Context Prediction of Mobile Users Based on Time-Inferred Pattern Networks: A Probabilistic Approach

Table 2

Effect of managing periodic data.

Target; context type; (location)	Probability
	Weekday case		Weekend case
	All	Periodic	All	Periodic
L1	0.18	0.16	0.06	0.04
L2	0.80	0.84	0.49	0.00
L3	0.00	0.00	0.00	0.00
L4	0.01	0.00	0.45	0.96
L5	0.00	0.00	0.00	0.00
Total sum	1.00	1.00	1.00	1.00

Values in each column are normalized for their sum to be 1.
Location L2 is the answer of our weekday test case.
Location L4 is the answer of our weekend test case.