I would like to buit a dynamic filtering engine based on document fileds and values distribution (decision tree for the next field to filter by)
I use mongodb mongoose and nodejs.
i.e :
The DB contains documents describing car characteristics:
{color:"blue", engine:"1598", numOfDoors:3, ...}
I would like to return the best field to filter by the documents. The goal is to return maximum 5 documents that are most compatibele with the characteristics entered.
For example if there are 2 blue cars and 8 red cars and 2 cars with 2 doors, 2 cars with 3 doors, 2 cars with 4 doors, 2 cars with 5 doors, 2 cars with 6 doors.
I would like return numOfDoors as the next field to filter by as the probability for returning the required number of documents(5) is 100% while filtering by color is only 20%.
I would really appreciate it if you could answer the following questions:
Thank you