从文本重建地图
Reconstructing Maps from Text
摘要:
先前的研究已经表明,分布语义模型(DSM)能够从新闻语料库(Louwerse & Zwaan,2009)和小说(Louwerse & Benesh,2012)中重建地图。再现地图的能力是令人惊讶的,因为DSM众所周知地缺乏感知基础(De Vega等人,2012年)。在本文中,我们调查所需的语言推断地图的统计来源,并由此产生的限制放在语义表示的机制。研究1将词共现置于实验控制之下,以证明语言中的直接共现对于传统DSM成功再现地图是必要的。研究2提出了一种基于实例的DSM,能够重建地图的城市名称的共现频率独立。
英文摘要:
Previous research has demonstrated that Distributional Semantic Models (DSMs) are capable of reconstructing maps from news corpora (Louwerse & Zwaan, 2009) and novels (Louwerse & Benesh, 2012). The capacity for reproducing maps is surprising since DSMs notoriously lack perceptual grounding (De Vega et al., 2012). In this paper we investigate the statistical sources required in language to infer maps, and resulting constraints placed on mechanisms of semantic representation. Study 1 brings word co-occurrence under experimental control to demonstrate that direct co-occurrence in language is necessary for traditional DSMs to successfully reproduce maps. Study 2 presents an instance-based DSM that is capable of reconstructing maps independent of the frequency of co-occurrence of city names.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。