Miscellaneous parameters

The following miscellaneous parameters can be used to further configure and refine your document form definition (DFDs) by enabling custom functionality. The following parameters can be added to the of the Misc Params column of the document form definition:

Parameter

Description

Example usage

TopStop

Stops assigning higher rows to the table when the specified word is reached (for example, TopStop=word). Multiple words must be separated by commas.

Use to define when the system starts reading table data. If configured, the system searches for a specific word near the top of the table and only reads the rows following the defined word.

For example, if the word is found on row three, Decipher IDP would only read from row four.

BottomStop

Stops assigning lower rows to the table when the specified word is reached (for example, BottomStop=word). Multiple words must be separated by commas.

Use to define when the system starts reading table data. If configured, the system searches for a specific word near the bottom of the table and only reads the rows preceding the defined word.

For example, if the word is found on row ten,Decipher IDP would only read rows one to nine.

BottomStopPhrase

Stops assigning lower rows to the table when the specified phrase is reached (for example, BottomStopPhrase=Total amount). This only works if BorderTable=On and the table has borders.

Similar to BottomStop, but a phrase as opposed to a single word can be used.

RowAreaExcludeValues

Excludes certain values from a row and can be used for single words only.

Where a table contains a word that should not be included in the output, for example, if N/A is used where a value is not available. Note other information in the field and row will still be read.

SubTableCategory=On

Ignores this column type during the standard table search and use the column as sub-table category holder.

Where there is a sub-table in a table, set this parameter to define columns containing the title or headings of sub-tables.

SubTableTotal=On

Ignores the column type during the standard table search and use the column as sub-table total holder.

Where there is a sub-table in a table, set this parameter to define columns containing subtotals.

FillToTheNext=On

Auto-fills each row with data from this row until another row with further data is reached.

Where a value in a column is repeated in subsequent rows until a new value is found in the column.

For example, if row one and row five both contain reference numbers, the row one reference will be copied to the output for rows two to four. The row five reference will be used until the next reference or table end.

VirtualRegions=Off

Prevents auto-fill with virtual regions for the individual sub-tables.

-

ExactRows

Defines table headers. Use with Form Tables (also known as Exact Tables), where each cell in the first column represents the header for its row. For example, ExactRows=BaseEmpl.Income,Overtime,Bonuses. The headers are filled automatically in the table grid, so they serve as a frame for the table.

Use to extract header values from the first row of a table.

TableTitle

Identifies tables. An additional identifier for tables with identical ‘ExactRows’, where many identical candidates could exist, and it is necessary to decide which candidate will take in which data. A table title can be used for an identifier, for example TableTitle=Schedule.

Use to assist Decipher IDP in identifying the correct table when operating with multiple, similar tables.

MainListColumn=On

Forces auto-correct to work regardless of limitation set by the logic. Use for form template fields (FTF) with a single column.

Use to populate tables which consist of one column, for example, a list of values.

CCL=number

The character recognition confidence level value. For example, if CCL=79 (range [0,80]) – all characters must be always in black (high confidence). If CCL=81 (range [80, 100]) – all characters must be in black under certain conditions. The field is validated against the formula, a list, or both. The confidence rating is based on the returned result being consistent from OCR engines, or if entered manually.

If set to a low value, fewer characters will be flagged as low confidence.

FCL=number

Field confidence level. Similar to CCL (character recognition confidence), but for field position confidence.

If set to a low value, fewer fields will be flagged as low confidence.

AutoAssign=Off

Disables auto-assign. The field will not be assigned automatically when the user manually types the field content (without clicking on the suggestion). If this parameter is not specified, then on manual content filled in for assignable fields the software will automatically search for the manually specified field content on the active page and if found the field will be assigned to the found region.

Use to disable the system from auto-assigning regions and allow for the manual mapping of regions during data verification.

Semantic=On

Specifies a semantic field, which is a multi-line field in free text or another form, and which cannot be contained within a rectangle.

You should assign only the text of the first line, and then manually append the second line's content in the field. It is recommended that this parameter is set before doing any training of the document.

Use to recognize non-rectangular regions for multi-line text, for example, if the first line of a multi-line element is removed and the shape is changed.

Multiple

Defines group values. Use for naming fields belonging to a single group. The value is the group name (for example, Multiple=Fruits). A number of values is received that is equal to the number of items in the group, and those values are distributed accordingly. The parameter is used for logical grouping of semantic fields.

For example, is set for fields FT_A and FT_B, they are excluded from the early analysis and are processed last.

Use to ensure all items in a group are captured correctly. Fields set with the parameter are not analyzed until the end of the process.

FormFields=On

Defines a static location. Use with fixed/structured forms, where the information for each field is expected to be seen at a specific location in the document. The user-defined regions will be memorized and they will not be changed afterwards, even if they happen to contain multiple auto-generated regions within.

Use with documents such as PDF forms that will not change structure. This parameter will only function where there is a single, and not multiple, form templates.

ResPhrase

Defines the location of non-standard elements. Use with sub-tables, when the values for sub-table totals are located in an unusual place. If the value is pointed out by a phrase, it can be specified here (for example, ResPhrase=Paid).

