A key challenge in clustering is to find the optimal number of groups in the data. Learn about the elbow technique for finding optimal cluster sizes.
- [Instructor] One of the challenges … of using k-means clustering … is to data mine the optimal cluster size. … Basically, how big do we let the clusters get? … So we can differentiate different groups in the data. … While there are a number of ways to do this, … the most popular technique is called the elbow method. … To use the elbow method, we execute k-mines clustering … for a given data set, iteratively from one … to 10 cluster groups. … For each of these cluster sizes, … we find the some of squared distances between the clusters. … As a number of clusters go up, … the sum of squared distances will go down. … We create a function called optimal cluster log … that will compute the sum of square distances … given the cluster size. … We also will plot the sum of squared distances … against the number of clusters. … Let's execute this code and see the plot. … This graph usually has an elbow shape. … The cluster value where the elbow occurs … is the most optimal cluster size. … In this case, the elbow occurs at three. …
- Creating a word cloud
- Analyzing sentiment
- Extracting emotions from text
- Clustering similar entities based on text
- Using classification for supervised learning
- Recommending items to users based on text data analytics
Skill Level Intermediate
Predictive Customer Analyticswith Kumaran Ponnambalam1h 37m Intermediate
1. Word Cloud
2. Sentiment Analysis
5. Predictive Text
- Mark as unwatched
- Mark all as unwatched
Are you sure you want to mark all the videos in this course as unwatched?
This will not affect your course history, your reports, or your certificates of completion for this course.Cancel
Take notes with your new membership!
Type in the entry box, then click Enter to save your note.
1:30Press on any video thumbnail to jump immediately to the timecode shown.
Notes are saved with you account but can also be exported as plain text, MS Word, PDF, Google Doc, or Evernote.