nlp   28629

« earlier    

Stop Using word2vec | Stitch Fix Technology – Multithreaded
When I started playing with word2vec four years ago I needed (and luckily had) tons of supercomputer time. But because of advances in our understanding of word2vec, computing word vectors now takes fifteen minutes on a single run-of-the-mill computer with standard numerical libraries1. Word vectors are awesome but you don’t need a neural network – and definitely don’t need deep learning – to find them2. So if you’re using word vectors and aren’t gunning for state of the art or a paper publication then stop using word2vec.

When we’re finished you’ll measure word similarities:

facebook ~ twitter, google, ...

… and the classic word vector operations: zuckerberg - facebook + microsoft ~ nadella

…but you’ll do it mostly by counting words and dividing, no gradients harmed in the making!
nlp  language 
yesterday by jotjotjes
Idioms in sentiment analysis
some nice small datasets for idioms
data  ml  nlp  sentiment 
yesterday by mootPoint

« earlier    

related tags

ai  algorithm  algorithms  api  article  artificialintelligence  audio  backlog  belgium  berkeley  bigdata  bigquery  bookmarks_bar  bots  chatbot  cheatsheet  chi2  clase  classification  clinicaltrials  cnn  coaching  coding  collaborativefiltering  computerscience  corenlp  corpora  corpus  data  data_sets  database  datascience  dataset  datasets  db  deep-learning  deeplearning  design  directory  discover-weekly  discovery  dl  doc2vec  echonest  edu  ehr  embeddings  entities  entityextraction  eu  examples  facebook  fast-text  fasttext  featureselection  fernando-pereira  gem  goals  godam  google  googlecloud  graphics  hypnosis  idk  ir  journal  junk  keras  language-detection  language  library  linearalgebra  linguistics  linux  logistic_regression  lstm  machine-learning  machine_learning  machinecomprehension  machinelearning  math  method  mining  ml  music  natural_language  neural_networks  neuralnetworks  nlproc  nltk  nlu  nn  ontology  opendata  opensource  overview  papers  parse  parser  people  productivity  programming  python  pytorch  q&a  qrnn  query  r  readlater  reco  recommendation_engine  recommendations  ref  reference  research  resource-list  rnn  robertdale  ruby  scripting  search  semanticweb  sentiment  shell  smoothing  software  speechrecognition  spotify  stanford  statistics  strings  summarisation  summary  t  tasks  technique  tensorflow  text  textanalysis  textgeneration  texttospeech  textual-entailment  toolkit  tools  training  tsne  tutorial  twitter  university  video  watchlist?  women  word-embeddings  word2vec  wordclouds  wordembedding  words 

Copy this bookmark: