Creating a Dataset

Creating a Dataset

1. Data Size

   • in deep learning : If you have too few data points _or_ your data is not evenly distributed between different categories that you want to distinguish, you could get some significant sampling bias in your end predictions;
   • Credit card fraud detection: most credit card transactions are valid ---  If all wolves are images with asnowy background, a machine learning model might mistakenly conflate snow with wolves,
   • Data Distribution and Pattern Detection
     • Credit card fraud detection: most credit card transactions are valid, and so these datasets often have thousands of valid examples and very few examples of fraudulent transaction data,
     • distinguish wolves from dogs. If all wolves are images with a snowy background, a machine learning model might mistakenly conflate snow with wolves,
   • If you are using a deep learning algorithm, as opposed to some traditional machine learning techniques, data size is even more important.
   • many examples of every possible category before they can learn to distinguish between different classes of data and find general patterns in some data

2. Confusion matrix

   • f1=pre*recall /(pre+recall)

3. Data Collection

   • How ( image | existing data and research | data from companies or customers ) -
   • Relevance ( match the real world and your problem )

4. Data Completeness

   • Represent all scenarios - relations and pattern, missing data
   • What problem ?

5. Data Annotation

   • These platforms will send unlabeled data to some human annotators who can classify or provide features for the data and send it back to you in a tabular format.

6. "In computer science, garbage in, garbage out (GIGO) describes the concept that flawed, or nonsense input data produces nonsense output or 'garbage'."

7. Planning for longevity

   • Static Model: No need to update
   • Dynamic model
   • frequently changing data

Bilder und Dateien der Mindmap