Decision Tree Plug-in (DTP) for Neo4j

A plug-in to create decision tree algorithms in Neo4j with the following splitting criteria

• Gini Index
• Information Gain
• Gain Ratio

About

Methodology

- Compile the JAR File of DTP Yourself

- Download the JAR File of DTP

Procedure Stack

Features

DTP comprises of 19 procedures, which read CSV-files, map nodes, split data, generate decision tree using 3 different metrics , perform k-fold cross validation, validate the classifier and visualize tree.

Procedures to

---> map data into training and testing

- Query data for training and testing

- Display training and testing data after mapping

---> create Decision Tree

- from Nodes in Neo4j

- from CSV file paths

- Confusion matrix (from nodes or CSV files without tree visualization)

---> Cross-validation

- Cross validation (nodes or CSV files)

Data (check wiki)

• Dataset 1 : Heart Failure Prediction

Citation
Davide Chicco, Giuseppe Jurman: 
"Machine learning can predict survival of patients with 
heart failure from serum creatinine and ejection fraction alone." 
BMC Medical Informatics and Decision Making 20, 16 (2020). 
(https://www.kaggle.com/andrewmvd/heart-failure-clinical-data)

License
CC BY 4.0

• Dataset 2 : Prediction of Inflammatory Bowel Disease from Microbiome

Citation
Lehmann, T et al. 
“Metaproteomics of fecal samples of Crohn's disease and Ulcerative Colitis.” 
Journal of proteomics vol. 201 (2019): 93-103. 
doi:10.1016/j.jprot.2019.04.009
(https://pubmed.ncbi.nlm.nih.gov/31009805/)

• Dataset 3 : H1N1 COVID19 Classification

Citation
Li, W., Ma, J., Shende, N. et al. 
"Using machine learning of clinical data to diagnose COVID-19: 
a systematic review and meta-analysis." 
BMC Med Inform Decis Mak 20, 247 (2020). 
https://doi.org/10.1186/s12911-020-01266-z
(https://github.com/yoshihiko1218/COVID19ML)

• Dataset 4 : Diabetes Health Indicators

Citation
Alex Teboul 
(https://www.kaggle.com/alexteboul/diabetes-health-indicators-dataset)
BRFSS 2015
(https://www.kaggle.com/cdc/behavioral-risk-factor-surveillance-system)