Portfolio Assignment 2

Python

Starting off I added the chapter 2 files to my Lib directory and then downloaded feedparser to the Lib directory as well before finishing chapter 2 exercises.


pydelcious imported correctly and get_popular(tags='programming') provided results in the correct fashion. I then continued to build the dataset by adding the user tsegaran to the list, filling his username with items and recommending similar users:

• u'wlyndon


• u'sillyputty1967


• u'shashashasha


• u'rufous

and then recommending similar links:

• www.rollingstone.com.news.story.5939600/steve_jobs_the_rolling_stone_interview


• http://www.woopra.com/


• www.youtube.com/watch?v=eRkmhJR10Ec


• zenhabits.net

I then used the function calculateSimilarItems in Recommendations and the results for the movies Lady in theWater, Snakes on a Plane, You Me and Dupree, and The Night Listener where displayed properly.

I attempted to download the Movie Lens Data Sets but could not get the file to work correctly to be accessed by the loadMovieLens() function.

exercise 1:

"The Tanimoto similarity score is a method of calculating the similarity between two ligand fingerprints. It is determined as shown in the equation below where:T is the Tanimoto ScoreNa and Nb are the number of bits set to 1 in fingerprint of ligand a and b respectively and Nc is the total number of bits set to 1 found in fingerprints of both ligand a and b. The Tanimoto similarity score is a method of calculating the similarity between two ligand fingerprints. It is determined as shown in the equation below where:T is the Tanimoto ScoreNa and Nb are the number of bits set to 1 in fingerprint of ligand a and b respectively and Nc is the total number of bits set to 1 found in fingerprints of both ligand a and b. "

http://www-mitchell.ch.cam.ac.uk/pld/background_simil_lig.html

my similarity function:

def tanimoto(prefs,p1,p2):
si={}
for item in prefs[p1]:
if item in prefs[p2]: si[item]=1
if len(si)==0: return 0
distance=sum([prefs[person1][item]*prefs[person2][item] / (pow(prefs[person1][item],2)+pow(prefs[person2][item],2)-prefs[person1][item]*prefs[person2][item]) for item in prefs[person1] if item in prefs[person2]])
return 1/(1+distance)

Im still working on the weka section, but I wanted to publish this part before midnight!