Short answer is probably yes (if we are talking about the next election). But not as it was done in this election cycle. But there is a lot of potential.

Here is what we can answer: we can tell what people are talking about; categorize all comments into meaningful contextual buckets that we call Topics and do trend analysis based on that. We can also accurately extract names, places and brands.

If we could do this again the results would be dramatically more accurate and actionable

This is how we would do this next time: First we would (automatically) deduce commenters political affiliation from their first comment. At least in this type of situation of two candidates, that would be fairly easy. Whether people like this is a whole other question but we could do it anonymously by giving each commenter an ID. (That is how we handle the feedback analysis. We take away all the personal information.) Knowing commenters political affiliation is important: it enables us to fix the “cross commenting” problem so that we could really analyze the supporters and antagonists compared to their political affiliation. This would enable us to find out what each candidates supporters are talking about in the candidate’s and opposition’s site.

I wish Facebook had a location API

One really cool thing to have would be commenters location by state or preferably city. With this type of volumes, it would make the geographic visualization meaningful. You could (maybe) see clusters of certain topics form in certain areas. This would of course mean that Facebook would make the location available through their API, which we think will NEVER happen (because it is one of their biggest assets). One option could be to develop a Facebook Political Analysis app within the FB ecosystem. However, I am not sure whether they make the location available even then.

With minor optimization work sentiment analysis would be much more accurate

Etuma sentiment analysis is very accurate for those industries (and languages) that we have optimimized it for. Because we did not have any data nor the resources to do the optimization, we never reached the sentiment analysis accuracy we would have like to. Next time we would use this data to run a few man working week sentiment optimization project. We have a clear process and tools for it. The sentiment accuracy would be in totally different level. Right now we estimate that it was about 70% (what this mean is that in 7 Topic mentions out of ten we can accurately say whether people are talking about the Topic negatively, neutrally or positively). With optimization it would get close to 90%. Then you could really rely on the statistical trend analysis and detect even minute changes in Topic specific sentiment.

And we would also include Twitter analysis. Not just sentiment but what people are talking about.

We are confident that with these improvements we could predict the outcome in the next elections.

Thanks for all the readers. This is the last Baramitter blog. If you are interested in following feedback text analytics business and technology, we will continue the discussion in www.etuma.com.

Visit etuma.com to learn more about the solution and technology behind Baramitter.

Education and Energy&Environment getting more discussion but Taxes rules

This time I focused on just few of the main topics. Now just a couple of days before the election it seems that the topics that are on the rise are education and energy&environment.

Taxes taxes taxes

Taxes have become the number one topic. Higher than economy. Naturally we could debate about whether taxes belong to economy but we made it its own Topic.

Visit etuma.com to learn more about the solution and technology behind Baramitter.

It is easy to push the Like-button. Much harder to actually express an opinion and post it. According to  Huffington Post  Mr. Obama has many more social media followers than Mr. Romney.

That does not seem to be true when looking at the Facebook data. Maybe Huffington Post means Like-button clickers and that is actually true but when you looking at people …talking about this…. and actual comments in Facebook walls, Mr. Romney supporters seem to be in the majority.

This just shows that it is so easy to click the Like-button and much harder to get active and say something. Therefore, clearly in relation to Likes, Mr. Romney supporters are much more active.

What makes this and other analysis quite difficult is the “cross-commenting” phenomenon.

Here is a an actual quote from a comment to which the answer is absolutely affirmative.

There is lot of “cross commenting” between the candidates. We have no hard data on the “cross-commenting” phenomenon as we are don’t know commenters’ political affiliation. But reading through hundreds of comments it is obvious that there is a lot of it. It could be even a majority because there are relatively few positive comments on the candidates Walls. So you actually go to the other candidates web page to say nice things about your own candidate.

Emotions still run high and raw

After the dent on week 41 volume is up as high as ever.

Visit etuma.com to learn more about the solution and technology behind Baramitter.

After the final debate on Oct 22, 2012, the discussion volumes on Obama’s and Romney’s facebook pages were evenly distributed. This week, the overall sentiment of comments has been slightly more negative on Romney’s facebook page.

The final debate, whose focus was foreign policy, generated more discussion regarding foreign policy and the military.

Here is a trend chart showing an increase in the relative proportion of the topic
foreign policy in Facebook discussions this week:

Here is a trend chart showing an increase in the relative proportion of the topic
military in Facebook discussions this week:

Visit etuma.com to learn more about the solution and technology behind Baramitter.

