AI and Agentic Automation

I am encountering an issue with completing the Document Extraction Reconciliation Task in the Acme Doc Classification and Extraction model from the "Automate Your Business Tasks with AI Skills" course. I have tried the following steps: Cleared the browser cache Tried on a different browser Re-checked if the PDF is valid Any advice or assistance would be appreciated.

Hi everyone, Does anyone know how model weights are stored in Appian after training the AI document classification wizard? Is it some internal, non accessible table or the developer has some access rigths over that data? Is that as far as it goes, or is there a way to fine-tune the model further with more detailed, low-level adjustments? Also, can you integrate or pair it with other AI tools to create more accurate models than what Appian’s skill offers on its own? Thank you.

I am using the Extract From Document AI tool to read data from a pdf. In addition to text fields, the tool is also extracting a data table, and I'm using the Reconcile Doc Extraction smart service to train the AI to find the data in the table. After running this a few times, the Extract From Document tool has started to bring in data extracted from the table as well. Using Extract From Document, the confidence scores for the text fields are easily visible as an output of the node, but there doesn't seem to be a way to view confidence scores for the table or its text values within it. Are confidence scores used in data table extract, and is there a way to view them?

Hi All,I am working on AI skill document extraction with semi structured documents. While working, I am getting the below attached error almost after every click.FYI, I have already tried: 1. clearing cache 2. login logout3. different Appian environment4. incognito windowThanks in advance.

Hi everyone, I’m facing a behavior in Appian that I wanted to confirm is expected or not.I have a List of Text String, and when I use it directly inside an if() condition, it seems to always returns the first item in the list.Here’s a simplified example: a!localVariables( local!boolenArr: { false }, local!texts: { "A", "B" }, if(local!boolenArr, {}, local!texts) ) Is this expected behavior in Appian? I was expecting it to return all the items in local!texts. Just want to confirm if I’m missing something or if this is something we should be cautious about in expressions. Thanks in advance!

I am facing the issue while accesing the reconciliation task. This is the error message "An unexpected error occurred while displaying the pdf for the run [docExtractionId=435]."

Hi Team, I'm using the Document Classifier to distinguish between Medical Certificates and other document types (e.g., doctor prescriptions, blood reports, etc.). My goal is to extract data only from Medical Certificates after classification. I am using the AI Classifier Skill, which returns a list of classified documents in the following format:  The output of this is a list  say : [classifiedItem=325673, type=Medical Certificate, confidenceScore=94, errorMessage=],[classifiedItem=325672, type=Medical Certificate, confidenceScore=95, errorMessage=]  To process the classification result, I wrote the following expression to match each classified document with the corresponding uploaded document and then extract data only from Medical Certificates: However, I have noticed that even when the classification result contains multiple values, Appian is treating the entire list as one single value, instead of iterating over each item properly. (Screenshot below) a!loca

Hi Everyone, I'm facing an issue with importing my application to target envt. I'm using manual export and import option here. I have even tried to put the ai skill object alone in a separate package and tried to import and getting the following error :  An error occurred while creating aiSkill [uuid=00c07df7-3153-40dc-a697-e8a8afa54afe]: An error occurred while calling another system. Try again later. (APNX-1-4569-008) (APNX-1-4071-007) Note : The source and target environments are in same version(24.4) Thanks in advance!

Hi Team, I’m using the Document Classifier to distinguish between Medical Certificates and a Statutory Document. I’ve trained the model with 21 Medical Certs and 12 Statutory Docs. The incoming documents are predominantly (80%) Medical Certs, followed by Statutory Docs and other unrelated documents. The goal is to classify incoming documents and extract data using AI only if the document is a Medical Certificate. However, I’ve observed that the model is classifying random, unrelated documents as Medical Certificates with high confidence.(note the one - "use arrays functions exercise :(an Appian training exercise) and "blood report" Has anyone encountered a similar issue? Any suggestions on improving classification accuracy and reducing false positives? The metrics are not helping too, I haven't seen them changing? Is IDP classification not meant for unstructured files?

