Dataset: Companion Plants

About this Dataset

Companion Plants

Introduction
Companion planting in gardening and agriculture is the planting of different crops in proximity for any of a number of different reasons, including pest control, pollination, providing habitat for beneficial insects, maximizing use of space, and to otherwise increase crop productivity. Companion planting is a form of polyculture.

Companion planting is used by farmers and gardeners in both industrialized and developing countries for many reasons. Many of the modern principles of companion planting were present many centuries ago in forest gardens in Asia, and thousands of years ago in Mesoamerica.

Dataset
Dataset comes from table on Wiki page "List of companion plants".
It was formatted with application to Graph-based tasks in mind, such as Link Prediction and Node Classification, often used in recommendation systems. Dataset has 996 rows and four columns in the dataset:

Source Node - plant name where link starts
Link - link type (helps, helped_by, avoid)
Destination Node - plant name where link ends
Source Property - class for the source node (vegetables, herbs, flowers, fruits)

While in the original dataset there are three link types, helped_by is just reverse of helps (logically only, dataset does not contain both links for each connected pair of Source and Destination Nodes). Some Destination Nodes are not present among the Source Nodes.

Tables

Companion Plants

@kaggle.aramacus_companion_plants.companion_plants

7.62 KB
995 rows
4 columns


CREATE TABLE companion_plants (
  "source_node" VARCHAR,
  "link" VARCHAR,
  "destination_node" VARCHAR,
  "source_type" VARCHAR
);