We’re having quite a busy fall here at Etuma, so none of us have had as much time as we had wanted to do in-depth analysis on the presidential candidates’ Facebook posts nor to write the Pulitzer-worthy posts we had hoped for. But there is a lot of interesting data there, and we want to make it available to you with or without our own brand of investigative journalism. So, today I decide to just throw up some of the analysis straight from our tools and let you draw your own conclusions. I picked the topic ‘Truthfulness’, because it has been quite active from the beginning and I wanted to see what kind of comments were behind it. And now, without further ado…

Weekly volumes for the topic ‘Truthfulness’ by week number

The topic peaked a couple of weeks ago, but keep in mind that the overall comment volume also peaked a couple of weeks ago.

Keyword tag cloud for topic ‘Truthfulness’

Cumulative volume and sentiment by recipient

Remember that +1 would be the result of all positive comments and -1 all negative. Also don’t forget that anyone is free to post a comment about either candidate on both Facebook walls. So, it is difficult to know if we can draw some conclusions from the fact that one candidate’s wall has a slightly more negative sentiment than the other. But it is clear that this is a hot topic on both sides of the electorate.

We’ll be back soon with more analysis and interesting findings, but in the meantime, let us know what you would like to see. Is there any specific topic you’ve wondered about? Post your suggestions in the comments section or tweet @baramitter.

Visit etuma.com to learn more about the solution and technology behind Baramitter.

Facebook Comment Volume is Decreasing Drastically

First, interestingly (hard to avoid that word and actually probably not so interestingly) the Facebook comment volume is going down. Last week there was a 35% reduction from the peak week. Does this mean that people are running out of interest or have moved to other venues? Because at the end of the day it is hard to be heard among over 100 000 comments per week. Nobody reads them; it is clear from reading hundreds of comments that it is about ranting and raving. There are extremely few feeble tries for some kind of objective dialogue and they always fail miserably because these polarized people shoot them down. Even worse, they are not read by anybody because everybody is so busy writing their own strongly biased comments. In short, there is no dialogue, and if there is no dialogue people get tired of screaming alone. While going out and attending rallies, they can at least scream among other like-minded people.

And the Interest in Energy and Environment is Also Going Down

Today I am going to focus in Energy and Environment. I picked out Topics Energy and Environment, Environment, Alternative Energy. Here are the volumes:

Interest in this topic is going down even faster than the general trend (more in upcoming blogs on what is replacing it). Here is the contextual tag cloud (on Topic Energy and Environment).

These are the things people were talking about. Here are the words that were mentioned more than 100 times. Overall there were 51 words, phrases and names “mapped” to Energy and Environment in this data: energy 2635, earth 1798, gas 1577, coal 1532, planet 1197, oil company 390, soil 376, fuel 204, gasoline 176, petroleum 148, ocean 141. Overall our system found 51 words in the conceptual Energy and Environment -basket that we call Topics in this data.

Both candidates had an amazingly similar number of discussions about Energy and Environment on their FB-pages. In this chart you can find the volume on the left and the relative sentiment of all comments on the bottom. -1 means all comments are negative and +1 that all are positive. Many Facebook comments are actually statements, with no strong value statement, which can be seen in the volume charts as the grey area. But these comments are clearly on the negative side but not as far as one would suspect. There were lot of comments in the data saying that there are no Environmental problems, which in Etuma360 analysis means a positive thing.

Visit etuma.com to learn more about the solution and technology behind Baramitter.

Photo illustration by Nick Baumann ; original photo by Krivosheev Vitaly/Shutterstock

There has been a lot of talk about political action committees, also known as Super PACs, and how they affect the presidential election. They often run TV ads that promote or bash candidates but also have other ways of influencing voters. The facebook pages have a lot of comments that bash the candidates’ political agendas, intentions or what they have done in the past. Browsing the Facebook page comments, I encountered a new use for the acronym OMG: Obama Must Go.  Therefore this post will discuss Obama’s mistakes according to the “Facebook Super PACs”.

Obama & Terrorism – 188 hits

Quite a big surprise is that Barack Obama is accused of mistakes when it comes to terrorism. Some claim that he is a terrorist or loves terrorists. Also, there are allegations that passing the law known as HR 1388 has given $20 million to Hamas in Palestine.

Golf – 500 hits

Obama is the first US president that is like a rockstar in sharing his personal life with the public on a daily basis. Many comments mention golf, basketball and Hollywood parties as things that Barack Obama prefers and should do when he is not re-elected. One person had even calculated how many days of work Obama has lost while playing golf. Most likely the bashing comments are a result of his sometimes rock star-like appearance to the public.

