Data analysis revealing behavior patterns

Advanced Systems Analysis (ASA) Program researchers apply contemporary approaches to analyze newly available data sets and find new insights and stylized facts.

© Hemeroskopion | Dreamstime © (c) Hemeroskopion | Dreamstime.com

© Hemeroskopion | Dreamstime

In 2014 ASA researchers published the results of data analysis from a massive on-line multi-player game Pardus; the data contained complete information on the social network of players, as well as their features.

Studying human behavior in virtual environments provides extraordinary opportunities for a quantitative analysis of social phenomena with levels of accuracy that approach those of the natural sciences. Researchers investigated the data set and found that the social network involved in the game had a hierarchically three-layered nested structure which appears to be very typical for human societies in general, given that it was found in hunter-gatherer societies [1] [2].

In the latter, a special algorithm, named the “generalized K-core algorithm,” was developed to identify so called “elite” subgroups in a social network, that is, subgroups of individuals with the ability and means to influence, lead, govern, and shape societies. It was revealed that members of elites are not only well-connected individuals (who can naturally impose their influence on many and quickly gather, process, and spread information), but also intermediaries connecting hubs to form a cohesive and structured elite subgroup at the core of a social network. Having found that the wealth distribution of players to be similar to that of Western countries, researchers investigated the relations between wealth of individuals and their local network positions [3]. This revealed that wealthier people have a higher degree of connections and lower clustering coefficients, while their nearest neighbors have relatively lower degrees of connections; individuals not organized within social groups are significantly poorer on average. Gender- and age-related differences in the behavior of individuals were studied by [4]. It was found that while males and females act similarly when performing positive actions, females are slightly faster for negative actions; more experienced (i.e., older) players are generally faster at making decisions about engaging in and terminating enmity and friendship.

Researchers studied the data set containing information on all Austrian patients suffering from diabetes in order to reveal determinants of diabetes; they found evidence of a negative impact of malnutrition during early life on metabolism in older age [5].

A data set of user-generated music taxonomy measuring the “instrumentational complexity” of each piece of music was studied by [6]. It was shown that album sales of a given style typically increase with decreasing instrumentational complexity, (i.e., music becomes increasingly formulaic in terms of instrumentation once commercial or mainstream success sets in).

Researchers studied data on economic and ICT development of countries and revealed bi-polarization between ICT growing economies and ICT advanced economies [7]. While the former enjoy a virtuous cycle between ICT advancement and productivity increase, the later have fallen into the trap of a vicious cycle between ICT advancement and productivity decrease.

References

[1] Fuchs B, Sornette D, Thurner S (2014). Fractal multi-level organisation of human groups in a virtual world. Scientific Reports, 4:6526

[2] Corominas-Murtra B, Fuchs B, Thurner S (2014). Detection of the elite structure in a virtual multiplex social system by means of a generalized K-core, PLoS ONE, 9(12):e112606

[3] Fuchs B, Thurner S (2014). Behavioral and network origins of wealth inequality: insights from a virtual world. PLoS ONE, 9(8):103503

[4] Mryglod O, Fuchs B, Szell M, Holovatch Y, Thurner S (2015). Interevent time distributions of human multi-level activity in a virtual world. Physica A: Statistical Mechanics and its Applications, 419:681-690 (1 February 2015) (Published online 22 October 2014)

[5] Klimek P, Leitner M, Kautzky-Willer A, Thurner S (2014). Effect of fetal and infant malnutrition on metabolism in older age. Gerontology, 60(6):502-507

[6] Percino G, Klimek P, Thurner S (2014). Instrumentational complexity of music genres and why simplicity sells. arXiv:1405.5057 [physics.soc-ph]

[7] Watanabe C, Naveed K, Zhao W (2014). New paradigm of ICT productivity - Increasing role of un-captured GDP and growing anger of consumers. Technology in Society, Article in press (Published online 4 December 2014)


Print this page

Last edited: 02 April 2015

CONTACT DETAILS

Elena Rovenskaya

Program Director and Principal Research Scholar Advancing Systems Analysis Program

Further information

Events

Staff

International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313