<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tigunova-etal-2020-charm">
    <titleInfo>
        <title>CHARM: Inferring Personal Attributes from Conversations</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Anna</namePart>
        <namePart type="family">Tigunova</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Andrew</namePart>
        <namePart type="family">Yates</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Paramita</namePart>
        <namePart type="family">Mirza</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Gerhard</namePart>
        <namePart type="family">Weikum</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2020-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Bonnie</namePart>
            <namePart type="family">Webber</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Trevor</namePart>
            <namePart type="family">Cohn</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yulan</namePart>
            <namePart type="family">He</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yang</namePart>
            <namePart type="family">Liu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Online</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Personal knowledge about users’ professions, hobbies, favorite food, and travel preferences, among others, is a valuable asset for individualized AI, such as recommenders or chatbots. Conversations in social media, such as Reddit, are a rich source of data for inferring personal facts. Prior work developed supervised methods to extract this knowledge, but these approaches can not generalize beyond attribute values with ample labeled training samples. This paper overcomes this limitation by devising CHARM: a zero-shot learning method that creatively leverages keyword extraction and document retrieval in order to predict attribute values that were never seen during training. Experiments with large datasets from Reddit show the viability of CHARM for open-ended attributes, such as professions and hobbies.</abstract>
    <identifier type="citekey">tigunova-etal-2020-charm</identifier>
    <identifier type="doi">10.18653/v1/2020.emnlp-main.434</identifier>
    <location>
        <url>https://aclanthology.org/2020.emnlp-main.434</url>
    </location>
    <part>
        <date>2020-11</date>
        <extent unit="page">
            <start>5391</start>
            <end>5404</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T CHARM: Inferring Personal Attributes from Conversations
%A Tigunova, Anna
%A Yates, Andrew
%A Mirza, Paramita
%A Weikum, Gerhard
%Y Webber, Bonnie
%Y Cohn, Trevor
%Y He, Yulan
%Y Liu, Yang
%S Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
%D 2020
%8 November
%I Association for Computational Linguistics
%C Online
%F tigunova-etal-2020-charm
%X Personal knowledge about users’ professions, hobbies, favorite food, and travel preferences, among others, is a valuable asset for individualized AI, such as recommenders or chatbots. Conversations in social media, such as Reddit, are a rich source of data for inferring personal facts. Prior work developed supervised methods to extract this knowledge, but these approaches can not generalize beyond attribute values with ample labeled training samples. This paper overcomes this limitation by devising CHARM: a zero-shot learning method that creatively leverages keyword extraction and document retrieval in order to predict attribute values that were never seen during training. Experiments with large datasets from Reddit show the viability of CHARM for open-ended attributes, such as professions and hobbies.
%R 10.18653/v1/2020.emnlp-main.434
%U https://aclanthology.org/2020.emnlp-main.434
%U https://doi.org/10.18653/v1/2020.emnlp-main.434
%P 5391-5404

Markdown (Informal)

[CHARM: Inferring Personal Attributes from Conversations](https://aclanthology.org/2020.emnlp-main.434) (Tigunova et al., EMNLP 2020)

• CHARM: Inferring Personal Attributes from Conversations (Tigunova et al., EMNLP 2020)

ACL

• Anna Tigunova, Andrew Yates, Paramita Mirza, and Gerhard Weikum. 2020. CHARM: Inferring Personal Attributes from Conversations. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 5391–5404, Online. Association for Computational Linguistics.