Platform v7.0-Feature Clarification

Item

Currently Supported

Testing Details

Observation

Future Plans

Browser Compatibility

Google Chrome

Verified with Chrome version 76.0.3809.100 and above

All features may not work properly with other browsers or versions

No current plan to support other browsers

Item

Currently Supported

Testing Details

Observation

Future Plans

File size limit for Upload & Prediction

File size up to 1MB used in upload and predict works without performance lag

Verified with files < 1 MB and > 1 MB

Performance drop for documents with size greater than 1 MB

No current plan to improve this

Concurrent users

Supports up to 5 concurrent users without performance hit

Verified with 5 and more users. Current hardware: 16 core 64 GB RAM for Vision App + 8 Core 32 GB for dependencies

Performance drop when more than 5 users access the application concurrently

Not in current plan. Increasing the hardware can be one option

Unstructured document extraction accuracy

70% test accuracy reported, but lower accuracy on unseen documents

Verified using Annual Reports from US companies

Solution needs HITL, auto extraction solutions doesn’t give desired accuracy

No current plan to improve this

File size limit for multi-page table extraction

Documents with size <= 1MB

Verified with files < 1 MB and > 1 MB. Current hardware: 16 core 64 GB RAM for Vision App + 8 Core 32 GB for dependencies

Multi page table extraction times out for documents with size greater than 1 MB

Not in current plan. Increasing the hardware can be one option

Preview for masked documents

Test page without preview

Verified test page by adjusting masking levels

No preview option available in the Test page to verify masked image

Not in current plan

Lambda & Throttler functions in Richly formatted document extraction

Provision to upload lambda and throttler functions in a file

Verified by uploading lambda and throttle functions via UI

Need pre-requisite knowledge on Fonduer framework and underlying data model

Not in current plan

Item

Currently Supported

Testing Details

Observation

Future Plans

Segregation Accuracy

40% accuracy in detecting cheques with handwritten characters

Verified with machine printed & handwritten cheques, blank and other type of documents

Accuracy issues with Cheque classification – Handwritten, printed, blanks and non-cheques

Improve with 3PP (Hyper Science) Integration

Identification of non-cheque documents

Non-cheque documents will be considered for segregation

Verified with cheque and non-cheque documents

Non-cheque documents are not classified

No current plan on this area

Handwritten data extraction

Not supported at present

NA

Handwritten data extraction from cheques are not supported

Support handwritten data extraction from cheques using 3PP (Hyper Science) integration

Item

Currently Supported

Testing Details

Observation

Future Plans

Field Extraction Accuracy Issues

Minimum resolution quality should be 300 DPI, Minimum dimension has to be 500 x 400 pixels, Non-blurred documents, document tilt not more than 5 degrees

Verified with machine printed and handwritten cheques

Field extraction accuracy issues for license & PAN documents due to insufficient training data

No current plans on this area

Supported Document types

Kerala drivers license in card format & PAN card

Verified with mentioned document types and others

Document types other than mentioned are not supported

No current plans on this area

KYC Application status updates

Submitted application shows as ‘Review Pending’ for 5-10 seconds in the listing

Verified submitting KYC application forms and validating status in listing page

Slight delay in overall status update when reviewer submits changes

No current plans on this area

Tilt & Skew correction of images

Document images which are not tilted or skewed

Verified with images tilted / skewed and proper images

Tilt & Skew correction of images are not available in this release

No current plans on this area

Handwritten data extraction from documents

Handwritten data extraction is available for application form available in the PWF

Verified with handwritten application forms and other types of documents

Handwritten data extraction will not work for document types other than specific application form available with PWF

Support using 3PP (Hyper Science) integration

Supported Document formats

Supports PDF, PNG and JPG formats

Verified with PDF, PNG and JPG files

Formats other than mentioned are not supported (e.g.: TIFF)

No current plans on this area

Item

Currently Supported

Testing Details

Observation

Future Plans

Data extraction accuracy

Field level accuracy below

Verified by uploading MSAs, SOWs, Addendums from different vendors

Accuracy issues in MSA/SOW/Addendum data extraction

Moving towards Vespa based solution in 7.x

Volume Rebates Recalculate

Volume rebates data extracted from MSA and applied in the total contract value to get rebate value

Verified by uploading MSAs with rebate details and SOWs linked to MSA

User will not be able to change wrong linking and recalculate rebate value

Not in current plan

Volume rebate table extraction

Supports rebates tables with following 4 columns - Tier, Minimum, Maximum and % of Level

Verified by uploading MSAs with rebate details represented in different forms

Rebate tables with difference in column titles or number of columns and rebate information represented in non-tabular format are not extracted

