Tag Archives: predictions

Gang “homicides spread like infectious disease”; other homicides do not

Posted on by

A new study adds to the social network analysis of gang activity by comparing clusters of gang homicides to other kinds of homicides:

Using police data from Newark, New Jersey, Zeoli and fellow MSU researchers Sue Grady, Jesenia Pizarro and Chris Melde were the first to show, in 2012, that homicide spreads like infectious disease. Similar to the flu, homicide needs a susceptible population, an infectious agent and a vector to spread. (The infectious agent could be the code of the street – i.e., guarding one’s respect at all cost, including by resorting to violence – while the vector could be word of mouth or other publicity, Zeoli said.)With the new study, the interdisciplinary team of researchers analyzed the Newark data to gauge whether specific types of homicide cluster and spread differently. In addition to gang-related murders, the researchers looked at homicide motives such as robbery, revenge, domestic violence and drugs. These other motive types were not directly connected to gang participation.

The study found that the various homicide types do, in fact, show different patterns. Homicides stemming from domestic violence and robberies, for example, show no signs of clustering or spreading out.

Gang-related killings were the only type of homicide that spread in a systematic pattern. Specifically, there were four contiguous clusters of gang-related homicides that started in central Newark and moved roughly clockwise from July 2002 through December 2005.

Such findings, adding to previous research showing a relatively small cluster of gang members in a big city can be responsible for a large number of homicides, should help lead to better prevention and policing efforts. All homicide is not alike as the root causes and people involved can differ.

Two other things are interesting about this coverage:

1. The medical analogy – an infectious disease that needs to be cured – is likely to be appealing to a broad number of people. This might work better than the rhetoric of needing to find the killers and lock them up.

2. The headline of the story is “Can sociology predict gang killings?” and one quote in the story might provide evidence for this: “Taken together, this provides one piece of the puzzle that may allow us to start forecasting where homicide is going to be the worst – and that may be preceded in large part by changes in gang networks.” However, forecasting where homicides are more likely to happen is not exactly the same as predicting gang killings.

Predicting riots using social media

Posted on by

In addition to the identified factors from research coming out of the 1960s and 1970s, one sociologist suggests social media activity can show how riots and protests spread:

The most promising method of “predicting” unrest might be through social media. Dan Braha, a professor at the University of Massachusetts and affiliate of the New England Complex Systems Institute, has studied unrest in hundreds of countries and the phenomenon of “contagion,” or how it spreads. In the past, printed newspapers, televisions, and other media played an important role, he said. “Today, the use of Twitter, Facebook, and other social media platforms is fundamental to the rapid self-organization and spreading of unrest activities—much like the spread of fire in a forest.” And the data from these media can be tracked. Riots, he claims, are certainly foreseeable, but “prediction regarding ‘when’ and ‘where’ becomes more precise on short time scales.”

It sounds like social media is just part of the puzzle here. There are certain underlying conditions mentioned in this article – such as hot weather or precipitating incidents (such as police violence) – but these do not always lead to riots. (In fact, given the inequalities present in many American cities, riots and protests could be considered relatively rare.)  Just as with the analysis of the Arab Spring activity, social media does not cause protests or riots but it can help facilitate it. This was reported in Egypt as protestors shared information through social media and even peer-to-peer options. This was also reported in Baltimore as protestors selected places to show up. This is not a new phenomena; riots in the 1960s spread in a contagion like manner and the dispersion could be tracked through news coverage in the New York Times. But, the availability of social media now makes it theoretically possible to watch things develop in real time, an advantage for both protestors and authorities.

Thinking to the future, what happens when protestors make use of non-public social media or peer-to-peer options that cannot be viewed by authorities?

“Using a Real Life SimCity to Design a Massive Development”

Posted on by

As a massive SimCity fan, I find this use of predictive urban models intriguing:

596 acres, 50,000 residents, $4 billion dollars and even a 1,500-boat marina: Everything about the proposed Chicago Lakeside Development, developer Dan McCaffery’s massive micro-city being built at the former site of the U.S. Steel Southworks Plant, is on a different scale. It follows that the design process for this mixed-use project requires a different set of tools, in this case, LakeSim, an advanced computer modeling program. Developed as part of a collaboration between the University of Chicago, Argonne National Laboratory, Skidmore, Owings & Merrill and McCaffery Interests, this program functions like a customized SimCity, analyzing and simulating weather, traffic patterns and energy usage to help architects and designers plan for a site that may eventually contain more than 500 buildings.

