Meta Analysis

The Chicago Web, Game and Social Media Analytics Meetup has been around several years and has developed a substantial membership. Now, the group has a new home. Thoughtworks, a global IT consultancy based in Chicago, will host meetings at their headquarters at 200 East Randolph. Last week, I presented “Crossing the Language Chasm: Extracting Information from Foreign-Language Text” for the group at the new location, and it was a pleasure. The space is roomy, comfortable and a great match for this use. The meeting was well attended, and I expect that the new space will help to build attendance.

If you didn’t get to attend the presentation, you can read the original article on Smart Data Collective:

Crossing the Language Chasm: Extracting Information from Foreign-Language Text

Social Media Analytics Summit, April 17-18, San Francisco
Capitalize on Multi-lingual Social Media Analytics

European Text Analytics Summit, April 23-24, London
Cross-lingual Text Analytics: A New Frontier in Linguistic Technology

Chicago Web, Game and Social Media Analytics Group, May 2, Chicago Free!
Crossing the Language Chasm: Extracting Information from Foreign-Language Text

Predictive Analytics World, June 25-26, Chicago
Cross-Language Text Analytics: Overcoming Language Barriers

Visits have increased by nearly a factor of ten since I added a blog to the site.

Pre-blog, activity varied little by time of day. Post-blog, there’s a clear spike in activity when posts appear in the morning.

A lot of people visit my site using the links left with comments on other sites.

Google’s spiders crawl the site with amazing frequency! (You’d think I was CNN or something.)

There was a dramatic spike in visits on February 28. (I have no idea why.)

Someone found me by searching for the term “douchegrammer.” (Wonder if he or she was pleased or disappointed?)

This is hilarious – a summary of research on the wants of men and women, as expressed in social media, from Netbase. We have a lot more similarities than differences! You must read this for yourself, but here’s a hint – we all want food!

BTW, I tweeted about this a few days ago, just once, and was stunned when I noticed that my tracking link was drawing hundreds of clicks. Turns out that the tweet got picked up by msn now. I would never have known if I hadn’t tracked. The moral: use tracking links and you may learn something!

Here are a couple of interesting comments that have come up for my recent article on Smart Data Collective: “Big Data Blasphemy: Why Sample?”

From Simon Geletta, Associate professor at Des Moines University, commenting in the SAS Analytics & BI group on LinkedIn:

“That (everything being equal) bigger samples result in narrower confidence intervals is not a matter of opinion… The argument for sampling as presented in this blog, would benefit (become worth following) if the blogger can demonstrate that estimations that are based on sampe (from a bad sampling frame) yield better results as compared to estimations that are based on the bad sampling frame itself.”
My thoughts:

Yes, it is a matter of fact that bigger samples result in narrower confidence intervals.

The case for sampling does not depend on samples from a data resource producing better estimates than using all of the data available. There are cases in which the sample data can be more carefully inspected, corrected and otherwise cleaned, and in those cases estimates may, indeed, be better than those which would be made using a larger, dirtier, data resource. However, even when there is no improvement in data quality for a sample, there is always the issue of balancing the resources consumed with the value of the information obtained. The question is not whether the estimate obtained using a sample is better than the estimate that would be obtained by using all available data. Rather, it is a matter of obtaining “good-enough” information to address the business problem at hand, and of doing so in a manner that does not waste resources which could yield greater returns if used in some other way.

From Blaise Egan, Lead Network Infrastructure Analyst at British Telecommunications PLC, commenting in the Predictive Analytics group on LinkedIn:

“This reflects my own experience encountering data miners with a background in computer science rather than statistical science. To some extent they have been hoodwinked by misleading sales material from vendors of large-scale computing systems, both hardware and software.

It’s an important message that you’re putting out.”

New post on Smart Data Collective today, “Big Data Blasphemy: Why Sample?”

The New York Times report about Target’s pregnancy prediction model has made a splash in 2012, but Predictive Analytics World had it in 2010, presented by Target’s own Andrew Pole. His talk is much clearer on just how hard and imperfect this process really is.See the video.

Lately I have been speaking out against translating text with the aim of feeding it into text analytics tools that were not designed for use with the original language of the text. On several occasions, some conscientious person has stepped up and announced that, on the contrary, this practice is producing good results. All of these have been respected colleagues, and I don’t care to pick on them individually.

But, speaking to all of you as a group, here’s something you need to know: your customers are coming to me behind your back and telling me your results are lousy.

The Asian Banker has established a working group focused on data and analytics. They offer specialized content and events for this audience, in a members-only environment. Today they are featuring my piece, “Rising to the challenge of cross-lingual sentiment analysis.”

Christian Arno’s Secrets to Social Media Success in Foreign Languages on Social Media Today explains some basics for global social media planning. One topic that it does not address – how will you evaluate your work?

No matter the language, if you’re going to invest in marketing of any type, you should have measurable goals and a plan to evaluate your results. This is often neglected, and so a lot of marketing resources are wasted. Much of the planning and analytics is the same no matter the language, but there are special issues when multiple languages and culture are involved.

Consider questions like these:

Do your goals need to vary for individual languages or countries?

Is text analysis for content and or sentiment useful to you? What process will make this possible for the languages you require?

What will it take to help a manager in your country able to develop a marketing plan appropriate for another, and evaluate the results?