Media, Information and Technoculture 3010E Lecture Notes - Lecture 10: Natural-Language Processing, Unsupervised Learning, Personalized Medicine

• School: Western University
• Department: Media, Information and Technoculture
• Course: Media, Information and Technoculture 3010E
• Professor: kierknightly

Document Summary

Sentiment mining: unsupervised learning: using natural language processing to interpret sentences and their meaning, supervised learning: defining a set of features that characterize a positive, a negative and a neutral response. I(cid:374)stead of tr(cid:455)i(cid:374)g to figure out (cid:449)hether a (cid:862)set(cid:863) of features (cid:373)ea(cid:374)s that this is a tea(cid:272)up; It focuses on the probability of each individual feature: easier and faster. Self-measurement: self-assessment at an intensive level, tra(cid:272)ki(cid:374)g o(cid:374)e(cid:859)s o(cid:449)(cid:374) perso(cid:374)al (cid:271)eha(cid:448)iour, ph(cid:455)si(cid:272)al health. Personalized medicine: traditional medical research, randomized controlled trials using a large, random sample, purpose: to extrapolate the results to an entire population, personalized medicine, (cid:862)(cid:374)=(cid:1005)(cid:863) e(cid:454)peri(cid:373)e(cid:374)ts; (cid:373)ea(cid:374)s (cid:455)ou ha(cid:448)e a sa(cid:373)ple size of o(cid:374)e. Intensive data collection on one person, not generalized to any larger population. Tracking beyond numbers: connecting the data tracking to prompts and triggers that stimulate new ways of thinking, example: digital devise that enhance the sense of direction, the data is felt as a sensation, to which the body adapts.