<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="takahashi-tanaka-ishii-2019-evaluating">
    <titleInfo>
        <title>Evaluating Computational Language Models with Scaling Properties of Natural Language</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Shuntaro</namePart>
        <namePart type="family">Takahashi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kumiko</namePart>
        <namePart type="family">Tanaka-Ishii</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2019-09</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>In this article, we evaluate computational models of natural language with respect to the universal statistical behaviors of natural language. Statistical mechanical analyses have revealed that natural language text is characterized by scaling properties, which quantify the global structure in the vocabulary population and the long memory of a text. We study whether five scaling properties (given by Zipf’s law, Heaps’ law, Ebeling’s method, Taylor’s law, and long-range correlation analysis) can serve for evaluation of computational models. Specifically, we test n-gram language models, a probabilistic context-free grammar, language models based on Simon/Pitman-Yor processes, neural language models, and generative adversarial networks for text generation. Our analysis reveals that language models based on recurrent neural networks with a gating mechanism (i.e., long short-term memory; a gated recurrent unit; and quasi-recurrent neural networks) are the only computational models that can reproduce the long memory behavior of natural language. Furthermore, through comparison with recently proposed model-based evaluation methods, we find that the exponent of Taylor’s law is a good indicator of model quality.</abstract>
    <identifier type="citekey">takahashi-tanaka-ishii-2019-evaluating</identifier>
    <identifier type="doi">10.1162/coli_a_00355</identifier>
    <location>
        <url>https://aclanthology.org/J19-3003</url>
    </location>
    <part>
        <date>2019-09</date>
        <detail type="volume"><number>45</number></detail>
        <detail type="issue"><number>3</number></detail>
        <extent unit="page">
            <start>481</start>
            <end>513</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Evaluating Computational Language Models with Scaling Properties of Natural Language
%A Takahashi, Shuntaro
%A Tanaka-Ishii, Kumiko
%J Computational Linguistics
%D 2019
%8 September
%V 45
%N 3
%I MIT Press
%C Cambridge, MA
%F takahashi-tanaka-ishii-2019-evaluating
%X In this article, we evaluate computational models of natural language with respect to the universal statistical behaviors of natural language. Statistical mechanical analyses have revealed that natural language text is characterized by scaling properties, which quantify the global structure in the vocabulary population and the long memory of a text. We study whether five scaling properties (given by Zipf’s law, Heaps’ law, Ebeling’s method, Taylor’s law, and long-range correlation analysis) can serve for evaluation of computational models. Specifically, we test n-gram language models, a probabilistic context-free grammar, language models based on Simon/Pitman-Yor processes, neural language models, and generative adversarial networks for text generation. Our analysis reveals that language models based on recurrent neural networks with a gating mechanism (i.e., long short-term memory; a gated recurrent unit; and quasi-recurrent neural networks) are the only computational models that can reproduce the long memory behavior of natural language. Furthermore, through comparison with recently proposed model-based evaluation methods, we find that the exponent of Taylor’s law is a good indicator of model quality.
%R 10.1162/coli_a_00355
%U https://aclanthology.org/J19-3003
%U https://doi.org/10.1162/coli_a_00355
%P 481-513

Markdown (Informal)

[Evaluating Computational Language Models with Scaling Properties of Natural Language](https://aclanthology.org/J19-3003) (Takahashi & Tanaka-Ishii, CL 2019)

• Evaluating Computational Language Models with Scaling Properties of Natural Language (Takahashi & Tanaka-Ishii, CL 2019)

ACL

• Shuntaro Takahashi and Kumiko Tanaka-Ishii. 2019. Evaluating Computational Language Models with Scaling Properties of Natural Language. Computational Linguistics, 45(3):481–513.