Frank Kane's Taming Big Data with Apache Spark and Python

Making movie recommendations

Believe it or not, the following code is all you need to replicate what was done with the Netflix prize and actually create what should be really good movie recommendations. However, as we'll see, your mileage may vary.:

data = sc.textFile("file:///SparkCourse/ml-100k/u.data")    ratings = data.map(lambda l: l.split()).map(lambda l: Rating(int(l[0]),    int(l[1]), float(l[2]))).cache()rank = 10numIterations = 20model = ALS.train(ratings, rank, numIterations)recommendations = model.recommendProducts(userID, 10)

Let's walk through it. All it's doing is opening up a text file from your hard drive, the same old u.data file you're used to all the time:

data = sc.textFile("file:///SparkCourse/ml-100k/u.data")

Next, it's ...