Question

Previously we reviewed supervised learning (based on pre-existing data patterns) & unsupervised learning (based on hidden patterns).

a. Share the industry and problem you previously identified.

b. Explain both the pros and cons of using (1) supervised and (2) unsupervised learning for your initiative.

When you respond to your colleagues, refrain from adding generic phrases from the text and/or other resources. Instead, target your response to their industry and/or problem with new content or insight.

Answer 1

Thanks for asking question here

Supervised learning is a learning in which we teach or train the machine using data which is well labeled whereas in unsupervised learning we do not need to supervise the model. your algorithm automatically group the unsorted information according to the similarities of the data.
Unsupervised learning classied into two categories of algorithm

• Clustering: A clustering problem is where you want to discover the inherent groupings in the data, such as grouping customers by purchasing behavior.
• Association: An association rule learning problem is where you want to discover rules that describe large portions of your data, such as people that buy X also tend to buy Y.

Let me explain above two with example.

Supervised learning - Let assume that we have created a model which tell us fruits name based on colour and shape.

I have written in the model that if shape is round and colour is red . then indenify that fruit as apple.

similarly , if shape is cyniderical and colour is yellow. then idenify that fruit as banana.

So, our model match the with shape and colour of the object and give us output. Let assume that an apple has space round and colour is green . then, our model failed to indentify here the correct fruit name. So, we can say that supervised model can be used only in known possible outcomes.

Unsupervised learning - Let assume that we have created a model which tell us animals type and we didn't define any our model. So the unsupervised model categories/groups the data based on the shape of the animals. Model will put cat type animals in one group and dogs type animal in another group.

This type of model is applied on unknown data set and we do not accept 100% accurate answer in this case.

Example-Our model may put cats and tigers in one group of data base on basic space of both animals.

(b)Pros of Supervised model - (1)This accuracy is very high about 99%. (2) Performance(Time taken to run) of this model is much higher than unsupervised model. (3) we can apply supervised in any other different data set with few changes in model.

Cons of Supervised model - (1) Cannot be applied in unknow data set. (2) Have limited scope of one Supervised model.

Pros of Unsupervised model - (1) Can be applied on unknow data set. (2) We can develop Supervised model based on ouput of Unsupervised model.

Cons of Unsupervised model - (1) Take much time to run. (2) Accuracy is not very high.