly in the graph. You smoothen the edges. – skymningen Aug 23 '13 at 6:12
  • Adjust the 'smoothing' parameter on an Ngram (eg the one you link to) to progressively higher settings to see how the 'jaggedness' (fine detail - which probably isn't too accurate and would be different if a slightly different sampling technique was being used) is progressively smoothed out. You could open one-over-the-other views to compare the different graphs directly. – Edwin Ashworth Aug 23 '13 at 7:20

RTM: http://books.google.com/ngrams/info

What does "smoothing" mean?

Often trends become more apparent when data is viewed as a moving average. A smoothing of 1 means that the data shown for 1950 will be an average of the raw count for 1950 plus 1 value on either side: ("count for 1949" + "count for 1950" + "count for 1951"), divided by 3. So a smoothing of 10 means that 21 values will be averaged: 10 on either side, plus the target value in the center of them.

At the left and right edges of the graph, fewer values are averaged. With a smoothing of 3, the leftmost value (pretend it's the year 1950) will be calculated as ("count for 1950" + "count for 1951" + "count for 1952" + "count for 1953"), divided by 4.

A smoothing of 0 means no smoothing at all: just raw data.

  • Thanks. By the way what does "RTM" mean? – Terry Li Aug 23 '13 at 7:50
  • 3
    Read the manual :) Sometimes written RTFM, but that is too rude... – mplungjan Aug 23 '13 at 7:58

It's a technical (but somewhat loosely defined) term in statistics and other data-processing fields.

Wikipedia summarizes it as:

In statistics and image processing, to smooth a data set is to create an approximating function that attempts to capture important patterns in the data, while leaving out noise or other fine-scale structures/rapid phenomena.

Essentially, it denotes techniques that are used to make data sets (usually represented as graphs or images of some form) look smoother, less noisy, less sharp.

The number (3) is a parameter to the smoothing algorithm used. Chances are this is a form of moving/sliding average calculation, with a "width" of 3 (i.e. the mean of 3 consecutive data points are considered rather than the exact value of each point).

Here's an example of smoothing (using Bézier curves, generated by gnuplot):

enter image description here

The red line represents data with a lot of "noise", the green one is the smoothed version: the peaks and valleys are evened out, to form a more fluid curve.

Not the answer you're looking for? Browse other questions tagged or ask your own question.