Talk the Talk
- Big data
- Data is usually classified as big data when it is impractical to manage with traditional software tools. What is classified as big data is always changing and is based on volume (how much data), velocity (how fast it is created), and variety (how many different data types).
- Classification algorithm
- Sort data into predefined labeled groups. First, the computer is given a training data set, which includes a set of data that has already been sorted, and then new data is given to be sorted based on the training data set. For example, email filters use a classification algorithm to sort newly received emails in a two groups: spam and not spam.
- Clustering algorithm
- Sort data into clusters without knowing the cluster labels in advance such that points in a cluster are similar to each other and points from different clusters are dissimilar. For example, to sort athletes by sport without knowing anything about the athletes in your data set, you might plot all the athletes based on height and weight, and label the clusters that are close to each other. Short and light athletes might be coxswains who steer boats, while tall and heavy athletes might be football players in defense positions.
- An open-source framework for storing and processing big data across many computers. Instead of using one powerful computer to process data, Hadoop is a way process data across many machines.
- Machine learning
- Creating and using a variety of algorithms that ...
Get Getting a Coding Job For Dummies® now with O’Reilly online learning.
O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.