Big Data & Machine Learning
November 8, 2021
What is Big Data and Machine Learning?
Big Data is data that contains a huge amount of information. Oracle defines it as having 3 V's: variety, volume, and velocity.
Variety describes the type of information that the data contains. For example, Facebook could collect data about your likes, shares, reposts, the time you spent watching a video, the types of posts you comment on, the groups you're in, etc. That's a huge variety of data that is collected and can be used by Facebook. Volume describes the sheer amount of data collected. If we stick with the Facebook example, think about how many people use Facebook. According to Statica, there are roughly 2.91 billion monthly users. That's a lot of data to sort through and analyze. Lastly, velocity refers to the rate at which this data is gathered. Facebook has roughly 1.93 billion daily active users, which means they collect data rapidly as people interact with the site.
After all this data is gathered, machine learning comes into play to aid in the analysis and lead to insights. According to IBM, machine learning stems from artificial intelligence and computer science, and it uses data to imitate human processing and learning to increase it's accuracy.
Big Data is data that contains a huge amount of information. Oracle defines it as having 3 V's: variety, volume, and velocity.
Variety describes the type of information that the data contains. For example, Facebook could collect data about your likes, shares, reposts, the time you spent watching a video, the types of posts you comment on, the groups you're in, etc. That's a huge variety of data that is collected and can be used by Facebook. Volume describes the sheer amount of data collected. If we stick with the Facebook example, think about how many people use Facebook. According to Statica, there are roughly 2.91 billion monthly users. That's a lot of data to sort through and analyze. Lastly, velocity refers to the rate at which this data is gathered. Facebook has roughly 1.93 billion daily active users, which means they collect data rapidly as people interact with the site.
After all this data is gathered, machine learning comes into play to aid in the analysis and lead to insights. According to IBM, machine learning stems from artificial intelligence and computer science, and it uses data to imitate human processing and learning to increase it's accuracy.
Spotify Case Study
I have to have music playing all the time. I love having a constant soundtrack to my life that can fit any mood I'm in. I have specific songs I listen to when I'm happy, sad, want to dance, and want to chill and all of these songs can be found on Spotify. This company is known to use the data they collect to enhance the user experience through machine learning.
Just by scrolling through their app, I can see the various ways they use big data and machine learning to enhance the listening experience.
Spotify tracks the songs, artists, and genre of music that I listen to and they create different Mixes that cater to the different types of music I listen to, as seen in Figure 1.
They also recommend playlists bases on songs that I've liked. For example, you can see in Figure 1 that they're recommending this playlist called "omw" and when I open it, I can see there is a song that I previously "liked" on here, indicated by the green heart (Figure 2.)
Just by scrolling through their app, I can see the various ways they use big data and machine learning to enhance the listening experience.
Spotify tracks the songs, artists, and genre of music that I listen to and they create different Mixes that cater to the different types of music I listen to, as seen in Figure 1.
They also recommend playlists bases on songs that I've liked. For example, you can see in Figure 1 that they're recommending this playlist called "omw" and when I open it, I can see there is a song that I previously "liked" on here, indicated by the green heart (Figure 2.)
Figure 1. Curated playlists based on my previous listening.
Figure 2. A recommended playlist containing a song I "liked"
They must collect a huge volume of data each time I open the app to spit out such good recommendations and curated playlists. I don't mind them monitoring my behavior when it comes to music, because it's led to me discovering so many amazing artists and songs.
What do you think about the amount of data they collect? Do you like the recommendations or do you think it's creepy? I'd love to hear your thoughts!