|
Here are some research problems we are interested in, for further examples and list of publications visit the page of the members. | ||
Word dynamics in online groups |
||
E. G. Altmann, J. B. Pierrehumbert, and A. E. Motter, "Niche as a determinant of word fate in online groups" Pre-print: [arXiv:1009.3321], (2010). |
There is a growing need for simple quantitative models of individual and collective behavior able to, e.g., describe statistical features of Internet activity, predict popularity of online contents, and understand statiscal properties of language. A driving force in this very active and multi-disciplinary field is the availability of new data, which allow us to address with an unprecedent quantitative precision old and new questions, stimulating the development of theoretical models. Natural language is a unique human activity that allows us to trace and quantify ideas and behavior. In the work mentioned in the left we have investigated the statistical properties of words in online discussion groups. We propose a new set of statistical measures, the dissemination measures, that quantify the degree into which a word is concentrated in specific topics and group of individuals. We found a strong correlation between dissemination and frequency increase of the words. Using these measures in our decade-long database we are able to give a fresh view on how the vocabulary of community is "invaded" by new words (e.g., slangs or products) "invade". |
Chaotic and stochastic dynamics in Hamiltonian Systems |
E. G. Altmann and A. Endler, "Noise-enhanced trapping in chaotic scattering" Phys. Rev. Lett. 105, 255102 (2010) or [arXiv:1008.3935]. |
Conservative (Hamiltonian) dynamical systems lie at the core of theoretical Physics and of applications ranging from optical microcavities to fluid dynamics. Typical Hamiltonian systems are neither fully chaotic nor completely integrable, but have a coexistence of regions of both dynamics in a so-called mixed phase space. In our group we investigate different problems related to chaotic Hamiltonian systems. The paper highlighted here focus on the interplay between the nonlinear (chaotic) dynamics and stochastic perturbations that are likely to be present in realistic models. It is shown that in very broad circumstances the stochastic perturbations (noise) plays a constructive role and enhances the trapping of trajectories in a chaotic scattering system. |