bezi11/ethiopian-banking-app-analysis

A data analysis project aimed at scraping and processing user reviews from the Google Play Store for leading Ethiopian mobile banking apps (CBE, BOA, Dashen Bank) to extract actionable insights regarding customer satisfaction, challenges, and potential feature improvements. # Ethiopian Banking App Review Analysis ## Project Overview This project focuses on a detailed analysis of **1454 customer reviews** collected for major Ethiopian banking applications. The reviews pertain specifically to **Commercial Bank of Ethiopia, Bank of Abyssinia, and Dashen Bank**. The primary objective is to process this textual feedback, extract sentiment and recurring themes, store the processed data efficiently, and derive actionable insights to recommend app improvements. ## What I Did (Project Tasks) This analysis was conducted through a structured approach, divided into the following key tasks: * **Task 1: Data Collection & Preprocessing:** Initial acquisition of raw review data, followed by comprehensive cleaning, normalization, and preparation for analysis. * **Task 2: Sentiment Analysis & Theme Identification:** Application of Natural Language Processing (NLP) techniques, including VADER sentiment analysis, to classify review sentiment (positive, negative, neutral) and identify prevalent themes and topics discussed by users. * **Task 3: Database Population:** Integration of the processed and analyzed data into a structured SQL Server database (`bank_reviews`) for efficient storage, querying, and future use. Tables `Banks` and `Reviews` were created and populated. * **Task 4: Insight & Visualization:** Deriving key insights from the sentiment and theme data, comparing bank performance, and creating various visualizations to present findings effectively. This also involved proposing practical recommendations for app enhancement. ## Key Features * **Automated Data Cleaning:** Handling missing values, duplicates, and text normalization. * **Sentiment Analysis:** Categorizing reviews using a robust lexicon and rule-based tool. * **Theme Extraction:** Identifying common topics and issues directly from customer feedback. * **SQL Database Integration:** Storing analyzed data in a relational database for structured access. * **Data Visualization:** Gener …