# Predict User App Deletion — Machine Learning Interview Question

Aced (formerly Exponent) publishes a public search API over its indexed pages; the guide is at https://www.aced.io/agents/agent-access.

Source: https://www.aced.io/questions/4328/predict-likelihood-app-delete
Source observed: 2026-11-05T01:51:24.601Z
Last checked: 2026-11-05T01:51:24.601+00:00
Indexed: 2026-11-07T04:04:13.451+00:00

---

# Predict User App Deletion

**Difficulty:** Hard  
**Format:** Machine learning interview coding question

We ask you to develop an ML solution that predicts whether a user will delete our app. The dataset includes user attributes, in-app behavior, context, app information, and an outcome indicating whether the user deleted the app.

## Instructions

Download the dataset and try this question in a local `.ipynb` environment. There are many valid ways to approach the task; the page provides six expandable hints and one challenge for additional guidance.

[Download the formatted CSV dataset](https://assets.ctfassets.net/x78yjrjc11pq/4y7C1yxEWeTPYS7rxWJ8Qb/75949ba4d778a6f9d8cd39bbbc986582/predict_user_app_deletion.csv)

## Dataset

Features are grouped into four semantic categories: User, User-Behavioral, Context, and App.

| Feature | Description | Data type | Category |
|---|---|---|---|
| F1 | Country | categorical | User |
| F2 | Gender | categorical | User |
| F3 | Age | numeric | User |
| F4 | How did the user download the app | categorical | User |
| F5 | Number interactions 7d | numeric | User-Behavioral |
| F6 | Number of interactions 14d | numeric | User-Behavioral |
| F7 | Number of interactions 30d | numeric | User-Behavioral |
| F8 | Time spent on app - daily avg | numeric | User-Behavioral |
| F9 | Time since last use | numeric | User-Behavioral |
| F10 | Time of day | numeric | Context |
| F11 | Day of week | numeric | Context |
| F12 | IOS or Android | categorical | Context |
| F13 | Version of OS | categorical | Context |
| F14 | App version | categorical | App |

Currently active users have `Label: 1`. Users who deleted the app in the last N days have `Label: 0`. The prediction target is `Label`.

## Formatted training-data example

The supplied example contains missing values, represented by blank cells.

| F1 | F2 | F3 | F4 | F5 | F6 | F7 | F8 | F9 | F10 | F11 | F12 | F13 | F14 | Label |
|---|---|---:|---|---:|---:|---:|---:|---:|---:|---:|---|---:|---:|---:|
| US | M | 17 | E | 1 | 3 | 10 | 45 | 2 | 17 | 1 | I | 17 | 2 | 1 |
| UK | U | 34 | E | 3 | 7 | 56 | 13 | 6 | 3 | 7 | A | 3 | 4 | 1 |
| AR |  | 19 | P | 1 | 9 | 12 | 45 | 7 | 12 | 4 | A | 5 | 2 | 0 |
|  |  | 87 |  |  |  |  | 4 | 5 | 1 | 5 | I |  | 2 | 0 |
|  | F |  | P |  | 6 | 19 | 9 | 12 | 8 | 6 |  |  | 2 | 0 |
| US | M | 17 | E | 1 | 3 | 10 | 45 | 2 | 17 | 1 | I | 12 | 2 | 1 |
| UK | U | 34 | E | 3 | 7 | 8 | 13 | 6 | 3 | 7 | I | 3 | 4 | 1 |
| BR |  | 19 | P | 1 | 9 | 2 | 5 | 7 | 12 | 4 | A | 5 | 2 | 1 |
|  | M | 6 |  |  |  |  | 4 | 5 | 1 | 5 | I |  | 2 | 0 |
| IN | F |  | P |  | 6 | 19 | 9 | 12 | 8 | 6 |  |  | 2 | 1 |

## Page controls and participation

The question has tabs for **Question**, **Solution**, **Mock Interview**, and **Community**. The question is marked **Hard**. The community section currently has no contributed answers. Answer submission requires logging in or signing up; submitting marks the question complete.

## Related Aced resources

We provide interview preparation across these areas:

- [Product Management](https://www.aced.io/courses/pm?src=nav) — Ace product interviews from strategy cases to technical skills.
- [Forward Deployed](https://www.aced.io/courses/forward-deployed-engineering?src=nav) — Beta preparation for FDE loops at Anthropic, OpenAI, Palantir, and more.
- [Engineering Management](https://www.aced.io/courses/engineering-management?src=nav) — Review leadership and people-management skills.
- [Software Engineering](https://www.aced.io/courses/software-engineering?src=nav) — Learn strategies for coding problems.
- [System Design](https://www.aced.io/courses/system-design-interviews?src=nav) — Define architectures, interfaces, and databases in a time crunch.
- [Data Science](https://www.aced.io/courses/data-science?src=nav) — Execute statistical techniques and experimentation.
- [Machine Learning](https://www.aced.io/courses/ml-engineer?src=nav) — Review building, evaluating, and deploying AI/ML models.
- [Data Engineering](https://www.aced.io/courses/data-engineering?src=nav) — Design data models and ETL pipelines.
- [Browse all courses](https://www.aced.io/courses?src=nav)

We also offer [Loops](https://www.aced.io/guides?src=nav), [Experiences](https://www.aced.io/experiences?src=nav), [practice](https://www.aced.io/practice?src=nav), and [coaching](https://www.aced.io/coaching?src=nav), including [mock interviews](https://www.aced.io/coaching?category=mock_interviews&src=nav), [resume review](https://www.aced.io/resume?src=nav), and [salary negotiation](https://www.aced.io/coaching?category=negotiation&src=nav).

---

[Read this page’s summary](https://www.aced.io/agents/api/public/v1/page-tldr?url=https%3A%2F%2Fwww.aced.io%2Fquestions%2F4328%2Fpredict-likelihood-app-delete&trace=t_679ff9ab98f3)

## Ask a follow-up about this page

Need a more specific answer? Ask the exact question you have. It returns a fresh, cited answer from Aced (formerly Exponent)’s knowledge base. No account or API key required.

Open this URL with your own question, URL-encoded. Include your question in q. All research inputs are optional; no account or setup.

`GET https://www.aced.io/agents/agent/faq?q={question}&trace=t_679ff9ab98f3`

[Tool guide and examples](https://www.aced.io/agents/agent-access) · [OpenAPI](https://www.aced.io/agents/api/public/v1/openapi.json) · [Page TLDR form](https://www.aced.io/agents/agent-tools/page-tldr?url=https%3A%2F%2Fwww.aced.io%2Fquestions%2F4328%2Fpredict-likelihood-app-delete). Opening a form does not execute a tool.