At processing, the Customer Data Parser generates the original and the final mask patterns in the Parser output area. It also produces the mask logic numbers that describe the status of mask separation. See Attributes for Mask Patterns. By examining those outputs, you can investigate the records that are not parsed correctly and enhance the tables to obtain better parsing results. The Parser uses the tables and masks to separate the data in the following order:
-
The Parser identifies the original mask pattern (pr_mask_orig) using the information in the standard definitions table. Any unknown token is set to ‘U’.
Example
Input (財） 新潟県 保険 センター 小林 英字 Mask Original T C C C S R U -
The Parser identifies the final mask (pr_mask_final) using predefined algorithms and the masks defined in the standard definitions table. The process ends here for the records that are determined as personal and the PNP outputs are generated. For a record that is determined to be a business, contact name is included in the business branch section (B token) until the contact separation is performed in Step 3.
Example
Input (財） 新潟県 保険 センター 小林 英字 Mask Original T C C C S R U Mask Final T N S B -
If the record is a business and you choose to separate contact information from the branch name, the Parser proceeds to parse the ‘B’ token in the final mask (pr_mask_final). The Parser first identifies the original contact mask pattern (pr_contact_mask_orig) using the clues and names in the standard definitions table. Any unknown personal token is set to ‘X’ .
Example
Input (財） 新潟県 保険 センター 小林 英字 Mask Original T C C C S R U Mask Final T N S B Contact Original T N S L X -
The Parser determines the final contact mask (pr_contact_mask_final) after converting the original pattern to the final pattern using the contact mask in the standard definitions table. The Parser then generates the final BNP outputs.
Example
Input (財） 新潟県 保険 センター 小林 英字 Mask Original T C C C S R U Mask Final T N S B Contact Original T N S L X Contact Final T N S L F