“A lot of the Big Data approaches tend to be statistical in nature, looking at past data,” says Argonne scientist Jonathan Ozik. “We’re modeling a complex system of interactive components, running the data forward, so what we end up having is your SimCity analogy, energy systems interacting, vehicles and people moving. What we’re doing here is using a complex systems approach to tackle the problem.”…

The challenge for planners is predicting how so many different systems and variables will interact. LakeSim gives them a framework to analyze these systems over long timelines and run millions of scenarios much quicker than past models — hours as opposed to days — asking “hundreds of questions at once,” according to Ozik. The program is a step forward from similar modeling software, especially valuable at a site that in most respects is being built from scratch.

This seems quite useful at this point but it will be necessary to look at this down the road once the site is developed. How much time did the model save? How accurate was the model? Did relying on such a model lead to negative outcomes? If this is a predictive model, it may be only as good as the outcome.

Interesting to note that the commenters at the bottom are wondering where all the people to live in this development are going to come from. I assume that demand is appropriately accounted for in the model?

LA to have the worst traffic on Thanksgiving getaway day

Posted on by

Add in more drivers this Thanksgiving and one firm says Los Angeles will have the worst getaway traffic on Wednesday:

Trips on Wednesday—deceptively nicknamed “getaway day”—are expected to take 36 percent longer on average, with peak times from 3 to 5 in the afternoon. Perhaps it would be more efficient for everyone to just Skype their family members while eating meals at home.

A rep for Inrix tells the LA Times that leaving either before or after the late afternoon peak times might spare drivers a little road pain, but that leaving the morning of Thanksgiving would be even better. The unfortunate folks picking up at LAX on Wednesday and traveling on the 110 south can expect the worst delays on their trip between 7 and 8 a.m., with delays of up to 15 minutes, which actually just sounds like regular holiday traffic at the airport.

Some interesting info – the data we have these days is pretty amazing – but this is such a limited issue. The two hour window in Los Angeles between 3 and 5 PM is expected to be worse and some of the other cities on the list have only one hour windows. If this is the case, then people should take note and try to leave at other times. Many people may not be able to help it given plane or work schedules but the traffic doesn’t have be as bad as suggested if people use the information well.

Predicting more Thanksgiving traffic due to a closed car-to-plane gap?

Posted on by

One way to predict traffic on the roads at Thanksgiving is to look at the car-to-plane travel gap:

Drivers will make up about 89.5 percent of holiday travelers this year, a gain of 0.1 percentage point from 2013, while air passengers will drop by the same amount to 7.5, forecasts prepared by Englewood, Colorado-based IHS Inc. show. A 0.1 point increase may not seem like a lot, but based on last year’s estimate that 39.6 million people traveled by car for Thanksgiving, that would roughly equate to at least another 40,000 people piling onto America’s highways.

The car-over-plane travel choice is made easier by the fact that airfares aren’t coming down like gasoline pump prices are. While the plunge in oil has driven down wholesale jet fuel prices 17 percent since August, almost matching the 18 percent drop in retail gasoline, airfares have risen 3.4 percent over that time, data compiled by industry groups show…

“Right now the airlines aren’t in the sharing mood,” Rick Seaney, chief executive officer of the Dallas-based travel website FareCompare.com, said. “They just went through six years of multi-megamergers and dividing the country up by city with little or no competition, so they’ll pocket whatever difference they may get for a while.”

Gasoline’s drop will save the average U.S. driver about $500 annually, helping boost consumer spending, according to IHS. U.S. auto sales have risen 5.5 percent to 13.7 million in the first 10 months of 2014, on pace to be the strongest in eight years, Woodcliff Lake, New Jersey-based data provider Autodata Corp. said.

A few thoughts:

1. Having 40,000 more people on the roads at Thanksgiving is going to complicate traffic all across the United States? Spread these people cross hundreds of metropolitan areas and assume they aren’t all leaving at the same time (Wednesday after work) and adding that kind of volume may not matter much at all.

2. The prediction of future traffic is interesting to me. This reminds me of Carmageddon fears, first in Los Angeles (twice) and then in Chicago earlier this year. This seems like the creation of news: get prepared for more Thanksgiving traffic now! It is the kind of fear-based reporting done by many local news outlets about things like weather or traffic, fairly mundane events that occasionally turn out to be horrible.

3. The Carmageddon cases hint at another piece of this prediction: making such claims could change future behavior. If Americans hear that there will be more drivers at Thanksgiving, even just a few of them changing their plans (not driving or changing their departure times) might go a long ways toward relieving the predicted traffic. Perhaps this forecast is all part of some plan to actually reduce Thanksgiving traffic?

4. Just from personal observation: plane tickets appear to be really high during Thanksgiving, Christmas, and New Year’s this year. As the article notes, airlines are looking to make money and haven’t budged much in their prices even with the recent gas price drops.

 

 

2014 Democrats echo 2012 Republicans in arguing political polls are skewed

Posted on by

Apparently, this is a strategy common to both political parties: when the poll numbers aren’t in your favor on the national stage, argue that the numbers are flawed.

