Download as:
Rating : ⭐⭐⭐⭐⭐
Price: $10.99
Language:EN
Pages: 2

Discuss the pros and cons each algorithm and suggest algorithm use

(a) Cancer/No cancer detection
(b) Part of Speech Tagging (In a given sentence, detect which words are nouns, verbs, adjectives etc)
(c) Which candidate will a voter vote for
(d) Order/rank webpages based on usefullness
(2) Comment on the performance of decision tree algorithm on the following data
(a) Sparse Data (A lot of data is 0, for example word count in documents)
(b) Features have many missing values
(c) Only 500 data points (Consider a regular dataset like for cancer detection with 5-10 features)
(d) Imbalanced dataset: Consider 2 classes A and B. There are 10000 samples of class A whereas only 500 samples of class B.

(3) We know that we can use continuous features in decision trees. Explain how to do this and how to overcome the increased complexity.

Logistic Regression

(1) Consider the following training set of four labeled data:

x1 x2
1 2
3 2
2 2
0 3

0

Using the Sigmoid function for the Logistic Regression problem,where xTθ = θ0 +θ1x1 + θ2x2. Estimate the parameters θ0, θ1, and θ2.

(5) A student conducted an experiment in which he claims to achieve a training accuracy of 93% and the test accuracy of 78%. Would you have any suggestions to improve on the obtained results? What would you have suggested if the training accuracy was 78% and

Copyright © 2009-2023 UrgentHomework.com, All right reserved.