Data comes in many forms and is disorganized. Whether we're talking about missing data, unstructured data, or data that lacks regular structure, there needs to be some way of cleaning the data before it can be processed to improve data quality. This series of tutorials explores the important issues of working with real-world data, and some of the methods that can be applied.
This tutorial series is divided into 3 parts: working with scattered data, gaining valuable insights from clean datasets, and visualizing data.
Part 1: Dealing with scattered data. Discover common problems and solutions related to cleaning data for validation and processing. You'll also find a custom tool for performing data cleaning and merging datasets for analysis. It mainly includes the following parts:
- What is stray data
- Data formats and schemas
- Data blending or fusion
- Data cleaning methods
- Data profiling
- Build a data cleaning tool
- Open source data cleaning tool
Part 2: Gain valuable insights from clean datasets. Learn about VQ and ART algorithms. VQ can quickly and efficiently cluster a dataset, and ART can adjust the number of clusters based on that dataset. It mainly includes the following parts:
- vector quantization
- Implement VQ
- Clustering by VQ
- adaptive resonance theory
- implement ART
- Clustering by ART
Part 3: Visualizing the data. Explore some of the more useful applications for visualizing data, and some of the methods you can use to create such visualizations, including the R programming language, gnuplot, and Graphviz. It mainly includes the following parts:
- Visualize the raw dataset
- Visualize the operation process
- Visual cluster
Hurry up and click " read the original text " to get the full article, and play with the data for accurate insights!