diff --git a/algorithms/german/stemmer.html b/algorithms/german/stemmer.html index febaff8..d6d2107 100644 --- a/algorithms/german/stemmer.html +++ b/algorithms/german/stemmer.html @@ -320,10 +320,11 @@

The stemming algorithm

(a) em (not preceded by syst [condition added in Snowball 2.3.0])
(b) ern   er
(c) e   en   es
- (d) s (preceded by a valid s-ending) + (d) s (preceded by a valid s-ending)
+ (e) ln   lns [added in Snowball 2.3.0]

- and delete if in R1. (Of course the letter of the valid s-ending is + and if in R1 then delete (for (a) to (d)) or replace with l (for (e)). (Of course the letter of the valid s-ending is not necessarily in R1.) If an ending of group (c) is deleted, and the ending is preceded by niss, delete the final s.

@@ -470,7 +471,7 @@

The same algorithm in Snowball

do ( [substring] R1 among( 'em' - ( not 'syst' + ( not 'syst' // don't remove -em from words ending -system delete ) 'ern' 'er' @@ -483,6 +484,9 @@

The same algorithm in Snowball

's' ( s_ending delete ) + 'ln' 'lns' + ( <- 'l' + ) ) ) do ( diff --git a/algorithms/german2/stemmer.html b/algorithms/german2/stemmer.html index d57bdd5..4cd9bae 100644 --- a/algorithms/german2/stemmer.html +++ b/algorithms/german2/stemmer.html @@ -58,15 +58,15 @@

Links to resources

We used to present a variant of the main German stemmer, termed "german2" which -was the same as the German stemmer but adjusted the first step to improve -handling of input text where the German letters ä, +was the same as the German stemmer except with the first step adjusted to +improve handling of input text where the German letters ä, ö and ü, were written as ae, oe and ue respectively.

Snowball 2.3.0 added these adjustments to the main German stemmer, so there -is no longer a "german2" variant - just used the "german" stemmer. +is no longer a "german2" variant - just use the "german" stemmer.