Next week we will take a look at what the Facebook Super PACs have to say about Mitt Romney. Stay tuned!

Visit etuma.com to learn more about the solution and technology behind Baramitter.

Today, I am solely going to focus on the fact that Mitt Romney’s name gets mentioned almost three times as often as Barack Obama’s. At first, I thought it was because Obama is referred to as “the President”, but this is only a part of the story.

It is curious that Obama’s first name does not really get used at all: Mitt Romney is “mitt” (yes, often with non-caps), but Barack Obama is “President Obama”. Media “experts” are saying that people can’t relate to Mitt Romney, that Barack Obama is more likeable. But knowing that people tend to use first names and nicknames only for those with whom they are most comfortable, this data would indicate otherwise.

Tell us your ideas and views on the affection behind the names we use for people. Does Barack Obama’s foreign-sounding name make people somewhat uneasy or at least not as attached as the attractive, even homey “Mitt”? (Of course, a mitt also happens to be a cozy winter accessory, a glove: our system first categorized the word “mitt” under the topic Accessories, but after a quick fine-tuning, it fell into the correct NER process.)

Visit etuma.com to learn more about the solution and technology behind Baramitter.

I don’t know much about what happened in the debate, because I decided not to watch it before looking at the data. The data from the two days after the debate shows clearly that Mitt Romney’s performance was seen much more positively than Barack Obama’s: there were more than twice as many positive comments on Romney’s page than Obama’s categorized under the topic Debate. This is a major change from what we have seen up to now, which is shockingly similar volume and sentiment levels on the two pages.

Interesting to note is that most negative comments on Barack Obama’s page seemed to be criticism from Romney supporters, based on the non-constructive and derogatory nature of the comments. The messages were relatively black and white, and it was difficult to draw much constructive information from them except that it is clear that this electorate is highly polarized.

This is what I did with today’s data: I filtered everything out except those comments that were categorized under the Topic Debate during the past two days. From this filtered data, it is clear that the impact of the debate on Mitt Romney’s Facebook wall was much more powerful than on Obama’s. Also of note was the fact that the discussion topics were quite different on each page and that Mitt Romney’s name was mentioned much more than Obama’s. In fact, Mitt Romney’s name was often mentioned on both walls.

Here is Mitt Romney’s Debate Analysis:

And Obama’s:

ON ANALYZING FACEBOOK PAGES

What is crucial to understand, and this is one of the biggest challenges in analyzing Facebook comments, is that both candidates’ supporters write on both pages. Because of this, you cannot draw strong overall sentiment conclusions about either candidate. What this means in practice is that Barack Obama’s pages have negative comments about not only Obama but also Romney and vice versa. The “cross discussion” is so widely conducted that it is dangerous to draw strong conclusions based solely on the analysis of either candidate’s Facebook page.

Visit etuma.com to learn more about the solution and technology behind Baramitter.

I guess that the headline is not big news. But what is interesting is how quickly – the next day – it dominated discussions. What is also worth noting is how quickly the discussion returned to its normal level.

But before we move forward, let me explain what a discussion Topic is in our system: Topics are sort of conceptual baskets used to gather the thousands of keywords from discussions into a manageable number of units for analysis. Each of our industry-specfic ontology segments has about 1,700 Topics.

One of the topics discussed in both candidates’ pages is Religion and Faith. Here is a contextual tag cloud that shows the words and concepts that are “mapped” to this topic.

The frequently mentioned keywords (frequency) were god (10574), religion (2881), Islam (2637), faith (1497), Mormon (1474), belief (1259) and bible (1047). There were also forty-two other words related to religion, each used at least thirty times.

The topic Religion and Faith was discussed about eight percent more often on Barack Obama’s Facebook wall, but such a difference in this type of system falls within the margin of error. In topic identification, the margin of error is typically about ten percent and sentiment detection a bit higher. The sentiment breakdown (i.e., level positivity or negativity) was approximately the same for each candidate.

But what was most interesting about this topic was the jump in volume during week 37 (September 10-16). The U.S. ambassador to Libya was killed on September 11, and after that, discussion about religion exploded on both walls. The following week was still at an elevated level, but it returned to w36 levels in about two weeks.

On September 12, one day after the Libya attack, Religion and Faith became the highest-volume discussion topic.

Visit etuma.com to learn more about the solution and technology behind Baramitter.