As long as we're on the subject of NPR, I thought I'd share another networks story I heard this morning on NPR. Researchers at Harvard Medical School and UC San Diego recently completed a study on the nature of happiness is social networks and found that happiness tends to be contagious. The authors found that people with happy friends tended to be happier, thus creating happy social clusters. In fact, the researchers found that a person who is close to someone who becomes happy has a 25% chance of becoming happier themselves. Additionally, the authors found that happiness tends to be more contagious within social networks than unhappiness is. You can see the social network of happiness clusters on the NPR website.
Image of the happiness network from CNN story. The nodes are colored for average mood. Yellow is happy, blue is sad, and green is in-between.
I know that most of you are on holiday but this is worth posting on the blog. During my trip to NYC last week, I was listening to NPR which happened to come across an interview with Mark Newman, a co-author of the book The Atlas of the Real World: Mapping the Way We Live. Newman says that the regular map that we see is not the "real world" representation. He was quoted:
"Maps can be misleading, absolutely...Your standard map of the world makes the North Pole look huge and the equator look very small. And we just accept it the way it is."
The book could be used for so many reason's including, which part of the world has more health expenditure versus which accumulate disease which kill millions of people, importers and exporters of automobiles, population densities and so on. All this things might have been already represented in different maps but Newman is applying to the whole globe which will help us see the world in different perspectives. So the largest countries that we know of could be invisible and vise verse:
The reality is that New York has more than 10 times the number of electoral votes, because its population is so much bigger. "If you just counted the amount of color, you might think the Republican Party won by a landslide," Newman tells NPR's Andrea Seabrook. "The way we do it is we change the sizes of the states to represent how many people are living in each one."
The paper “Complex food webs prevent competitive exclusion among producer species” looks at the effects that nutrient supply and predation have on the survival of multiple producer species. Brose did this by randomly assigning nutrient intake efficiency to five producers. The same producers were also put into a food web formulated based on the niche model. Then using a nutrient intake model that included two limiting nutrients a simulation was run to see what would happen to the network with just producers and what would happen to the food web including producers and consumers. There were also food web models that differed in terms of predators. Some webs only had predators that were generalist so they ate the producers in proportion to their biomass and other webs where predators are given random preference for a particular producer.
After the simulations were compared it was found that 99.7% of the time in the networks with just producers only one producer dominated and all others became extinct. This extinction is thought to be caused by competition between producers where the most efficient nutrient obtainer eventually excludes all other species. When the food webs with predators are simulated then a majority of the time more producers survive than in webs without predators. 86.7% of simulations with predators that had random preferences had more producers than number of limiting nutrients and 91.9% of simulations with predators, with random food preferences, had more producers than the number of limiting nutrients. This suggests that there is big reduction on the influence of competition when there are predators but that without predators, producers will engage in competition and the dominant competitors will drive the other producers to extinction.
As mentioned in the article these networks have assumptions built in or effects not taken into account. Although not mentioned the model also does not take the environment or disturbance. (It has been proposed by Connell (1980) that if there is intermediate harshness in environmental conditions the effectiveness of predators would be reduced allowing completion to occur. It has also been proposed in ecology that disturbances could allow inferior competitors to exist.) However the point of this model seems to have been to illustrate that just predation by itself can limit competition and it does illustrate the general concept of predation influencing competition.
I read another paper, “Compilation and Network Analyses of Cambrian Food Webs” , looking at food webs as networks. This one was comparing modern food webs to those derived from fossils during the Cambrian period (time period between about 542 to 488 million years ago). It analyzed S (taxa), C (connectance), L/S (links per species) and 17 other separate components of networks to try and see if there were any differences between ancient and more current food webs and what might be learned from these differences. Dunne et al. analyzed 17 features including the fraction of species that are herbivores, the fraction of species without consumers and the mean length of food chains.
Dunne et al. tried to make sure that the representation of the Cambrian food web was as realistic as possible by ranking the surety of the links between nodes from 3 (highest) to 1 (lowest). They then tested their results to make sure the lack of certainty about connections in the food web did not skew their results. They did this by removing the 10, 25, 50,75 and 100 percent of number 1 links from the food web and then removed the same number of links randomly from the original food web and found the 17 conditions, S, C, and L/S for both these networks (each reduced food web was taken 100 times except the 100 percent removal of # 1 links because it would be the same each time and conditions were averaged) . When the two networks conditions were compared Dunne at al. found very little difference between the two results suggesting that the Cambrian food web is not being skewed by lower certainty links.
