issues
search
ArdalanM
/
kaggle_avito
123rd of 548
https://www.kaggle.com/c/avito-duplicate-ads-detection
1
stars
1
forks
source link
Features
#1
Open
ArdalanM
opened
8 years ago
ArdalanM
commented
8 years ago
Train w2v on raw/clean data
Train doc2v on raw/clean data
ArdalanM
commented
8 years ago
Distance between string:
leven raw/clean
dl raw/clean
jaro raw/clean
jaron winkler raw/clean
hamming raw/clean
Distance between two list:
Dice (uni/bi/tri grams)
Jaccard (uni/bi/tri) -compression (uni/bi/tri)
edit (uni/bi/tri)