Jump to content

Lennon or McCartney? Can statistical analysis solve an authorship puzzle? - General Hangout & Discussions - InviteHawk - Your Only Source for Free Torrent Invites

Buy, Sell, Trade or Find Free Torrent Invites for Private Torrent Trackers Such As redacted, blutopia, losslessclub, femdomcult, filelist, Chdbits, Uhdbits, empornium, iptorrents, hdbits, gazellegames, animebytes, privatehd, myspleen, torrentleech, morethantv, bibliotik, alpharatio, blady, passthepopcorn, brokenstones, pornbay, cgpeers, cinemageddon, broadcasthenet, learnbits, torrentseeds, beyondhd, cinemaz, u2.dmhy, Karagarga, PTerclub, Nyaa.si, Polishtracker etc.

Lennon or McCartney? Can statistical analysis solve an authorship puzzle?


Recommended Posts

 

Stylometry -- the use of statistical techniques to determine authorship -- is best known for identifying the Unabomber as Theodor Kaczynski and revealing that Shakespeare collaborated with Christopher Marlowe on the Henry IV play cycle. In textual analysis, it is not the unusual word choice that betrays the hidden voice, but the habitual -- the recurring patterns of common words, such as prepositions, that mark the probable identity of one person alone.

It was a mutual Beatles passion -- discovered at a conference on Prince Edward Island -- that led Mark Glickman, senior lecturer in statistics at Harvard, and Jason Brown, professor of mathematics at Dalhousie University, to wonder whether a stylometric approach could answer the burning question: Lennon or McCartney?

As Glickman explains, for most Lennon-McCartney songs, it is well-known and well-documented which of the two wrote the song. However, a surprisingly large number of songs (or portions of songs) have disputed authorship. As an example, no one knows who wrote the music for "In My Life," a track from the 1965 album Rubber Soul, which is ranked 23 on Rolling Stone's The 500 Greatest Songs of All Time. Both Lennon and McCartney remembered differently. "So, we wondered whether you could use data analysis techniques to try to figure out what was going on in the song to distinguish whether it was by one or the other," says Glickman.

With help from former Harvard statistics student Ryan Song, Glickman and Brown "decomposed" each Beatles song from 1962 to 1966 into five representations. Each representation consisted of the frequency of occurrence of a set of musical features within each song. "The basic idea behind our approach," says Glickman, "is to convert a song, whose musical content is difficult to quantify in any direct way, into a set of different data structures that are amenable for establishing a signature of a song using a quantitative approach." Glickman continues, "Think of decomposing a color into its constituent components of red, green and blue with different weights attached. We're doing the same thing with Beatles songs, though with more than three components. In total, our method divides songs into a total of 149 constituent components."

"The first representation simply consists of the frequencies of different commonly played chords, along with aggregations of uncommon chords," says Glickman. "We were able to form 11 chord categories." Then, they characterized melodic notes -- notes sung by the lead singer. Third, they recorded the frequencies of occurrence of chord transitions, that is, one chord followed by another chord. Again, certain uncommon chord transitions were aggregated into single categories. Fourth, they recorded the frequencies of consecutive melodic note pairs.
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
  • Customer Reviews

  • Similar Topics

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.