# Design tips when modelling for machine learning

**URL:** <https://community.neo4j.com/t/design-tips-when-modelling-for-machine-learning/42673>\
**Category:** Modeling\
**Created:** [August 10, 2021, 3:35am UTC](https://community.neo4j.com/t/design-tips-when-modelling-for-machine-learning/42673 "2021-08-10T03:35:06Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![samchgm](https://sea1.discourse-cdn.com/flex021/user_avatar/community.neo4j.com/samchgm/32/3868_2.png) [@samchgm](https://community.neo4j.com/u/samchgm)\
**Post date:** [August 10, 2021, 3:35am UTC](https://community.neo4j.com/t/design-tips-when-modelling-for-machine-learning/42673/1 "2021-08-10T03:35:06Z")

</div>

Hello,

What are good references for designing a graph and applying machine learning? I am looking for:

1. Tips on when an entity should be a node vs property of a node to best facilitate ML?
2. Walk-through of dataset to problem to DB design to ML

Any pointers? Thanks.