Filter By
Filter Reset
Semester
Class Notes (1,031,120)
US (404,596)
ND (2,748)
CSE (32)

Class Notes for Computer Science and Engr. at University of Notre Dame

CSE30246 Lecture 9: Lectures 9 and 10 Notes Part 4
Premium

Lectures 9 and 10 Notes Types of cross validation: o K-fold o Random o Stratified o Leave one out Bootstrapping: o Use sampling WITH REPLACEMENT o Potential problem = if imbalanced dataset sampling with replacement so you ...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture 9: Lectures 9 and 10 Notes Part 3
Premium

Lectures 9 and 10 Notes Estimating Future Performance: o Resubstitution error = accuracy of model training process when model is being trained, it tries to predict the data based on the training data several times internal...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 9: Confusion Matrix
Premium

Lectures 9 and 10 Notes How to evaluate and improve upon the performance of a model Measuring Performance: o Goal is to extrapolate model to future cases how will model today do in the future o Important to put context beh...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 7: Lincoln Near-Earth Asteroid Research, Logistic Function, Maximum Likelihood Estimation
Premium

Lectures 7 and 8 Notes Simple Linear Regression: o Goal is to estimate alpha and beta trying to find a function for the slope (beta) and intercept (alpha) that MINIMIZES the error closed form calculation o Chart interpreta...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 7: Additive Smoothing, Association Rule Learning, Poisson Regression
Premium

Lectures 7 and 8 Notes Additive Smoothing: o Add a pseudo count (alpha) to my calculation in order to change the expected probability so as to avoid zero-frequency problems o Example: d = number of features Likelihood valu...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 7: Association Rule Learning, Conditional Probability, Additive Smoothing
Premium

Lectures 7 and 8 Notes Nave Bayes: o Probabilistic learner = use prior information to predict the most likely class gives class and probability Different than k-nearest neighbor because that looks at what is closest TO ME ...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 9: Hyperparameter Optimization
Premium

find more resources at oneclass.com Lectures 9 and 10 Notes 1. What model to choose? o Need to bring model to data and bring them together 2. Which parameters do I want t...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 9: Odds Ratio, Receiver Operating Characteristic
Premium

Lectures 9 and 10 Notes Kappa: o Adjustment of accuracy based on chance factor in possibility that I could randomly choose the right values o Use this to adjust accuracy value so it is now adjusted based on possibility of ...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 9: Random Forest, Adaboost, Machine Learning
Premium

Lectures 9 and 10 Notes Meta-Learning: o Combination Function = how do I bring predictions together in order to learn Approach in which you use another model to learn the outputs to figure out how to combine together If I ...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 7: Logistic Regression, Generalized Linear Model, Logit
Premium

Lectures 7 and 8 Notes Logistic Regression: o Generalized Linear Models = try to create a relationship between linear values and a function similar to the logistic function do this through thetransformation function (in th...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 4: Association Rule Learning, Predictive Power, Sparse Matrix
Premium

find more resources at oneclass.com Lecture 3 & 4 Notes Association Rules: o Support: How did he get the 0.4 there are two tran...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 2: Unsupervised Learning, Central Tendency, Data Science
Premium

Lecture 2 Notes Class 1 Review: o Finding the best algorithm to solve the problem you want to accomplish o Goal of machine learning = be a matchmaker o Predictive models = supervised learning o Descriptive models = unsuper...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 2: Data Deduplication, Signal-To-Noise Ratio, Centroid
Premium

find more resources at oneclass.com Lecture 2 Notes Data Cleaning: o Noisy Data: Can use smoothing to fix it trying to minimiz...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 5: Logistic Regression
Premium

find more resources at oneclass.com Lectures 5 & 6 Notes K-Means Clustering: o Elbow Method: if I plot all possible combinations of k against WCSS, see d...

Computer Science and Engr.
CSE30246
Fred Nwwanganga
CSE30246 Lecture Notes - Lecture 5: Euclidean Distance
Premium

Lectures 5 6 Notes Lazy Learners: o Learn verbatim look at training data and do not estimate a function o Also known as instance based learners or rote learners Rote Learner: learned based on repetition just learning on r...

Computer Science and Engr.
CSE30246
Fred Nwwanganga

Log In


OR

Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit