Download Computational Social Science Lecture Notes

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Enactivism wikipedia , lookup

Community development wikipedia , lookup

History of social work wikipedia , lookup

Computational linguistics wikipedia , lookup

Anthropology of development wikipedia , lookup

Postdevelopment theory wikipedia , lookup

History of water supply and sanitation wikipedia , lookup

Third Way wikipedia , lookup

Tribe (Internet) wikipedia , lookup

Social Bonding and Nurture Kinship wikipedia , lookup

Social psychology wikipedia , lookup

Unilineal evolution wikipedia , lookup

Social network analysis wikipedia , lookup

Inclusive fitness in humans wikipedia , lookup

Social theory wikipedia , lookup

Social rule system theory wikipedia , lookup

Six degrees of separation wikipedia , lookup

Social perception wikipedia , lookup

Sociological theory wikipedia , lookup

Origins of society wikipedia , lookup

Social group wikipedia , lookup

Social computing wikipedia , lookup

Social history wikipedia , lookup

History of the social sciences wikipedia , lookup

Transcript
Computational Social Science
1. A Knowledge Discovery and Data Mining workshops in Anaheim in 1991. Now, there is
much more to it with two journals.
2. Why such success in KDD?
a. Exploaratory
b. Ecumenical
i. Not fixated on a single class of techniques
c. Externally Focused
d. Computational Social Science is the interdisciplinary study of complex social
systems and their investigation through computational modeling and related
techniques
e. CSS is a convergence of
i. New data collections and availability
ii. New methods, languages and formalisms
iii. New questions and theory that have associations with many of the social
sciences
f. Social Science is central
i. Science- how are social systems formed?
ii. Engineering- How do we build systems that make sure the people
interactions with the system go well
iii. Business- what do the customers want and how can we best determine
their purchasing patterns?
iv. Government
g. Examples of existing work
i. Six degrees of Separation with Milgram’s work
ii. TRANSIMS- a city level transportation simulator in the 90s
3. London in 1854
a. Very filthy place
b. Cholera Epidemic in the 1850s.
i. Could not find reason for cholera but John Snow thought it was because of
the sewage water interacting with drinking water. This contaminated water
was then used by people and they got sick.
ii. Theory was not popular.
iii. John Snow attempts to spatially map the data he collects about the deaths
related with cholera.
iv. He did not have computation to really work with the data
4. Themes
a. Relational, Temporal and Spatial Models
i. What does snow need to represent?
1. Entities
2. Attributes
3. Relationships
4. Spatio-temporal extent and variation
ii. What can we respreset
1. Probabilistic models
2. Issue: Modeling social systems
3. Explain algorithm behavior
a. Association rules
4. Issue: acceptance by social scientists
b. Casual Analysis
i. Statistical association alone is insufficient to distinguish among different
casual models
c. Performing Quasi-Experiments to find causality
i. Snow performed this on two different water companies that were
supplying water to a specific region
ii. One company’s water proved to be eight times more deadly
d. Local Methods and Global Models
5. CSS and Privacy
a. Snow had much freer access to data due to the lax of privacy laws
b.