Open source recommendation system based on time-series data and statistical analysis. Written in TypeScript and Node.js using Redis for storage. The recommendation system uses the Jaccard index to calculate the intersection between two sets. One set is represented by the maximum possible sum of tag score and the other set is the score sum of user events per tag. The higher the Jaccard index, the higher the recommendation. It uses numbers to represent sets to increase performance.
Features
Use tag score and Jaccard index
Content-based filtering
Event-driven powered engine
Naive exploration of new tags
Suitable for product and content recommendation
Fine-tuning of tag weights
Minimalist and lightweight
Written in TypeScript and Node.js
How it works
How the data is stored:
Actors are stored in Redis as simple String keys with create date timestamps as value.
Items are Set type with tags as members. The item may have multiple tags.
Events are String type with actorId:id:tag:timestamp:ttl and an expire attribute set to ensure freshness of recommendations.
How the data is accessed:
Get the actor with events
Check if the actor exists with EXISTS actor:${id}
Get all user events with SCAN ${loop cursor} MATCH actor:${id}
Delete a single actor
Scan for each related to actor key SCAN ${loop cursor} MATCH actor:${id}*
For each key delete with DEL ${key}
Add a single actor
Scan for each related to actor key SCAN ${loop cursor} MATCH actor:${id}*
For each key delete with DEL ${key}
Add new actor with SET actor:${id} ${Date.now().toString()}
Add a single event
Check if actor exists if flag is set using EXISTS actor:${id}
Add event with SET actor:${id}:${tag}:${date}:${ttl} ${score}
If TTL has been provided, set expiration for event with EXPIRE actor:${id}:${tag}:${date}:${ttl} ${ttl}
Get all items with tags
Get all items with SCAN ${loop cursor} MATCH item:*
For each item get all tags with SMEMBERS ${itemKey}
Open source recommendation system based on time-series data and statistical analysis. Written in TypeScript and Node.js using Redis for storage. The recommendation system uses the Jaccard index to calculate the intersection between two sets. One set is represented by the maximum possible sum of tag score and the other set is the score sum of user events per tag. The higher the Jaccard index, the higher the recommendation. It uses numbers to represent sets to increase performance.
Features
Use tag score and Jaccard index
Content-based filtering
Event-driven powered engine
Naive exploration of new tags
Suitable for product and content recommendation
Fine-tuning of tag weights
Minimalist and lightweight
Written in TypeScript and Node.js
Overview video
How it works
How the data is stored:
Actors are stored in Redis as simple String keys with create date timestamps as value.
Items are Set type with tags as members. The item may have…
Replaces time series database for high volume input of actor (e.g. user) events and OLAP operations to calculate recommendations. Have in mind the time limitation of hackathon so take this statement with a pinch of salt ;)
