How To Build An A.I. Model In 5 Minutes With Predictive Insights

Apteo

Everyone Can Build An A.I. Model With Apteo

For most businesses, the concept of using your data to build and refresh A.I. models to surface insights for the KPIs you care about is a pipe dream. Most companies don't have a team of business analysts or data scientists who can whip up answers to every question by aggregating, transforming, and modeling their data with ease. Instead, the responsibility to answer those big business questions falls upon whoever asked the question to begin with.

Today, instead of having to know how to code, write SQL queries, or be an Excel wizard, you can use Apteo's Predictive Insights to automatically surface the key drivers of your KPIs, predict future trends, understand potential churn, and forecast sales. Sound too good to be true? Keep reading to see how you can build an A.I. model in less than five minutes. We've pre-loaded data into all Apteo accounts so anyone can follow along.

What Are Predictive Insights?

Apteo’s new Predictive Insights is a no-code solution for professionals who want to use data to run forecasts or make predictions using the latest artificial intelligence (A.I.) techniques but may not have a background as a data scientist or machine learning engineer.

Use Case

Let's build an A.I. model together in 5 minutes using data. We'll define our objective and then follow 4 steps to create a model for free on Apteo.

Background

You work as a human resources manager at a company and you want to use data to find what’s causing your employees to leave, commonly referred to as an HR attrition analysis.

Dataset Used

Your Apteo account comes pre-loaded with a sample HR dataset to generate insights about your workforce to forecast attrition and surface other insights. The dataset has attributes about current and former employees such as title, role, department, age, income, tenure, job performance, and if they have left the company or not.

Objective

Use the analyses to prioritize employee retention efforts or post new employee job listings.

Step 1: Connect Your Data

Open a new window in your browser, login to Apteo, and click New Workspace. Name your workspace HR Retention Analysis, you can write a description or invite friends if you’d like, and click Create A Workspace.


You will then be brought to the Home dashboard for your new workspace, a place where you can share data and analyses with teammates. On the left hand side, click Data and click the dataset Apteo Employee Attrition Sample Dataset. Review the dataset and columns.

Step 2: What Do You Want To Predict?

Click Predictive Insights. Let’s select the KPI we want to analyze, and let’s get started.

Click Get Started and click Select Your KPI, and use the dropdown to select the KPI you care about, which is employee attrition, or did the employee leave the company or did not leave the company. You will see all the columns in the dataset are listed and you can click the last variable Attrition since we want to determine who is most likely to leave the company based on data from current and former employees.

Once you select your KPI, you will see the KPI you are looking to predict on the right hand side (column Attrition with 1,470 rows) and you can see there are 1.2K records for people who did not leave the company (equals NO) and 224 records for people who did leave the company (equals YES). Click Next.

Second, you will Select Your KPI Details. We suggest you choose Auto and let our A.I. determine the nature of your variable, but you have two basic categories, Number (represents a numerical value like prices or customer counts) or Category (represents a categorical prediction like fraudulent transaction, active customer). Click Next.

Third, we will automatically selected some appropriate attributes that can help us analyze your KPI. You can modify or remove one of the existing attributes, or add more attributes. We’ll leave as is, but you can click Edit Attribute to remove a variable. Click Next.

Step 3: Build Your A.I. Model To Generate Insights

Finally, we will Review the A.I. model before creating it. If there are any changes, click Back or Reset. Otherwise click Analyze KPI.


Once you click Analyze KPI, Apteo’s cutting-edge technology does the heavy lifting of training 10 different machine learning models, everything from linear regression to neural networks, and will choose the best machine learning model to analyze your KPI.

Our Predictive Insights will automatically surface trends for your KPI and generate A.I. forecasts with custom machine learning models we’ve built for you.

Step 4: Generate Predictions With Your Built A.I. Model

Let’s use the A.I. model we just built to generate a single prediction. Click Predict Now.

We can select the attribute we want to modify and click Generate Prediction to see how the attribute changes KPI you are trying to predict.

Best of all, once you’ve built an A.I. model, you and all your workspace members can always go back and revisit your insights or past predictions by adding the analysis to your dashboard.

Add team members to your workspace and even control whether they can read-only, edit, or if they have Admin rights. With Apteo, you control access and user permissions with a few clicks.

Watch A Video

If you prefer to watch a video, check out this 5 minute clip to walk through these steps on Apteo.

Shanif

Shanif Dhanani is the co-founder & CEO of Apteo. Prior to Apteo, Shanif was a data scientist and software engineer at Twitter, and prior to that he was the lead engineer and head of analytics at TapCommerce, a NYC-based ad tech startup acquired by Twitter. He has a passion for all things data and analytics, loves adventure traveling, and generally loves living in New York City.