Use when capturing columns of sub-tables whose header is not above but instead to the left of the value. The header is specified as the parameter value.

ForceBestTemplate=On

Forces Decipher IDP to use the best-matching available template for documents loaded with this FTF.

If a page has not been trained a new template file is created by default. Enabling this parameter ensures the best suited existing template is used.

FilterContent=word

Removes specified values from captured data. Use for fields and table columns where you need to validate a certain value that is invalid. For example, if the type is Money, but the prices in the table have a suffix “ft”, we filter that suffix by specifying FilterContent=ft. This works in case-insensitive mode. The data we want to filter can be anywhere, including as a prefix or suffix. We can also specify several values, as in FilterContent=aaa,bbb,ccc. We can also specify a combination of words, e.g. FilterContent=filter this phrase. If we specify to filter a certain value and the assigned region has that exact same value, the value will not be filtered.

Use to capture money values which may include additional text characters and allow the system to recognize the data as a valid money value.

CheckMode=On

Disables table detection on landscape pages, such as the first page in an elns check. It is only used in the Online version because the merge mode is unknown. Can be specified in any assignable field.

Use to exclude line items on pages with landscape orientation.

AggressiveListSearch=On

Enables collecting of more than one list item candidate as the search doesn’t stop when perfect candidate is found. It might be slow for large lists so it’s recommended to be used only in specific cases.

Useful when two or more fields have mutual list items and the logic must collect all available candidates and determine later which candidate corresponds to which field.

IcrDigit=True

Returns only digits for any region assigned to this field.

-

IcrAlpha=True

Returns only letters for any region assigned to this field.

-

IcrPunctual=True

Returns only punctuation signs for any region assigned to this field.

-

DelayNameChanged=True

Signifies that the field will be captured last. The purpose of this parameter is to invoke the callback for CAP_RCC_NAME_CHANGED at the end of all region callbacks. This way we can afford triggering a formula after the item table is filled. The parameter name and value are not case sensitive.

If a formula depends on line items, such as a subtotal field, this parameter can be enabled to ensure the field is captured last so the formula is correctly executed.

AUTOCALC_LAST_ROW=True

Populates the last table row with a calculation results. Use for table fields. By default, it’s set to false and is case sensitive.

Use when setting a formula in a table field to insert the calculation result in the last table row.

CalcInVerify=True

Populates the field on batch load, regardless of whether it’s Autocalc or not. If the parameter is enabled for a DynamicList, the field will also get populated with the first value from the result list.

Use to add a time stamp of when the batch was loaded in Verify, or if used with a dynamic list, which user loaded the batch.

MultiplePriority

Determines the priority of fields that are within the same group defined by the “Multiple” parameter for machine learning. The smaller the number, the higher the priority. The priority can be any integer: positive, negative or zero. The candidates will be assigned to the fields starting from the highest-priority field, and then the lower-priority fields. This makes sure that the highest-priority fields have the highest chance of getting assigned. If there are several fields but fewer candidates, the candidates will be assigned to the fields with the highest priority.

If there are multiple candidates, this parameter can be used to instruct the machine learning to search for specific candidates first that are more likely to be present in the document.

BorderTable=On

Activates table extraction mode by existing border tables. It can be registered to any FTF column, it’s a table-scoped functionality.

Use to assist Decipher IDP in automatically recognizing field separations by table borders.

MaxSubTableSumDiff=1.0

Allows change to the maximally allowed difference between the sub-table line total sum and the sub-table subtotal. It can be registered to any FTF column, and is a table-scoped functionality.

To ignore small differences between the subtotal of sub-tables and the whole subtotal, this can be used in instances where an invoice includes an error.

RowAreaExcludeValues=total,position

Excludes rows containing any of the registered keywords. It can be registered to any FTF column, and is a table-scoped functionality.

Use where you want to exclude rows from a table containing specific words. This parameter should be added to the table column where the word appears.

TemplateMatchWith

Defines the ID of a template to use with the document. If there is no matching of a template the specified one will be tried and if it totally does not match the page it won’t be used but new SFT won’t be generated. The GUID is not case-sensitive. For example, TemplateMatchWith = 45fed1c5-bc10-4122-8777-34311181925e.

Where the page does not fit the SFT it has been matched with, you can provide the ID of a different SFT.

TemplateMinMatchPercent

Determines the threshold for template matching. Use to set the threshold for similarity between SFT and page. If a SFT and a page have similarity lower than the specified threshold, they will not match. Accepts values between 0 and 100. The default setting is TemplateMinMatchPercent=60.

Setting to a low value will cause templates to match more easily, meaning fewer templates are generated and the system will be less flexible. Higher values cause the system to be more flexible, but will require a greater amount of training.

StrictPosition=On

Defines individual fields with static position. This is specific to the region template and can move for each layout. The document quality and DPI can impact field positioning.

Use where you want a field to be locked to an exact region of the document and not affected by other elements in the region.

RFT=Value

Defines a region fullness threshold (RFT) value for the given field. This specifies what percentage of the region needs to be black in order for a checkmark field to be detected as ticked.

Use to assist Decipher IDP in identifying enabled check boxes.