Article Instance

HTTP 200 OK
Allow: GET
Content-Type: application/json
Vary: Accept

{
    "pk": 28861,
    "title": "Modelling semantics by integrating linguistic,\nvisual and affective information",
    "subtitle": null,
    "abstract": "A number of recent models of semantics combine linguistic\ninformation, derived from text corpora, and visual information,\nderived from image collections, demonstrating that the\nresulting multimodal models are better than either of their\nunimodal counterparts, in accounting for behavioural data.\nHowever, first, while linguistic models have been extensively\ntested for their fit to behavioural semantic ratings, this is not\nthe case for visual models which are also far more limited in\ntheir coverage. More broadly, empirical work on semantic\nprocessing has shown that emotion also plays an important role\nespecially for abstract concepts, however, models integrating\nemotion along with linguistic and visual information are\nlacking. Here, we first improve on visual representations by\nchoosing a visual model that best fit semantic data and\nextending its coverage. Crucially then, we assess whether\nadding affective representations (obtained from a neural\nnetwork model designed to predict emojis from co-occurring\ntext) improves model’s ability to fit semantic\nsimilarity/relatedness judgements from a purely linguistic and\nlinguistic-visual model. We find that adding both visual and\naffective representations improve performance, with visual\nrepresentations providing an improvement especially for more\nconcrete words and affective representations improving\nespecially fit for more abstract words.",
    "language": "eng",
    "license": {
        "name": "",
        "short_name": "",
        "text": null,
        "url": ""
    },
    "keywords": [
        {
            "word": "language; vision; emotion; distributional models;\nmultimodal models; similarity/relatedness; concreteness."
        }
    ],
    "section": "Papers with Poster Presentations",
    "is_remote": true,
    "remote_url": "https://escholarship.org/uc/item/51r6h6n5",
    "frozenauthors": [
        {
            "first_name": "Armand",
            "middle_name": "S.",
            "last_name": "Rotaru",
            "name_suffix": "",
            "institution": "University College London",
            "department": ""
        },
        {
            "first_name": "Gabriella",
            "middle_name": "",
            "last_name": "Vigliocco",
            "name_suffix": "",
            "institution": "University College London",
            "department": ""
        }
    ],
    "date_submitted": null,
    "date_accepted": null,
    "date_published": "2019-01-01T18:00:00Z",
    "render_galley": null,
    "galleys": [
        {
            "label": "PDF",
            "type": "pdf",
            "path": "https://journalpub.escholarship.org/cognitivesciencesociety/article/28861/galley/18732/download/"
        }
    ]
}