Deep Learning

Google recently introduced a potentially very interesting open source tool called word2vec. It is software designed to understand the relationship between words without human intervention.  How? Through the use of "deep learning" - basically neural network models on steroids.  Essentially these models understand the features of an input (in this case words) and how those inputs relate to each other.


You don't have to understand the ins and outs to see how potentially useful software that can do this is.  One potential application is better understanding of tweets.  Typical sentiment analysis, for example, has a difficult time with tweets due to their short nature and the fact that they are often infused with symbols and sarcasm.  With deep learning techniques it may be possible to decipher the sentiment of tweets more accurately.  Can you imagine?

Data and Movies

Even movie scripts can't escape the data analysis trend. Mr. Bruzzese of Motion Picture Group, a former statistics professor, analyzes scripts for movie studios. He relies on statistics and survey results of other movies. His findings show things like bowling scenes are present in movies that flop, so best to avoid them.

He doesn't currently use algorithms but I think it's only a matter of time.

Sentiment Analysis

The bane of sentiment analysis is sarcasm. It's very difficult for a computer to know if "that was way fun" is actually someone exclaiming enjoyment or being sarcastic. Heck, it can even be tough for humans to tell the difference. A USC Annenberg lab out of Los Angeles is trying to solve the problem. They built their model using human annotation.

Only time will tell if they can crack the code. Some folks on Twitter simply use (*S) to indicate sarcasm, but it hasn't caught on universally. And that's a shame (*S).

99 Problems

An awesome legal analysis of Jay-Z's hit song, "99 Problems" by associate professor of law at Southwestern University, Caleb Mason. Lyrics: In my rearview mirror is the motherfucking law/I got two choices y'all, pull over the car or/Bounce on the double put the pedal to the floor

"The calculation Jay-Z has to make is whether, knowing that the car contains concealed contraband, he's better off trying to flee or hoping that the police won't find the drugs during the stop. This may be the hardest choice perps face (until they have to decide whether or not to cooperate), but there's only one answer: you are always better off having drugs found on you in a potentially illegal search than you are fleeing from a potentially illegal search and getting caught. The flight will provide an independent basis for chasing and arresting you, and the inadequacy of the quantum of suspicion supporting the initial attempted seizure will not taint the contraband discovered if there is an intervening flight. Law students: practice explaining the preceding sentence to a layperson. Smugglers, repeat after me: you have to eat the bust, and fight it in court."