Unlock insights into Moroccan banking customer experiences! 🇲🇦

This dataset contains scraped and cleaned Google Maps reviews for banks across all cities in Morocco. Collected as part of a collaborative student/freelancer project, it’s perfect for sentiment analysis, market research, or academic projects.

What’s Inside?

• 2 Versions:
  • Raw Data: As scraped from Google Maps.
  • Cleaned Data: Filtered to exclude non-bank businesses (e.g., cash services, unrelated entries).
• Columns:
  City, Business Name, Address, Phone Number, Website, Google Map ID, Review Text, Timestamp, Stars.
• Cities Sourced from data.gov.ma: Ensured comprehensive coverage of Moroccan regions.

Methodology:

1. City Identification: Used official data from data.gov.ma to target cities with banks.
2. Search Strategy: Queried “bank in [city name]” on Google Maps to compile business links.
3. Scraping: Extracted business details (name, address, etc.) and latest reviews using Python + Playwright (automation) and BeautifulSoup (parsing).
4. Cleaning: Removed duplicates and non-bank entries for accuracy.

Potential Use Cases:

• 📈 Sentiment Analysis: Analyze customer satisfaction trends.
• 🗺️ Geospatial Visualization: Map bank ratings by city/region.
• 🔍 Competitor Analysis: Compare bank reputations.
• 🎓 Academic Projects: Practice NLP, data cleaning, or visualization.

Tech Stack:

• Python 🐍
• Playwright (for browser automation)
• BeautifulSoup (HTML parsing)
• Pandas (data cleaning)

Why This Dataset?

• First-of-its-kind: Focused on Moroccan banks.
• Ready-to-use: Cleaned version requires minimal preprocessing.
• Transparent: Raw data included for reproducibility.

License: CC0: Public Domain (Free to use, modify, and share).

This dataset is a combination of Stanford's Alpaca (https://github.com/tatsu-lab/stanford_alpaca) and FiQA (https://sites.google.com/view/fiqa/) with another 1.3k pairs custom generated using GPT3.5 Script for tuning through Kaggle's (https://www.kaggle.com) free resources using PEFT/LoRa: https://www.kaggle.com/code/gbhacker23/wealth-alpaca-lora GitHub repo with performance analyses, training and data generation scripts, and inference notebooks: https://github.com/gaurangbharti1/wealth-alpaca… See the full description on the dataset page: https://huggingface.co/datasets/gbharti/finance-alpaca.

This dataset contains large-scale user reviews collected from the Google Play Store for five leading mobile banking applications in Türkiye: İşbank (İşCep), YapıKredi, Garanti BBVA, Akbank, and Ziraat Bank. The dataset includes more than 250000 user reviews, covering multiple dimensions of user experience such as satisfaction, complaints, feature requests, and performance feedback.

Each record provides detailed information, including:

• package_name (unique identifier of the mobile banking app)

• review_id (unique review identifier)

• user_name (anonymized or pseudonymized user name)

• content (review text)

• score (star rating, 1–5)

• thumbs_up_count (number of likes/upvotes)

• app_version and review_created_version

• timestamps (UTC and Istanbul local time)

• bank_name (associated financial institution)

The dataset was collected in August 2025 using the Google Play Scraper library, ensuring systematic extraction of publicly available app store data. All reviews are provided in Turkish (scrape_lang = "tr"), with precise timestamps for temporal analysis.

This dataset can support research in:

• User experience analysis in digital banking

• Sentiment analysis and opinion mining

• Topic modeling and service quality evaluation

• Time series forecasting of user satisfaction trends

• Comparative studies across multiple financial institutions

Researchers, practitioners, and developers can use this dataset to explore trends in digital banking adoption, analyze service quality signals, and develop machine learning models for predicting user satisfaction in mobile financial technologies.

