Deep Forest-Based E-Commerce Recommendation Attack Detection Model

Table 1

Training set data overview.

Field name	Description	Train 0 range	Train 1 range
Id	Unique identifier for each data sample	1–500000	1–500000
visit_time	Timestamp of user accessing the product	0–86395	0–86396
user_id	Identifier for each user	2–1150958	4–1150951
item_id	Identifier for each product	7–434620	3–432626
Features	Product and user features	1–72, 73–152	1–72,73–152
Label	The value “1” indicates that the data sample represents malicious behavior, “0” represents normal behavior, and “−1” indicates that the data is unlabeled	0: 40054	0: 4476
		1: 9938	1: 520
		−1: 450008	−1: 495004