Corpus ID: 127598

Controlled Experiments for Word Embeddings

  title={Controlled Experiments for Word Embeddings},
  author={B. Wilson and Adriaan M. J. Schakel},
  • B. Wilson, Adriaan M. J. Schakel
  • Published 2015
  • Computer Science
  • ArXiv
  • An experimental approach to studying the properties of word embeddings is proposed. Controlled experiments, achieved through modifications of the training corpus, permit the demonstration of direct relations between word properties and word vector direction and length. The approach is demonstrated using the word2vec CBOW model with experiments that independently vary word frequency and word co-occurrence noise. The experiments reveal that word vector length depends more or less linearly on both… CONTINUE READING
    15 Citations

    Figures, Tables, and Topics from this paper

    Intrinsic Evaluations of Word Embeddings: What Can We Do Better?
    • 62
    • PDF
    Discrete Word Embedding for Logical Natural Language Understanding
    • PDF
    NTUA-SLP at SemEval-2018 Task 3: Tracking Ironic Tweets using Ensembles of Word and Character Level Attentive RNNs
    • 25
    • PDF
    A Dual Embedding Space Model for Document Ranking
    • 95
    • PDF
    PIC a Different Word: A Simple Model for Lexical Substitution in Context
    • 15
    • PDF
    Norm-Based Curriculum Learning for Neural Machine Translation
    • 11
    • PDF
    SAO2Vec: Development of an algorithm for embedding the subject–action–object (SAO) structure using Doc2Vec
    • 2
    Extended Association Rules in Semantic Vector Spaces for Sentiment Classification
    • 1
    Identifying lexical relationships and entailments with distributional semantics
    • PDF
    Spherical Regression under Mismatch Corruption with Application to Automated Knowledge Translation
    • 12
    • PDF


    Glove: Global Vectors for Word Representation
    • 16,735
    • PDF
    Efficient Estimation of Word Representations in Vector Space
    • 16,738
    • PDF
    Random Walks on Context Spaces: Towards an Explanation of the Mysteries of Semantic Word Embeddings
    • 50
    • PDF
    Measuring Word Significance using Distributed Representations of Words
    • 49
    • PDF
    Evaluation methods for unsupervised word embeddings
    • 393
    • PDF
    Distributed Representations of Words and Phrases and their Compositionality
    • 20,831
    • PDF
    Don't count, predict! A systematic comparison of context-counting vs. context-predicting semantic vectors
    • 1,179
    • PDF
    Natural Language Processing (Almost) from Scratch
    • 5,841
    • PDF