---
title: "Faith Mwende (@mwendekiindu) | Hashnode"
description: "Learning in public, one post at a time. Read the latest articles by Faith Mwende on Hashnode."
url: "https://hashnode.com/@mwendekiindu"
---

@mwendekiindu

# Faith Mwende

@mwendekiindu · Nairobi, Kenya · Joined September 2026

Learning in public, one post at a time.

### [What I Learned Analyzing 20,000 Airbnb Listings](https://beginnerslog.hashnode.dev/what-i-learned-analyzing-20-000-airbnb-listings?utm_source=hashnode&utm_medium=feed)

My first market analysis: capping price outliers, why location matters at neighbourhood not borough level.

5 min read · [beginnerslog.hashnode.dev](https://beginnerslog.hashnode.dev/what-i-learned-analyzing-20-000-airbnb-listings?utm_source=hashnode&utm_medium=feed)

[2](https://hashnode.com/posts/what-i-learned-analyzing-20-000-airbnb-listings/6ad0dc5cafbfd10d12158c93) [Oct 15](https://hashnode.com/posts/what-i-learned-analyzing-20-000-airbnb-listings/6ad0dc5cafbfd10d12158c93)

### [My Final Fraud Model (Day 7)](https://beginnerslog.hashnode.dev/my-final-fraud-model-day-7?utm_source=hashnode&utm_medium=feed)

Test AUC was 0.9571, lower than validation's 0.9765. Here's why the test number is the honest one, and why threshold mattered more than model choice.

5 min read · [beginnerslog.hashnode.dev](https://beginnerslog.hashnode.dev/my-final-fraud-model-day-7?utm_source=hashnode&utm_medium=feed)

[Oct 8](https://hashnode.com/posts/my-final-fraud-model-day-7/6ac769cdd3ee440fcb5c9d32)

### [Picking a Threshold Is a Business Decision (Day 6)](https://beginnerslog.hashnode.dev/picking-a-threshold-is-a-business-decision-day-6?utm_source=hashnode&utm_medium=feed)

The chart doesn't decide for you. I had to.

4 min read · [beginnerslog.hashnode.dev](https://beginnerslog.hashnode.dev/picking-a-threshold-is-a-business-decision-day-6?utm_source=hashnode&utm_medium=feed)

[Oct 4](https://hashnode.com/posts/picking-a-threshold-is-a-business-decision-day-6/6ac4176b75eb84c85c7c3fd1)

### [XGBoost Caught More Fraud - But There's a Catch (Day 5)](https://beginnerslog.hashnode.dev/xgboost-caught-more-fraud-but-there-s-a-catch-day-5?utm_source=hashnode&utm_medium=feed)

My recall jumped from 59% to 81%. My precision dropped from 79% to 57%.

4 min read · [beginnerslog.hashnode.dev](https://beginnerslog.hashnode.dev/xgboost-caught-more-fraud-but-there-s-a-catch-day-5?utm_source=hashnode&utm_medium=feed)

[Oct 4](https://hashnode.com/posts/xgboost-caught-more-fraud-but-there-s-a-catch-day-5/6ac22b2afc58d52f4165a7ca)

### [I Finally Understand AUC, Confusion Matrix, and Classification Report (Day 4)](https://beginnerslog.hashnode.dev/i-finally-understand-auc-confusion-matrix-and-classification-report-day-4?utm_source=hashnode&utm_medium=feed)

My first fraud detection model scored 0.9571 AUC and 99.9% accuracy but missed 40% of frauds.

4 min read · [beginnerslog.hashnode.dev](https://beginnerslog.hashnode.dev/i-finally-understand-auc-confusion-matrix-and-classification-report-day-4?utm_source=hashnode&utm_medium=feed)

[Sep 29](https://hashnode.com/posts/i-finally-understand-auc-confusion-matrix-and-classification-report-day-4/6abbc36881e777c6e43370c8)