The [Democratic] party is stoking skepticism in the final stretch of the midterm campaign, providing a mirror image of conservative complaints in 2012 about “skewed” polls in the presidential race between President Obama and Republican Mitt Romney.

Democrats who do not want their party faithful to lose hope — particularly in a midterm election that will be largely decided on voter turnout — are taking aim at the pollsters, arguing that they are underestimating the party’s chances in November.

At the center of the storm, just as he was in 2012, is Nate Silver of fivethirtyeight.com…

This year, Democrats have been upset with Silver’s predictions that Republicans are likely to retake the Senate. Sen. Heidi Heitkamp (D-N.D.) mocked Silver at a fundraising luncheon in Seattle that was also addressed by Vice President Biden, according to a White House pool report on Thursday.

“Pollsters and polling have sort of elbowed their way to the table in terms of coverage,” Berkovitz said. “Pollsters have become high profile: They are showing up on cable TV all the time.”

This phenomenon, in turn, has led to greatly increased media coverage of the differences between polling analyses. In recent days, a public spat played out between Silver and the Princeton Election Consortium’s Sam Wang, which in turn elicited headlines such as The Daily Beast’s “Why is Nate Silver so afraid of Sam Wang?”

There are lots of good questions to ask about political polls, including looking at their sampling, the questions they ask, and how they make their projections. Yet, that doesn’t automatically mean that everything has been manipulated to lead to a certain outcome.

One way around this? Try to aggregate among various polls and projections. RealClearPolitics has a variety of polls in many races for the 2014 elections. Aggregation also helps get around the issue of celebrity where people like Nate Silver build careers on being right – until they are wrong.

At the most basic level, the argument about flawed polls is probably about turning out the base to vote. If some people won’t vote because they think their vote won’t overturn the majority, then you have to find ways to convince them that their vote still matters.

Using social media data to predict traits about users

Posted on by

Here is a summary of research that uses algorithms and “concepts from psychology and sociology” to uncover traits of social media users through what they make available:

One study in this space, published in 2013 by researchers at the University of Cambridge and their colleagues, gathered data from 60,000 Facebook users and, with their Facebook “likes” alone, predicted a wide range of personal traits. The researchers could predict attributes like a person’s gender, religion, sexual orientation, and substance use (drugs, alcohol, smoking)…

How could liking curly fries be predictive? The reasoning relies on a few insights from sociology. Imagine one of the first people to like the page happened to be smart. Once she liked it, her friends saw it. A social science concept called homophily tells us that people tend to be friends with people like themselves. Smart people tend to be friends with smart people. Liberals are friends with other liberals. Rich people hang out with other rich people…

On the first site, YouAreWhatYouLike, the algorithms will tell you about your personality. This includes openness to new ideas, extraversion and introversion, your emotional stability, your warmth or competitiveness, and your organizational levels.

The second site, Apply Magic Sauce, predicts your politics, relationship status, sexual orientation, gender, and more. You can try it on yourself, but be forewarned that the data is in a machine-readable format. You’ll be able to figure it out, but it’s not as pretty as YouAreWhatYouLike.

These aren’t the only tools that do this. AnalyzeWords leverages linguistics to discover the personality you portray on Twitter. It does not look at the topics you discuss in your tweets, but rather at things like how often you say “I” vs. “we,” how frequently you curse, and how many anxiety-related words you use. The interesting thing about this tool is that you can analyze anyone, not just yourself.

The author then goes on to say that she purges her social media accounts to not include much old content so third parties can’t use the information against them. That is one response. However, before I go do this, I would want to know a few things:

1. Just how good are these predictions? It is one thing to suggest they are 60% accurate but another to say they are 90% accurate.

2. How much data do these algorithms need to make good predictions?

3. How are social media companies responding to such moves? While I’m sure they are doing some of this themselves, what are they planning to do if someone wants to use this data in a harmful way (say, affecting people’s credit score)? Why not set limits for this now rather than after the fact?

Congressional Research Service estimates foreign-born population in 2024 of 58 million

Posted on by

One projection of the foreign-born population in the United States is that it will rise steadily to 58 million by 2024:

 

CRSForeignBornProjections2024

The full report has some interesting data on immigration as well as the methodology behind estimating some of the figures. And, the report notes that these projections assume consistent rates of growth with no policy changes, both unlikely occurrences. Predictions like these are hard to make. At the same time, this is a reminder that large flows of immigrants into the United States is not just a historical fact kids learn in history class but rather is an ongoing phenomenon.

Predicting the ongoing rapid urbanization of the South

Posted on by

The American South is known for its sprawling cities and one new model suggests this will continue in force in coming decades:

New predictions map the future spread of urban sprawl in Dixie, and it is immense. Basing their model on past growth patterns and locations of existing road networks, researchers at North Carolina State University projected the region’s expansion decades into the future. According to their forecast, the Southern urban footprint is expected to grow 101 percent to 192 percent.

