15min solution to missing data problem in RPA

99.8% accuracy to predict a product category from invoice data. Done in 15 minutes. See how, and try yourself!

Missing data is everywhere

Dataset

How to implement missing value predictor with Aito

Prerequisites

Upload the dataset with CLI


mv Train.csv InvoiceData.csv aito database quick-add-table InvoiceData.csv

Review schema


curl -X GET \ https://$AITO_INSTANCE_NAME.api.aito.ai/api/v1/schema/InvoiceData \ -H "x-api-key: $AITO_API_KEY" \ -H "content-type: application/json"
{
"columns": {
"GL_Code": {
"nullable": false,
"type": "String"
},
"Inv_Amt": {
"nullable": false,
"type": "Decimal"
},
"Inv_Id": {
"nullable": false,
"type": "Int"
},
"Item_Description": {
"analyzer": "english",
"nullable": false,
"type": "Text"
},
"Product_Category": {
"nullable": false,
"type": "String"
},
"Vendor_Code": {
"nullable": false,
"type": "String"
}
},
"type": "table"
}

Run a prediction


curl -X POST \ https://$AITO_INSTANCE_NAME.api.aito.ai/api/v1/_predict \
-H "x-api-key: $AITO_API_KEY" \
-H "content-type: application/json" \
-d '
{ "from": "InvoiceData",
"where": { "GL_Code": "GL-6101400",
"Inv_Amt": 55.93,
"Item_Description": "Arabian American Development Co Final Site Clean Up 2008-Oct General Requirements General Contractor Store Construction",
"Vendor_Code": "VENDOR-1254"
},
"predict": "Product_Category",
"limit": 3
}'
{
"offset": 0,
"total": 36,
"hits": [
{
"$p": 0.9998137062471193,
"field": "Product_Category",
"feature": "CLASS-1522"
},
{
"$p": 7.345473788575557E-5,
"field": "Product_Category",
"feature": "CLASS-1828"
},
{
"$p": 3.118734024876525E-5,
"field": "Product_Category",
"feature": "CLASS-1983"
}
]
}

Evaluating accuracy

{
"test": {
"$index": {
"$mod": [3, 0]
}
},
"evaluate": {
"from": "InvoiceData",
"where": {
"GL_Code": {"$get": "GL_Code"},
"Inv_Amt": {"$get": "Inv_Amt"},
"Item_Description": {"$get": "Item_Description"},
"Vendor_Code": {"$get": "Vendor_Code"}},
"predict": "Product_Category"
},
"select": ["trainSamples", "testSamples", "accuracy", "error"]
}
{
"trainSamples": 3711.0,
"testSamples": 1856,
"accuracy": 0.9983836206896551,
"error": 0.001616379310344862
}

Operationalise


curl -X POST \
https://$AITO_INSTANCE_NAME.api.aito.ai/api/v1/data/InvoiceData \
-H "content-type: application/json" \
-H "x-api-key: $AITO_API_KEY" \
-d '
{
"GL_Code": "GL-9900990",
"Inv_Amt": 39.00,
"Inv_Id": 34001,
"Item_Description": "Predictive database monthly subscription, developer tier",
"Product_Category": "CLASS-9999",
"Vendor_Code": "VENDOR-9999"
}'

Comparison

Summary

--

--

--

Aito.ai decision automation in the cloud. #ML for #nocode and #rpa operators.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Bayesian Optimization, Striking Gold

Deep Learning: Which Loss and Activation Functions should I use?

Comparative Analysis of Three Open-Source Automatic Speech Recognition (ASR) Neural Network Models

Parts to transcribing audio to text as well as conversational AI

A gentle intro to Reinforcement Learning

CDS Members Organize and Present at the NYU AI School

Supervised machine learning for consultants: part 3

Understanding Music Genre Classification — A Multi-Modal Fusion Approach

Learn All About Deep Learning Techniques in 2022

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
aito.ai

aito.ai

Aito.ai decision automation in the cloud. #ML for #nocode and #rpa operators.

More from Medium

Explaining what learned models predict: In which cases can we trust machine learning models and…

Projected Scores: Reliving childhood memories through Data Science

Recess in the AI Playground with Ian Kelk from Clarifai

Data Science at LINE MAN Wongnai