# improve logistic regression model

**URL:** <https://forum.knime.com/t/improve-logistic-regression-model/76225>\
**Category:** KNIME Analytics Platform\
**Tags:** machine-learning\
**Created:** [December 23, 2023, 6:56pm UTC](https://forum.knime.com/t/improve-logistic-regression-model/76225 "2023-12-23T18:56:40Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![mohammad\_alqoqa](https://forum-cdn.knime.com/user_avatar/forum.knime.com/mohammad_alqoqa/32/55687_2.png) [@mohammad\_alqoqa](https://forum.knime.com/u/mohammad_alqoqa)\
**Post date:** [December 23, 2023, 6:56pm UTC](https://forum.knime.com/t/improve-logistic-regression-model/76225/1 "2023-12-23T18:56:40Z")

</div>

hi everyone, it’s a basic question but I’m new to machine learning, I’m trying to improve the logistic regression model but the best score I got is less than 75%, I saw many videos on YouTube but nothing new in the score. I uploaded the workflow if anyone can help with that. and how you do it it’s important to me.  
what parameter must be changed?

 ![lr](https://forum-cdn.knime.com/uploads/default/original/3X/f/2/f221d843bb251818ef661de84f28e58edd706ec9.png)

[logistic regression.knwf](https://forum.knime.com/uploads/short-url/8AEWLwTfhjSyrceEPlAEhq9SDcC.knwf) (25.8 KB)

I use this dataset  
[heart-csv.csv](https://forum.knime.com/uploads/short-url/59bRGA1133zXQzPs32P1zdCpkNE.csv) (51.2 KB)

---

<div class="post-metadata">

**Author:** ![izaychik63](https://forum-cdn.knime.com/letter_avatar_proxy/v4/letter/i/e9a140/32.png) [@izaychik63](https://forum.knime.com/u/izaychik63)\
**Post date:** [December 23, 2023, 8:57pm UTC](https://forum.knime.com/t/improve-logistic-regression-model/76225/2 "2023-12-23T20:57:27Z")

</div>

First of all sample has to be proportional. Also, look at node description and other examples here:

> **[Logistic Regression Learner — NodePit](https://nodepit.com/node/org.knime.base.node.mine.regression.logistic.learner4.LogRegLearnerNodeFactory4)**
>
> Performs a multinomial logistic regression.

---

<div class="post-metadata">

**Author:** ![mohammad\_alqoqa](https://forum-cdn.knime.com/user_avatar/forum.knime.com/mohammad_alqoqa/32/55687_2.png) [@mohammad\_alqoqa](https://forum.knime.com/u/mohammad_alqoqa)\
**Post date:** [December 23, 2023, 9:15pm UTC](https://forum.knime.com/t/improve-logistic-regression-model/76225/3 "2023-12-23T21:15:21Z")

</div>

> [@izaychik63](#):
>
> First of all sample has to be proportional

I need more explanation here (First of all sample has to be proportional) if you have time.

---

<div class="post-metadata">

**Author:** ![izaychik63](https://forum-cdn.knime.com/letter_avatar_proxy/v4/letter/i/e9a140/32.png) [@izaychik63](https://forum.knime.com/u/izaychik63)\
**Post date:** [December 23, 2023, 11:42pm UTC](https://forum.knime.com/t/improve-logistic-regression-model/76225/4 "2023-12-23T23:42:46Z")

</div>

Please see below

> <https://stats.stackexchange.com/questions/171983/logistic-regression-the-relation-between-sample-proportion-and-prediction>

---

<div class="post-metadata">

**Author:** ![system](https://forum-cdn.knime.com/uploads/default/original/1X/ab3ccf34482a0329361734a18199390177204f15.png) [@system](https://forum.knime.com/u/system)\
**Post date:** [March 22, 2024, 11:43pm UTC](https://forum.knime.com/t/improve-logistic-regression-model/76225/5 "2024-03-22T23:43:07Z")

</div>

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.