Who will die in Game of Thrones S7?
he new season of Game of Thrones is almost upon us and fans are excited about what it may bring. I am probably
not alone in wondering which of my favourite characters are going to meet their ends, and which will live on to the
next season. So I decided to come up with a ranking for the characters based on how likely it is they will die. Game
of Thrones is a complex world in which social position and true friends seem to be quite important, so I quantified each
character’s social interaction patterns using the tools of network science. I then predict their fate using machine learning methods.
Read the full report here or the corresponding press releases here:
Times Higher Education, Tech Xplore, Futurism, GQ, Economic Times, Huffingtonpost, 444, Gizmondo Australia, International Business Times, Fossbytes, Yahoo news, Marketwatch, , The Hindu, Inverse, india.com, CNET, International Business Times UK, Newsbyte, Storypick, Catchnews, Interesting Engineering, Inverse, Leicester Journal, Motherboard, CetusNews, Wow Science, Morningstar, NDTV, Express, Perk Science, Marketplay, Times Now, Tech Every Eye, ScienceX, DN.no, Kumparan, Rusreality, Buffed.de, Metro.ni, latam.ign, usbeketrica, Agerpres, TVMag, linfo
Game of Thrones: network science and machine learning predicting the death probabilities
Quantifying and Modeling Artistic and Scientific Career Success - PhD thesis project
The existing work on career development psychology and career modeling has studied many aspects of creative careers and success,
however, the field is still lacking a universal understanding of the behavior of the key components of career success. Therefore in
the proposed research I plan to answer the question how we can untangle the stochastic and deterministic features of the individual's
(e.g. scientists, movie directors, music producers) career success on creative fields, and how this relates to the underlying social structure.
I intend to tackle these questions by developing a quantitative methodological framework to describe the evolution of creative careers and test
the framework on large scale data about artistic and scientific careers from various disciplines.
During the research I am going to collect large-scale data on artistic and scientific careers and reconstruct the careers as time series,
where each time event is a creative product with a certain impact. I plan to use these data to study the properties of the success measures
both on the level of the field, and during the career of the individual. I will use tools from time series analysis and stochastic processes
to untangle the probabilistic and deterministic components of success. I also plan to build the collaboration networks of the individuals,
study these networks' main properties, and find their relation to the individuals' success.
To read more, see the blogpost based on a research seminar presentaiton and get preaparad for the publications!
The career trajectory of George Lucas reconstructed by using the IMDb
The success of the movies is measred by the number of ratings they got, and only the movies with at least a thousand ratings are shown.
Data mining on various social media platforms, in collaboration with Maven7 US and Mesura. Big data analysis, social network analysis and visualization. Geospatial data and the identification
of key opinion leaders. Mostly for marketing and business purposes.
Mapping social media in Mexico
#IstandwithCEU campaign on Twitter
Monitoring and analysing the #IstandwithCEU campaign - the support campaign for the Central European University (CEU) using
the publicly available data in Twitter. The studies include the network analysis of the co-occurance of the various hashtags,
estimating the potential and effective social reach of the campaigne, determining the key opinion leaders, and understanding their success.
The reports and press releases can be found here:
Article on CEU.edu
Finding Key Opinion Leaders
Determining the potential social reach
Hashtag co-occurrace network
#IstandwithCEU campaign and hashtag co-occurrence network on Twitters based on the first 10k tweets
Group chasing tactics: how to catch a faster prey? - MSc thesis project
We propose a bio-inspired, agent-based approach to describe the natural phenomenon of group
chasing in both two and three dimensions. Using a set of local interaction rules we created a
continuous-space and discrete-time model with time delay, external noise and limited acceleration.
We implemented a unique collective chasing strategy, optimized its parameters and studied its
properties when chasing a much faster, erratic escaper. We show that collective chasing strategies can
significantly enhance the chasers’ success rate. Our realistic approach handles group chasing within
closed, soft boundaries—in contrast with the periodic ones in most published literature—and
resembles several properties of pursuits observed in nature, such as emergent encircling or the
escaper’s zigzag motion. For more, check out the publication and the supplementary video.
Corresponding press releases:
Institute of Physics
The Science Times
Group chasing tactics: how to catch a faster prey?