Moving towards Vespa based solution in 7.x

First Party

Second Party

Effective MSA Date

Payment Due Days

Renewal Term

Automatic Renewal

Venue

Jurisdiction

MSA End Date

100

98

89

83

78

78

97

92

88

Payment

Confidential

Warranties

Indemnification

Termination

Volume Rebate Table

97

73

71

71

66

100

Contract Value

Effective MSA Date

Effective SOW Date

Expiry Date

First Party

Second Party

81

81

77

72

96

81

Effective Addendum Date

Effective MSA Date

Effective SOW Date

First Party

Second Party

83

57

22

92

87

Item

Currently Supported

Testing Details

Observation

Future Plans

Data extraction accuracy for Annual Reports

Field level accuracy shown below

Verified by uploading annual reports from multiple companies

Field accuracy needs to be improved for Annual Reports extraction

Not in current plan

Field selection change

Supported fields are pre-selected for extraction

Verified by creating DU projects for annual reports and invoices

Fields selected for extraction cannot be changed

Not in current plan

Orientation Correction

Orientation correction is supported only for non-searchable PDFs

Verified by uploading PDFs rotated clockwise, anti-clockwise and upside down

Orientation correction is not applied on searchable PDFs

Not in current plan

Company Name

Report Year

CEO

Total Assets

Total Liabilities

73

93

86

66

20

Item

Currently Supported

Testing Details

Observation

Future Plans

Invoice Extraction Time

Extraction time for a single page invoice is 1 minute

Verified by uploading a single page invoice and calculating time based on UI status

More than 6 seconds for a single page invoice can slow down multi page invoice extraction in real scenario

Not in current plan. Higher speeds would need GPU

Field data accuracy

Field level accuracy below

Verified by validating against Tolas and PP invoices for Navistar

Field wise extraction accuracy is not good for all 14 fields

HITL feedback-based model training will improve field suggestions and accuracy

Line item data accuracy

Line item extraction works for tables with strong closed borders containing in a single page

Verified by validating against invoices containing line items in tabular and non-tabular format

Line item extraction will not work for invoices with partially closed bordered tables, multi-page tables and non-tabular representations

Fully fledged Smart Extract features will help to address these

Accuracy for fields with dark background

Fields with light or no background

Verified by validating against Tolas and PP invoices for Navistar

Extraction is not possible for FOI data in dark background

Not in current plan

Extraction time calculation

Extraction time in analytics is total time in DU - which includes queue wait time and execution time

Verified by validating extraction time in Admin Analytics section

Extraction time for an invoice will not be consistent; as the queue time can vary

No current plan for this

Orientation correction accuracy

Landscape oriented images will be converted to portrait mode

Verified by validating with a landscape-oriented invoice

2 out of 75 invoice pages had wrong orientation correction applied

No current plan for this

Extraction Traceability

Basic traceability is available from Run list, Trace logs and Batch summary

Verified by running extraction against multiple types of invoices

The extraction process is not fully traceable end-to-end

End-to-end traceability is planned for future releases

Delay in updating invoice count & flags in Run list

For a batch of 10 invoices, 5 minutes will be taken to update the run list

Verified by running batch of invoices

When a batch is processed in DU there is a delay in reflecting the invoice count & flags in Run list

No current plan for this

Delay in updating Accuracy Analytics

It takes 5-7 minutes to get the details updated in analytics

Verified by making corrections to a batch of invoices

Line item accuracy graph and Insight accuracy graph is not updated immediately after making corrections

No current plan for this

Advanced section performance

Advanced page works fine for a project with invoices less than 700

Verified by running 700+ invoices in a project

Advanced page hangs when project is having more than 700 invoices (Intermittent Issue)

No current plan for this

Company Address

Purchase Order Number

Ship To Code

Vendor Code

Company Name

Freight Amount

Invoice Date

Invoice Number

Miscellaneous Charges

Scan Number

Sub Total

Total Amount

51

64

90

97

79

94

83

63

34

97

73

72

Part Number

Quantity

Customer PO Number

Unit Rate

Extended Cost

58

49

30

54

52

Item

Currently Supported

Testing Details

Observation

Future Plans

Alert Clustering

Alerts gets clustered 100% if they come in lesser volumes for a given time

Verified by sending multiple alerts continuously for a given timeframe

Certain alerts are not getting clustered when multiple alerts are received one after another

To be fixed in upcoming releases

Email alert template format

Alert fields need to be in the same order and number as in the template

Verified by sending multiple alerts after removing fields, values and adding fields with values in Defined template

If the alert fields are not matching to the template, data will get wrongly mapped to fields

To be fixed in upcoming releases

Alert re-processing

Alerts get lost due to exceptions / errors while processing

Verified by sending batch of email alerts for longer duration

No mechanism is available to re-process messages if lost

Re-processing mechanism is going to get implemented in future releases

Recovery policy setup

Recovery policy is setup via APIs

Verified by setting recovery policy via Postman interface

No user interface available to set recovery policy details

UI will be available in future releases

Auto closure condition setup

Auto closure conditions are setup via APIs

Verified by setting auto closure conditions via Postman interface

No user interface available to set auto closure settings

UI will be available in future releases

Display issues in screens with different resolutions

Works good only if standard resolution with zero scaling applied

Verified by setting different resolutions and scaling ratios

Page contents will not be fully visible for resolutions with scaling applied

Responsive Web Design based implementation in the future releases will address this

Device inventory data import

Data import for device inventory is available only via APIs

Verified by importing device inventory data through Postman interface

No user interface available to import device inventory data

UI will be available in future releases

Item

Currently Supported

Testing Details

Observation

Future Plans

Supported file size of PWF project for import

Import functionality works with a file size limit of 1 MB

Verified by importing PWF projects

Import functionality will not work with files above 1 MB

Not in current plan