Data mining 10,000 porn actors

There are many myths and misconceptions around the adult film industry. For example, many tend to think that the typical actress is a curvaceous blonde. In fact, this is far from the case. Jon Millward carried out the data mining of the cinema base of the Internet Adult Film Database and analyzed information on 125 thousand films that contain information on more than 115 thousand actors. 10,000 people were sampled for statistical analysis.

John Millward analyzed the appearance of the actors, places of birth, the most popular first and last names in pseudonyms, skin color, hair color, age of beginning and end of a career. Revealed some interesting facts.

For example, the layout of hair color in women showed that only 32.7% of them have light hair color. Dark-haired more than about twice.
')


The distribution of actors by skin color completely coincides with the normal distribution in society.



It also turned out interesting statistics on weight. The weight of male actors by 12 kg is different from the average weight of men in the United States, and for women actresses, the weight is 22 kg less than that of ordinary women.

The number of tattoos for women is more than that for men, but most of the actors do not make tattoos.



Thus, the average adult industry actress is a girl with dark hair, 166 cm tall, 53 kg in weight, without a tattoo, with a chest size of 34B (second size), who was born in California and begins her career at the age of 22. The age of starting a career for women has not changed in the last 40 years, when they began to keep a record in the database, but for men it decreased from 29 to 24 years. The average length of a career for women is 3 years and 19 films. Now the market has become very competitive. Fees dropped due to free content on the Internet. The actresses are trying to actively communicate on Twitter and social networks in order to form an audience of fans and secure their future.

John Millward made the morphing of the faces of the 10 most popular actresses, this information is not of scientific value, but is given simply as an illustration. Morphing is first implemented in pairs, and the final result is shown on the right.



For other research results, see Millward's blog or infographics .