Opened 17 years ago
Closed 16 years ago
#57 closed request (fixed)
implement Naive Baysian Classifier
Reported by: | Peter | Owned by: | Peter |
---|---|---|---|
Priority: | major | Milestone: | yat 0.3 (Public release) |
Component: | classifier | Version: | |
Keywords: | Cc: |
Description
Change History (10)
comment:1 Changed 17 years ago by
Milestone: | later → 0.2 |
---|
comment:2 Changed 17 years ago by
comment:3 Changed 16 years ago by
Milestone: | 0.2 → later |
---|
comment:4 Changed 16 years ago by
Milestone: | later → 0.3 (Public release) |
---|
Since it exists in not complete form we have to decide what to do with it before the public release (see NBC.h and NBC.cc files)
comment:5 Changed 16 years ago by
Ok, if there is time I will look into this before 0.3 rel. else I think we move it to a branch.
comment:7 Changed 16 years ago by
Status: | new → assigned |
---|
comment:8 Changed 16 years ago by
comment:9 Changed 16 years ago by
The core of the prediction is the calculation of -log(P) = sum[sigma_i] + sum[(x_i-m_i)2/(2*sigma_i2)]
The latter term looks very much like the NCC predict. In fact when the sigmas are the same for all features NCC and NBC should be identical. Therefore in teh weighted case check out NCC implementation.
comment:10 Changed 16 years ago by
Resolution: | → fixed |
---|---|
Status: | assigned → closed |
Note: See
TracTickets for help on using
tickets.
(In [662]) refs #57 added class for Naive Baysian Classification. Predict function not yet implemented though.