The projected map in 2060:

Read the full paper here. As the discussion section notes, the model doesn’t really account for future decisions in opposition to current patterns. In other words, such a model is not deterministic: it is based on past data but communities could make decisions that continue down this path (and even intensify urban growth beyond the predictions here) or pursue different patterns of urban growth (say if New Urbanism catches on in a big way and exurban growth slows quite a bit).

Put another way, is it possible to imagine an American South that in 50 or 100 years wouldn’t be noted for its sprawl?

Are NFL fans now better off with all the draft knowledge they can access?

Posted on by

The NFL draft process has been drawn out even further this year and it leads to an interesting question: is a better-informed fan a more-in-control fan?

For many Americans, football fandom is a knowledge contest, an anxious dedication to information gathering that drives us to consume the NFL’s human-resources wing as entertainment. Last year, more than 7.9 million of us watched the draft and another 7.3 million viewed some portion of the scouting combine. This year, the draft moved from April to May, a transition attributed to a scheduling glitch: Radio City Music Hall, the draft’s venue in recent years, booked a Rockettes Easter special during the NFL’s big weekend. But it’s a favor, really: We need more time for recreational panic, more time for our 11-year-olds to prognosticate with radio hosts…

When Mayock started his work, most information about prospects was relegated to team officials and media members. But now, anyone could develop informed opinions about someone like Landry. Anyone who wants to can study six of his games and learn about his perceived value on mock draft sites. Walter Cherepinsky, the founder of one such site, tells me it gets 40 million visits per month. (One of his recent mocks has Landry going to the Carolina Panthers with the 92nd selection.) For the most committed students, there are draft guides such as Matt Waldman’s Rookie Scouting Portfolio, more than 1,200 pages about offensive prospects. Waldman writes that Landry blocks and runs routes like a reserve player, but he catches passes like an NFL star.

While the adage tells us knowledge is power, though, it’s less clear how all of this information empowers draft-obsessed fans. That 11-year-old from the sports talk show wanted his team to select a receiver, but wanting that or having an argument in favor of it won’t make it so. What erudition of this sort provides is a sense of autonomy, in terms of identity, a guard against power abused. NFL insiders tend to whisper the same general stat: that one-third of the league’s general managers have no business overseeing personnel decisions—they’re either misguided in the way they evaluate players or they don’t bother to put in the requisite research. Draft savvy, then, lets fans separate their outcomes (the success of their favored college prospects) from those of their favorite teams (the players chosen by their teams and the team’s outcome on the field); fans can timestamp their opinions and later say, “I told you so.”

But does this kind of autonomy relieve fans’ helplessness, or does it make them feel more like pawns beholden to the real draft-day outcomes they want to control but can’t? Let’s say you’re sure, after months of research, your team should use its third-round pick on a quarterback, but the team instead drafts a punter—a punter—and the quarterback selected five slots later goes on to win a Super Bowl within two seasons. Besides a conniption, this could also give you a grudge to unleash on team executives, message board commenters, and media members who disagree with your football opinions.

The evidence seems clear: the draft is popular and the NFL can afford to drag it out when people keep watching. But, do people really enjoy it? More broadly in sports, if fans know even more about potential players (college, minor leagues, developmental leagues, overseas prospects, etc.), does this lead to feeling more in control?

Having more information is generally seen as a good thing in today’s world. The more input you can gather, the better. Yet, this doesn’t necessarily lead to better outcomes or more perceived control. (Read The Paradox of Choice for a good introduction.) I would argue that much of the appeal of sports is the unpredictably, the odd things that can happen on a playing surface at any point. All the information in the world can’t easily explain some of these events – and would we want it to or would we rather see unpredictable things happen in games?

The draft is a good example of this unpredictability and how we might perceive information as a way to limit this. Think about all of the mock drafts. All of the talking heads. Stretching out the draft even longer. Yet, there are still things that happen on draft day that are hard to predict, even for all the experts. (I’m particularly intrigued by recent mock drafts that incorporate more complicated draft-day trades.) Assessing the results of drafts can take years or even decades. Sports Illustrated had a recent story about the Tampa Bay Buccaneers making a disastrous pick in the 1980s that led to 10+ years of ineptitude – but this wasn’t visible for years.

All together, football players make choices, teams make choices, fans respond to all of this with more or less information, and it all collides in a “sports experience.” I suspect sports fans don’t really want to know everything (stronger predictive abilities would reduce the uncertainty about outcomes) even if they often want to immerse themselves in the sports experience. At some point, the return on having more and more sports knowledge likely decreases enjoyment though this curve could easily differ by person.