Week of 09/21/09

Content:

The first lecture was on machine learning.  The difference between supervised and unsupervised learning by machines was covered.  Another topic that was taught was a few tasks that machines use for learning.  Those tasks are classification, clustering, regression, and semi-supervised learning.  Classification can be performed by using a model for a class attribute.  That model is created by using a record training set to find a class attribute as a function of the values of other attributes.  Some applications and algorithms for classifications were explained.  A definition of clustering and some applications of clustering were instructed.  Machine learning seems like a useful but complicated topic.  At what point to set k values in the knn classification method seems complicated.  That complication is from finding a balance between avoiding having that k value be sensitive to noise points and including points from other classes in its neighborhood.  I think the presenter did a nice job at including a diverse number of applications of machine learning in the lecture.

The second lecture covered a variety of study designs.  The importance of knowing about an assortment of study designs was taught.  Knowledge of study designs can help with interpreting studies and the statistics they contain.  That knowledge can also help a researcher to select a study and understand its practicality.  Some key study design dimensions were instructed.  Those dimensions are purpose, a questionnaire's effect on subjects, the subjects' and researchers' views of time, and feasibility.  Classifications of study designs were also explained in the second lecture.  In epidemiology and medicine study designs can be classified as descriptive, observational, or experimental.  In behavioral science and evaluation research study designs can be classified as experimental or quasi-experimental.  Additionally, subtypes and examples of the studies were taught.  The study design lecture was interesting to me.  Using interrupted time series in quasi-experimental design seems like a particularly helpful type of study design to know about.  Researchers using a interrupted time series timeline can avoid spending their resources to collect middle data in a timeline if that data is not useful to them.  That could be a valuable way to save resources in a study.

The Journal of Machine Learning Research offers a large variety of articles on machine learning.  That journal's site can help anyone who is trying to gain an indepth understanding of machine learning.  The site provides an article search engine for anyone who is interested in specific machine learning topics.  The link for the journal's site is: http://jmlr.csail.mit.edu/ .

Posted by:

Nate