Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

dawit-hopes/fintech-cx-analytics

Domain:

natural language processing

Record type:

project
Creator:
daw
Host:
Web scraping, sentiment analysis, NLP theme extraction, and Postgres-based data storage for Google Play reviews of Ethiopian fintech apps. ## 🚀 Customer Experience Analytics for Fintech Apps This project addresses a real-world data challenge in the rapidly evolving Financial Technology (Fintech) sector. The core objective is to utilize data engineering and Natural Language Processing (NLP) techniques to **quantify and enhance the customer experience** for mobile banking applications. Acting as a Data Analyst for Omega Consultancy, the goal is to provide three major Ethiopian banks—Commercial Bank of Ethiopia (CBE), Bank of Abyssinia (BOA), and Dashen Bank—with **actionable, data-driven insights** to improve their app features, manage customer complaints, and boost overall user satisfaction and retention. ### **Project Scope and Methodology** The project is structured around a complete data pipeline, from raw data acquisition to final business recommendations: 1. **Data Acquisition:** User reviews, ratings, and associated metadata are scraped from the **Google Play Store** for the target mobile banking applications. A minimum of 1,200 reviews are collected to ensure robust analysis. 2. **Data Engineering & Storage:** The raw text data is preprocessed, cleaned, and then engineered into a structured format. This refined dataset is stored persistently in a **PostgreSQL relational database**, simulating a professional data warehousing workflow. 3. **Advanced Analytics (NLP):** * **Sentiment Analysis:** Reviews are classified (e.g., positive, negative, neutral) using advanced models (like a fine-tuned DistilBERT) to gauge the emotional tone of customer feedback. * **Thematic Analysis:** Keywords and phrases are extracted (using techniques like TF-IDF or spaCy) and clustered into overarching themes (e.g., 'Login Errors,' 'Slow Transfers,' 'User Interface') to pinpoint common satisfaction drivers and critical pain points. 4. **Business Intelligence:** Insights derived from the data are translated into concrete recommendations aligned with specific business scenarios, such as retaining users, enhancing …

Visit

github.com

Tasks

sentiment analysistext classification

Languages

Amharic

Similar

nathanaeldereje/fintech-app-cx-analyticsdawit-hopes/ethiopian-food-classifierdawit-hopes/insurance-risk-segmentation-mlKalalayu/playstore-fintech-cxE-KB-N/betekbn-ghana-cx-analyticsRahemetGisho/fintech-review-analytics

nathanaeldereje/fintech-app-cx-analytics

Customer Experience Analytics for Ethiopian Mobile Banking Apps | Scraping, NLP sentiment & theme an

dawit-hopes/ethiopian-food-classifier

A deep learning project that classifies traditional Ethiopian dishes from images using FastAI and Py

dawit-hopes/insurance-risk-segmentation-ml

End-to-end insurance risk analytics project using real claims data from South Africa. Includes A/B t

Kalalayu/playstore-fintech-cx

Customer Experience Analytics for Ethiopian Fintech Apps # Fintech Customer Experience Analytics -

E-KB-N/betekbn-ghana-cx-analytics

An end-to-end Power BI analytics solution designed to evaluate customer onboarding friction, identit

RahemetGisho/fintech-review-analytics

Customer review analytics pipeline for Ethiopian mobile banking apps, using NLP, sentiment analysis,