Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

EVE: explainable vector based embedding technique using Wikipedia

EVE: explainable vector based embedding technique using Wikipedia We present an unsupervised explainable vector embedding technique, called EVE, which is built upon the structure of Wikipedia. The proposed model defines the dimensions of a semantic vector representing a concept using human-readable labels, thereby it is readily interpretable. Specifically, each vector is constructed using the Wikipedia category graph structure together with the Wikipedia article link structure. To test the effectiveness of the proposed model, we consider its usefulness in three fundamental tasks: 1) intruder detection—to evaluate its ability to identify a non-coherent vector from a list of coherent vectors, 2) ability to cluster—to evaluate its tendency to group related vectors together while keeping unrelated vectors in separate clusters, and 3) sorting relevant items first—to evaluate its ability to rank vectors (items) relevant to the query in the top order of the result. For each task, we also propose a strategy to generate a task-specific human-interpretable explanation from the model. These demonstrate the overall effectiveness of the explainable embeddings generated by EVE. Finally, we compare EVE with the Word2Vec, FastText, and GloVe embedding techniques across the three tasks, and report improvements over the state-of-the-art. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Intelligent Information Systems Springer Journals

EVE: explainable vector based embedding technique using Wikipedia

Download PDF
29 pages

Loading next page...
 
/lp/springer_journal/eve-explainable-vector-based-embedding-technique-using-wikipedia-M5BlJBfgD4
Publisher
Springer Journals
Copyright
Copyright © 2018 by Springer Science+Business Media, LLC, part of Springer Nature
Subject
Computer Science; Information Storage and Retrieval; Data Structures and Information Theory; Artificial Intelligence; IT in Business; Natural Language Processing (NLP)
ISSN
0925-9902
eISSN
1573-7675
DOI
10.1007/s10844-018-0511-x
Publisher site
See Article on Publisher Site

Abstract

We present an unsupervised explainable vector embedding technique, called EVE, which is built upon the structure of Wikipedia. The proposed model defines the dimensions of a semantic vector representing a concept using human-readable labels, thereby it is readily interpretable. Specifically, each vector is constructed using the Wikipedia category graph structure together with the Wikipedia article link structure. To test the effectiveness of the proposed model, we consider its usefulness in three fundamental tasks: 1) intruder detection—to evaluate its ability to identify a non-coherent vector from a list of coherent vectors, 2) ability to cluster—to evaluate its tendency to group related vectors together while keeping unrelated vectors in separate clusters, and 3) sorting relevant items first—to evaluate its ability to rank vectors (items) relevant to the query in the top order of the result. For each task, we also propose a strategy to generate a task-specific human-interpretable explanation from the model. These demonstrate the overall effectiveness of the explainable embeddings generated by EVE. Finally, we compare EVE with the Word2Vec, FastText, and GloVe embedding techniques across the three tasks, and report improvements over the state-of-the-art.

Journal

Journal of Intelligent Information Systems – Springer Journals

Published: Jun 4, 2018

References

  • A latent variable model approach to pmi-based word embeddings
    Arora, S; Li, Y; Liang, Y; Ma, T; Risteski, A
  • Distributional memory: a general framework for corpus-based semantics
    Baroni, M; Lenci, A
  • A neural probabilistic language model
    Bengio, Y; Ducharme, R; Vincent, P; Jauvin, C
  • Evaluating wordnet-based measures of lexical semantic relatedness
    Budanitsky, A; Hirst, G
  • A dendrite method for cluster analysis
    Caliński, T; Harabasz, J
  • Cluster analysis
    Everitt, B; Landau, S; Leese, M
  • Euclidean embedding of co-occurrence data
    Globerson, A; Chechik, G; Pereira, F; Tishby, N
  • Distributional structure
    Harris, ZS
  • Mathematical structures of language
    Harris, ZS
  • A peek into the black box: exploring classifiers by randomization
    Henelius, A; Puolamäki, K; Boström, H; Asker, L; Papapetrou, P
  • Explaining qualitative diagnosis
    Hunt, J; Price, C
  • Feature-based approaches to semantic similarity assessment of concepts using wikipedia
    Jiang, Y; Zhang, X; Tang, Y; Nie, R
  • An introduction to latent semantic analysis
    Landauer, TK; Foltz, PW; Laham, D
  • Improving distributional similarity with lessons learned from word embeddings
    Levy, O; Goldberg, Y; Dagan, I
  • Dykor: a method for generating the content of explanations in knowledge systems
    Lopez-Suarez, A; Kamel, M
  • Introduction to information retrieval
    Manning, CD; Raghavan, P; Schütze, H
  • Pharmacovigilance from social media: mining adverse drug reaction mentions using sequence labeling with word embedding cluster features
    Nikfarjam, A; Sarker, A; O’Connor, K; Ginn, R; Gonzalez, G
  • Introduction to modern information retrieval
    Salton, G; McGill, MJ
  • Visualizing data using t-sne
    Maaten, L; Hinton, G
  • Semantic expansion using word embedding clustering and convolutional neural network for improving short text classification
    Wang, P; Xu, B; Xu, J; Tian, G; Liu, CL; Hao, H
  • Reconstructive expert system explanation
    Wick, MR; Thompson, WB