- Clone this repo
- Run
npm install - Modify
.envwith your database settings
If you want to play with some data, I've written a script to pull down handset data from the Samsung website (AEM) so you can import it into your Neo4j DB.
From your project run npm run seed.
Once the command is complete, run the below query in Neo4j desktop. Make sure you change the devices.json file path to your project folder.
CALL apoc.load.json('file:///path/to/project/data/devices.json') YIELD value AS device
MERGE (product:Product { id: device.code, name: device.name, thumb: device.thumbUrl, price: device.price, released: datetime({epochmillis: apoc.date.parse(device.released, 'ms', 'yyyy-MM-dd')}) })
MERGE (category:Category { id: device.category.id, name: device.category.name })
MERGE (color:Color { name: device.color })
MERGE (product)-[:BELONGS_TO]->(category)
MERGE (product)-[:HAS_COLOR]->(color)
WITH device, product
UNWIND device.features AS feature
WITH feature.key AS feature_key, feature.value AS feature_value, product
MERGE (feature:Feature {key: feature_key, value: feature_value})
MERGE (product)-[:HAS_FEATURE]->(feature)
TODO
TODO
TODO
TODO
This will give you devices that are relevant to a device model
MATCH (o:Product {id: 'SM-G975FCWGBTU' })-[:HAS_FEATURE]->(f:Feature)<-[:HAS_FEATURE]-(p:Product)
WHERE p.price > (o.price / 1.3)
WITH p as product, count(*) AS relevancy, p.released as released, collect(distinct f.value) as features
RETURN product
ORDER BY relevancy, product.released DESC