Short-Term-Debt Time Series for GMO Payment Gateway Inc. GMO Payment Gateway, Inc., together with its subsidiaries, provides payment related and financial services in Japan and internationally. The company offers online payment system, such as PG multi-payment service, a payment system that allows to select payment methods, such as credit card, carrier, bank transfer, payment after delivery, and CVS payment services; Ginko Pay Base System, a smartphone app that enables payments to be made by an immediate debit from the bank account; and GMO-PG processing platform, which helps financial institutions and financial service providers in the business of payment-related services by enabling payment infrastructure building. It also provides face to face use services, including cashless platform, infrastructure of payment, and cooperation; pay payer system, such as paypay, d payment, Rakuten pay online payment, amazon pay, merpay, and AEON pay services; and business to business and buy now pay later services. In addition, the company offers payment agency services in the online and recurring billing, and face-to-face field; banking as a service; lending; remittance; and instant salary receipt services. Further, it provides marketing support services for listing ads that use Yahoo! Promotional advertising, and Google AdWords; and administrative services for Facebook Ads, Google Analytics, etc. Additionally, the company offers website analysis support, consulting, and other support services. The company was incorporated in 1995 and is headquartered in Tokyo, Japan.

Current-Deferred-Revenue Time Series for GMO Payment Gateway Inc. GMO Payment Gateway, Inc., together with its subsidiaries, provides payment related and financial services in Japan and internationally. The company offers online payment system, such as PG multi-payment service, a payment system that allows to select payment methods, such as credit card, carrier, bank transfer, payment after delivery, and CVS payment services; Ginko Pay Base System, a smartphone app that enables payments to be made by an immediate debit from the bank account; and GMO-PG processing platform, which helps financial institutions and financial service providers in the business of payment-related services by enabling payment infrastructure building. It also provides face to face use services, including cashless platform, infrastructure of payment, and cooperation; pay payer system, such as paypay, d payment, Rakuten pay online payment, amazon pay, merpay, and AEON pay services; and business to business and buy now pay later services. In addition, the company offers payment agency services in the online and recurring billing, and face-to-face field; banking as a service; lending; remittance; and instant salary receipt services. Further, it provides marketing support services for listing ads that use Yahoo! Promotional advertising, and Google AdWords; and administrative services for Facebook Ads, Google Analytics, etc. Additionally, the company offers website analysis support, consulting, and other support services. The company was incorporated in 1995 and is headquartered in Tokyo, Japan.

Dataset Name: Cymbal Investments Trade Capture Report

Description: This dataset, derived from the bigquery-public-data.cymbal_investments.trade_capture_report in BigQuery, provides a comprehensive view of trade capture reports for financial transactions. The data is presented in CSV format with various columns capturing essential information about each trade.

BigQuery description: Dataset in BigQuery About Cymbal

The Cymbal brand was created to make storytelling consistent across Google Cloud. Datasets are synthetic, and >provided to industry practitioners for the purpose of product discovery, testing, and evaluation. Cymbal Investments

Cymbal Investments is a US-based, global investment and asset manager. Founded in 1925, the boutique investment >banking firm’s mission is to provide meaningful financial opportunity to veterans. After nearly a century of >consistently positive returns and smart bets, it has grown into a global institution and has acquired multiple funds and >smaller institutions. In 1986, Cymbal Investments was acquired by Cymbal Group. Today, the company holds $850B >under management, employs 49K+ people and, in 2019, reported $35B in revenue. Cymbal Investments is digitally transforming legacy financial services institutions.

CSV Columns:

1. SendingTime

   • Type: TIMESTAMP
   • Mode: NULLABLE
   • Description: Time the message was sent.

2. TargetCompID

   • Type: STRING
   • Mode: NULLABLE
   • Description: Assigned value used to identify the firm receiving the message.

3. SenderCompID

   • Type: STRING
   • Mode: NULLABLE
   • Description: Assigned value used to identify the firm sending the message.

4. Symbol

   • Type: STRING
   • Mode: NULLABLE
   • Description: Trading symbol of the asset.

