Cross-Project Defect Prediction Using Transfer Learning with Long Short-Term Memory Networks

Table 2

List of the metric attributes.

ID	Metrics	Meaning	Type
1	wmc	Method weights in a class	Integer
2	dit	Inheritance tree depth	Integer
3	noc	The number of direct subclasses of a class	Integer
4	cbo	Coupling between objects	Integer
5	rfc	The size of the response set of the class	Integer
6	lcom	Lack of methodological cohesion	Integer
7	ca	Depends on the number of classes in the current class	Integer
8	ce	The number of classes that the current class depends on	Integer
9	npm	The number of public methods in the class	Integer
10	lcom3	Cohesion metrics	Float
11	loc	The number of lines in the Java binary of the class	Integer
12	dam	The ratio of all private attributes in the class	Float
13	moa	The number of fields in a class that a user defines as a class	Integer
14	mfa	The percentage of methods in a class that are inherited	Float
15	cam	The proportion of arguments in a method that are of the type of the argument	Float
16	ic	The number of inheritance coupling classes of a class	Integer
17	cbm	Number of new methods coupled to all inherited methods	Integer
18	amc	Average size of the methods in the class	Float
19	max_cc	The maximum cyclomatic complexity of all methods in a class	Integer
20	avg_cc	The arithmetic average of the cyclomatic complexity of all methods in the class	Float