One option is to specify user-defined variables as a target for a fiel, in the place of a column name in the column list at the end of the LOAD DATA statement.
For example (omitting most of the syntax):
LOAD DATA ...
...
(@dummy1, @dummy2, c, d, e, @dummy3)
The first field from a line will be assigned to user-defined variable @dummy1, the second field to another user-defined variable, the third, fourth and fifth fields will be assigned to columns c, d, and e (respectively), the sixth field will be assigned to another user-defined variable.
If you want to provide a value for a column that's the result of an expression, you can use the SET clause...
LOAD DATA ...
...
(@a, @b, c, d, e, @dummy3)
SET a = UPPER(@a)
, b = STR_TO_DATE(@b,'%m/%d/%Y')
, f = CONCAT(@a,@b)
UPDATE
If you only have three fields in the csv, and you want to assign the first field to column c, the second field value to column d, and the third field value to column e...
LOAD DATA ...
...
(c, d, e)
It's the position of the column name in the (col_or_user_var,...) list that determines which field value will be assigned to which column. To "skip" a field, assign it to user-defined variable... the field in the file has to be assigned to something, and if you don't want to assign it to a column, the assign it to user-defined variable.
It wasn't clear in your question how many fields were in the file, and which field should be assigned to which column.