# Problem with Term Frequency

**URL:** <https://forum.knime.com/t/problem-with-term-frequency/9381>\
**Category:** KNIME Extensions\
**Tags:** text-processing\
**Created:** [September 23, 2017, 1:30am UTC](https://forum.knime.com/t/problem-with-term-frequency/9381 "2017-09-23T01:30:56Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![PacoAlbano](https://forum-cdn.knime.com/letter_avatar_proxy/v4/letter/p/a9a28c/32.png) [@PacoAlbano](https://forum.knime.com/u/PacoAlbano)\
**Post date:** [September 23, 2017, 1:30am UTC](https://forum.knime.com/t/problem-with-term-frequency/9381/1 "2017-09-23T01:30:56Z")

</div>

Hi,&nbsp;  
  
I have a text and I want to extract the term frequency. Aftrer applying a stemmer node and other preprocessing (stop word filter, punctuation erasure, etc.), I used the BoW node and the term frequencies node on the follwoing text:

- Reviews of Apple's new iPhone 8 and 8 Plus have been laudatory. However, the reviewers can't seem to get their minds off &nbsp;the jewel of the Apple universe, the iPhone X. Both the iPhone 8 and 8 Plus are "awesome" and better than last year's models -- but iPhone shoppers who want to be part of the future will save their money and buy an iPhone X later in the year, suggested reviewer David Pierce.

The output regarding the absolute term frequency is strange. The stemmed term **review** has a an absolute term frequency of 6 and the stemmed term **iPhone** &nbsp;has a frequency of 10.&nbsp;  
I only count **review** 3 times and **iPhone** &nbsp; 5 times.&nbsp;  
  
Someone has an idea?&nbsp;  
  
Attached, you find the output table of the term frequency node.&nbsp;

&nbsp;

Thank you for help  
&nbsp;

---

<div class="post-metadata">

**Author:** ![PacoAlbano](https://forum-cdn.knime.com/letter_avatar_proxy/v4/letter/p/a9a28c/32.png) [@PacoAlbano](https://forum.knime.com/u/PacoAlbano)\
**Post date:** [September 23, 2017, 6:53pm UTC](https://forum.knime.com/t/problem-with-term-frequency/9381/2 "2017-09-23T18:53:54Z")

</div>

Hi,&nbsp;  
  
have found an explaination to my problem in another forum post:&nbsp;  
  
  
https://www.knime.com/forum/knime-textprocessing/understanding-the-bow-tf-and-idf-nodes-in-textprocessing  
  
&nbsp;

---

<div class="post-metadata">

**Author:** ![ipazin](https://forum-cdn.knime.com/user_avatar/forum.knime.com/ipazin/32/806_2.png) [@ipazin](https://forum.knime.com/u/ipazin)\
**Post date:** [November 23, 2020, 7:49am UTC](https://forum.knime.com/t/problem-with-term-frequency/9381/4 "2020-11-23T07:49:21Z")

</div>


