Revisiting Algorithms, This Time, For Fun.

In our constantly connected world, where the prefix “smart” is being latched onto more and more devices (smart TVs, smart watches, smart homes…) almost everything we do is tracked and analyzed by countless algorithms that try to predict or affect what we’ll do next.

As we learned on a recent Tech Tuesday, these algorithms can be very helpful, like when companies like Netflix and Spotify use them to predict what movie or song you’d likely enjoy.

But algorithms can also be troublesome and discriminatory: a study from Harvard University found that ads for arrest records were more likely to show up when people searched for historically black fraternities or characteristically black names. In another case, researchers at Carnegie Mellon found that men were much more likely than women to be shown an ad for high-paying, executive jobs after visiting employment websites.

All to say: algorithms are complex, and they present ethical challenges that will only become more vexing as the formulas weave themselves into every aspect of our lives. But not all algorithms are cause for debate. Here, a few algorithms that are just for fun. But keep in mind: rules- and algorithms- were made to be broken.

Find The Perfect Scary Movie

(es + u + cs + t)2 + s + (tl + f) / 2 + (a + dr + fs) / n + sin x – st

es = escalating music u = the unknown
cs = chase scene
t =sense of being trapped
s = shock tl = true life
f = fantasy
a = character is alone dr = in the dark
fs = film setting
n = number of people x = blood and guts
st = stereotypes

A team from King’s College London spent two weeks watching horror flicks in the name of mathematical research. The result? A formula to measure how scary a movie is. To calculate, simply assign a numerical rating for each factor they considered important. Using their algorithm, they concluded that The Shining is the ideal horror movie. How does your favorite measure up? (BBC)

Predict How Long Your Relationship Will Last

L = 8 + .5Y – .2P + .9Hm + .3Mf + J – .3G – .5(Sm – Sf)2 + I + 1.5C

L = the predicted length in years of the relationship
Y = the number of years the two people knew each other before the relationship became serious
P = the number of previous partners of both people added together
Hm = the importance the male partner attaches to honesty in the relationship
Mf = the importance the female attaches to money in the relationship
J = the importance both attach to humor (added together)
G = the importance both attach to good looks (added together)
Sm and Sf = the importance male and female attach to sex
I = the importance attached to having good in-laws (added together)
C = the importance attached to children in the relationship (added together)

MSN surveyed 2,000 men and women about their past relationships to come up with a formula for how long love will last. To calculate, assign each variable a rating from 1 to 5, from not at all to extremely important. Some takeaways: The more you care about looks, the shorter your relationships will be, and children are the strongest indicator of a long-lasting relationship. (Business Insider)

Predict the Killer in an Agatha Christie Novel

k( r, δ, θ, c) = f{r k+ δ + θ{P,M}, c (3≤4.5}

k = identity of the killer r = relationship to the victim
δ = Primary means of transport associated with the novel
θ =method of murder and detective characterized in the novel
P = Poirot
M = Marple
c = chapter of introduction of the killer f = female

To celebrate Agatha Christie’s 125th birthday, a team from Queen’s University in Belfast, developed an algorithm that will accurately predict the killer in one of Christie’s mysteries. They also identified a number of trends that will give you clues: if the book takes place in a country house, there is a 75 percent chance the killer will be female. If the primary mode of transport in the novel is by air or sea, the killer’s likely to be male. This formula is complicated to calculate, so you may just want to stick to the formula my mom, who has read every Agatha Christie novel, uses: killer = the person you’d least expect. (Vulture)