#Some User Statistics
ratingTable = []
print "Building Review Table"
print "There are %d reviews" % len(loader.reviews)
for id, review in loader.reviews.iteritems():
paper = review.paper
reviewer = review.user
ratingStats = {
"rating": review.overallRating,
"time": np.datetime64(review.time),
"pastPaperSimilarity": calcFeatures.computePaperReviewerSimilarity(
tfidf, paper, reviewer),
"authorReviewerSimilarity": calcFeatures.getAuthorReviewerSimilarity(
tfidf, paper.primaryAuthor, reviewer),
"maxAuthorReviewerSimilarity": calcFeatures.getAuthorReviewerSimilarity(
tfidf, paper.maxAuthor, reviewer),
"specificCommonSubjects": calcFeatures.computeSpecificCommonSubjects(
paper, reviewer),
"generalCommonSubjects": calcFeatures.computeGeneralCommonSubjects(
paper, reviewer),
"samePrimaryGeneralSubject":
reviewer.primaryGeneralSubjectArea == paper.primaryGeneralSubjectArea,
"samePrimarySpecificSubject":
reviewer.primarySpecificSubjectArea == paper.primarySpecificSubjectArea,
"paperId":
paper.id,
"userId":
reviewer.id,
"rating":
review.overallRating,
"revPaperCount":
len(reviewer.pastPapers),
"revTopPaperCount":
reviewer.topPastPapers,
"revKDDPaperCount":
reviewer.topKDDPast,
###Similarity
"maxSimilarity":
calcFeatures.getAuthorReviewerSimilarity(tfidf, paper.maxAuthor,
reviewer),
"primarySimilarity":
calcFeatures.getAuthorReviewerSimilarity(tfidf, paper.primaryAuthor,
reviewer),
"maxJacSimilarity":
calcFeatures.getAuthorReviewerSimilarity(tfidf,
paper.maxAuthor,
reviewer,
jaccard=True),
"primaryJacSimilarity":
calcFeatures.getAuthorReviewerSimilarity(tfidf,
paper.primaryAuthor,
reviewer,
jaccard=True),
###Co-Author Distance
for id, review in loader.reviews.iteritems():
paper = review.paper
reviewer = review.user
reviewTable.append({
"paperId": paper.id,
"userId": reviewer.id,
"rating": review.overallRating,
"revPaperCount": len(reviewer.pastPapers),
"revTopPaperCount": reviewer.topPastPapers,
"revKDDPaperCount": reviewer.topKDDPast,
###Similarity
"maxSimilarity": calcFeatures.getAuthorReviewerSimilarity(
tfidf, paper.maxAuthor, reviewer),
"primarySimilarity": calcFeatures.getAuthorReviewerSimilarity(
tfidf, paper.primaryAuthor, reviewer),
"maxJacSimilarity": calcFeatures.getAuthorReviewerSimilarity(
tfidf, paper.maxAuthor, reviewer, jaccard=True),
"primaryJacSimilarity": calcFeatures.getAuthorReviewerSimilarity(
tfidf, paper.primaryAuthor, reviewer, jaccard=True),
###Co-Author Distance
"minDist": review.minDist,
"avgDist": review.avgDist,
"revCountry": reviewer.country,
