Trends and Outliers
TIBCO Spotfire's Business Intelligence Blog
2012
Is Social Analytics Better at Tracking Disease?
There’s evidence that real-time data from social media is a better source for tracking disease outbreaks when compared to traditional data sources.
For instance, two years after the devastating earthquake in Haiti, a new study shows that Twitter and HealthMap (a project at Children’s Hospital in Boston that creates a worldwide data visualization of disease outbreaks on an interactive map in nine languages) tracked the cholera epidemic, which killed 7,000 people and affected over 500,000, up to two weeks faster than traditional reporting methods.
The research team led by Dr. Rumi Chunara, a research fellow at HealthMap and Harvard Medical School, had a simple goal – to determine if social media analytics could tell us more about epidemic disease faster than traditional reporting.
Continue reading »
2012
Is Sentiment Analysis a Subset of Text Analytics?
Typically, industry analysts lump sentiment analysis and text analytics together, particularly when they talk about how to find value from social media conversations.
But is that the right way to view these two technologies? Is sentiment analysis a component of text analytics or is it an application on its own? And either way, what about the human element?
Recently, analytics visionary Seth Grimes (@sethgrimes) indicated that sentiment analysis draws on, but isn’t a subset of, text analytics. “Strong sentiment analysis relies on semantic analysis — on application of natural-language processing (NLP) techniques to identify sentiment objects (entities, topics, and concepts), opinion holders, and the sentiment, attitudes, and emotions that the opinion holders attach to the sentiment objects. But sentiment can also be inferred without any semantic analysis. For instance, the Consumer Confidence Index (CCI) and the Michigan Consumer Sentiment Index.”
In a post earlier this week, Grimes continued to break down why sentiment analysis doesn’t depend on text analytics.
Continue reading »
2012
Analytics is Key to Energy Industry
Since energy was a top issue in last night’s State of Union address, we wanted to emphasize that analytics will play a key role. As deregulation in various states helps to open up competition, energy companies have tremendous opportunities to expand their customer and revenue bases. On the flip side, deregulation also ushers in competition for many companies that have enjoyed monopolies or near monopolies and suddenly find themselves competing to retain the same captive customers they’ve had for years.
Energy companies in states where deregulation has occurred can use analytics to better understand and target customer needs in a number of ways. For instance, decision makers and data scientists at energy companies can use analytics to better understand how residential, commercial, and industrial customers are using energy and to help tailor rate offers geared to meet their specific needs. For instance, some manufacturers that operate during the graveyard shift (11 p.m. to 7 a.m.) could be offered more attractive rates to power their machinery.
Continue reading »
Filed under: Big Data, Data Analytics, Data Scientist, Energy
2012
Collaborative Analytics and the Benefits of Local Language Support
Enterprise companies rely on data scientists from a range of geographies and backgrounds. To help keep analytics teams on the same page, it’s useful for analytics team members to use the same analytics tools. This can and should include tools that offer local language support for data scientists who speak Chinese, Russian, Japanese, etc.
The ability for companies to build and draw upon geographically-dispersed collaborative analytics teams has become essential to business success, if not survival. A Frost & Sullivan report reveals that collaboration is a cornerstone of business performance. For instance, 36% of a company’s business performance is tied to its “collaboration index,” according to the report. By comparison, this is more than two times the impact of a company’s strategic direction (16%) and more than five times the impact of market and technological turbulence influences (7%).
Continue reading »
2012
The Data Analytics of the NFL Playoffs
Saints fans are sad. Everyone is pondering Tim Tebow’s 2012 fate after he defied the odds of making it to the playoffs. The Packers and Aaron Rogers did not make it to back-to-back Super Bowls.
Tom Brady is obviously a machine, but his 55.9 completion percentage against the Ravens in five career games is his lowest against any other team, writes Jeff Reynolds of The Sports Xchange in a USA Today article. Will this matter this weekend?
Eli Manning’s elite standing as a quarterback is still making waves in the sports gossip circles. San Francisco’s offense, led by quarterback Alex Smith, got hot in the postseason. The Giants’ defense also came alive.
As we move into the conference championships, we can note that the statistics may be used to determine polls, odds and the Pro Bowl contenders, but they can’t always predict the games. Especially in the playoffs.
Before we move into more of the data analytics of the NFL playoffs, I have to issue a little disclosure and give you a fun stat about this NFL postseason.
I went to college with New York Giants quarterback Eli Manning (we weren’t close friends, but we did graduate at the same time). Now for the fun stat. Manning and four other former University of Mississippi players are starters on the four teams playing in this weekend’s conference championships. Six other colleges nationwide can also claim between three and five players on each of the championship teams. Now, on to why this postseason is an anomaly in more ways than one.
Do Stats Really Matter in the NFL Outcomes?
