Algorithm of classification of text separators groups
The algorithm needs several groups of text separators and tokens as input parameters. So, to understand this step, algorithm of dividing text separators into groups should be understood first. This step also uses some definitions from the previous step.
Connections graph like this
. connected with:
- I - 0.3
- M - 0.3
- T - 0.3
is built for each unique character that ends a token and is not from any separator groups. For example:
Input text: "Hello! My name is: Max. I live in Kiev. My age is 18 years. This summer i'll go to Paris! And what about you?"
The following graphs are built:
'y' connected with:
- n - 0.5
- a - 0.5
'e' connected with:
- i - 1.0
'I' connected with:
- l - 1.0
'n' connected with:
- K - 1.0
and so on.
All these graphs are merged into one graph (one group).
Each separator group graph is compared with the result from step 2.
The closest graph to the graph from step 2 represents separator characters from Group 2, other graphs represent separators from Group1.