Recommendation Model
While building this model I was facing problems to find a appropriate dataset. And then I came up to a conclusion to create own dataset to solve this problem.
How the Dataset is arranged
1. For every entry in x_train there is a corresponding entry in y_train
2. Codes are gives for the entry of the comoditie
Every comodity has a code of three integers which are calculated as follows
i. First Integer represents its type. You have to choose a integer from the options given below
a. food-item : 1
b.
medicine : 2
c. self care : 3
d. electric : 4
e
Study : 5
f. cleaning : 6
g. decoration : 7
ii. Now the next two integers represents the subcategory (as it can be in two subcategories).
For '1' (food-item)
a. vegetable :1
b.
fruit : 2
c. liquid : 3
d. packed: 4
e.
chinese: 5
f. dairy : 6
For '2' (medicine)
a. fever : 1
b.
family-planning : 2
c. for acute : 3
d. for chronic : 4
e. daily nutrients : 5
f. syrup : 6
For '3' (self-care)
a. teeth : 1
b. hair
: 2
c. face : 3
d.
body : 4
e. grooming : 5
f. sanitizing : 6
For '4' (electric)
a. hair : 1
b.
computer : 2
c. laptop : 3
d. mobile : 4
e.
other : 5
For '5' (study)
a. writing : 1
b.
paper : 2
c. measuring : 3
d. tools : 4
For '6' (cleaning)
a. body : 1
b.
surrounding : 2
c. electric : 3
For '7' (Decoration)
a. self : 1
b.
surrounding : 2