| Rank in Wordlist | Word | Rank in Wordlist | Word |
|---|---|---|---|
| 1 | un | 26 | tas |
| 2 | ir | 27 | jo |
| 3 | ar | 28 | tika |
| 4 | ka | 29 | gada |
| 5 | par | 30 | valsts |
| 6 | no | 31 | vēl |
| 7 | arī | 32 | pie |
| 8 | uz | 33 | Latvijā |
| 9 | kas | 34 | Rīgas |
| 10 | bet | 35 | var |
| 11 | kā | 36 | – |
| 12 | Latvijas | 37 | kad |
| 13 | bija | 38 | Eiropas |
| 14 | jau | 39 | taču |
| 15 | nav | 40 | tiek |
| 16 | līdz | 41 | Kā |
| 17 | pēc | 42 | savu |
| 18 | to | 43 | latu |
| 19 | gan | 44 | eiro |
| 20 | vai | 45 | kurš |
| 21 | lai | 46 | viņš |
| 22 | tā | 47 | tad |
| 23 | laikā | 48 | pret |
| 24 | būs | 49 | ko |
| 25 | tikai | 50 | ļoti |
The table shows the top-50 words of the corpus. Usually we see stopwords.
Language: Afrikaans
This list is a good candidate for a first stopword list for a language.
Usually a small, balanced corpus is enough to get a good list of high frequent words. But if the small corpus has some very prominent topic, this will be visible even in the top word lists.
select w_id-100 as rank_in_wordlist, word from words where w_id>100 order by w_id limit 50;
3.4 Sample words for different frequency ranges