5. Quantity

   • Type: INTEGER
   • Mode: NULLABLE
   • Description: Overall/total quantity (e.g., number of shares).

6. OrderID

   • Type: STRING
   • Mode: NULLABLE
   • Description: Order identifier.

7. TransactTime

   • Type: TIMESTAMP
   • Mode: NULLABLE
   • Description: Time the transaction occurred.

8. StrikePrice

   • Type: FLOAT
   • Mode: NULLABLE
   • Description: Price at which the Contract for Difference (CFD) closed.

9. LastPx

   • Type: FLOAT
   • Mode: NULLABLE
   • Description: Price at which the CFD was entered.

10. MaturityDate

    • Type: TIMESTAMP
    • Mode: NULLABLE
    • Description: Date of contract expiry.

11. TradeReportID

    • Type: STRING
    • Mode: NULLABLE
    • Description: ID of this trade report.

12. TradeDate

    • Type: DATE
    • Mode: NULLABLE
    • Description: Date the trade was executed.

13. CFICode

    • Type: STRING
    • Mode: NULLABLE
    • Description: Financial instrument classification code.

14. OrderID

    • Type: STRING
    • Mode: NULLABLE
    • Description: Order identifier.

15. PartyID

    • Type: STRING
    • Mode: NULLABLE
    • Description: Counterparty identifier.

16. PartyIDSource

    • Type: STRING
    • Mode: NULLABLE
    • Description: Party ID Source

17. PartyRole

    • Type: STRING
    • Mode: NULLABLE
    • Description: Counterparty role.

Potential Analyses: - Trade Pattern Analysis: Explore patterns in trading behavior over time, identifying common trends or anomalies. - Risk Assessment: Evaluate the risk associated with different trades based on quantities, prices, and counterparties. - Market Impact Analysis: Examine how trades impact the market, considering factors like liquidity and price movements. - Time Series Analysis: Analyze the temporal aspects of trade data, identifying seasonality or recurring patterns.

Possible ML Tasks: - Predictive Modeling: Develop models to predict future trade quantities or prices based on historical data. - Anomaly Detection: Implement algorithms to detect unusual trading activities that deviate from the norm. - Counterparty Risk Assessment: Build models to assess the risk associated with specific counterparties in trade transactions. - Market Sentiment Analysis: Utilize natural language processing to analyze textual data related to trades and assess market sentiment.

Note: - The dataset provides a comprehensive view of trade capture reports, including information about the trade itself, the entities involved, and crucial timestamps. - The CSV format facilitates easy integration and analysis using various data analysis tools and platforms.

Context

Typically e-commerce datasets are proprietary and consequently hard to find among publicly available data. However, The UCI Machine Learning Repository has made this dataset containing actual transactions from 2010 and 2011. The dataset is maintained on their site, where it can be found by the title "Online Retail".

Content

"This is a transnational data set which contains all the transactions occurring between 01/12/2010 and 09/12/2011 for a UK-based and registered non-store online retail.The company mainly sells unique all-occasion gifts. Many customers of the company are wholesalers."

Acknowledgements

Per the UCI Machine Learning Repository, this data was made available by Dr Daqing Chen, Director: Public Analytics group. chend '@' lsbu.ac.uk, School of Engineering, London South Bank University, London SE1 0AA, UK.

Image from stocksnap.io.

Inspiration

Analyses for this dataset could include time series, clustering, classification and more.

IMPORTANT NOTE

This dataset contains mostly indonesian reviews

Overview

Livin' by Mandiri is a digital financial service platform developed by Bank Mandiri, one of the largest banks in Indonesia. The platform is designed to provide users with a range of financial services and features, including the ability to make payments, transfer money, and manage their finances on their mobile devices. Livin' by Mandiri is available as a mobile app for both Android and iOS devices.

Data source

This dataset collected by scraping reviews on Google Play Store

Usage ideas

EDA and Sentiment Analysis