6 Matching Annotations
 Sep 2015


Second,we considered the paper’s 10thpercentilezscore.The left tail allows us to characterize the paper’smore unusual combinations, where novelty mayreside.
The highest value in the lowest 10% of zscores in the article.

First, to characterizethe central tendency of a paper’s combinations, weconsidered the paper’smedianzscore
Median zscore: the middle z score for all the journals cited in the paper. I wonder why median?

Zscoresbelow zero indicate pairs that appear less oftenin the observed WOS than expected by chance,indicating relatively atypical or“novel”pairings.
Interesting! So the more random the pairing appeared the more novel the original paper was deemed to be?

In this study, we examined 17.9 million research articles in the Web of Science (WOS) tosee how prior work is combined. We present factsthat indicate (i) the extent to which scientific papers reference novel versus conventional combinations of prior work, (ii) the relative impact ofpapers based on the combinations they drawupon, and (iii) how (i) and (ii) are associated withcollaboration
This is a tall order!

In hisPrincipia,Newton presented his laws of gravitation usingaccepted geometry rather than his newly developed calculus, despite the latter’s importance in developing his insights (22)
The importance of framing your work in work that has already been done.

The highestimpact science is primarily grounded in exceptionallyconventional combinations of prior work yet simultaneously features an intrusion of unusualcombinations
It will be interesting to see how they measured this in so many articles.