Of all the 17 components found, only a few proved to vary between the Cambrian food webs and modern food webs. The Cambrian food webs were found to have higher variability in total species links than modern webs. It is possible that this is due to some species having a large amount of predators. Since it is thought that the Cambrian food webs were possibly transitioning into more modern-like food webs during this period Dunne at al. suggest that the large amount of predators could be because certain species had not adapted to predation and that given time they would either become extinct or evolve defenses against the predators. One of the two ancient webs also differed from the other food webs in having a longer mean shortest path and a higher amount of loops. The longer mean shortest path length suggests a higher amount of separation between species and therefore a lack of influence between species. A high amount of loops in a food web are often thought to make the web less stable and therefore not to last very long. Both these conditions suggest that this food web is unstable and may be transitioning into a more modern-like food web.
Some of the methods performed here were a bit above my knowledge. Dunne et al. seem to have tested the assumptions in connecting the Cambrian food web quite thoroughly and the value of this model lies not in being 100% accurate but more in showing general trends. It would be interesting to see what the same analysis done on an older food web, thought to be before the transitions stage of the Cambrian food network, would look like compared to modern networks. It may be that there is not a food web available for this comparison but if there was it might show a different structure than the transitioning food web of Cambrian and the modern food webs.
I was not planning on blogging at the end of week ten but this article in the New York Times caught my eye. It discusses the Netflix challenge to create an algorithm to improve their recommendation algorithm. The part that I found particularly interesting is that certain movies are really hard to classify. Bertoni, the computer programmer in the article, says that his algorithm is really accurate for the fast majority of movies but there are a few movies that are really hard to predict. Napoleon Dynamite for example. It is one of those movies that people either love or hate and it is hard to say why. He says that other polarizing movies such as Lost in Translation, Fahrenheit 9/11, and Kill Bill are also hard to predict. The difficulty of predicting specific movies and relative ease of predicting others adds an interesting dynamic to the network of movie preferences.
There is an interesting article in today's NYTimes that outlines how Google is using search queries to model flu outbreaks. Google Flu Trends watches for key search terms that could indicate someone has the flu: thermometer, flu symptoms, muscle aches, congestion, etc. By knowing when and were the searches originate they can model the spread of the flu.
They have tested their models against Center for Disease Control (CDC) data--they claim they can see the start of a flu strain 7-10 days before the CDC. Google has also tested their historical data against the CDC's and found very high correlations. There is a study in the works that will be published in Nature.
From the article:
“This seems like a really clever way of using data that is created unintentionally by the users of Google to see patterns in the world that would otherwise be invisible,” said Thomas Malone, a professor at the M.I.T Sloan School of Management. “I think we are just scratching the surface of what’s possible with collective intelligence.”
Oh, and our friend Hal Varian is quoted in the piece.
It seems obvious that diseases such as colds and STDs are spread through social networks, but it now seems that obesity may also be. The Framingham Heart Study, a longitudinal study looking at health, specifically weight, among people and their friends, shows that obesity tends to cluster in social networks.
It does not seem surprising that obese people may tend to be friends with other obese people, but the Framingham study shows that if someone becomes obese in a given time interval, their friends have a greatly increased chance of also becoming obese. Possible reasons for this could be that friends may simultaneously adopt similar lifestyle choices such as diet, exercise or smoking, which could effect their weight. There could also be changing attitudes towards weight that could spread through social networks and make one more or less inclined to gain/lose weight because of social pressures. Similar effects were seen among siblings, where if one gained weight, another sibling is also likely to gain weight.
The effects both among siblings and friends are strongest among same sex relationships. They also depend primarily on social distance, ie. closer friends have stronger impacts, and the effects appear to be independent of geographic distance.
Another article on the spread of obesity through social networks, looks at how obesity, and other diseases, should be looked at as a hierarchy of networks (see figure below). On the bottom, there are all of the molecular networks such as gene regulation, protein interaction and metabolic networks which can be studied to look at the molecular basis of disease. Then the paper discusses a disease network where diseases such as obesity are also linked to other diseases such as heart disease and diabetes. Diseases could be connected because they frequently co-occur, one leads to another, they involve similar molecular components or other types of connections. Lastly, there is the social network. Diseases such as the flu or HIV could be passed directly through a social network or it could be a transfer of attitudes or lifestyle. The relation between the networks adds an interesting angle to the subject. For example, genetics. One's genetics most obviously effect the molecular aspects of disease, but the social network has a big impact on genetics because within an obesity cluster, genetic predispositions for obesity are likely to be passed on to children from both parents. The combined factors of genetics, lifestyle and social pressures could make these children highly prone to obesity.
The interacting levels of networks may also lead to new strategies for treating obesity. The Framingham study emphasizes that their findings suggest that the social network provides an important resource in treating obesity as well and that perhaps social support groups could be a helpful strategy. It seems so human ecological to say it, but issues like obesity and other public health problems, need to be addressed on many levels, and the hierarchy of networks provides a model for how those levels interact.
This blog is associated with "Theory and Applications of Complex Systems," a class taught at College of the Atlantic in Fall 2008. The instructor of the course is Dave Feldman. Students will be posting content throughout fall term.