We have a requirement to filter the Dashboard grid based on child table data. For each Company, multiple forms exist, categorized into different types (1, 2, 3). However, only one form is considered active. To determine the active form, we refer to the Form Uploads table, where: Status ID = 2 Source ID = 17 (Only these records should be considered.) The following filter is not working, likely due to an issue with related records filtering: a!queryFilter( field: Company.form.FormUploads.statusId, operator: "=", value: 2 ), a!queryFilter( field: Company.form.FormUploads.referenceId, operator: "=", value: 17 ) The query record type is Company, and we need to display Company data in the grid.

Hi Team, I have a requirement where the default date and time field should display the nearest rounded time. For example: If the current time is 3:54 PM, the default time should be set to 4:00 PM. If the current time is 4:01 PM, the default time should be set to 4:30 PM. Is there any simple way to acheive this

Hi Team, Hi Team, I’m using the Document Classifier to distinguish between Medical Certificates and a Statutory Document. I’ve trained the model with 47 Medical Certs and 11 Statutory Docs. The incoming documents are predominantly (80%) Medical Certs, followed by Statutory Docs and other unrelated documents. The goal is to classify incoming documents and extract data using AI only if the document is a Medical Certificate. However, I’ve observed that the model is classifying random, unrelated documents as Medical Certificates with high confidence.(note the one - "use arrays functions exercise" and "blood report") Has anyone encountered a similar issue? Any suggestions on improving classification accuracy and reducing false positives? The metrics are not helping too, they are all reported as 1.000

I am working on an expression rule. In that rule, I need to get rows with null values as output. I am providing some map inputs like {username: test, fname: test1, lname: test2, email: test@gmail.com, username: test, fname: test1, lname: test2, email: ""}. I want to get the output as: {username: test, fname: test1, lname: test2, email: ""}.  I am getting only null.However, I am unable to get the exact output. Can anyone help me out?.Thanks for your help in adavncea!localVariables( local!users: a!forEach( items: ri!users, expression: a!map( username: fv!item.username, firstname: fv!item.firstname, lastname: fv!item.lastname, email: fv!item.email ) ), local!nullValueUsers: a!forEach( items: local!users, expression: if( or( isnull(fv!item.username), isnull(fv!item.firstname), isnull(fv!item.lastname), isnull(fv!item.email) ), fv!item, null ) ), lo

Hey, good evening! I’ve got a big question, and I’m kinda stuck. So, I’ve been working with Appian skills lately, and everything else has been going fine, but I’m having some trouble with the email part. I’ve been running some tests, trying to set up email classification. From what I understand, you need at least 10 emails to get started, but they should be different types. So, I started with two sets of 10 different types of emails, but when I tried to upload them to the AI, it didn’t work. I got this error: “Training failed because there are too few files that meet the requirements. Collect and upload more files that meet the requirements to train a model.” Okay, fair enough. Then, I decided to create a Python script to generate some dummy emails. I uploaded 72 emails across 4 different categories, but I’m still getting the same error. What am I missing here? Do I need even more emails? :/ I’ve attached some screenshots for

Hi Guys, I have the following values in Appian: {a!map(year: 2023, value: 15), a!map(year: 2024, value: 20), A!map(year: 2023, value: 100)} and I want to get an output as a map where the values are summed by year. For example: the years are aggregated and the value is the sum {a!map(year: 2023, value: 115), a!map(year: 2024, value: 20)} thanks

We are using 24.2 on premise but not seeing any option for "AI Copilot for Developers". As per documentation, it is available for 24.2 on premise. Do we need to take some further steps to enable this?

Do we have any function to sort an array of type integers in Ascending or Descending order (except a!sortInfo(), which did not help)?   For instance: I have an Array {1, 4, 6, 8, 5, 3} & I want to sort it in Ascending. So the O/P will be {1, 3, 4, 5, 6, 8}  Thanks in advance!

Hi Appian Community, Quick question: Is there any way for Appian Gen-AI to tell us how many tokens are used for a particular task, from input to output? I believe there isn’t, but just in case, I wanted to ask. The purpose of this is to analyze token usage so developers can be better informed. Thank you!

