Dataset is a language that a computer understands. It is the only it can communicate with human beings and human beings can communicate with computers. It acts as a basic building block to the foundation of computers. Based on these datasets machine learning happens. AI technology is based on these small data sets.

This post provides links to the datasets useful for the computer science field. Three pieces of information are provided in the following order.

1)        Name

2)        Description

3)        Link to the dataset page

Importance of Datasets:

  • It is one of the biggest dataset collections, about 240.
  • The datasets are classified into different types based on the task for which they are used. E.g. Classification, Regression, Clustering and others etc.
  • The areas to which these datasets belong are life sciences, engineering, games, business, social science etc.
  • The datasets are freely available to download with detailed descriptions of attributes types and values.
  • One can also donate datasets.
  • As a whole, this is a great source for researchers.

Types of Datasets:

There are different datasets used for different purposes. Let us discover them.

Image Datasets:

  • They are important for computer vision tasks.
  • Examples include Image detection and Segmentation of images.
  • Image datasets include
    • MNIST is a type of dataset that includes handwritten images.
    • CIFAR-10 is a type that includes small images from the 10th standard.
    • ImageNet is a large database and includes a lot of images.

Text Datasets:

  • These are a type of datasets that are used in machine learning stuff.
  • Examples include Language translation and summarization of text.
  • Such datasets include Wikipedia which consists of millions of text. Another example is the IMDB review which includes a review of each movie.

Audio Datasets:

  • Such datasets use audio for the recognition of commands. It includes music search with voice and other voice-related commands.
  • Famous Example of the Audio dataset includes TIMIT and Librosa.
  • Both of these are big databases and include millions of kilobytes of data.

OPPORTUNITY Activity Recognition:

  • These are types of datasets that are used in advanced sensors that we use in our day-to-day gadgets.
  • Mobile phone giving use maps is based on these datasets.
  • Digital watching showing our step count while wearing it is another example of such a type of dataset.

Conclusion:

This is just the tip of the iceberg that we discussed here and it is a complete speciality that is being taught to students around the world. The best dataset is the one that suits your needs and can give you the results that you desire.