Bonnie's Ramblings bout Web Analytics & Technology

SI 618 Wk1 Blog

I’m excited to work with R, more visualizations, and building interesting R applications.

What I’ve found helpful in learning an application

instead of going through all the different functions quickly which washes over my head… If you provide me a slide of all the important functions, I can look up how different functions work.

It’s more useful to walk through 3 common/meaningful R examples and then going through the motivations of why certain things occur the way they are

Is there a R application you’ve built? Is there a simple R script that you’ve put together to solve a problem that you could share?

Why is c our new best friend?

What is the most common pitfall of uses matrix?

Try using the whiteboards perhaps too?

SI 601 Week 6 Blog Post

I find the project very exciting to work on. There are many interesting challenges to extracting data. In my project, we are also working on geocoding information so the Google API is very relevant to that. I actually required to translate lat/long to location information aka reverse geocoding and found a useful library called geopy to do this task.

I feel that it would have been better to NOT have any lab/hw this week so we can devote our time to insuring the quality of our project. In fact, I feel that perhaps having a lab AND different hw every week is too much. There is a lot of overhead costs to handling a different data set and new code task.

Perhaps having a way to extend the lab or lab/hw once every 2 weeks makes more sense? 

SI601 Week 5 - Breadth vs Depth?

While it’s nice to cover lots of different topics, I’ve found that less is more. I benefited from the extra time we spent on regular expressions for example. During the next half, we’ll be doing visualizations. I believe that’s an area of research for you. It’ll be exciting if we can hear your perspectives on

- how to visualize a large set of data?

- how do we communicate these ideas to a normal user?

- how do we leverage data visualizations to generate features?

- beyond the normal “data” visualization questions… how do we illustrate “concepts”

It’ll be great if we can go deeper into a few of the topics and generate discussions and debates! I think that will be fun :)

List Comprehensions and Matrices

Double Post today - i know odd for me.

List comprehensions is one of those things I never use. I tend to be very explicit. I decided to implement Levenshtein’s edit distance algo with Python to play with it.

editMatrix=[[0 for i in range(len(b)+1)] for j in range(len(a)+1)]

That’s where I initialized a matrix of 0s… It’s nice that it can be done in one line. Without the list comprehension, the code is like this:

newList=[]

            for i in range(len(b)+1):

                test=[]

                for j in range(len(a)+1):

                    test.append(0)

                newList.append(test)

            print newList

SI 601 Week4

It’s challenging to teach a broad spectrum of concepts to a broad spectrum of skillsets :( The lab and hw has been key to disseminating the information. Otherwise adding links to good tutorials may be best…

For the hws, can you please provide a small sample output of what you are looking for? That would help me understand what is required for the assignment.

Also, what is the difference between 618 and Eytan’s Information Visualization class in the Fall? 

SI 601 Week3

I liked this week’s lab and lecture much better. I enjoyed the subtle examples and the fact that we got to code!

It would be nice if you shared what you thought was the most efficient solutions to this lab and homework.