The input "Records" cannot insert or update a record which is referring to different referenced records via propostaId at the same time. To create a reference to a related record, include either a relationship reference or the required common field value for "propostaId". Location of conflicting changes: GC Document [1] = a!forEach( pv!documentos, a!update( fv!item, 'recordType!{GC Documento .fields.{389973d7-a3f2-4ce3-aafb-9c2021e72ce9}propostaId', pv!idProposta ) ) I can't find anything about this error at all

Hi Team, We have a requirement to place a signature in PDF. We created two ways of solutions. Please find the below details, 1.Signature Component and PDF Utilities plugin. 2.Docusign integration As per my organization suggestion, Docusign needs to purchase a license. But here we want a solutions for R&D purpose. Hence we are not able to purchase any third party license. We used signature component and PDF utilities(Image to PDF) to merge the signature into PDF. Question: But we need to send X and Y coordinates to place a signature in right place in PDF. Is there any solution can User picks X and Y coordinates in Appian else share your suggestion too. Thanks, Pradeep Sridharan

Hi Team, I was using the below rule and trying to fetch the data with the condition that selecting one ID from a grid should get one CaseID: 50324 and if selecting more than one ID should get CaseID: Multiple ID. can anyone please help me on this  a!localVariables( local!data: a!queryEntity( entity: cons!BEMS_ENTITY_PAYABLE_CASE_VIEW, query: a!query( /*selection: a!querySelection(a!queryColumn(field: "clientName")),*/ aggregation: a!queryAggregation( a!queryAggregationColumn(field: "clientName", isGrouping: true) ), filter: a!queryFilter( field: "caseId", operator: "in", value: a!flatten({ ri!caseIds }) ), pagingInfo: rule!TCO_defaultPagingInfo() ) ), { if( local!data.totalCount > 0, if( local!data.totalCount = 1, left( "Standard Template - " & proper( joinarray(local!data.data.clientName, " & ") ) & " - " & "CaseId:" & joinarray(ri!caseIds, ", "), 80 ), if( local!data

Hi Team, I was unable to get the data with the below condition that passing one ID should get CaseID: 1234 and passing more than one than one ID should get CaseID: Multiple ID a!localVariables( local!data: a!queryEntity( entity: cons!BEMS_ENTITY_PAYABLE_CASE_VIEW, query: a!query( aggregation: a!queryAggregation( a!queryAggregationColumn(field: "clientName", isGrouping: true) ), filter: a!queryFilter( field: "caseId", operator: "in", value: a!flatten({ ri!caseIds }) ), pagingInfo: rule!TCO_defaultPagingInfo() ) ), { if( local!data.totalCount > 0, if( local!data.totalCount = 1, left( "Standard Template - " & proper( joinarray(local!data.data.clientName, " & ") ) & " - " & "CaseId:" & joinarray(ri!caseIds, ", "), 80 ), if( local!data.totalCount >= 2, left( "Standard Template - " & proper( joinarray(local!data.data.clientName, " & ") ) & " - "

local!data: { { name: "John", status: "Active", test: 10 }, { name: "John", status: null, test: 4 }, { name: "Alice", status: "Active", test: 5 }, { name: "Alice", status: null, test: 6 }, { name: "Bob", status: null, test: 8 } }This a the above data now my requirement is remove null status only for the duplicate name and transfer the test value to the non-null status.Can any one please help me with this.

Hi everyone, I have a use case where a due date will be the 2nd Wednesday of the next month of given month and year. 

Hi,Having the below scenario,Having 2 different object lists as below. Now A type of object having 1,2,3 fields, similarly for B object list as well.A - 1,2,3 B - 1,2,3 Now in the first for each the Items will be A object list. foreach( items: ri!A, expression: foreach( items: ri!B, expression: { If(fv!item[B.2] = fv!item[A.2], "Hi", "Hello" ) } ) )Here, I am getting an error on below line for referencing the fv!item of A's object list in second for each.fv!item[B.2] = fv!item[A.2]Can anyone help me out this? Thanks in advance.