Anomaly Detection in Discrete Manufacturing Systems by Pattern Relation Table Approaches

. 2020 Oct 12;20(20):5766. doi: 10.3390/s20205766

Subroutine 1Candidate Patterns Extracting (

D, | E |

)

Input: Data set D, event cardinality K

Output: Ordered candidate pattern set C

1:
$C \leftarrow \emptyset, M a p = s e t (s e q, i m p o r t a n c e)$
2:
for $d_{i} \in D$ do
3:
for $w \leftarrow$ sub-string of $d_{i}$ with $| w | \in [2, min {| d_{i} |, K}]$ do
4:
$n u m \leftarrow$ frequency of w in $d_{i}$
5:
if $n u m > 1$ then
6:
$M a p$ .add(w,0) if $w \notin M a p . k e y s ()$
7:
$i m p o r t a n c e \leftarrow i m p o r t a n c e + l e n (w) \times n u m$
8:
end if
9:
end for
10:
end for
11:
Order the $M a p$ based on $i m p o r t a n c e$ in a descending order
12:
return $M a p$ .keys()