Don't forget to hit the :star: if you like this repo.

About Us

The information on this Github is part of the materials for the subject High Performance Data Processing (SECP3133). This folder contains general big data information as well as big data case studies using Malaysian datasets. This case study was created by a Bachelor of Computer Science (Data Engineering), Universiti Teknologi Malaysia student.

📚 Big data processing

Big data processing involves the systematic handling and analysis of vast and complex datasets that exceed the capabilities of traditional data processing methods. It encompasses the storage, retrieval, and manipulation of massive volumes of information to extract valuable insights. Key steps include data ingestion, where large datasets are collected from various sources, and preprocessing, involving cleaning and transformation to ensure data quality. Advanced analytics, machine learning, and data mining techniques are then applied to uncover patterns, trends, and correlations within the data. Big data processing is integral to informed decision-making, enabling organizations to derive meaningful conclusions from their data, optimize operations, and gain a competitive edge in today's data-driven landscape.

Notes

• Essential Skills for Big Data Processing with Google Colab: A Beginner's Guide
• The Challenges of Working with Big Data in Data Science
• Strategies for Efficiently Handling Large Datasets in Data Science
• Nowogrodzki, A. (2020). Eleven tips for working with large data sets. Nature, 577(7790), 439–440. doi:10.1038/d41586-020-00062-z

1. Big Data: Pandas

Big Data processing with Pandas, a powerful Python library for data manipulation and analysis, involves implementing strategies to handle large datasets efficiently. Scaling to sizable datasets requires adopting techniques such as processing data in smaller chunks using the 'chunksize' parameter in Pandas read_csv function. This approach facilitates reading and processing large datasets in more manageable portions, preventing memory overload. To further optimize memory usage, it's essential to leverage Pandas' features like data types optimization, using more memory-efficient data types when possible. Additionally, utilizing advanced functionalities like the 'skiprows' parameter and filtering columns during data import can significantly enhance performance. By mastering these strategies, one can effectively manage and analyze vast datasets in Python with Pandas, ensuring both computational efficiency and memory optimization in the face of Big Data challenges MORE 💡.

2. Big Data: Alternatives to Pandas for Processing Large Datasets

This topic delves into the challenges encountered when using Pandas, a popular Python library for data analysis, in handling large datasets. Recognizing the limitations of Pandas, the article explores alternative solutions specifically designed for efficient processing of extensive data. It examines cutting-edge libraries such as Dask, Modin, Polars, Vaex, and others, showcasing their unique features and advantages. From parallel and distributed computing to out-of-core processing and GPU acceleration, the article provides insights into how these alternatives address the scalability and performance issues often faced when dealing with big datasets, offering readers a comprehensive guide to navigate the complexities of large-scale data processing beyond Pandas MORE 💡.

3. Comparison between libraries

• Faster Pandas with parallel processing: cuDF vs. Modin
• Scaling Interactive Data Science with Modin and Ray
• Scaling Pandas: Comparing Dask, Ray, Modin, Vaex, and RAPIDS

4. Big Data: Case study

• 7 Amazing companies that really get big data
• Data Science Case Studies: Solved using Python
• 10 Real World Data Science Case Studies Projects with Example
• Top 8 Data Science Case Studies for Data Science Enthusiasts
• Big Data in Practice: How 45 Successful Companies Used Big Data Analytics to Deliver Extraordinary Results
• Unveiling Instagram's Engagement Magic through Machine Learning
• Unlocking Spotify's Musical Enchantment with Machine Learning

Lab

Comparison between libraries

• Lab: Modin vs Pandas
• Lab: Large datasets (100MB to 1TB+) Pandas_Modin_Dask_Vaex

Contribution 🛠️

Please create an Issue for any improvements, suggestions or errors in the content.

You can also contact me using Linkedin for any other queries or feedback.