An assignment for my “Intro to Data Science” course was to estimate which state was “happiest” based on twitter sentiment analysis. Words such as “adore, admire, fun, love, reassure” score positive, words such as “despair, harmful, mediocrity, upset” score negative.

I analyzed 300,000 tweets and came up with the following result. (Note: Darker red means happier sentiment.) Hawaii was the clear winner, but UT-WY-CO-KS-OK form a very strong region of positive sentiment tweets.

In my analysis, LA and AL were the only two states that had a negative twitter sentiment, meaning, on average, there were more negative sentiment words than positive.

I started a new class from Coursera.org – Introduction to Data Science by Bill Howe from the University of Washington – and I expect the 8-week course to be both challenging and rewarding.

The course covers method of handling and manipulating, analyzing, and visualizing massive data sets.

The first class assignment was to use Python script to analyze the “sentiment” of a large number of Twitter posts. The assignment was not intended to be particularly sophisticated; rather, just to get a taste of what we can do with data science.

I captured about 175,000 tweets over a period of 30 minutes and used the 2,477 words and phrases found in the AFINN-111 list to score the sentiment of each tweet.

After removing all of the tweets with a sentiment of zero (i.e., they didn’t contain any words from AFINN-111), I plotted them in the following histogram. Positive sentiment tweets are marked in green, and negative in red. The further from the middle of the graph, the stronger the sentiment.

Again, this is not a sophisticated analysis (i.e., I didn’t clean up words by removing punctuation, etc.), but you can see that for the 170,000 tweets I analyzed, sentiment is more strongly positive.

The assignment has multiple parts. I’ll continue to post the highlights for this and other assignments from the course.

We were in Houghton, MI, (the UP – Upper Peninsula) last Thanksgiving visiting our friends. Here are two family photos that were taken during the traditional “get the Christmas tree” outing.

The first has a classic photobomb from their youngest. Which made us smile for the second shot.

I learned a new word today – Paraprosdokian!

From wikipedia:

A paraprosdokian is a figure of speech in which the latter part of a sentence or phrase is surprising or unexpected in a way that causes the reader or listener to reframe or reinterpret the first part.

Here are some favorites I’ve pulled from various sources:

• Some people are like Slinkies … not really good for anything, but you can’t help smiling when you see one tumble down the stairs.
• The voices in my head may not be real, but they have some good ideas!
• When tempted to fight fire with fire, remember that the Fire Department usually uses water.
• I haven’t slept for ten days . . . because that would be too long.
• If I could say a few words, I’d be a better public speaker.
• Nostalgia isn’t what it used to be.
• I like going to the park and watching the children run around because they don’t know I’m using blanks.
• She looks as though she’s been poured into her clothes, and forgot to say ‘when’.
• I’ve had a perfectly wonderful evening . . . but this wasn’t it.
• I want to die peacefully in my sleep, like my grandfather, not screaming and yelling like the passengers in his car.
• The last thing I want to do is hurt you. But it’s still on the list.
• Knowledge is knowing a tomato is a fruit; Wisdom is not putting it in a fruit salad.
• To steal ideas from one person is plagiarism. To steal from many is research.
• A bus station is where a bus stops. A train station is where a train stops. On my desk, I have a work station.
• I didn’t say it was your fault; I said I was blaming you.
• A clear conscience is usually the sign of a bad memory.
• Change is inevitable, except from a vending machine.

As part of our trip to Buffalo, NY, we took a hike to Eternal Flame Falls, which is located in Chestnut Ridge Park.

A small cave at the waterfall’s base emits enough natural gas to create an on-going 6-inch flame. Here’s a picture of the flame and then a Photosynth panorama I took with my iphone (using the Photosynth app).

Here are a few other pictures I took along the way.

Our family spent the Independence Day weekend in Houghton, MI. We stayed with our long-time friends, the Shaws. It’s always a party with 4 adults, 6 children, and 2 dogs staying in one house!

This year, we took a hike up to Hungarian Falls. It was a nice hike, with lots of sharp drops-offs that provided the kids plenty of opportunities to startle the adults.

After the hike, we watched the fireworks at Lake Linden. It was quite an impressive show for an economically-depressed small town in Michigan. Stimulus money?

Almost all online campaigns need to be optimized. While there are a number of software packages available to assist you with this, most beginning internet marketers do it manually.

One of the most common questions is, “When do you know whether a test was successful or not?”

The precise answer is, “It depends.”

Obviously, the more data you collect, the more information you’ll have and the better decisions you can make.

But this is expensive.

You can run very short tests to save money, but the chances you’ll make bad decisions increases considerably.

There is a free online tool you can use to get a better feel for the statistical significance of a test you run. In other words, it will estimate the range of your result, allowing you to decide if it makes sense to continue a test.

The website is the Exact Binomial and Poisson Confidence Intervals page. (Don’t let the title scare you. It’s easy to use.)

Here is how you can use the page to determine whether you should stop or continue a test.

First, scroll to the bottom of the page and change the confidence interval from 95 to 90%. While 95% might be good for scientific research, it is just too expensive for most of the internet marketing you’ll be doing. (A 90% confidence interval means you’ll only be wrong about your decision 1 time out of 10.)

The actual calculation is easy. Go to the Binomial Confidence Interval section of the page. Enter the number of conversions you got from your landing pages in the top box, and the number of visitors to your page in the bottom box. Then click compute.

You’ll get back the exact ratio (in this case, 0.030, or 30%). The important numbers are below that. They represent the upper and lower estimate of how your page actually converts.

In this case, if you get 3 conversions after 100 visitors, your actual conversion may be as low as 0.8% and as high as 7.6%.

So, should you continue your test? It depends!

If you’ve calculated an acceptable ROI at 4% conversion, it totally makes sense to continue the test. On the other hand, if you need a 7% conversion for an acceptable ROI, it probably makes sense to stop the test.

Have any questions? Leave them in the comments and I’ll get back to you!

Yes, both of these are real fears for those of us heavily involved in a lean transformation.

When we hit the wall one of two things happens.  Sometimes the lean program gets a back seat and dies a slow death and in other cases the CEO wants lean spread fast with direct accounting for bottom-line results.  Then we get cookie-cutter lean (e.g., spread the kanban, standardized work sheets everywhere, etc.) and managers who are accountable for results or else.  Often that means lean audits of various kinds.

Jeff Liker: Lean Has a Short Half-Life Without Intense Involvement Of The CEO » The Lean Edge.

Twisted Dazzle, the band for which my daughter Emily is the drummer, got to play outside in downtown Appleton, Wisconsin, to help “motivate” the runners of the May 7, 2011 Sole Burner up the hill by Lawrence University. It was a perfect day, and they sounded great! Here’s one of their new cover songs – Let’s Get the Party Started by Pink.

During the end of their performance, they had a fun surprise visit from the Harmony Cafe Coffee Cup mascot!