According to Sam Farmer, a sports writer for the Los Angeles Times, this year is different because stats don’t tell the whole story. He writes: “In an unusual twist, these playoffs follow a noteworthy pattern: Most of the NFC participants have top offenses; most of the AFC teams are more defense-minded.”
And he says this year’s Super Bowl may give us the answer to the burning question: “What’s better – a great offense or defense?”
Jeff MacGregor, a senior writer for ESPN.com, made a good point this past week: “If stats, science, and analysts really predicted the outcome, would we watch?” His piece centered on quarterbacks and the fact that no one predicted the “Year of the Quarterback” would have Eli Manning beating Aaron Rodgers.
He brings us back to an observation we’ve made before on the Spotfire blog – the human element. Without that, analytics can’t tell the whole story. And would we really want them to? To this point, MacGregor writes:
“The propositional knowledge of 21st century football is now so incredibly complex it’s impossible to predict the outcome of any single game using only statistics. The numbers just don’t mean much. Too many people and too much rage and too much chaos to account for. The sample size is too small and the stage is too big and the ball is too pointed and too much depends upon momentum and bad chance.”
This begs the question – should we just leave the analytics to our fantasy football teams?
The Business of America’s Obsession with Football
While this postseason shows how hard it is to predict game outcomes and who will make it to the final game, no one can deny the impact of the game on the economy. This infographic from IBM shows the data analytics of the impact a lack of the 2012 season would have had if there had been a NFL lockout in 2011.
Next Steps: Tweet us your predictions for the conference champions this weekend.
Amanda Brandon
Spotfire Blogging Team
Filed under: infographic, Unconventional Uses
2012
The Business Value of Uncovering the “Unknowns” in Data
One of the greatest benefits of using analytics is that it can help decision makers and data scientists spot new trends or anomalies in data that, in turn, can generate new business opportunities or help mitigate risk. Or, stated another way, using analytics to spot the “unknowns” in data.
This is something that the healthcare industry has been doing for years. For instance, researchers at Children’s Hospital Boston recently created a new method for assessing and monitoring drug safety that combines multiple forms of widely-available data to help predict adverse drug reactions.
Continue reading »
Filed under: Advanced Analytics, Big Data, Data Scientist, Social Analytics
2012
As Chief Yahoo Resigns, Can Big Data Save Stumbling Company?
Yesterday, Yahoo co-founder Jerry Yang - aka the “Chief Yahoo” – resigned from the board of directors and all other positions at the troubled company. This comes on the tail of Yahoo recently naming a new CEO, Scott Thompson. What may save this stumbling company is the new CEO’s belief in big data. So much so that he says big data is the “key to Yahoo’s long-term health.” In fact, Thompson is counting on big data and analytics to help Yahoo compete with Facebook and Google.
For Thompson, formerly the president of PayPal, making use of Yahoo’s data is crucial to building the best products for consumers and advertisers. In his interview with AdAge, Thompson explains that PayPal was able to harness its data to create “an unbelievably compelling business because the company used data to understand risk and fraud better than anyone on earth.”
Continue reading »
2012
Data Scientists Gain Business Value from Data Mashups
Data scientists who do mashups from disparate data sources are able to gain deeper insights into problems and develop more comprehensive approaches toward attacking them by having a richer data set to work from. In turn, data mashups can help companies become more agile and speed time to market.
Consider the use of customer data. Customers interact with companies using a wide range of channels, including online, email, mobile, voice, SMS, and social. Companies collect heaps of data from customer interactions, including structured data amassed through online surveys, unstructured data such as sentiments that are shared in social media, as well as preferences and needs that are captured in recorded contact center discussions.
Continue reading »
2012
Applying Big Data to 2011 Holiday Shopping Information
By most measures, 2011 holiday sales saw solid gains. Overall retail sales rose 4.1%, according to the National Retail Federation. And while 2011 holiday sales didn’t achieve the 5.2% gains made during the 2010 holiday season, they still outpaced the 2.6% growth over the past decade.
But a closer examination shows that not all categories fared well. For example, sales of “mature” consumer electronics devices dipped 5.9% during the Nov. 20 to Dec. 24 sales period, according to The NPD Group. Yet not all consumer electronics suffered. For example, 3DTV sales grew more than 100%, according to The NPD Group.
Continue reading »
2012
The Data Analytics of the 2012 Election . . . So Far
This week, Mitt Romney became the front-runner in the 2012 Republican nomination bid for the White House with his “unprecedented wins” in Iowa and New Hampshire. According to a CNN Report, this was the first time a non-incumbent Republican candidate won in both states.
The Iowa caucuses came down to just eight votes between the former Massachusetts governor and Rick Santorum, a former Pennsylvania senator.
The CNN report showed that with 95% of precincts reporting, Romney led with 40% of the vote in New Hampshire, followed by Rep. Ron Paul (R-Texas) who received 23%. Jon Huntsman, the former Utah governor, received 17% of the New Hampshire votes for third place. Santorum placed fifth.
Continue reading »
Recent Comments