Machine Learning

Unsupervised Learning an Angle for Unlabelled Data World

Share this

In Unsupervised Learning; data have no target attribute. In this learning algorithm takes as training examples the set of attributes/features alone.

This is our second post in this sub series “Machine Learning Types”. Our master series for this sub series is “Machine Learning Explained”. First post about Supervised Machine Learning is available here

Unsupervised Learning; is one of three types of machine learning i.e. Supervised Machine Learning, Unsupervised Machine Learning and Reinforcement Learning. This post is limited to Unsupervised Machine Learning to explorer its details.

Unsupervised Learning

AILabPage’s – Machine Learning Series

Unsupervised Machine Learning

  • A technique with the idea to explore hidden gems / patterns.
  • To find some intrinsic structure in data.
  • Something cant be seen with naked eye requires magnifier (UML)

In Unsupervised Learning available data have no target attribute. Machine Learning Algorithm takes training examples as the set of attributes/features alone. The purpose of unsupervised learning is to attempt to find natural partitions in the training set. The most common unsupervised learning method is cluster analysis at the same time two general strategies in UML includes:

  • Clustering – Partitions data into distinct clusters based on distance to the centroid of a cluster
  • Hierarchical Clustering – Cluster tree is build with multilevel hierarchy of clusters. No assumptions on the number of clusters
    • Agglomerative – In this technique its start with the points as individual clusters as it move forward; at each step, merge the closest pair of clusters until only one cluster left.
    • Divisive – Here its start with one, all-inclusive cluster. At each step, split a cluster until each cluster contains a point.

System does self-discovery of patterns, regularities and features etc. from the input data and relations for the input data over output data. Discovering similarities and dissimilarities to forms clusters i.e. self-discovery is main target here. Since the examples given to the learner are unlabelled, there is no error or reward signal to evaluate a potential solution. This distinguishes unsupervised learning from supervised learning and reinforcement learning.

Unsupervised Learning

AILabPage’s – Machine Learning Series

Unsupervised learning – Pros & Cons

Since no labels are given to the learning algorithm, leaving it on its own to find structure in its input. Unsupervised learning can be a challenging goal in itself. The training data consists of a set of input vectors x without any corresponding target values; hence known as learning / working without a supervisor.

  • Pros
    • It can detect what human eyes can not understand
    • The potential of hidden patterns can be very powerful for the business or even detect extremely amazing facts, fraud detection etc.
    • Output can determine the un explored territories and new ventures for businesses. Exploratory analytics can be applied to understand the financial, business and operational drivers behind what happened.
  • Cons
    • As seen in above explanation unsupervised learning is harder as compared to supervised learning.
    • It can be a costly affair, as we might need external expert look at the results for some time.
    • Usefulness of the results; are of any value or not is difficult to confirm since no answer labels are available. 
Unsupervised Learning

AILabPage’s – Machine Learning Series

 

Unsupervised Learning Categories

  • Parametric Unsupervised Learning
  • Non-parametric Unsupervised Learning

Though in parametric algorithms, despite having not required much data to train, it however  does  also cause overfitting. It is more common that parametric under fit and non-parametric overfit. Both types of algorithms can over and under fit data though.

Unsupervised Learning

AILabPage’s – Machine Learning Series

Frequently used algorithms in unsupervised machine learning

Common clustering algorithms include:

  • Hierarchical clustering: In this technique the algorithm builds a multilevel hierarchy of clusters by creating a cluster tree
  • k-Means clustering: Here data gets partitions into k distinct clusters based on distance to the centroid of a cluster
  • Gaussian mixture models: Algorithms builds a model in which model clusters a mixture of multivariate normal density components
  • Self-organizing maps: This one gets super simplified by using neural networks that learn the topology and distribution of the data
  • Hidden Markov models: Simply uses observed data to recover the sequence of states
Unsupervised Learning

AILabPage’s – Machine Learning Series

The number of cluster seekers can be chosen adaptively as a function of the distance between them and the sample variance of each cluster. The best use for unsupervised is around exploratory analytics to understand the financial, business and operational drivers behind what happened.

 

Unsupervised Machine Learning (UML) uses cases for real businesses

UML is used to find anomalies in data or cluster data items to groups that humans can’t assume themselves.  Since output variables are unspecified here so algorithms looks for structures in the data to describe and hidden distribution or structure of data. Customer segmentation in different groups for specific interventions, product targeting, market categorization and recommendations are few examples here.

Unsupervised Learning

AILabPage’s – Machine Learning Series

 

Points to Note:

All credits if any remains on the original contributor only. We have covered Unsupervised machine learning in this post, where we find hidden gems from unlabelled historical data. Last post was on Supervised Machine Learning. In the next upcoming post will talk about Reinforcement machine learning.

 

Unsupervised LearningConclusion –  Collecting and labelling a large set of sample patterns can be very expensive. How this type of learning helps business to see some potentials which is usually hidden normally. The goal in such unsupervised learning problems may be to discover groups of similar examples within the data, where it is called clustering, or to determine how the data is distributed in the space, known as density estimation. In contrast with sequence mining, association rule learning typically does not consider the order of items either within a transaction or across transactions. Association rules mining is another key unsupervised data mining method, after clustering, that finds interesting associations (relationships, dependencies) in large sets of data items.

 

#MachineLearning 

 

Books Referred

 

============================ About the Author =======================

Read about Author at : About Me

Thank you all, for spending your time reading this post. Please share your feedback / comments / critics / agreements or disagreement. Remark for more details about posts, subjects and relevance please read the disclaimer.

FacebookPage    ContactMe      Twitter         ====================================================================

Advertisements

8 replies »

  1. Nice complied work … pls add some real business use cases which your key or golden point and I read your posts for that you are now becoming my mentor

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.