# Tag Cloud - Split Words

**URL:** <https://discuss.elastic.co/t/tag-cloud-split-words/287493>\
**Category:** Kibana\
**Created:** [October 23, 2021, 3:13pm UTC](https://discuss.elastic.co/t/tag-cloud-split-words/287493 "2021-10-23T15:13:38Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![mazenrb](https://avatars.discourse-cdn.com/v4/letter/m/f0a364/32.png) [@mazenrb](https://discuss.elastic.co/u/mazenrb)\
**Post date:** [October 23, 2021, 3:13pm UTC](https://discuss.elastic.co/t/tag-cloud-split-words/287493/1 "2021-10-23T15:13:38Z")

</div>

Hello all,

i am trying to analyze spotify data. Therefore i would like to create a Tag Cloud for analyzing the most used words in song names.

When i set up a Word/Tag Cloud and choose my field. The hole song name is counted. But i want to count each word and over the hole dataset i got. How could i do that?

The song name type is keyword

[Bildschirmfoto 2021-10-23 um 17.09.46|690x374](https://discuss.elastic.co/uploads/short-url/wGy7Oi0718boWlTpoEehSLhHokx.png)

Thanks for helping !!

---

<div class="post-metadata">

**Author:** ![Felix\_Roessel](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/felix_roessel/32/41623_2.png) [@Felix\_Roessel](https://discuss.elastic.co/u/Felix_Roessel)\
**Post date:** [October 23, 2021, 3:25pm UTC](https://discuss.elastic.co/t/tag-cloud-split-words/287493/2 "2021-10-23T15:25:09Z")

</div>

In order to get each word you need to split the field first into multiple keywords. There is a split processor in the ingest node pipelines to do this.

---

<div class="post-metadata">

**Author:** ![mazenrb](https://avatars.discourse-cdn.com/v4/letter/m/f0a364/32.png) [@mazenrb](https://discuss.elastic.co/u/mazenrb)\
**Post date:** [October 24, 2021, 8:44am UTC](https://discuss.elastic.co/t/tag-cloud-split-words/287493/3 "2021-10-24T08:44:10Z")

</div>

Hello Felix,

thank you very much. That worked for me. Just one thing, i want to remove the special characters from these keywords. I didn't find any possibility in the ingest node pipelines. How can i achieve this?

Sincerely  
Mazenrb

---

<div class="post-metadata">

**Author:** ![legoguy1000](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/legoguy1000/32/54301_2.png) [@legoguy1000](https://discuss.elastic.co/u/legoguy1000)\
**Post date:** [October 24, 2021, 1:48pm UTC](https://discuss.elastic.co/t/tag-cloud-split-words/287493/4 "2021-10-24T13:48:31Z")

</div>

U can use a `gsub` processor to remove characters. You'll have to define what characters you want to remove.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [November 21, 2021, 1:48pm UTC](https://discuss.elastic.co/t/tag-cloud-split-words/287493/5 "2021-11-21T13:48:42Z")

</div>

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.